tube amp project.

[Anonymous No.690267]

Hey.
for some time i wanted to build a tube amp.
I've been messing with it for a while, but i wanted to have some results before posting.

I'm an electronics technician, but i have little experience with tubes or audio for that matter.
After having my attempt at a Daisycutter shelved for a solid two months, i got it out again.

I was having some crazy oscillations last time and i didn't know here to start.
Not having a scope at my residence didn't help.
I figured I'd just remove the tonestack and see, i had a hunch i messed that up somehow.
Sure enough, it worked. a lot of hum though.
No shielding what-so-ever and the heatersupply runs right by the input, still a prototype i suppose.

I'm not really knowledgeable when it comes to guitar, tone or amps. but i like what I'm hearing so far.
I'll get some of my guitar friends to check it out later.

the parts is whatever i had from my failed marantz7c build, components from work and of course,Ebay.
Nothing was chosen for tonal quality.

tubes: 2x Chinese nobrand 7ax12's. just reads: 7ax12 china.
2x russian NOS 6p14p with CCCP branding.

Powertransformer: Toroidal 55W supply. Hugely overdimensioned for a marantz preamp. still over overdimensioned for an 18W daisycutter.

output transformer: OUTPUT TRANSFORMER 22W 6600 CT.
the ebay listing mentioned fender but it looks brandless.

the rest is vishey polyester film caps and resistors of different origins, sockets and screwterminals ect. from ebay.

I'm powering it through a fusebox I built with 2x 5A automatic fuses, for projects involving mains.
when I'm done it will have the regular glassfuse, obviously.


Now i will have to read up on building a chassis.
I'm not sure i even want a tonestack at this point.
Maybe a single pot one, with a pullout switch to bypass. that would be cool.

any comments and suggestions are welcome.

[Anonymous No.690268]

schem. form ax84.com

[Anonymous No.690270]

form before i got it working.

[Anonymous No.690271]

preamp tubes.

[Anonymous No.690272]

in communist russia
amp plays YOU!

6p14 NOS tubes power pentodes.

[Anonymous No.690276]

[Anonymous No.690282]

https://www.youtube.com/watch?v=yMDfPXOm07g

[Anonymous No.690406]

>>690267
The basics: When you think vacuum tubes, think FETs. Similar principles; both are voltage amplifiers, not current amplifiers like BJTs.
More to the point tubes have extremely high input impedance, just like FETs, and as such are susceptible to interference and crosstalk. Signal routing is important (keep outputs away from inputs, inputs away from noisy power wiring, etc). Get some foil-over-braid shielded cable for the line-level signals. Filter the crap out of your B+ supply, too (LC filtering is good), so the amp is *quiet*. Toroidal transformer = GOOD.

Chassis? Two words: CAKE PAN. Not kidding you. Flip it over and use Greenlee punches to make the tube socket holes, drill the front and back for controls and jacks. Make a cover for the bottom.

[Anonymous No.690415]

>>690406
>The basics: When you think vacuum tubes, think FETs. Similar principles; both are voltage amplifiers, not current amplifiers like BJTs.
i post a working amplifier project and you felt it was necessary to tell me this?

i probably won't go for the cakepan, but rather a regular metal box chassis.
i want it to look nice.

is a tonestack strictly necessary?

[Anonymous No.690431]

As a guitarist, having the tone stack bypass-able is best. This gives you a switchable boost since the stack attenuates a share of your signal.

Your layout has decent space. Try to keep all of your leads parallel and perpendicular. I do everything I can to keep preamp lead dress clear and clean. The lack of a chassis and your lead lengths are likely giving you hum and oscillation.

Filament cables are a primary hum concern being unfiltered and low impedance. These must be routed cleanly. I see the schematic says heater elevation, though their notes aren't clear to me. This is a very wise implementation if you've not done so. This will persuade cathode/heater leakage towards DC and is a tried and true hum squashing method.

[Anonymous No.690441]

>>690431
>heater elevation
i don't really know for sure, but isn't heater elevation for increasing the heater voltage relative to ground so you don't exceed the tubes maximum heater-cathode voltage?

my heater feed has its own separate winding, so its galvanicly isolated from ground, making elevation unnecessary afaik.

am i missing something?

thanks for the replies

[Anonymous No.690463]

>>690415
>i post a working amplifier project and you felt it was necessary to tell me this?
All I know is what you told us: You're a tech. I've been a tech for a little over 30 years; there's techs, then there's Techs, if you know what I mean, and nothing personal. You can do quite a bit with just basic knowledge. Besides which if you didn't notice that led into other things I was talking about. Don't get all bent out of shape, just trying to help. Besides you seemed to indicate you were having some issues with it.

