/diy/ - Do It Yourself

>>1407670

Looks like your dressing isn't up to snuff, and/or the gibs on your machine need some adjustment.

>>1407670
Is your dresser sharp? I've heard of people melting the braze and turning the diamond 45 degrees to get back on a sharper face. Or if your dresser is a separate cylinder from the base, you can undo the set screw and turn it. Then, from touch off, go in about one thou, and crank across decently fast, and that's it. Don't do a spark out pass. The fast dress leaves the wheel grit sharp. That will allow it to cut without excessive heat buildup. I am assuming you also use no coolant, which would make this more critical. Also, throw the table fairly fast.

I am not an expert on grinding, but this has worked well for me. I only occasionally run the grinder, usually when I'm making tools for myself (or sometimes just to make stuff shiny for looks).

>>1407696
>>1407693
Sounds good, yes I can dress it better, I've recently started turning the diamond a little for a different edge. I did do a skim pass on the dress too.

Correct not using the noga minicool I have set up for the machine.

I wonder if because it isn't level front to back there is some twist in the bed (Y axis) and the wheel is actually hitting the work at a slight angle? I wonder what else would cause that, probably lots of things on a surface grinder.

>>1407701
But one would think dressing the wheel with that twist would make it level to the work...

Unless I dressed it up near the front where there is ostensibly less out-of-squareness with the column, then ground the part closer to the operator?

>>1407670
update, leveled the machine front to back, would like rigid feet for it to stand on, but here's the difference

>>1408100
The weird thing about those lines is, they don't match up with my in-feed, at least I don't think.

>>1408101
at least it looks better than ripples going across it, it's basically less than 0.0001" on the surface, I think

>>1408102

no offense, but I hate threads like this one. "hey look guys, I have an expensive tool that none of you will ever own, please watch me use it incompetently"

>>1408110
Surface grinders aren't that expensive used compared to some other things.

https://www.youtube.com/watch?v=5kUWScmbJQM

>>1408110
and the imperfections on the surface there are microscopic...

It looks good IRL and feels smooth as silk, but it's not a good mirror finish yet so I'm still trying to improve it.

If you would call that incompetent well, I don't think you know how bad other people really are at this stuff.

pic related is the surface grind from the Tormach demonstration video, trying to sell their tormach surface grinders

>>1408102
The finish pass was like 0.0003" btw, and I sparked it out afterward by zig-zagging across the surface.

>>1408144
The piece also isn't hardened.

I believe it's 4140.

>>1407670
maybe /diy/ is more interested in lapping?

>>1408110
>I have an expensive tool that none of you will ever own

https://www.ebay.com/itm/Norton-Hydraulic-Surface-Grinder-with-6-x-18-Magnetic-Chuck-model-687-A/183177288684

>$700 all day long on ebay, just gotta find one close!

>>1408153
Where's your optical flat, boy?

>>1408179
Bought a soviet one on ebay, it works.

>>1408193
Good on ya

Damn, surface grinders going for $350 these days.

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

Looks like my grind might not be all that bad after all.

Even a master's grind isn't too perfect.
https://youtu.be/DVLXsq7pi9Y?t=942

This is the tailstock from a cheap chinese lathe. I took it off to see why it was .008" higher than the spindle.
it's rough and scratched up, like somebody just used an angle grinder on it. The top of this piece that mates with the tailstock is ground flat. it's like they just forgot to grind it at the factory.
I don't have a surface grinder. I want to clean this up but I don't want to risk putting the tailstock out of parallel, even in this condition it's parallel within .001 over the whole quill travel.

>>1408855
you could try lapping it, it will improve the surface finish without removing much material

see: >>1408153

>>1408855
as to how to approach lapping it, well if you don't want to run it up and down the slideways of the machine (understandable) you can buy/make/scrape in a cast iron straight-edge to use as a lap.

http://www.mini-lathe.com/lapping/lapping.htm

http://www.mini-lathe.com/Lapping/mt_lap.htm

>>1408855
I suspect the quill doesn't fit the bore of the tailstock perfectly either, so see if there's some play there.

Take the quill out and use a micrometer to see if it's actually round

>>1408919
The tailstock bore is actually pretty good. The quill is round and fits close, the surface finish on the bore is nice.

>>1408906
I have a granite surface plate. Do you think it's better to lap with sandpaper on the plate, or mark high spots and scrape? I don't have much experience with scraping.
And is there a trick to ensuring it stays parallel? Or should I just frequently clean it, reassemble, install on the machine and indicate the quill?
In the past with sandpaper lapping small parts, I've clamped them to a bigger block, but the tailstock is too big to do that with.

>>1409311
well... if you don't lap the V groove too it's gonna lean to one side.

scraping would probably be a bit more difficult

any chance you have a piece of aluminum that will fit into the V groove properly to lap it with?

aluminum laps work well, if it's 90 degree V you could try using the edge of a surface plate with sandpaper or a sheet of aluminum foil + lapping compound over it to reach down into the groove and lap it. I'm thinking you could hold the piece up against a square edge of your surface plate if the V groove is 90 degrees.

honestly the lapping would give great results, but you might still need to flake some oil grooves into it, and you will have to check it to see what you're doing either way.

0.008" is a lot to remove! Especially by lapping, if you're ballsy enough you could machine 6 or 7 thousandths off of it first.

>>1410080
hmm I suspect that scraping would be the way to remove up to a thousandth or so of material, grinding for 5 or 10 thou, and machining for more than that, and lapping for under a thousandth.

