/diy/ - Do It Yourself

>>2570913
it seems so obvious now that you've said it, thank you. i don't know how i missed that. one iteration i built did actually have pull-down resistors on the data lines, but i guess i somehow totally missed the address lines this entire time.
i also clearly have been ignorant to the fact that no current is absolutely NOT the same as a current to ground.
i appreciate the brutal honesty, obviously i'm a huge noob to all this stuff and i've been studying fundamentals but also doing stuff like this is fun practical experience that makes it a lot easier to apply those concepts. i'll give it another shot and report back in a day or two.

>>2570943
>>2570913
>Also, can CMOS chips source enough current to light a LED?
high voltage output according to the datasheet is >2.4V, so i believe it should be able to unless i'm totally mistaken. other CMOS chips i've been using have had no issues lighting an LED

Anyone know of a good reverse engineering guide for finding and accessing serial console pads on linux-based embedded devices?
I've got a WiFi access point that I want to console and I'm out of my depth.

>>2570944
>enough current
>>2570944
>2.4V
anon, I...

>>2570969
>finding and accessing serial console pads on linux-based embedded devices?
the trick is, you post pics of the pcb to /ohm/, and then /ohm/ tells you where the pads are

>>2570976
>tells you
argues about*

>>2570976
>where the pads are
Yeah, but after that? I just hold probes to them with helping hands? To a USB RS-232 dongle based on DB9;pinout? Which is DCE and DTE?
Do I have to care about common grounds somehow? Worth soldering leads? Or even a connector?
I seek the wisdom of masters that I may lessen the chance of destructive mistakes.

>>2570975
>outputting 2.4V
>LED has forward voltage of 2.0V
>running current through 470R
you could be onto something... i will investigate further and check my understanding

>>2570983
>hold probes to them with helping hands?
If you can access a pin on an chip, or a through-hole part or connector, you can get what's called IC hooks or test hooks, which clip on. Another good option is to find a flex circuit ribbon cable with the right pitch and number of leads, then scrape off some of the insulation and mate it to all the pads at once.

>To a USB RS-232 dongle based on DB9;pinout?
they aren't always RS232, but that will work 90% of the time. Working out which pin is which, may indeed prove difficult without a scope or something.

>>2570944
Always use a buffer (mosfet/BJT switch) at ICs output, never drive any load from them directly is usually the best advice. I also don't think a memory module has enough current to drive LEDs unless it's some kind of educational memory chip intended for driving LEDs. Datasheet says maximum 50 mA, doesn't specify if it's 50 mA total or per pin.

>>2570771
It's not isolated though. Maybe you referenced the output improperly with respect to the input?

>>2570989
Aren't they TTL? Or are you not implying ±12V signalling when you say RS232?

>>2570996
>>2571007
thanks for the detailed breakdown, i'm working on a redesign that should take all this into account. i think this is a case of me overcomplicating something that is fundamentally simple.
>>2571014
i'll have to take a closer look, i totally missed the output voltage being so low on those pins, the CMOS logic chips I've been using output >4.8V

>>2571031
The issue is not just the voltage, but the output impedance on those pins. The output impedance is not 0, it's actually usually high enough to be significant. By drawing too much current, the voltage drops. By the time you reach anywhere close to the required amount, the voltage might be close to 0V. It's not enough to just look at the voltage. The amount of current that the GPIO can source at that voltage is also important.

>>2571014
so something like picrel is better?

>>2571034
Do common-collector instead of common-emitter, assuming you have the option.

>>2571034
Not exactly. Whenever you're driving a load with an NPN or NMOS transistor, you put the load on the high side (if you drive the load with a PNP or PMOS, you put it on the low side). You can read more about it here (I didn't read it, but it was first search result, try to dig deeper if it's not very clear):
https://www.baldengineer com/low-side-vs-high-side-transistor-switch.html

Once you've installed the load properly (and used the right type of transistor depending on if you're switching high/low), you have to calculate the base resistance This might be a bit too complex for you, so you could also try it with trial and error. Here is another link that might help you do that:
https://www.electronics-tutorials ws/transistor/tran_4.html

So I have this reel of LED strip I bought a while back. I completely forgot to write down any specs and I'm trying to figure it out. Power LEDs didn't exist when I studied this stuff so I have questions :

The strip is in 2 inch sections. Each section has 3xLEDs and 3x 150 ohm resistors. I cut off a 2 inch piece and fed it 12.15v and read 42 mA. So that's 520mW per 2 inches or 3.06 W per foot, right?

Also, assuming that each LED/R pair is in parallel, that means each R is passing 14mA for 2.1 volt drop. So the LEDs are 10.05 V each (which seems high) and are dissipating 140.2 mW. Does this make sense?

>>2571046
Bah. I just measured current over a 4 foot length and it was 329mA. I was expected 42*12*4/2= 1008 mA. Now I don't understand anything.

>>2571035
running this through the simulator works a lot closer to what i want than common-emitter. looks like i'm going to have to go through >>2571045
this material and revise my understanding of transistors.
again, thanks guys.

>>2571056
1/2
Well put it this way. You want the BJT to be in saturation. When does that happen? When the collector-base NP junction is reverse biased and when the base-emitter PN junction is forward biased.

When does that happen? Well the N-P collector-base junction is reverse biased when the voltage at N is larger than the voltage P. This translates to Collector voltage > base voltage.
The P-N base-emitter junction is forward biased then the P junction is higher than the N junction, i.e. base > emitter.

Therefore to operate as a switch, Vc > Vb > Ve. Great!

Now think about what happens when you wire the load the way you did here >>2571034.
What is Vc? Well it's clearly tied to the 5V rail, therefore Vc = 5V.

What about Ve?. Well Ve = Vled + V(30 ohm resistor).
Let's assume that you want 20 mA of current through the LED. Assuming a red LED, the forward voltage drop is ~2V. Therefore Ve = 2V + 20mA*30ohms = 2V + 0.6V = 2.6V.
so Ve = 2.6V. What about the base voltage? Woops, it's tied to 2.4V and a 3.3k resistor. Whatever current flows from +2.4V and the 3.3k resistor will induce a voltage drop. Therefore Vb < 2.4V. Ve = 2.6V and Vb = <2.4V. REVERSED BIAS!
READ NEXT COMMENT
1/2

>>2570937
One anon from the last thread was talking about the an870 19999 counts and I fucking took the bait, it was an impulsive buy. Now I looked around the store more and found the Q1. It doesn't have a rotary switch that craps out, like my old one did and it's got this sick ass screen. Tell me how is a half digit going to make it better or I'm killing myself, the an870 is a 19999count and the Q1 is only 9999count.

>>2571035
>common-collector instead of common-emitter
you got those backwards m8

>>2571056
You want to do common-emitter. What this guy posted >>2571035 is common emitter, not common collector. Common-emitter means emitter is tied to common (i.e. ground).

>>2571075
>>2571058
okay, i think this is making a lot more sense now. i've taken some time to go back over BJTs and revise the relevant maths.
this is what my data line display looks like now, and i feel like I mostly understand what each value here actually represents.
picrel should allow the proper current to drive the LEDs, as the 1.7k resistor attached to the data lines will hit the 0.7V forward voltage of the transistors when the base is fed >0.4V from the chip.

here's tonight's last schematic for the whole thing, it's looking a lot better i've managed to cut down on two unnecessary components lol.
i've run through each connection and from what i understand this should have the desired result

>>2571062
why not just get a fluke 179 or 289? are you poor?

>>2570937
Soldered an LQFP-48 IC for the first time on a board
Took some tries, but I finally got it.
Also that cheap solid brown solder flux is the worst shit ever. It leaves dark stains which are impossible to clean.

>>2571062
One instrument for everyting won't work. At some point you need different devices for different purposes. Stop counting the digits and experiment and learn to use what you have. Learn what you need.
WTF is 19999 counts anyway?

>>2571182

>>2571165
nah pretty standard to burn the gel flux, get a strong alcohol cleaner. like turps or grease cleaner (isopropyl is for fags) drop some on there, use a toothbrush to clean the pcb. itll blast through it in seconds. better yet, get liquid flux, it takes a lot more to burn it brown

>>2571105
This looks much better, although there are still a few things you should double check.

First of all, what kind of LEDs are you using? If you have the datasheet, it would be good to post it, otherwise I'll assume typical values (~2.2V for yellow LEDs). According to the datasheet, you can expect a Vce(sat) of 90 to 250 mV for the BC548 NPN transistor. Therefore you have the following equation:
Ve = 0V (common emitter)
Vc = Ve + Vce(sat) = 0.09V (best case) or 0.25V (worst case)
Therefore, the voltage drop across R12,13,14,15 assuming similar component tolerances is:
Vr = Vcc-Vled-Vce(sat) = 5V-2.2V - 0.25V = 2.55V (worst case)
Vr = Vcc-Vled-Vce(sat) = 5V-2.2V-0.09V = 2.71V (best case)
Now that you know Vr, figure out the LED current you need. Most datasheets use 20mA, but that doesn't mean you have to run them at 20 mA. You can run them lower, although brightness might change. Let's just assume you want to run them at 20mA, to find the value of R12-R15, we use ohm's law, we have:
R = Vr/Iled = 2.55V/20mA = 127 ohms (worst case)
R = Vr.Iled = 2.71V/20mA = 135.5 ohms (best case)
As you can see, 470 ohms is a bit too high, if you want 20 mA of current,you'll only get:
2.55V/470 = 5.42mA (worst case)
2.71/470 = 5.76 mA (best case)
5mA might be bright enough for you, so it's really your choice if you want to change to a lower resistance or not.