There are some nice-looking cake pans out there. :-)

>tonestack
I guess if you were for instance using an external graphic equalizer or preamp then no, you wouldn't necessarily need it, you'd end up setting the treble/bass/mid to 0dB, but it is a guitar amp, and they usually have them, up to you really.

[Anonymous No.690466]

>>690406
Oh and I just remembered you said 'I don't know too much about tubes', and who would these days? A fossil like me played around with them when I was a kid and TVs were still around that were all tubes, someone younger doesn't necessarily know jack about them and may need to know what they're analogous to in solid-state. Hell, these days I work with engineers all the time who couldn't design a Class-AB discrete transistor audio amp if their lives depended on it.

[Anonymous No.690471]

>>690441
You'd think so. Some of the old Fenders toe right up to that heater-cathode max without too much of an issue. Our purposes for elevating heaters would be to keep the filaments constantly above cathode voltages rather than swinging above and below. With enough elevation, the leakage to the cathodes will approach even dc. This is instantly noticeable for pulling down hum.

You definitely want a center tap or artificial center tap on your filament supply. All the info you need is right here.
http://www.valvewizard.co.uk/heater.html

Just anchor to your power tube cathode resistor for some elevation (or add another point to your B+ for a custom voltage).

[Anonymous No.690474]

>>690463
>>690466
I'm not mad :) it just read very spoonfeedingly condecending. my mistake^^

>there's techs, then there's Techs
oh god do i know that feel.
I service old systems on ships. old, repairable stuff like engine exhaust temp alarms from the 70's and stuff.
you wouldn't imagine how retarded the average chief engineer and ships electrician is.

so my experience is mostly systemvoltage stuff. sometimes simple analog stuff. sometimes digital logic of varying complexity. but little audio.

i got a fair bit of tube theory down, but all the theory in the world is useless without practical application. so i'm a bit of a noob, i will concede that.

thanks you for your input.

[Anonymous No.690484]

>>690474
>it just read very spoonfeedingly condecending
Remember where you are: There are lots and lots of young guys on here who need their hands held just making an LED light up without screwing up, and they want to build guitar pedals and stuff like that, don't even own a soldering iron yet..

>you wouldn't imagine how retarded the average chief engineer and ships electrician is
LOL yes I would.
Short story: Some years ago I had a sad epiphany: The reason I would have a tough time getting a decent job sometimes? Was because there were too many total incompetent hack 'techs' (with a small 't') out there who were only good at lying through their teeth about what they knew/could do. Then I come along (I don't fuck things up) and I sound just like them -- except I'm not lying. Welcome to the dystopia!
I find most mechanics are pretty damned stupid, too. Fix my own cars/trucks/motorcycles/bikes whenever possible, too..

Look on Craigslist for old used analog oscilliscopes. You don't need anything with massive bandwidth to work on audio, you just need to be able to see the signal without loading things down. Cheap! A simple signal generator wouldn't hurt, either, and that you can build.

[Anonymous No.690489]

>>690471
ok thanks, so like this?
sorry it this is stupid.

the schem has some values labeled heater elevation, what does this mean?

>>690484
>Remember where you are
fair point.

i have an old analog scope, but not at my apartment atm.
my birthday is tomorrow, if i don't get a scope from my brother, i'm buying a DSO.
functiongenerator is also on my purchasing list.

im in Norway so buying used is difficult. people want several hundred dollars just for the effort of making an ad and not just throwing stuff away.

[Anonymous No.690491]

>>690489
omg, 2 seconds on google and i found it.

R1 = 220K
R2 = 27K
C1 = 10uF

at work now. will try this when i get home

[Anonymous No.690501]

Tips : to reduce the likley hood of rf oscillation make sure your grid stopper resistors are as close to the valve bases as possible and keep your input leads as far away from your output leads as possible.