>>1409311
also, I hope you don't clean your plate with water, when the humidity changes the water will rise up to the top of the plate and you'll be trying to work in a puddle. It also rusts anything metal you leave on the surface of the plate and the rust particles stick just below the surface leaving a nasty looking mark (you can clean it, but it's not worth the trouble mostly).

most inks and dies will stain your plate too, if it isn't black granite. Please try some of this stuff, it's supposed to not stain up the plate and actually comes off your hands unlike "prussian blue" which is grease-ish and will stain up a non-black plate.

Canode Blue (also comes in red and yellow), one of the big bottles is probably a lifetime supply, and if it dries up add some water to it.

http://www.artcotools.com/Canode-Blue-Die-Spotting-Ink.html


The "trick" to remaining parallel is checking it on the surface plate, if you have a good surface plate don't use it for lapping. Use one of the imports for that, and the good plate to check your work with an indicator and one of these: https://www.ebay.com/itm/LUFKIN-520-ST5009598/391960667448 (or you can make your own, it holds an indicator and you can slide it around the plate instead of sliding the object around under the indicator, a mag-base indicator stand would work too)

>>1410358

>>1410358
>>1410361
Tom Lipton's "square off" videos show some good surface plate action.

https://www.youtube.com/watch?v=HVTuWEjqCo8&t=190s

and here's how lapping an OD goes, veryyy slowly

https://youtu.be/HRkZtZYJ0DM?t=1095

>>1410413
and notice he uses "timesaver" lapping compound on the sliding parts, it goes inert and stops being abrasive after a few minutes in case any abrasive is embedded into the part.

>>1410080
>I'm thinking you could hold the piece up against a square edge of your surface plate if the V groove is 90 degrees
That's what I was thinking too. unfortunately the groove isn't 90 degrees. it's about 88 degrees.
why would they choose something so arbitrary I have no idea.
Milling an piece to fit the groove is easy enough, though.

Facing off a few thousandths might not be a bad idea. I can get it very square on the mill.

>>1410358
Never with water, always with isopropanol. the pure stuff, lab stuff, not first-aid rubbing alcohol.
Haven't tried putting dye on the plate. the few times I've tried scraping, I sandpaper lapped a block flat, then put the blue on that.

>>1410579
In the OD lapping video Lipton is removing a tenth at a time via "power lapping" in the lathe.

ID lapping is similar and can be done on the mill.

Maybe starting with sandpaper and finishing up with lapping compound would go faster. Kind of wonder how you will check your progress on the V groove on the surface plate though, I guess you could drop a precision round rod (gauge pin or something) into the groove and measure the height it makes the tailstock plate sit at, somehow.. Probably better to avoid measuring an angled surface, and lay the plate on it's top/bottom and have the round rod sitting in the groove facing upward, then measure the top of the round rod with the indicator.

For the "measurement" well as long as you have something the same height as where you started and can compare it to where you are, a relative measurement would be fine. The alternative is a height gauge for a more accurate measure of distance. A gauge block stack would work well for comparative measuring.

Some way to measure if you're lapping it straight/perpendicular to the sliding action of the tailstock (offsetting it left/right) would be good too, or you could try to machine the slot square to the sliding of the plate afterward (the tailstock offset sliding groove).

Chances are the slot is all over the place right now anyway though, it might involve adding a new key to the top part of the tailstock.

but at the end of the day, the measurements you probably care about are at the tailstock quill, does it extend in-line with the bore of the headstock or is it crooked, etc.

making the sliding portion square and parallel and lapped with oil flaking might be futile if you have to scrape/grind/lap the part that sits on it too.

>>1410592
and be sure to check for good contact with the ways of the lathe too, lol

>>1410594
>>1410592
>>1410579
it almost sounds like scraping would be the right method to use, because of the difficulties created when not lapping two pieces together, but needing to do slight modifications to the piece after checking it.

tl;dr, try your hand at scraping it, the measuring and checking is mostly the same tho.

>>1410595
>>1410594
>>1410592
>>1410579
and with 0.008" to go, some crappy scraping at first wouldn't be too awful.

>>1410579
but be sure the machine is level (very level) before you try measuring contact with the machine ways, they will be bowed out or twisted if it isn't.

>>1410579
another thing you might be concerned with is the leveling of your mill, and if the head swivels then the tram of it too.

Put an indicator on the spindle and spin it around against the table making a 6" circle, see how flat the table is to the head/spindle.

variation of 0.0005" or more would not be unexpected, if it's a bed style mill then leveling will absolutely affect that value.

>>1411112
bed mill is the same style as Tormach mills

>>1410595
What do you think about putting dye on the ways and using that to mark the tailstock? Seems like it would be a lot less labor-intensive if it would work.
I just realized milling the tailstock base at all would be very difficult, because of that <90 V angle. I don't have the right kind of cutter.
Marking with the ways and scraping seems best, that way I could be sure it matches the ways, unless there's some kind of hidden difficulty here.

>>1410599
This machine at least, I am sure is level. I installed it.

>>1411165
Yeah that would work, it would only compare it to that one piece of the lathe bed so you might want to move it up and down once in a while.

You could probably use a fly cutter to mill the V surface, or maybe even a dovetail cutter.

>>1408145
Try a less aggressive cross-feed like .005/.010 and rotating it after each pass. With the less aggressive cross-feed you allow more of the wheel to blend/smooth out the area you've already covered. Also bear in mind you are also going to be generating a lot more heat doing that so if you are looking for really fine .0001/.0002 flatness you're going to have to deal with the lines.

Also try a finer grit wheel. Theres like 20 different reasons your shit could be all fucked up.

Can't you just take the tailstock apart where it adjusts from side to side? That part should be just flat with a keyway. just surface grind that down or something.

>>1411354
ok, I'll give it a try