Now you want to find the base resistance. According to the datasheet, you can expect a Vbe(sat) of around 0.7V for 10 mA. Taking that value, we have V in = D1 = 2.4V and you have Vb = Ve + Vbe = 0V + 0.7V = 0.7V
Therefore, we have
Rb = (Vin-Vb)/Ib = (2.4-0.7)/Ib = 1.7/Ib

Here Ib is selected to give a collector current of 5mA (in your case). the Hfe in saturation is usually lower, but couldn't find it in the datasheet. honestly if you just take Ib = 1-5 mA, you should be good, the only downside is higher power usage than if you were to find the right Ib current. This gives Rb = 340 ohms. You might be fine with 1.7k

>>2571105
Another thing which is important is switch Sw3. although theoretically, there is nothing wrong with it, in practice, switches or not ideal and when you press the switch, the contacts tends to bounce, which creates multiple on/off cycles before stabilizing. Depending on your circuit, this could be a big issue. You therefore need to debounce the switch (look up how to do a debouncer circuit). There are many ways to do this, but putting a 100 nF capacitor in parallel with Sw3 and ground should usually do the trick. You might have to tune it a bit (up and down) and see what works best.

Another thing i've noticed is that you removed 3 resistors from Q5,Q6,Q7. You might think this is a good way to save a few components, but you have to be careful. Remember, Q5,Q6,Q7 do not necessarily have the same value (they might in your circuit, since it looks like they are permenantly grounded), but if for whatever reason whatever is stored in Q5-Q7 does not match, then you end up once again shorting one of them through the other. For instance, if Q5 is 0V and Q6 is 5V, then when mode is set to read, Q6 shorts through Q5. So a safe bet would be to keep those pins separated with their own pull down resistor. But like I said, if you're absolutely certain that the memory module always has Q4-Q7 at the same level, then you can skip it but it would be poor design

Another issue is Chip Select is tied to Q4-Q7. If at anytime those outputs are high, it will disable the chip select (since it's negative logic). Have CS defintely on it's own path to ground. Since it's an input pin, you can tie it directly to ground.

Everything else looks good to me, although you should definitely test it and come back if you're still having trouble.

Also on schematics, it's typical to connect the power flags on standalone gnd/vcc pins instead of tying it on components like you did, but it's not important at all, just good practice.

I need a lamp that needs to fit the following LED in:
GY6.35 10W LED COB AC-DC12V

Any idea where can I find one online?

>>2571214
https://www.mpja.com/10-Watt-LEDs/products/578/

>>2571217
thx

>>2571105
>>2571207
>>2571212
One last thing, you still end up shorting pins Q0,Q1,Q2,Q3 to ground/vcc if your dip switch is on when the chip is in read mode. You haven't fixed that, but it's a simple fix. all you have to do is put a bunch of resistors in series with the switch. That way if you accidentally leave one of the dip switch open while it's in read mode, you don't short it to ground but have a resistor there.

>>2571219
>GY6.35 10W LED COB
Sorry, Anon. I posted incompatible LEDs because I'm an illiterate tard. Those won't work without modification to the lamp.

>>2571222
Oh bummer, well still thx
I guess i can try to ask locally

>>2571224
2 for $10, but it's fleabay
https://www.ebay.com/itm/174643752330

>>2571024
>Aren't they TTL?
could be, guess we'll see when he opens it.
Most of the newer usb serial devices I've used can do that though Probably not 90% like I initially said...more like ~60%.

>Do I have to care about common grounds somehow?
as for this, nope, thats the point of an interface. they either use differential signaling (a bit like AC or current-loop signaling), or carry their own ground to the driver.

>>2570983
>Which is DCE and DTE?
It's a DTE. No one has made DCE since the mainframe era.

>>2571024
>Aren't they TTL? Or are you not implying ±12V signalling when you say RS232?
If they output 5v or 3.3v they are generally called a UART. If they are called RS-232 and have DB9, they should be +/-12v or more commonly 10v.

>>2571341
look at this bullshit in pic related

>>2571341
>If they are called RS-232 and have DB9, they should be +/-12v

please dont make shit up.

My Keurig has a bug apparently: if there is not enough water, then most of the time it triggers an alert and it won't work until you add water. but *sometimes* it starts brewing anyway and then stops prematurely and goes into an "out water mode" and only pours maybe 1/2 of a cup. that is annoying. is it possible to hack the firmware so it makes a better estimate based on the cup size selected?

>>2571391
The best hack for these types of sticky situations is to buy a coffee pot that just brews coffee. If you can't do that, then bypass all the "smart" shit with a big rocker switch. Get one with a light in it if you wanna be a fancy prick.

>>2571391
>is it possible to hack the firmware
Better question is if your skill level is adequate for task
>obviously no
so no

I encourage you to prove random autist wrong tho

>>2571391
>that is annoying.

of course, they made a simple process super complicated, which created untold horrors to society.
you might try to scotch-tape a thin magnet next to the sensor.
or do what i did: returned to my 25-yo Braun.
it's just a heater and a pipe, as simple as you can get. no firmware bugs ever.
one of my big peeves about the Keurig is you cant empty the reservoir.
if you go on vacation for 2 weeks, your first few cups contain a toxic cocktail.

>>2571391
dissassemble, see what sensor it is for water level and decide what to do.

>>2571407
9 out of 10 Keurnig owners agree that good coffee tastes like mold water. I bet they all have irritable bowels and watch Oprah reruns.

>>2571401
>>2571407
Are you suggesting disabling the sensor? That doesn't make sense. The process is not super complicated, and checking if your water tank is empty makes perfect sense. Otherwise it will cause bigger problems.

>you cant empty the reservoir
Yes you can, it is detachable. Not sure maybe it is not on some models but that would be retarded. Or do you mean there is still some water inside the machine that you cannot drain?

>>2571414
Meh, it is an acquired taste anyway. I can code switch from the office swill to fine coffee at an Italian coffee shop in NYC european/Latin american type they brew at the Caribbean resorts. It is great and starbucks tastes like shit for a while. Then you get used to it. Meh whatever. I am not a huge coffee fan anyway. I'd rather have a Kale smoothie with artisan phat beets or some green juice with matcha green tea doughnuts.
And, no Oprah for me. I prefer day time Jerry Springer when the chairs start flying.

>>2571429
>Are you suggesting disabling the sensor?

no.
on my model, water level is detected by a floating magnet.
you can use an external magnet to make it seem it's always full.

>Yes you can, it is detachable

there's an internal heated reservoir.
you cant empty it without disassembling the machine.
it's the reason you can brew a second cup fast, without waiting for the heating cycle.

>>2571429
>Or do you mean there is still some water inside the machine that you cannot drain?
This.

is this tip shot? i tried sanding but its not heating up wire or melting solder at all

>>2571463
>is this tip shot?

it's metal, so it cant ever be shot.
but obviously, it's not gonna work when it's jet black.
you have to sand it again, but then tin it as it heats.
then it'll should stay shiny for some indeterminate period.
then repeat until you have enough money to get a new brand-name one.
actually, i just make my own from heavy copper wire.
it doesnt last long, but i can make a new one in a minute.

>>2570937
Does anyone know, why LTSpice shows a negative current? I am following FesZ's tutorial and he is getting the correct results.

>>2571466
Oh, is it because of Kirchhoff's "knot"-rule? So the current through R1 is the incoming current, noted negative and the current at R2 is the outgoing current and noted positive?

>>2571465
>but then tin it as it heats.
this is the important part. I'd sand it in a bath/puddle of tin and keep it totally covered in a layer the entire time.

>>2571465
>heavy copper wire.
are there any other materials that are nice and oxide-free, to actually last the lifetime of the rest of a solder gun? platinum maybe?

>>2571466
the resistor is upside down bruh, look at your annotation (10k) versus his.

>>2571481
Holy shit, thanks.

>>2571478
No. This is a single series circuit with current in one direction.

This is, direction is arbitrary in a part like a resistor or (non-electrolytic) capacitor. You could put it in a circuit one way or 180 degrees the other way. The current "of" that device would incur that sign reversal, if it differs from the overall direction of current convention in your circuit.

>>2571480
> oxide free
Most soldering iron tips are plated with iron.
You can do this yourself with your bottle of ferric chloride and electroplating.

>>2571487
>your bottle of ferric chloride and electroplating.
anon, you might be overestimating my shop.
anyway, I meant something that would stay nice and clean even after sanding or grinding it down

How to blink an LED without microcontroller with a precise timing?

>>2571494
R(L)C circuit, digital timer/clock, or a premade PWM/LED driver chip

>>2571489
Give the soldering gun to a homeless mechanic or something. Then buy a proper soldering iron like a civilized non-monke.

>>2571495
>digital timer/clock
https://en.wikipedia.org/wiki/Clock_generator

>>2571494
>>2571495
>>2571497
the most common one is probably
https://en.wikipedia.org/wiki/555_timer_IC#Astable

>>2571494
>precise timing?

if you're looking for 0.03% kind of precision, start with a crystal oscillator and then a divider chip like the 4020, 4040, 4060, 4536, etc.

>>2571207
thanks for this, i am using yellow LEDs, yes. i didn't even consider that you could use those values in the data sheet to calculate resistance per current, thanks for pointing that out.
5mA has been adequate so far, the values have mainly been chosen to match what resistors i have on hand while still hitting the target needed to display the data lines. your breakdown is greatly appreciated nonetheless.

>>2571212
i have considered de-bouncing SW3, but since worst-case it will just trigger multiple writes at the same time, i figured it wouldn't be an issue given that the data/address lines wouldn't be changing during the potential bounces. for something like an EEPROM where there's a finite number of writes available, i would probably consider it.
>Remember, Q5,Q6,Q7 do not necessarily have the same value
very good point, the timing diagram does actually indicate that address lines don't all necessarily change at the same time, so it does make sense that very slight delays could cause unwanted inputs to those pins.
>Chip Select is tied to Q4-Q7
very true, thank you.
>>2571221
i need to take a closer look at this, because i think the switches i've got are actually totally ineffective, thanks.

>>2571581
>>2571221
so from what i understand, if the DIP switch is closed while the chip is in read mode, then the data line, for example Q0, will be grounded via the DIP switch + SW_Mode. if the DIP switch is closed, then Q0 would be grounded through R8.

if SW_Mode is open (write-mode) and the DIP switch is also open, then Q0 would ground through R8, which is intended.

however, the issue is when the DIP switch is closed: Q0 will actually read 2.5V because current will be flowing from 5V through R5, through the switch, and then through R8 to ground, with the voltage being divided at the junction.

while the data sheet says that minimum input high voltage is 2.4V, having the line sit at 2.5V is definitely too close for comfort. would adjusting the resistor values be a good solution?

how to adjust bias on this old bitch

>>2571463
buy some tip tinner

I have a 5V usb net, 5V battery net and 24V battery net on a PCB. Should I have three different grounds for each net or I ground everything together?