[Anonymous No.690509]

The purpose of heater elevation is to ensure you do not violate the Max cathode to heater voltage. In this circuit the only place this is likely to happen is on the phase splitter. You should be OK providing the cathode voltage is below 100v ish.

If your heater winding is centre tapped tie the centre tap to ground. If it hasn't got a centre tap use a couple of 330ohm 1%nfrom each leg of the heater to ground.

[Anonymous No.690515]

>>690509
this will also reduce hum according to the valve wizard.
as far as i understand, the AC will float on the DC offset. giving a constant current with little fluctuation compared to a 6v3AC crossing 0v, with going off and on every phaseshift.

i got no centertap. so i'll try a 100ohm pot with a 50ohm resistor on each lug with wiper to my elevated voltage as per >>690491

thanks for all the input and links guys!

[Anonymous No.690519]

The cake pan people are referring to is this... http://www.amazon.com/Parrish-Magic-Line-Inch-Oblong/dp/B001GCUCQS

Heavy aluminum with just a slight lip. Perfect for mounting wood sides to. I'm actually in the process of rebuilding this Maggie 9302 console pull as we speak.

[Anonymous No.690522]

>>690519
ok, cool.

though maybe you meants something like this..
http://retrothing.typepad.com/photos/uncategorized/2008/07/29/bakingpanamp.jpg

[Anonymous No.690548]

>>690515
If you want to reduce heater hum, why not feed them with regulated DC instead. It's year 2014, after all.

[Anonymous No.690563]

>>690548
It's an option. There's a design on Merlin's page linked earlier. Elevation is just extremely simple and effective.

OP built a Daisy Cutter, not a Top Hat.

[Anonymous No.690564]

>>690548
i'm considering it. for the preamp tubes atleast.
one 7805 with a few diodes to shunt it up to 6.2v maybe.

i'll try the humdinger and elevation first, since that's new to me.

[Anonymous No.690630]

bodged my elevation and humdinger on there.
the elevation just about cut the hum in half.
the humdinger doesn't do much.
i can just about hear changes in the hum when i mess with it. but not much i terms of volume reduction.


considering DC; for running dc to be worth it i assume that using DC for all the heaters would be best.
not having to chain any AC at all around the amp.

all the heaters amount to about 2 amps. the heater voltage is at 6.19v at idle.
6.19 x 1.41 = 8,65v ish.
giving me about 2.2 volts for regulator dropout.
should be fine for a pair of lm319's.

i considered using them for constant current regulation.
but that would require either one per tube, or run the risk of having all my tubes burn should one fail(parallel)

come to think of it. my neighbor has a bench PSU in the basement!
i think i just might try it wit that dialed to 6v.

[Anonymous No.690704]

>>690630
i tried the DC. seems solid.
i still had hum, but i realized that is from the input having a good 20cm stretch with no shielding.
i.e. an antenna. (facepalm)
i'll lift some shielded cable at work tomorrow. maybe some for the heaters as well.
i can only assume not being in a grounded metal box is also a factor.

i put the psu back in the basement, sticking with the elevation + humdinger for now
i can still go either way later on..

something like this would make dc a simple option.
http://no.farnell.com/ideal-power/25hk-ab-060a300-d5/power-supply-ext-plg-in-3a-6v/dp/2334611
maybe lowpass filter it though.

[Anonymous No.690782]

>>690630

Just passing by this thread, and I have no idea whats going on, but that is a really cool picture. Very cozy feel there.

[Anonymous No.691450]

>>690415
I scratch built an AC30 when I was like 17 and had almost no idea about basic electronic principles... Only problem was slight hum due to ground loops and lead dress that I fixed now that I have a /slight/ idea about basic electronic principles.

[Anonymous No.691462]

>>691450
cool, post some pic if you want to.
i'll try out some shilded cable after work today.

i guess i should be researching a bit on this amp.
it clips on the gainstages by design, which is cool.
but it would be fun to try some to ad switched to change the gain and have a clean-ish sound... maybe?

thanks for the replies guy^^

[Anonymous No.691942]

I'm currently working on getting an amp completely hum-free myself - it's a transistor circuit, but the basics should remain the same:
-Always run the signal return wires as close as possible (shielded twisted pair is optimal) to their corresponding hot wire.
-Twist all AC-carrying wires and their return wires as tightly as you can. One twisted bundle should carry a total of 0 current to minimize the EMI it gives off.
-Stick as close as you can to a "true" star ground topology. This means that every part of the circuit gets its ground reference from the same physical location ONLY ONCE. If you can remove a ground wire and the corresponding part of the circuit does not lose its ground reference, you've got a ground wire too many somewhere, which will cause a ground loop.
-Star ground and the first (and to some extent also the second) point I mentioned can come in conflict with each other. If you have to choose between minimizing signal loop area at the cost of grounding several return wires at the same point (which of course will have only one wire to star ground - that doesn't change), minimize signal loop area first.
-Keep signal wires/cables as far away from anything involved with AC to minimize pickup noise.
-Keep signal paths as short as you can.
-Obviously, maximize the distance between power supply and transformers and the rest of the circuit - especially from small-signal wires/input stage etc.

Cont. in next post due to character limit

[Anonymous No.691948]

>>691942
(cont.)
Since your amp is point-to-point, there's gonna be a lot of stuff you're gonna have to pay attention to with regards to wire and component placement.

Note that the above principles were collected when trying to get rid of hum (60Hz) and buzz (120Hz) in an integrated amp consisting of two power amp boards and a preamp plus an unregulated PSU for the power amps and a regulated one for the preamp. I had fucked up pretty much everything I could in terms of component placement and am now fighting to get rid of the last bit of buzzing picked up by one of the volume pot->power amp wires. Your situation (p2p tube amp, presumably with no separate preamp circuit) is obviously quite different and there may be (lots of) things I haven't mentioned that may play just as big a role. I'm pretty sure the above principles always apply though.

I've only ever built transistor/chip amps with PCBs (yeah I'm a pleb), and one thing I've noticed is that on good PCBs, they always keep the input stage as small and compact as possible to minimize noise pickup. You might want to pay special attention to that. Also, for easier twisting I heartily recommend silicone-insulated wire - that shit is amazingly flexible, you can twist 16G wires around like it's nothing.

And one final note (sorry if this is obvious to you, happened to me a few times and made me wtf): Don't forget to ground your shields if using shielded wiring for anything! Neglecting that will result in disastrous buzzing for reasons not quite clear to me.

[Anonymous No.691960]

>>691948
i got it almost hum free now.
heater hum and input is now fixed.

still there is a slight hum but that is poor placement of the powertransistor.
that will be fixed when i put in a chassis

i'm very happy with this so far!

[Anonymous No.691963]

>>691960
>powertransistor
power transformer
just woke up^^

[Anonymous No.691965]

>>691960
What was causing it? Just the noisy heater supply?
Also, can you explain to me the principle at work in that B+ supply? Why are they using only two diodes instead of just putting a regular bridge rectifier there and connecting the negative output to ground?

[Anonymous No.691975]

>>691965
because it has a centertap in the schematic.
i just use a regular bridge rectifier.

i think this is centertap rectification business is a remnant practice form the days of vacuumtube diode rectification.
afaik; those have 2 plates and 1 cathode. having a centertapped transformer in this way only needed 1 tube, saving cost and power.

the noise:
-noisy heater, fixed with heater elevation discussed earlier.
-input noise, fixed with shielded cable on input.
-transformer noise. not too much, but i will fix it with smarter layout when i mount this in a chassis someday.

[Anonymous No.692115]

>>691450
Where did you start? I'm reading these threads and trying to inform myself as much as possible (casually) just so that one day i could do this shit and experience lovely tube overdrive toness

[Anonymous No.692127]

>>691462
I hammered the chassis over angle iron. Made it from scrap sheetmetal and pop rivets. Its a pretty accurate replica of the originals. I'd recommend high heat engine enamel for the paint.

Tubes are all early '60s British that i pulled from junk Hammond organs. The wiring is pretty sloppy, I keep wanting to pull the transformers and rebuild the chassis and circuits to much higher standards, but there's a million other things I'd rather build, and it sounds great.

Definitely shield the things that the schematic tells you to shield.

[Anonymous No.692128]

>>692127
pic 2. The through-chassis wire pass-throughs aren't the best. Use rubber grommets. I had to go in with layers of heat shrink and electrical tape for extra safety.

[Anonymous No.692129]

>>692115
I dove in head-first and built the thing. Trial by fire/electrocution. Learning how to do it beforehand helps a ton, I highly recommend it. Check out Morgan Jones' Valve Amplifier building books. I may or may not have, but definitely did pirate PDFs of them somewhere.