>>2571609
Batteries float so you can ground them together, and to the overall gnd from the source/wall or whatever too.

Does a typical signal generator frequency stepping depend on the range? For example is it practical to have a 1Hz resolution or even 1K resolution when setting the frequency in 10Mhz range? That wouldn't be a problem if one could punch in the exact frequency from a keypad. But let's say there is no keypad and all you have is a single rotary encoder with multiple ranges. It would be probably very tedious to have 1Hz increments.

>>2571726
Yes. Even on the digital ones you lose precision at the high end of the range. It's probably consistent around some %, and the dial has some absolute ratings for each range.

>>2571609
I'd tie them together unless there's some reason not to (e.g. USB being floating for some sort of safety/protection reason), but tie them to a central star ground point. Prevents nodes being dragged about by high-current on another line. Probably not a big deal unless you're dealing with particularly high currents (more than ~10A) or if you have analog stuff going on, but it's good practice and shouldn't impede your general circuit design. For star grounding I make a footprint for a ground node connector, it's just two/three pads connected by copper without the normal solder mask gap over the pads.

For anyone wondering, there's a new microcontroller general here:

>>2571738

I let it archive again lmao

Hey diy,
retarded mechfag here, I bought this generic speed control box off ebay and tore it down with the intent to swap the voltage control/speed knob out for digital control via Arduino. The board is getting 24V and the only way I know how to manage >5V applications with the Arduino is via relays, which to my understanding are binary 0/1 switches only. Am I fucking retarded? What is the proper way to swap this dial out so I can have digital control?

>>2571365
I have a few of those.

>>2571746
>A: replace the potentiometer with a digital potentiometer or vactrols
>B: replace the existing PWM signal with one coming from the arduino
A probably won't work with digipots since they're usually 5V only, but vactrols could be funny.

B is probably what I'd go for, but I'd reverse engineer it first to see where the power FET(s) are getting their gate signal from. If it's a directly out of a 24V timer of some sort (e.g. NE555), then you might be able to just use the timer as-is by giving the comparator inputs out-of-phase square waves from the arduino. If that isn't possible, then removing the connection to the timer and adding your own level shifter and/or gate driver would be the way to go.

Post IC numbers, that includes the 8-pin chip and the three 3-pin power packages. Then trace the circuit if you dare.

>>2571600
is that different from solder?

>>2571746
8pin says NE555, 91M, DXHL
3pins, assuming you're talking about the shit on heatsinks, each say:
SI080NI018 PYA01971 2114 for the largest base one
SR20010LCT for second
7612CV GF536 V6 CHN11
Only results I'm getting are for strange industrial applications, and there's no way this chinesium brick is rated for anything like that.
To be completely frank I'm not very electrically competent, a lot of what you're throwing sounds like voodoo black magic. I'd honestly like to avoid fucking with the board in the controller if at all possible as I have nowhere near delicate enough of a hand to swap components there, and would prefer to just nigger rig something instead. For additional context, the CON1 port was previously occupied by the speed controller pictured above, but its toast and the reason I got the box for a few bucks. If there's a way to mount something up that allows the Arduino's 5VDC to control the 24VAC going through the board, I'd be sitting pretty.

>>2571752
>>2571765

>>2571765
>I'd honestly like to avoid fucking with the board in the controller

use arduino to turn a servo motor attached to speed pot using rubber tubing.

>7612CV

prob says 7812.
a 12V regulator.
coz 555 chip cant work above 16V.

>>2571775
>use arduino to turn a servo motor attached to speed pot using rubber tubing.
you jest but i was thinking about it, but the seller wasn't joking, the dial is fried
i could find a replacement fairly easily, but while it's in pieces i wanna try and pursue a more technically sound solution if possible
>7612CV
you're right, it's an 8, the etching on that one is a bit faded

>>2571594
>however, the issue is when the DIP switch is closed: Q0 will actually read 2.5V because current will be flowing from 5V through R5, through the switch, and then through R8 to ground, with the voltage being divided at the junction.
Correct
>while the data sheet says that minimum input high voltage is 2.4V, having the line sit at 2.5V is definitely too close for comfort. would adjusting the resistor values be a good solution?
Yes, if you increase the values of R1, R3, R6 and R8, you'll get a higher vin.

>>2571765
Ok it is an NE555. Chances are the circuit is like this:
There is an extra resistor and a zener diode, but R3 on your board looks to just be a 0Ω jumper, and ZD1 is probably just there to prevent overvoltage on the MOSFET gate.

If you just remove the potentiometer, then drive a PWM waveform to the middle pin (yellow wire) of the connector, and tie a ~4.7kΩ resistor from the CV pin to ground, you should get what you're after. For that resistor, it shouldn't be too hard to solder to since pin 5 is on the corner of the package (the one opposite the dot), and also is directly connected to what I assume to be C3 as written on your board (C2 on my circuit) so you could probably have a better time just soldering it across C3. Don't forget to tie the 0V of the arduino to the 0V of the PWM circuit also. Breaking/desoldering off the cap written as C4 on your board would be advised.
This is a hack that uses the NE555 just as a MOSFET gate driver. It does invert the PWM signal.

That SR20010LCT is probably just a diode IC package, for freewheeling. Can't find a datasheet, maybe the name is off slightly.

>>2571842
So what I'm getting is
>leave CON1 potentiometer port empty
>tie a resistor across the C3 cap
>send arduino PWM output to yellow input wire
>remove C4 cap
Is that about correct?

>>2570937
does anyone know of any cheap and "user friendly" CLPDs that i could program with an arduino? i basically want to dick around with a CLPD to replace logic gate ICs. i see microchip has a bunch of cheap CLPDs, some of which are even in DIP form factor, but i haven't really found out how they're supposed to be programmed.

>>2571971
Cplds like GAL/PAL are usually programmed with a special machine called a programmer (previously a "fuse burner") which has a DIP socket and takes a netlist or logic description file, and alters the chip accordingly. Modern CPLDs use flash/eeprom instead of fuses, but still typically require a programmer

>>2571847
>about correct?

there's the small matter that your 'duino is at 5V and the 555 is at 12V.
so it may just look at your puny signal, laugh, and demand a BeagleBone Black.

>>2571983
... Or a FET

Bought pic related v3 piece of crap from sunsky. worked fine on arrivel until USB connection started to drop, and now it does nothing with usb, except charge itself.

Will replacing the ch9102 on it fix it?

>>2571989
It seems my efforts will be fruitless cause I don't know how the original one was configured and ch9102 needs to be programmed to be used.

There are several really known accelerometers, like MPUs, that people use in their projects. But do you know some reasonably priced accelerometers, that are fast and precise?

Kinda autistic question i do not hope to find an answer but still interesting to ask:

Im trying to peel of the layers of chips to get to the correct layer to see how it is structured. I tried grinding little little with sandpaper but that's way to unprecise and random

See pic related and the marked layers, for example

What are the best methods to achieve that? Preferable not chemically because my commie government banned a lot of those to buy

>>2572001
Chemically. Dichloromethane, Nitric Acid, or at least Sulfuric Acid.
You'd have to be in a real third-world shithole to not be able to get your hands on one or all of those.

If you want to do it manually, you need an obnoxious degree of precision to actually reveal layers, and even then you'll need luck to take it evenly enough to see single layers.
You need something to hold the chip rigidly, and you need to sand against the smoothest surface you possibly can.
A thick sheet of glass is the ghetto option, a small surface plate would be better.
If you're holding the chip with your fingers, you're going to have a bad time. It needs to be rigidly mounted to something flat and stiff. Glue it to a brick if you don't have a better option.
From there, you're going slow, incredibly painfully slow, with incredibly fine grit sandpaper. At 2000 grit, the grains of aluminum oxide are basically boulders compared to the details you're sanding, you're aiming for more like the 10k-20k range.
Slow, even, smooth, you're trying to achieve retarded tolerances with sanding in your hut in Nigeria, it's going to be slow and painful to do it well.

>>2572009
Thx
>You'd have to be in a real third-world shithole to not be able to get your hands on one or all of those.
Atleast two of those are banned in germany

>>2572010
boil it until it fumes lmao

>>2571989
Open the bitch up, bypass the port, test it directly before you start considering replacing that tiny bitch of a chip.
I'd really be surprised if it was the bridge itself, that'd certainly be unusual.

how would you connect two thick solid copper wire for submersible/wet environment conditions? about 8 gauge

>>2572017
the water will connect them electrically with no modification at all

>>2572017
steel wool and duct tape

>>2572017
They make kits for this to make it easy.
https://www.homedepot.com/p/IDEAL-Thermo-Shrink-AL-CU-Underground-Splice-Kit-46-401/203761582
https://www.lowes.com/pd/Gardner-Bender-Butt-Splice-Kit-8-2-AWG/1000790568
https://www.amazon.com/Gardner-Bender-HSB-28-Cable-Splice/dp/B001AWP7OG/

Don't waste your time with the wrap-around gel seal ones, they're shit for this and usually half the length they need to be.

>>2571983
through a fet* like >>2571985 said, to condom the arduino

>>2572009
>third-world shithole
all are heavily regulated in EU...

>>2572114
The fuck, why? DCM isn't even flammable.
God damn Europe really is a hellish crumbling dystopian bog.

What the fuck is 5 items supposed to mean?
I have an analog meter that can display 0-20mA.
Does it mean it can handle 5 times the rated current ie. 100mA?

>>2571847
>leave CON1 potentiometer port empty
Yes.
>tie a resistor across the C3 cap
The resistor should be 4.7k, though 3.3k should work fine.
>send arduino PWM output to yellow input wire
That's the middle pin of CON1, not VIN-.
>remove C4 cap
Yes.
In addition, connect the 0V/GND pin of the arduino to VIN-. That or tie the arduino 0V/GND pin to the bottom-most pad of C4 / the bottom-right corner of U2.