[Anonymous No.692167]

>>692115
>Where did you start?
OP here, i guess i started with school^^.
but i learned most from repairing all kinds of shit as a maritime repair guy for almost 4 years now.

like i said, i'm no audio guru. just made a load of pedals.
i guess that's a good place to start.
add a powerstage to a pedal and you have an amp^^

I've made a tube stereo from a kit. its pretty nice and has been used for hours every day for over 2 years with no issues.
but being a kit with no real design input form me, i learned little from it.
other than that I've made a 7au12 valvecaster pedal.

i'm sure there a kits covering what you are looking for, shouldn't be too hard for the layperson.

>>692127
>>692128
>>692129
that's very cool man!
looks good to me. but then again, my amp is on an old shippingcrate lid^^

yeah, i have an old tube radio from '46.
i'd feel bad slaughtering it, restoring it won't happen.
besides people throw these on the heap every day.

I guess i should be hoarding them from the scrapshop for transformers,tubes and nice boxes.
what do you think?

OT: I'm considering making a single knob tonestack alá big muff with some component adjustments to get rid of the loss of mid.
i wont be able to turn the lows as bar down as the highs, but it don't think its will be a problem.
the pic. is a from Duncans Tonestack calculator. very cool piece of freeware.

i have this potentiometer with a shaft you can pull out to actuate a dpdt contact set on its back.
i was thinking i'd use that to bypass the tonestack should i want it.
being semi sick, i don't feel like it now and i will be away for the weekend.
i'll update next week sometime, i suppose.

i will probably be lurking this thread from time to time.

[Anonymous No.692884]

>>692167
I don't want to just buy gear and a kit, make the kit and be done with it. I'm just interested in learning the mechanics of it all. (I'm a fairly new guitar player and want to know what's going on)
>>692129
Thanks, I'll see if I can find myself some treasure on the sea now!

[Anonymous No.693196]

>>692884
http://www.hnsa.org/doc/neets/mod06.pdf

>>692167
TSC rules. I'm a fan of seeing what the stacks actually do.

Your shielded leads can be anchored to the tube plate rather than ground. This increases capacitance and could lower hum/add stability at possibly a small cost of high end. There are inherent safety concerns for doing this correctly (100ish volts on the shield).

[Anonymous No.693861]

>>692884
that's a good attitude, and i wish for you to go for it.
but electronics is a big field and can be a bit overwhelming for someone without prior experience.
don't get discouraged if it's a bit much at first.

>>693196
nifty trick, but it seems like a bit of a hail mary for just a few hundred nF's.

if HT noise was an issue, i got a lot of 6800µf 450v caps i could use.
way overkill and potentially lethal amounts of capacitance at that voltage though.

no need for them now, but i'll probably use them on other tube amp projects.
maybe for a highvoltage bench supply or something.

[Anonymous No.693863]

>>692167
filthy pedalphile!

[Anonymous No.693925]

Is it just a guitar person thing?
Why does no one say Valves?

[Anonymous No.693972]

>>693925
I always thought referring to vacuum tubes as valves was strictly a british thing...

grew up in the U.S and always heard them as Vacuum tubes until I got into guitars where most amp manufacturers shorten it to tubes... the vast majority of guitarists I know that care could tell you that tube amps have a different sound than digital, but other than saying that tubes are analog and that some models of tube sound different than others, most don't know that the tube acts as a valve in electronics terms... or at least thats how I understood it... and really I only learned it when I started trying to wire my own pedals... was interesting to learn that tubes were the original diodes, and how they actually worked... other than guitar amps and some stereophile and radio stuff I don't think you run into tubes in much else... so it makes sense that the most common term for it is the one used by the largest market... as in guitaramps...

then again pretty sure you can get a valvecaster pedal.. so their is that...

[Anonymous No.694011]

>>693925
I prefer "valves" and live in the Obama nation.

>>693972
Voltage controlled current devices. Seems about right.

[Anonymous No.694187]

isn't it thermionic valves/tubes?

[Anonymous No.694234]

>>693861
Luckily I pick up things quickly and whenever I did electronics before it always made a lot of sense to me. The books/pdf recommended to me here look really valuable and have already read a few chapters here and there. Cool stuff.