>>2571983
That's what the resistor across the CV cap is for. It pulls the threshold voltages of the two comparators such that the upper-most threshold is lower than 5V.

>>2572010
DCM can be distilled from common paint stripper.
Sulfuric acid can be electrolytically produced from copper sulfate, which should be available. If not, I think you can use distillation and ammonium sulfate.
Then with a distillation apparatus and a nitrate salt you should be able to make nitric acid. Nitrate salts themselves can be tough to get hold of though, you may want to try electrochemical oxidation of ammonium ions into nitrate ions in the presence of a copper catalyst. That or making a Birkland-Eyde reactor.

>>2572129
>Does it mean it can handle 5 times the rated current ie. 100mA?
I'd guess so. Maybe you can find an actual datasheet for it?

>>2572133
Would I need voltage control via a FET to avoid the 24V fucking the Arduino or is the potentiometer port already stepped down, and direct wiring the Arduino PWM input to CON1 is safe?

>>2572133
>I'd guess so. Maybe you can find an actual datasheet for it?

That is the fecking data sheet unfortunately.
It's multicomp pro brand which is basically Farnell / element14 / newark's own brand range where everything has shitty data sheets

>>2572129
>items
>times
>times of rate current
>emo cunt firestarter
>Within 5 emo cunt firestarter
It's clearly just Chinese gibberish, see if you can find a datasheet.

There's an amazing NRF24L01 2.4Ghz transceiver module. But is there something similar with higher frequency?

>>2572221

>>2572016
>I'd really be surprised if it was the bridge itself,
Aah! You are God damn right Anon.
It was just a loose B micro plug.

>>2572141
>That is the fecking data sheet

it seems pretty clear to me.
it's guaranteed to survive a long-term 500% overload, but may burn out (or catch on fire) beyond 501%
i usually wire a geranium diode in parallel to save the day.
but that's on 50-250uA meters.
i'd have to redo the math for a 0.1A meter, however my calculator's on strike.

>>2572369
Usually the case; if you were even considering replacing the chip then I'm sure replacing the port would be a breeze by comparison.

>>2572128
>The fuck, why?

>>2572385
Don't stick yourself with needles filled with solvents Euro-cucks.

>>2572010
I got nitric from concrete efloresence cleaner

>>2571746
>>2571752
MCP41HV31-104E is a 100kOhm digipot that can directly replace the analog pot and gets steered by 5v in while allowing up to 35v or so across. Good luck finding one though, does anyone know of any equivalent replacement?

any fun circuits with a 1k:8ohm audio transformer?

I'm confused about wheatstone bridges. I thought the balance condition was r1/r2 = r3/r4, and ive also seen it written as r1/r3 = r2/r4. Now looking at this i get why theres a voltage difference across the middle resistor, but how can the equations make sense then, if accordingly
r1 = 500
r2 = 1000
r3 = 500
r4 = 1000
so 500/1000 = 500/ 1000 or 500/500 = 1000/1000??

>>2572387
>Don't stick yourself with needles filled with solvents Euro-cucks.

>tfw using hot air to harvest components off dev-boards is cheaper than ordering them from digikey
What even is this universe?

>>2571463
Take your little wrench and tighten the jamb nuts.Repeated heating oxidizes the tip's surface and causes a high resistance lack of connection.

Hello everyone.
I'm looking for something specific :
An audio amplifier that inputs an audio signal from a jack (mic) and outputs to a jack.
Do you know the name of such hardware ?
What should I buy in your opinion ?

>>2572521
Don't know about off-the-shelf, but you're talking about a mic preamp and headphone amp which you can buy separately and patch together, or stuff them both into a project box and break out the pots and switches.

>>2572484
it's just comparing between two voltage dividers

>>2572484
wheatstone bridges usually connect one or more variable resistive loads. the voltage difference (where you put that middle resistor) is usually connected to a form of differential amplifier with massive gain. it doesnt matter what the resistance is its all about the difference in voltage. a good example is a single load cell or strain gauge. r1 = r2 = r3, then r4 is the variable (coming from the strain gauge). changes on the resistive element on the load cell (a kinetic force upon it) mean a linearly changing voltage. so we see one side of the wheatstone bridge forming a sort of voltage reference (voltage divider) and you compare that against the voltage on the other side. ive never seen a wheatstone bridge with a resistor in the middle. hope that helps

>>2572521
I have a steinberg UR22mk2..not sure if you're thinking of something like that, at very least you could use it with an amp that has none probably

>>2572531
That helps. Thanks.

>>2572557
Yeah it looks like what I want. Thanks

>>2572543
>>2572541
yeah i get that, and i know r4 is usually the variable load (the resistor in the middle is simply to illustrate current flow). but what i dont get is how the equation for a balanced load can be correct if as shown, there is current flow when the equations say there should be none.

>>2570937
I'm designing a PCB in kicad 6, I've copied the requirements from jlcpcb's website when I run DRC I get warnings about silkscreen to pad clearance.
This board isn't that big so I can go and modify the footprints manually so the silkscreen is further away but on bigger projects it'll be hard to go through every single component and modify it's footprint.
My question is is the 0.15mm silkscreen clearance actually required? And if so is there an easier way to maintain that clearance than manually editing all the footprints?

>>2572576
My preferred way is to not print silkscreen at all

Alternatively, just ignore those warnings as MOST manufacturers will just remove required silkscreen clearance around exposed areas
>there are some retarded manufacturers that print board as supplie

>>2572589
>>there are some retarded manufacturers that print board as supplie
I read on some forum that jlcpcb printed silkscreen on somebody's pads because they didn't understand a layer that was supposed to subtract from the silkscreen so I was wondering if it's a requirement that will make them reject the board or if it's for the user not to accidentally print over the pads

>>2572576
I use eagle and oshpark and I've never had to worry about that. They just delete anything that's over the exposed copper or holes.

Lmao I thought KiCad is a meme and everyone in real life uses hacked Altium.

>>2572620
nah, imo Altium is actually a huge pain in the ass unless you are doing near-RF or cutting edge stuff that really, really cares about parasitics and star grounds and shit like that .
KiCad isnt any better, but its also hardly worse for 90% of jobs

>>2572622
So what's better?

>>2572625
KiCad for hobbyists, Altium for EEs.

EasyEDA is the best for hobbyists.

>>2572633
Do you, Anon, take JLCPCB to be your lawfully wedded wife, to have and to hold, to love and to cherish, etc.?

>>2572634
Niggers hate chinese, I know.

>>2572633
>authored by Dillon "Klebold" He and Eric "Harris" Cui

>>2572636
Chinks are high yellow niggers though.

>>2572637
Do we actually know more about the girl to the right of dylan doing the ahegaho face?

>>2572638
No, they are at least ten floors above, they are able to use brains to make something like EasyEDA.

>>2572663
The one next to Dylan is Robyn Anderson (bought them guns), but I don't know who the other chick is.

>>2572633
>in-app advertisements

>>2572663
Her name is Kelli Brown.

Did any of you play with STM32F7 chips? I want something in 200Mhz range with 256k of ram

>>2572707
Out of stock

>>2572708
Ok and is it going to be in stock in a year? Because I am only starting and it would legit take a year for me to finish developing the thing.

>>2572720
I personally expect chip shortages to mostly be fixed in 2023. But it is impossible for us to know for sure because none of use work for the chip fabs.

>>2572723
Okay but what about usage, are F7 chips any good? I've been working for F1 mostly and doing a step up

>>2572728
>are F7 chips any good?
They have a 0R resistor.

>>2572731
ok mr. shitposter, not into giving advice, so be it

>>2572734
Just put a 0R on it, breh.

>>2572735
I wish we had pcb assembly machines at home for the price of 3d printers.

>>2572748
We're getting there, DIY ones are getting better and better but they are still too slow and too shitty for too much money.

>>2571842
I got access to some test equipment and believe the 7812 voltage reg on U1, top left sink, is faulty as the leftmost input pin has no continuity with the COM or OUT pins. Is this retarded or did I find the fault with the circuit?

>>2572708
Oh, you're right, shit is out of stock. But I need a few for development, and there seem to be some on chinkexpress. They are not going to send me a fake one are they?

>>2572762
Never mind its working as expected. No clue whats wrong, so gonna just call it quits

>>2572748
Pick-and-place can basically be done with a modified 3D printer. Same for solder paste extrusion, though a 3D-printed stencil might be better for that. The vacuum pump and hoses / syringe aren't going to be that difficult to connect up in place of the hot-end, but if you want automatic tool-changing you'd be in for a fair bit of extra complexity.

It's the chemical processes of etching (and especially electroless plating) that are the tough parts to automate.

>>2572762
>>2572779
What is the fault? Have you modified the circuit yet? Can you test the potentiometer to see if it was dead in the first place?

Hello,
Is there some devices I can plug into the wall plug to watch or give any ratio of PLC signal in it ?
I see weird things happening with my LED bulbs and wonder if it's not some PLC related noises.
Is there a a way to just cut stranger's PLC entering my home, using some sort of capacitors and filter ? I'd like a total cut off of that illegal alien shit.

>>2572782
What's PLC?

>>2572782
>PLC
Programmable Logic Controller?
If there's noise on your lines then adding line filters may well help. They're usually made of a common-mode choke with X and Y caps, but adding some normal-mode chokes can make a big difference if the noise is normal-mode.

>>2572637

>>2572788
Power Lines Communication
They are in the process to install this shit all over the world
https://www.academia.edu/9313801/Modeling_of_In_House_CENELEC_A_Band_PLC_Channel_using_Fritchman_Model_and_Baum_Welch_Algorithm

So I recently got this hand-me-down microwave that I'm trying to fix for experience as it stopped being fun a while ago, so far the front panel was busted (removed the whole membrane and placed 3 buttons; one for reheat, start and cancel) but the thing still won't turn on. It's an old samsung circa mid 00s.
Magnetron seems fine, it's got continuity between both high voltage leads but not its case itself, capacitor seems normal as it's got the right microfarads, diode and fuses too.
Front panel I tested unmounted for all of the button combos after figuring out which was which, worked too (pretty fun exercise doing all the combinations.)
I quickly tested the HV transformer leads with just my multimeter's resistance setting and none were open, though I did notice the leads going to the door switches had this black grime on them which I assume is tarnishing, same for the diode.
So suffice to say the 3 buttons I connected worked fine, plate turns but the thing won't still heat stuff up, what could it be? Seems like it's the high voltage part since magnetron's not doing its job.
There's no going back at this point.