[Anonymous No.694271]

slapped together that 1 knob tone controll today.
seems to work fine

[Anonymous No.694272]

you can pull out the shaft to bypass the control.
this should boost quite a bit relative to the loss in the circuit.

if its to drastic i might put a little trimpot on there to reduse the bypassed signal a bit.
i'd be a shame if i blew out my ears just by pulling a knob.

i'll try it on the amp when i get home.

[Anonymous No.694801]

>>694272
>>694271
i tried it last night.
the sound is pretty muted.
i can dail for frequency but the sound is a bit dull.
with the bypass sounding so much better. i dont see the point of having it at all.

maybe i'll try to make a 3 knob tonestack later on.for now i like it with only gain and vol.

[Anonymous No.695484]

>>690267
I built a P1 off ax84.com. I double and triple checked my wiring she I first built it. Had crazy oscillations (motorboat) when first turning it on late at night, no audio. Turned it off, went to bed, slept on the issue.
In the morning, I immediately checked if the ground rail on the pegboard was connected to the chassis. This immediately fixed the problem, and beautiful tone came out instead of oscillation.

[Anonymous No.695590]

>>695484
cool. post pics, man.

[Anonymous No.696096]

>>695590
1/2
Here's an external shot. I'll have to crack it open to get some insides

[Anonymous No.696105]

>>696096
2/2
The AX84 P1 was a good starter project for a first attempt.
Biggest thing I learned in the process...never again to use pre-drilled epoxy perfboard with DIP socket pins shoved through them.

[Anonymous No.696115]

>>696096
that transformer...
why no toroid, sheilding or r-coil?
you are going to have some terrible noise problems not to mention an electrocution hazard.

the flux produced by that transformer and being so close to the tubes will cause problems.

[Anonymous No.696433]

>>696115
haha

[Anonymous No.697594]

>>696115
You're probably right. I didn't consider (or even know about) the flux geometry of the iron to tubes location. I could take the time to pot the iron. I also need to investigate some potential ground loops inside the chassis that I noticed in my above photo >>696105
But otherwise it is a no-nonsense basic design, and a good number of successful builds convinced me to give this design and layout a try. Having since seen a good number of other people's builds of admirable quality
(even god-tier design quality)
there are a number of things I would do differently. I So far, it's been pretty noise-free through the speaker, it makes fun twangy noises with some attached strings.

I am scared to death of electricity, and I am still standing because of that healthy fear and respect. There's definitely a shock hazard if you happen to touch the very wrong spots.
So, since this beast is not in a cab yet, I keep it locked up in the basement.

[Anonymous No.697624]

>>697594
eh, you'll get shocked eventually. everybody who's ever done anything with hv circuits has. like the shop teacher with three fingers. such is life.

[Anonymous No.697699]

just a thought.
would having a switch to couple in a diode on the heater supply be a decent starve?
it'd cut the heaters in half. which might be a bit much though.

anyone try something like this?

[Anonymous No.698026]

>>697699
re:thought.
add diodes in antiparalell to drop the voltage down, but not rectifying it. starving a few volts could be interesting. no?

[Anonymous No.699699]

alright, i've been away for a while.
i tried the one pot tone knob again and i kinda like it now.
straight up bypass is retardedly high gain though, having gain passed 1/3rd i need to be really careful with the volume.

since its not so bad with the loss in the toneknob i thought i'd put on a resistor with a bit less resistance than the lower impedance in the toneknob.

i found about 26k at 1kHrz using an lcr meter.
so i put on a pot and dailed it to 20k.

i still want a boost when bypassing, just full on sunn))) level of noise.

the point is to adjust to a level i like, then i'll consider a fixed resistor.

[Anonymous No.699719]

>>696096
>>696105

hey man, can you tell me about that cabinet?
looks sturdy and fit fr my purpose.

[Anonymous No.700055]

>>699699
Check out the Tiny Terror schematic.
It has both a nice post phase inverter master volume and a one knob treble cut.

Both are extremely simple. The Daisy cutter has two volume controls that could easily be wired to one dual pot.

>>698026
This has been suggested for bringing down filament supplies that are on older transformers coiled for lower voltage mains. This is to correct the over 6.3 voltage they will transfer from today's mains' voltage.

Would be interesting to hear what you come up with designing and using an adjustable starve.