Ive made a few projects with KiCad, and almost invariably the library/footprint setup fucks me somehow. Its just baffling at times, even using LibraryKing or whatever the shit Mouser makes you sign up for. I see that EasyEda uses Jlcpcb's library directly which would make fab super simple. Dare I jump off the FOSS ship?

>>2572812
>Magnetron seems fine, it's got continuity between both high voltage leads but not its case itself,

magnetrons have 2 circuits, with 3 or 4 wires.
one circuit for high voltage @ low current
one circuit for low voltage @ high current
gotta check 'em all.
also, one of the high voltage wires *should* connect to the case.
which is a comforting thought.

>>2572562
theres only current flow because you put a damn resistor there

>>2572576
silk gets removed by the manufacturers when its not over the solder mask

>>2572844
never had a problem with it myself. maybe youre just a retard

>>2572892
so the balance equation is literally for a bridge without a load? that seems a bit retarded.

>>2572924
Why so? The only load you'd ever put there would be the input of an instrumentation amplifier, which is so high impedance it's a non-issue.

>>2572932
i suppose im thinking about it in exact terms. i guess for actual electrics its just good enough hey

>>2572812
> not heating
Inside the copper magnetron there is a tungsten filament which gives off the electrons. It’s basically an incandecent light bulb. Those burn out.

As you can imagine this is not fixable, but it’s by FAR the most common problem. For an “inverter” microwave, it’s the inverter that is the most common problem… those are hard to fix,

>>2572939
its not about good enough, the flow of current in this case doesnt matter. its about the differential voltage as the other anon and i mentioned. unless the top resistances are the same, and the bottom ones are too, the system is not balanced. in the case that it is. in this instance, the differential voltage will be 0V at your differential output. as far as im aware, the balance equation you keep mentioning, just stipulates the case where the ratio of resistancd is equal on each side. its nothing to do with current.

I've got a couple of (partially) dead headphones that I want to repair. I've got a few questions.
>Hurr just Google it
Faggot every time I search "headphone repair" it gives me shit to placate normies, like using THIS CHEAP TRICK to ""fix"" them. I want some real answers.

1. Is there a good chance that the break is with the jack and not the cable? I get some very faint sound if you twist it the right way and angle your head enough. Like definitely fainter than normal. I just plug them into my phone so its not an amp problem or anything.

2. If the break is in the cable, how am I supposed to identify where the break in the wires is? I've tried that trick where you bend the cable and run your thumb along the length until the sound cuts out, but it doesn't work. I don't want to keep cutting the wire at random places to find the break.

3. Can I get away with not using flux? They're tiny wires anyway.

hey guys, total noob here.
kicad is definitely better than tinycad, i can tell you that much for free.
you're welcome :^)

>>2572889
>>2572949
Well shit I'll check on it again after work. Thanks.

>>2573005
>I want some real answers.
If the headphones are hard wired, open them up and replace the whole cable. If the headphones have an input jack, check/reflow the solder joints. If that doesn't help, replace the jack. You should use flux, even if your solder is flux-core. Doesn't matter if the wires are tiny.

Can anyone tell me what connector this is?

Im not so sure this is /ohm/ related but I cant see where else it should go.

Does anyone have experience with PLC's and using TIA to programme them? I'm trying to write an ultra basic programme that utilises the clockrate but I'm getting nothing from the system when I upload the programme.

The ladder programme is litereally just

---] [ --- [1hz] --- ( )
when I remove the clock it works perfectly, the wiring is correct, and the clock is correctly mapped. there was origianlly a second tag that shared the same name with the clock but it has been removed. the analysis shows the signal reaching the switch and then just stopping. The clock is enabled in the PLC settings tab.

>>2573005
just replace the cable bro. your phones DO have replaceable cables, right?

>>2573005
Replace whole cable.
Safer, better, easier option than partial replacement/repair,

>>2573071
Looks like a Molex Micro-fit connector.
The Micro-fit family is big, don't be surprised if the first ones you find don't look like yours, keep digging.

>>2573164

>>2573166
There's a great deal of physics and chemistry involved there, not to mention the difficulty in keeping the materials from spoiling or getting contaminated during manufacture. How hard would it be to print PNP or NPN junctions on the same board with consistency and accuracy? i.e. the base is 10nm thick

>>2573176
Closest thing to proper DIY transistors I've seen are ones printed with inkjet printers, using nanocellulose and carbon fiber nanotubes.
Not exactly home-gamer friendly. The "headline" with those was that they were biodegradable, but it was just research paper shit.

>>2573181
>biodegradable
That's what I look for in all electronics. If it doesn't melt when I sneeze on it then it's trash.

>>2573190
What a horrible waste it would be if the things we made didn't rot and return to the earth.

>>2573196
Everything rots. It came from the Earth and so it shall return, including the circuit boards and nuclear warheads. Dust in the wind.

>>2573176
>>2573181
and you could always buy bulk, ~50nm litho surface mount transistors for cents, which would probably be cheaper than your ink

I think it'd be much more reasonable to make a 'desktop' PCB printer/etcher, pic-n-place, and (maybe) solder oven combo. But making multilayer nets and vias would require cutting and layering, something that would double or triple the cost.

>>2573181
>biodegradable PCBs and transistors
That shit would be good for mobile phones.

How do I stop getting anxious when dealing with high power shit?

I've been working with power electronics for years and been fine but last year I had a DC link converter catastrophically fail.
Basically all of the IGBTs in the unit get hi-pot tested at 1.7KV to make sure they can withstand it before the unit itself is brought up to 1200V but someone failed to do the hi-pot test and one of the IGBTs blew out at 1150V.

Also the DC link capacitors were connected so 17000 microfarads at 1150V worth of energy instantly discharged through the IGBT.

Needless to say, there was a massive bang. I had the unit in a test chamber and was controlling the power supply from my PC outside but the blast fried my USB to serial converter and so I couldn't turn the thing off. The supply kept charging the unit up to about 700V whereby the caps would arc over with another bang and rinse and repeat.

I'd every cunt in the building over trying to ask me what was going on while I was trying to figure out a way to stop it and eventually I had to wait until it arced over and then went in and manually switched it off before it charged up again. Everywhere was covered in dust from the ceiling tiles jumping when the explosions went off.

Anyway, now if I'm doing anything at more than 300 volts I start panicking even though I'm completely sure everything has been checked and I've calculated everything correctly. I may only be doing something at 10mA and I still get anxious.
If someone walks up behind me I jump like fuck.

Will this ever go away?

>>2573218

>>2573218
Back when I was just a wee lad, I thought it would be a good idea to pull the anode off the back of a neon sign display in a grocery store. Then my arm fired backwards and continued in a Popeye style wind-up punch motion which made my buddies laugh, but my arm was pins and needles for a while afterwards. The key is to build up immunity, same as with bullets.

ran out of solder, so i went to the local shop where buy shit like this but since EU doesn't allow leaded solder (fucking cunts) he didn't have any local made ones left so he sold me some that had chinese writing on it .. and guess fucking what.. what a piece of useless fucking shit. I soldered a micro ship and that piece of shit solder just made bridges i can't even clear up, it just won't melt properly even at 400c.
As a bench mark i set my iron to 240C which is enough to melt any decent solder and this thing would not even budge.
Serves me right for getting chinked i guess.

>>2573234
>EU doesn't allow leaded solder (fucking cunts)
Bullshit. You're just not allowed to sell electronic products using leaded solder commercially. Selling leaded solder is allowed.

>>2573236
No it's not, shops are only allowed to sell leaded solder to companies but not to regular people.

>>2573234
Cut the solder into sections and pack it into a pipe filled will black powder. Place fuse, cap pipe, light fuse, throw through store window. The customer is always right.

>>2573229
I remember being like 8 or 9 and my da showing me how to wire a plug.
About a week later I decided to wire the little 6V DC motor from my mecanno set to a plug and plug it in.
The socket and half the wall in the kitchen was stained black for months before my parents redecorated

>>2573242
Holy shit. You got lucky. Reminds me of the time I peed on a wall outlet in 1975. lmao

>>2573237
Pretty sure you're wrong. They sell it even on Amazon.

>>2573242
>be kid highly interested in electronics
>never get shocked or do anything dangerous
>some parents friends visit with young kid
>puts screwdriver into plug
Fortunately, nothing happened to the kid.

>>2573245
How much money was made out outlet covers?

>>2573243
Yeah, I done some stupid shit when I was younger and don't know how I didn't die.
My town was only a small village when I was growing up but it's a commuter town now and there were always loads of housing estates being built when I was a kid and me and my friends used to play on the sites.
Once I found one of the triangular keys for opening service panels and the like and we used it to open the panel on a street light.
I seen this big block of shiny metal and screws (The live bus bar) and stuck my hand in to grab them. I still remember the feeling of that shit going through my arm.

When I was a teenager I was always fucking around with fire and chemicals too and stole loads of sulphuric and nitric acid from the school lab.
Last year I was out fishing one day and bent down to duck under a tree branch and a stick went right into my eye and I couldn't see for hours and vision was blurry for days.
I couldn't believe that all the stupid dangerous shit I did when I was a kid and I ended up doing my eye in while fishing.

Eye got better after a few weeks anyway

>>2573248
It was an extension cord, but yeah, I don't get why safety mechanisms are standard, maybe even part of regulation. When I was very young, my mother told me not to put shit into plugs, and I never did it.

>>2573249
>don't know how I didn't die
It's a miracle that anybody made it through, really. The bus bar incident could've easily been the end of you, but God wanted to keep you around longer so you poked your eye on a stick in 2022. I think He's telling you to buy some sweet skiing goggles.
Glad your eye is okay.
>>2573254
>my mother told me not to put shit into plugs, and I never did it
Sometimes kids actually listen, others have to touch the hot stove to learn firsthand what "burn" really means.

Some more nice memories:

A ruin of a house that was almost completely demolished. Only the basement or something was left. The adults told us not to play there, because it was dangerous. Holes in the ground, maybe live wires still around, apparently some kids died there once. But we played there anyway. It wasn't even remote, but located on a central street. Nothing happened.

Disassembling CRT TVs. I found them so interesting. Adults told us they could still hold dangerous electrical charge, or that the tubes could implode when opened. Nothing happened. I threw away the electron gun only a year ago.

Creating a simple circuit with a feedback loop with a capacitor and a relay to create a low frequency high voltage signal, low voltage input. Fun to shock other kids.

>>2573067
>>2573153
>>2573155
>replace the cable
Is it possible to buy like a spool of aux cable? The only ones I can find have jacks on the end and have a huge mark up.

>>2573265
You sure can on Aliexpress, in a variety of styles and colors too. Don't know if it'd be easy to find on ebay or Amazon.
If you're going from scratch, it's also super easy to add sleeves and shit while you're at it, if you want to make things fancy.

Then again, most dollar stores carry headphones, earbuds, and audio cables that you can cannibalize for $1.25.

>>2570937
Has anyone ever worked with diy sringe pump equipment lkke pic related?

Are therre any off the shelf kits you recomend that can hit up to 150ccs? And is robust enough to take some movement.

Or hell, one that a non techie can at least follow along with?

...full discolure, this is for a bedroom application, so its gonna be full of lube

>>2573290
I've only ever come across ones that use compressed air.
Usually an adapter sits on the back of the syringe and constantly keeps the plunger under pressure.
Then the front of the syringe is screwed into a pneumatically actuated valve.
When the valve is opened, the goo comes oozing out (in my case, it's usually conformal coating)

>>2573265
Yeah def
https://www.bhphotovideo.com/c/buy/bulk-microphone-cables-audio-cable-spool/ci/14255

>>2573034
Ok, update:
Most likely the magnetron as it sounds like there's something loose inside after lightly shaking it around a few times.

>>2572889
>which is a comforting thought.
Oh yeah wanna know something funny? Guy who owned it before had removed the AC ground pin because it couldn't be plugged into his kitchen's outlet, and you know, pretty much every appliance got grounding in the case so I just put 2 and 2 together.
Gonna get another handmedown magnetron soon to see if it works, don't care at this point, I've sunk so much time into this thing might as well just fix it all.

>>2573290
I can’t believe people are still using 2-stroke fucking machines. Lube consumption is way lower with a 4-stroke, and the exhaust isn’t as dirty.

>Made two circuits in multisim testing out relays
>Identical values and components in each
>Somehow only one has a working relay system
Is multisim that unreliable or have I tarded up

>>2573420
multisim does suck, but also make sure your 2 circuits grounds are tied together

I need a (musical) keyboard for a synthesizer. It doesn't seem like there's any off-the-shelf solutions and I also want the freedom of my own design, so I think I need to roll my own.

What are the general ways to build a keyboard? IIRC it's typically done with something like diodes and multiplexing but I don't know enough about the specifics. Any resources would be appreciated.

>>2573356
> sounds like there's something loose
Yes, thats the burnt-out tungsten filament. I open the copper casings with a pipe cutter just to confirm, but it’s almost always that.
Don’t touch that pink ceramic material, and don’t grind it up into a powder and snort it. Some say that’s a berylium based ceramic. Which is bad.

>>2573218
I blew up an el cheapo harbor freight meter with 240V by not paying attention while trying to diagnose a water heater.
It goes a way but it takes some time

>>2573422
these are the two circuits, if there's something obvious I'm missing I cannot see it.

>>2573424
>What are the general ways to build a keyboard?

Same as any other interface with a gorillion keys. If it doesn't matter that you can't push all the buttons at once, multiplex them. If it does, hook each one up to a dedicated I/O. Depending on what you're doing and how many free lines you have on your MCU, that "dedicated I/O" may either be an actual GPIO on the chip, or you could use shift registers to save on cost.

In your case, I'd say it depends on how many keys you want. You're probably stuck with monitoring each key with no multiplexing, since you'll need to use multiple keys at once. If you want a mini keyboard of only like 1.5-2.5 octaves, you might be better off just using an MCU with enough inputs to read all of the keys at once. Much more than that and it'll be more cost effective to use a more minimal MCU and some shift registers.

>>2573434
This looks like veritassium’s schematic showing how wires aren’t needed to carry “energy”

Why my LED does want to blink and voltage is all over the place?
Here is the schematics I used https://elonics.org/sites/default/files/inline-images/flashing-blinking-led-circuit-using-555-timer-schematic.png

>>2573452
That's not "all over the place" look at the y-axis. It's within a thousandth of a volt

>>2573456
Yeah but it has to oscillate from 0V to 2V with square wave.

>>2573456
> 2.037 kV is deadly
Oh, it’s 2.037 euro volts.

>>2573476
5 microseconds isn't enough to see that, you need a longer period. You are zoomed way to far in both dimensions.

>>2573443
It helped. I will never be able to do electronics properly.

this is probably the wrong place to ask but idk where else to ask. i popped the top off of a thinkpad mechanical dock and found some blue paste on random parts in it. anyone know what it is and what it's for? this was refurbished, if it helps identify it.

>>2573506
It looks like lithium grease. It's not supposed to be used on plastic.

>>2573506
It’s silicone grease. You see the same shit on plastic gears in various throw-away inkjet printers.

>>2573265
I've been having this same problem, the wire inside the headband went bad.
I absolutely cannot find another substitute wire that's micro thin like that.

Why do you always have to put a ground in simulators, even if it's simple battery powered device hanging in the air? It's stupid as fuck.

>>2573531
math

>>2573426
You mean this thing? Cause when it took out of the case it was exposed; Don't think there's a way to unmount the magnetron without exposing the pink part.
I didn't bang it up or anything though.

>>2573434
>something obvious I'm missing I cannot see it.

if you drew your circuits like a human being, it'd be easier to spot the error.
which is:
1 - when S1 energizes relay, the relay shuts off power to lamp2
2 - when S1 does not energize the relay, S6 cuts off power to lamp2.
3 - since S1 and S6 turn on/off using the same keystroke, lamp2 never goes on.

How is programmable output level implemented in signal generators since DDS output level is unpredictable and depends on frequency? VGA plus some kind of sensing/feedback?

>>2573434
>something obvious I'm missing I cannot see it.

also, i dint mention the obvious.
coz it's obvious.
that one circuit uses the NO contact, while the other uses the NC.
so, no, they're no more identical than your left hand is identical to your right.

>>2573567
Both relays are NC fucktard

>>2573563
I've been operating the switches individually with the mouse, so never needed to change the binds.
>>2573567
Both are NC fucktard, that's why there's a problem.
Though anons would of guessed by the matching relay names

>>2573139
Try /mcg/, think I saw some people talking about PLCs there before.

>>2573166
Because making a sufficiently conductive filament is basically impossible. Even with a bunch of carbon or metal fill, the conductivity is pretty ass, and you wear through nozzles real quick. Making such large transistors is going to be useless for any integrated circuit replacement. Plus there's the matter of heat tolerance. Closest you'll get will be some sort of micro cement/ceramic printer for the substrate, depositing metals (and maybe semiconductors) via powder that gets sintered. Kiln required. There may be electroless CNC methods for depositing metals onto an insulating substrate via a solution. Otherwise you could maybe do it all via masking and vacuum deposition, but that's a pain in the ass to automate since you need to go in and out of vacuum in order to apply and remove masks. Maybe you could do it by steering charged matter beams, idk.

>>2573249
Brought a MOT into science class and make arcs with it. Tried to cook something over the arc once. Glad I didn't kill someone / myself.

>>2573424
Do you want to measure key velocity, or any other analog way of modulating individual keys? The traditional way is to have two switch contacts for each key, and measure the time it takes between them hitting each other, which gives you an estimation of the velocity at which they key was pressed. For a piano simulator, this allows you to "strike" the virtual string with varying strength, influencing the initial loudness and the time it takes to peter out simultaneously. Personally I think modern synths should have key force sensing as well or instead, allowing you to modulate the loudness and/or timbre of the key by modulating the force with which your fingers are pushing down on the keys after they're first pressed. But I don't know if the MIDI standard supports anything like this.

I've experimented a bit with variable-force capacitive keyboards, but the variable frequency output of each key proves difficult to interface with, and even using a bunch of filters followed by ADCs doesn't give a particularly good result. If you have the room for such a mechanical design, I'd recommend load cell force sensors. Then you'd need an ADC input for each key, or array of analog switches for shovelling all the analog values into less ADC inputs. The RP2040's processor speed, memory, and ADC sampling rate makes it somewhat appealing to use for this.

If you just use double switches then wire them up like a normal row*column matrix to minimise the number of I/O pins required.

>>2573438
You don't need to avoid multiplexing, in any musical situation it's trivial to sample keystrokes fast enough that it's effectively real-time. Diode keyboard matrices allow N-key rollover.

>>2573564
>DDS output level is unpredictable and depends on frequency
Why would this be the case?

>>2573536
if (ground_present == false) {
set_node(001) = GROUND
}

>>2573265
Have you seen how thin those shits are? Do you know where to buy connectors for both ends? One of those males has a locking mechanism and isn't 3.5mm. Have you factored in shipping costs?

I know it's jewish as hell, but paying the $10 markup in this case is the better choice.

>>2573555
> this thing?
Yes, that thing.
You can take out the whole tube (it’s hermetically sealed) if you are careful. There is two ceramic ring magnets in there as well.

The part sticking out is where the microwaves come out, like an antenna (but more of a wave guide).
The other end has the tungsten filament.

>>2573527
> micro thin wire
Probably two conductor shielded.
Got any old ear pods around, just use that wire.
Or get some at the dollar store, throw out the phones and keep the wire. Wire is wire.

Broskis...

I got a wilfa coffee grinder with issues. Basically it only runs well for the first half a second after I press the button, then it sorta chugs in pulses.
https://files.catbox.moe/rpkrsq.m4a

I figured it was an issue with the brushes in the engine but I didn't see anything obvious after taking it apart. Pls help I need my cuppa in the morning

>>2573617
>Try /mcg/, think I saw some people talking about PLCs there before.
ok thanks

I have never designed a PCB. And I want to get into KiCAD.
I would also need some circuit simulation and found "QUCS" for that purpose.
Is it a good idea to jump into this without any prior experience? I don't really know where to start.

>>2573678
>QUCS
isn't what I'm looking for apparently. What is some (free as in beer or freedom) live circuit simulator, preferably for loonix?
I don't really know what it is I'm looking for. Basically I want to simulate an equalizer like picrel, test that everything works as it's supposed to.

>>2573683
LTSpice
The kids hate it.

>>2573610
heres the problem, relays are polarity sensitive and the 2nd circuit is the wrong way around

I want to setup a ventilation system in my room for when I'm soldering.
At the top left in my room I have a vent that leads straight to the outside world.
Thinking of putting a shelf up there and buying a ventilation fan from ali, to which there are 100's of variations of but I assume they all do the same exact thing and dont have any big differences in specs.
Then I'll have some sort of flexible duct pipe around 1-2 meters in length to reach my table.
Am I forgetting anything here? Do I need to worry about filters if they go straight to the outer world?

>>2573690
>relays are polarity sensitive

come on.
a coil of wire will produce a magnetic field irrespective of which way the electrons are bunny-hopping.

>>2573699
>a ventilation system in my room

ridiculous overkill unless you're doing industrial assembly.
just use a 6-inch fan to blow it away from your face.

>>2573703
I suppose it is. I rather something a bit more elegant than jerry rigging a desktop fan.
I don't mind spending 50 euro on something that means it'll be a bit more convenient and easily adjustable without having to worry about my limited desk space and where to put the actual fan

>>2573699
You probably want speed control unless you don't care about noise, and a filter is unnecessary if exhausting outdoors.

I only solder inside of a fume hood

>>2573702
sounds like bullshit but that's what happens in multisim.
I donno myself either.

Where are those fuckers used? I need to buy one but i can't do it directly, i need to salvage it from something else. Don't care where, i have money

>>2573756
Why can't you buy it yourself?

>>2573621
>Why would this be the case?
reconstruction filter. the higher the frequency the lower the amplitude.

>>2573763
embargo
posting from my 8 proxy

>>2573804
what kind of shithole embargoes neon sign transformers?

>>2573804
You can build a Cockcroft-Walton voltage multiplier out of high-voltage capacitors and diodes. If you have any broken laptop screens then you could salvage the backlight inverter.

>>2573806
neons are for neonazis

>>2573786
Isn't that just the anti-imaging filter at the end of a DAC? You can easily take the transfer function of that filter into account in your DSP algorithm. Also a properly chosen filter would have a wide enough passband that it wouldn't be noticeable in the desired frequency range anyhow.

>>2573814
sieg heil

>>2573756
They are used in negative ion generators and those cheap bug zappers.
I mean obviously, you could take a nail and wrap copper wire around it by chucking it into a drill until you had the required ratio of turns….

>>2573634
Tried that, only wire thin enough is ipod headphone wire, but it's not long enough. Also it's really fuckin' hard to solder to, even after burning off the insulation

>>2573964
They put nylon strands alongside the copper for strength, you have to seperate them. They also insulate the individual strands, you have to burn the enamel off them before solder will stick, use plenty of flux and a high temperature.

What’s the problem with using a thicker cable? Just file or drill out the hole.

I have the PCB and big iron transformer from a 1kVA UPS. If I connect it to a 100Ah deep-cycle battery, will it still work fine?

>>2573622
Isn't this assuming the first node/net I drew is ground, vs hot or signal?

HAM is dead, so I'll ask here.

Can you give me a link to Digikey or Mouser section, where radio receiver chips are sold? Even better would be an Information about common receiver chips, that were used a lot previously on different frequencies and have examples.
For example a chip that capable of receiving on 1575.42 MHz.

>>2574010
Get a SDR dongle.

>>2574050
Fuck off retard.

>>2574051
Sit on my face and whisper your callsign into my balls.

What overhead lights are you guys using?

I was thinking of doing similar setup to what adam savage is doing, but I am a cheap bastard.

>>2573978
> UPS guts
Yep, it should. Please be aware that a lot of the bigger ones actually need 24v (the rack-mounts with multiple batteries in them)

>>2574054
what an incredibly specific niche for an insult

>>2574010
>chips
most of them aren't integrated into a single IC. Usually they require an antenna (duh) and inductor/coil along with some smaller discretes. Best you will find usually is a 'module' instead of 'chip' https://www.digikey.com/en/products/filter/rf-transceiver-modules-and-modems/872

But if you really do want to print your own RF PCB (you don't) and attach your own antenna, you can buy just the core part https://www.digikey.com/en/products/filter/rf-receivers/870

>>2574070
It's just a normal under-desk UPS. I think it had two batteries in it, but I'm not sure if they were two 12Vs or 6Vs, or if they were in series or parallel. I'll reverse engineer it a bit to be sure.
The 100Ah battery is free because it's got some leakage around the top though, we'll see if it's damaged or not sooner or later.

Also I'd probably want to connect a solar charge controller to the battery, that can't go wrong can it?

>>2573290
use a peristaltic pum you coomer, theyre piss cheap and way smaller that that abomination

>>2573980
ground is relative

Making an AB add in pcb for my motherboard. The intent is to let me flick a switch to choose which nvme drive is active. I'm looking through the documentation for ngff m key, but theres so many power pins that im not sure if such a toggle can be done with just a single SPDT switch like ive shown here.

The original plan was to have every other pin split out, and just have one be controlled by the switch, but i dont think it can be made this simple? pin outs for reference https://pinoutguide.com/HD/M.2_NGFF_connector_pinout.shtml

>>2574326
38 DEVSLP
>Device Sleep, input. If driven high the host is informing the SSD to enter a low power state.

54 PEWAKE#
>PCIe WAKE# Open Drain with pull up on platform. Active Low.

>>2574326
Beware that you might have to delay connecting power and data pins. They have different contact lengths, so some pins connect first. This matters for hot plugging and was intentionally designed.

>>2574326
I would be very surprised if works with just switching power. Even if you sleep one drive then wake the other >>2574333 you have to make sure the sleeping drive is in tristate mode. At the very least, you'll have to put one drive to sleep. Wait. Then wake the other drive

>>2574185
lol you are a dipshit. Ground is ARBITRARY, not relative

>>2573531
What is the ground in this picture and how does it 'relate' to the two (unconnected) "ground" terminals? Nothing. Completely your choice. Ground is just a spot you pick to call zero voltage. You can pick a sensible node, or the first node you drew, or one chosen at random.

>>2574337
>>2574335
>>2574333
hmm, this is more complex then expected. In that case, what about a way to toggle all pins? or is that not possible? or perhaps turning the system off first? although that would be less (word for ideal state)

>>2574348
Interesting question. What does the system do at boot time? All pins are already connected when power goes on. Anyway, I just wanted to mention this as a warning, I don't know how much it matters.

>>2574349
>What does the system do at boot time? All pins are already connected when power goes on.
i dont know, its just a normal computer. I know pcie is hotswappable, but i dont know much beyond that.

>>2574352
What OS?

>>2574326
hdparm

>>2574341
it's both

>>2574355
arch and windows10

>>2574392
ok, then what are these two ground terminals 'relative' to each other? Undefined. >>2574343

>>2574326
is there no situation where the data being written gets interrupted by the switchover? can that corrupt data?

>>2574422
first eject disk?
There would be no issue if the system was off, which seems to be the only solution so far.

>>2574422
>>2574440
You should read the specification, certainly it will have a lot on hot plugging and init/shutdown. I assume the worst that could happen is that the host thinks the device didn't change. Or maybe that doesn't even work, because the protocol is something really complex that needs link training etc.

>>2574411
https://wiki.archlinux.org/title/hdparm

I figured an AI gives better answers about electronics than anyone else on this board. It even gives the references. Thank me later perplexity.ai

>>2574349
>What does the system do at boot time?
At boot, the PCIe->board pins will be floating. The chipset/bios will then do put some of the pins into output mode (system->board) to do some negotiation with the device. Then all the system->board pins will be in output mode or floating depending on how the negotiation goes.

>>2573678
I want to design a PCB.

Should I jump straight into KiCAD or is there a better way?

What is a good simulation software? I tried QUCS.
It was pretty confusing and wouldn't offer "live" simulation.

>>2573683
So LTSPICE it is? Is there no noob friendly FOSS simulator?

in4 muh reddit formatting

>>2574734
>Is there no noob friendly FOSS simulator?
No, it's all 50 year old SPICE with different GUI. If you want pretty color and animations you can try circuitjs, but it's pretty limited for anything serious. You can always go old school and use ngspice with text file netlists. LTSpice is the least painful way to do simulations. Even if it has old crusty 90s interface is still way ahead of other clunky software in terms of user friendliness.

I found a tweeter not working on my loudspeaker set. Swapped cables to be sure it wasn't some audio signal not using high freqs on Left speaker. Still nothing.
How can I test a tweeter with different electronic tools ?

>>2574847
Are the other speakers on the left channel functioning?

>>2574847
Test the coil resistance with a meter, or connect the tweeter directly to the amp and play something at low volume.

shieeet
went to download kikead and version 7 already dropped?
what is going on? we waited like 100 years for 5 to go to 6 and now new major version is dropping every week. i don't want have to reimport all my customs libraries every fucking 3 months

>>2574870
>kicad
Do you need a link to free last version of altium? It's just better.

>>2574871
no thanks
kicad is great and open source on top
the particle accelerator was literally designed in it
altium is for people who like to larp as engineers

I have a bachelor's degree in electrical engineering and don't understand anything in this thread.

What I don't understand is why are those cheap laser engravers not been used to make PCBs at home.

Humility is something that AI can never learn.

>>2574895
congrats on learning that you learn nothing in university in regards to actual application

>>2574895
you don't go to college for experience, you go there for that piece of paper that will open doors to interviews to you.
You have to learn all of the useful skills for your chosen discipline on your own

anyone have any experience with converting canbus protocols between each other? im planning on doing an engine swap on my truck and the different ecus use different can protocols, so i figured to avoid having to custom wire everything (obd port, cluster, transmission, etc) my plan was to just use the new engine ecu to run the engine, then pipe that data into the old ecu in a language it understands and use it to control the electronics on the vehicle.

>>2574850
yep
>>2574855
I tried to connect wires to an Owon table multimeter, measure with ohm set and it says overload..
it's some old KEF T27 SP1032 from 1978

I found that video on Youtube about cleaning the inside.
https://www.youtube.com/watch?v=rGd2XiuecJs

is the transluscent solution he uses some isopropanol ?

>>2574983
sometimes I ask dumb questions..

>>2574985
Disembodied from its tweeter, I also tried to plug multimeters on V AC to the speaker's cables left dry, and got voltage evolving with the sound. I think it's the correct way to check, right ?

>>2574900
Because having them made professionally is dirt cheap, less bother and produces better results.

>>2574988
The amplifier output is okay. You need to disconnect the tweeter from the amp AND the other speakers, then test the resistance of the coil. In short, the tweeter is malfunctioning. The amp and other speakers are fine.

>>2575059
Yep and the resistance is out of limit. Sadly.
I checked another tweeter in spare parts and this one have a correct impedance.

I'd like now to learn if it's doable to repair it, if it's something like it's demagnetized or something..

>>2575061
There's probably a break in the voice coil. You'd have to wind another one or swap it with another tweeter. Not worth it in my opinion.

i am tired of googling one simple thing: how do i measure a pot or a rotary encoder shaft length? if it says 15mm or 20mm, is it the total length of the threaded part and the serrated part? basically whatever is sticking out from the body, as if you measured your dick?

>>2575114
also what should be the fitting knob? for example if it says
>1 pcs 25x13mm 6mm
does that mean the depth is 13mm? In which I should subtract the threaded part of the pot to get an estimate of how high it is going to sit or am i overthinking it and the "average" knob would fit the "average" pot just fine (as long the diameter is correct)?

>>2575114
Doesn't it say on the datasheet?

>>2574900
They are, but you need a decent focus on them. Marco Reps easily does it with doped fibre lasers and Nd:YAGs, but shitty $10 laser diodes have a pretty wide spot at the best of times. You can probably find people doing it with $20-50 laser diodes.

With a small spot your laser engraving cuts can just be a linear rasterisation, but when it’s wider or you want really tight tolerances, you need to need to follow the contours of the traces and pads, with multiple concentric layers to fill in the gaps. I’m working on an Inkscape shell script to do that, but even then I’ll probably need to upgrade my shitty laser diode.

Apparently Altium 365 is free. It's just like EasyEDA but better, why nobody told me it's that good?

>>2574872
kicad is dogshit, you can't even do signal integrity analysis

>>2575257
>you NEED signal integrity analysis

>>2575200
>falling for the autodesk jew

Do you guys talk about career stuff in here?

>>2575257
Good SI-PI simulator costs as much as a car and you still need to calibrate it against real hardware if you want to get any kind of useful answers from it.

Kicad is dogshit because of dogshit user interface and dogshit user experience. There's literally no need to use it unless you just started a company and have no money, but forced to use licensed software.

For all others there's EasyEDA and Altium 365 for noobs and Altium Designer Enterprise from torrents for hobbyists.

What do you think about this EQ circuit?
Are there better tone control circuits out there?

>>2575411
*with a similar level of simplicity

Noob here. In the oscilloscope you connect the signal and the ground wires. But what of you want to connect not to ground but another signal? Will it short my device? Why nobody explains it on youtube...

>>2575417
the ground wire of the scope and probe is connected to earth so yes it could. what i did was cut earth pin off the AC wall plug so it was always floating. that way thr ground is a reference and wont fuck with your shit

>>2575411
>>2575412
https://sound-au.com/p-cat.htm#equ

>>2575457
I want to use it as a preamplifier for a linear amplifier. Which one of these is best suited for that purpose?

>>2575417
>But what of you want to connect not to ground but another signal?
You use 2 probes and subtract the signals.

>>2575445
Neutral is tied to ground.

Im trying to test the mA of an unknown fuse-type bulb. the marking say 10V so using a 9v battery directly it lights up ok

but using my dmm like
>red probe -- bat+ bat- to bulb side 1 >> bulb side to black probe
yields nothing. what am i doing wrong

>>2575594
AC or DC?

>>2575679
bulb inside the unit was powered by 8.5VAC
(unit has two bulbs one is burned out so i pulled the working bulb to test)

>>2575594
>what am i doing wrong

could be:
- the current fuse in your DMM is blown
- the knob is not on current setting
- if you use the 10A setting, you have to use a diff jack
- even if not 10A, you have to use a diff input terminal to measure current

if you show a pic instead of a drawing, we may be able to figure out more
and you have to clarify what ''yields nothing'' means.
does it mean (1) no reading, light is off, (2) OL reading, light is on (3) something else

>>2575700
of course. by no reading i mean 0.00 when hooked up

>>2575726

wiring looks good
probably 200mA fuse is burnt
put red probe into 10A jack
put knob on Amps DC (one click to the right)
light should come on (if it's actually good) and a reading should appear
remember to remove probe from 10A jack afterwards or else bad things will happen in the future

>>2575734
fuses are ok, using 10A and the A dial gives the same result, 0.00
light does not light, very strange.
again running the battery directly to the light works ok, bulb lights up, not as strong as the AC charge in the unit but still enough to be noticable. im at a loss

>>2575749

could be broken alligator clips
move red probe to input
put knob on DC volts
should read 9V if all wires are good, and bulb is good.

>>2575755
yep everything checks out okay, putting it in a full loop with bulb reads 9.01VDC
bulb doesnt not light however

>>2575760
https://astroai-user-manual.s3.us-west-1.amazonaws.com/sku/user-manual/ASIMT5000A-EN.pdf?version=1622695363
Did you RTFM?

>>2575760

all out of ideas.
all i can surmise is that the current reading circuit is dead in that meter.
that's why i have around 8 meters. (not even sure how many)

>>2575765
oh well thanks for trying to help

>>2575774

one last idea.
take probes out of meter and short them together.
if lamp turns on, fuses in meter are dead, or traces are burnt or broken.

>>2575778
interesting, that worked. now i looked at the fuses and they are both opaque white ceramic, nothing visually looked off. is there a way to tell a burned out ceramic fuse?

>>2575779
>way to tell a burned out ceramic fuse?

you can prob use the ohms function
it may not need the fuses
or use another meter

>>2575782
ohms reading on the 400ma600V fuse reads OL
i tested a regular 120v 4A clear fuse and it reads fine. ill hunt around for another MM if i have one.

>>2575484
>Neutral is tied to ground
And? The power supply inside the scope is isolated, there’s just a ground connection to improve common-mode noise rejection, and arguably also safety. You can definitely get a slight buzz off an exposed metal piece of an unreferenced appliance, due to capacitances. This is generally considered undesirable, but it’s not really dangerous.
There’s no part in the LV side of the scope’s power supply where there’s any reference to neutral or live, except through small capacitances in the transformer. Any class-Y caps might also couple your 0V/shield of the isolated side to mains somewhat, which is a bit of a recipe for noise. Personally I’d have a dedicated switch to enable the ground connection.

I'm building a laser guidance system. Any suggestions?

>>2575782
i had a shitty centech i got from HF years ago. the best readings i could get was in battery DCA mode.
with the battery connected only i get 24mA. with the bulb connected as well I get 22.5mA. Does that mean the bulb draws 1.5mA?

>>2575792
Why don't you ask a laser?

>>2575417
depends whether it's earthed or floating. If it's a floating supply, it's fine, but if it's grounded you could burn out your scope or your probes (at least one of them will act as a fuse).
Subtracting two signals, or getting an op-amp to do it for you is safer.

>>2575795
Am doing it for real. Here is my IR sensor, it's a soviet PbS photoresistor used in Igla manpads.

>>2575808
I have more of them. And I guess it's not laser guidance but rather IR signature guidance.

>>2575794
>battery DCA mode.

no idea what you're talking about.
show me pic.

>>2575812
centech dmms have a little section to test batteries
but actually that looks more like a capacitance dual function. anyways testing a 9V battery gives me 24.0 which i can assume mA based on the setting label

>>2575835
>i can assume mA based on the setting label

no.
it'll read battery voltage on that setting.
the mA thing in parenthesis tells you how much current is gonna be drawn from the battery.
i.e. it puts a 360-ohm resistor across the battery, coz reading battery voltage without a load usu gives false (too-high) results.
to read current, either use the 10A setting (using dedicated 10A jack)or 200mA setting (using regular jack)
it's always a good idea to use the 10A first so you can judge whether it's safe to use the 200mA setting.

>>2575810
Do you mean thermal infrared? I’ve been wondering about using pyroelectric motion sensors for temperature measurement. Each has three elements but outputs some linear combination of them, so you’d probably need like 4 or 5 for thermal tracking. Probably quicker than PbS though. Or just use 1 with a spinning aperture like an old AIM-9.

>>2575835
I assume the 24 implies 24% life? Measure the battery voltage and compare it to the appropriate battery curve.

>building a battery powered widget
>needs to run for a few years at least on a single 3.6V battery
>can't find low dropout or low quiescent current regulators in my country
>somehow find a chinese part HT7333 on very shady website
>2uA quiescent current, 90mV typical dropout voltage
>jackpot.png
>buy 20 of said parts
>measure sleep current consumption of my gadget using my nrf power profiler, 2uA
>solder on the regulator, sleep current is suddenly 1.5mA wtf
>read datasheet again
>"2uA Iq on no load"

I got chink'd or not?

>>2575936
>>2575936
>>2575936
NEW THJERADS

>>2575681
Why are you running an AC bulb on DC current?