twist in the top of a cross slide

Hi Richard,

This one's for you. As you remember, at the end of your class I was working on the cross slide of my Studer cylindrical grinder. I had to put it to one side for the past weeks, but am getting back to it over the holidays.

I spent some time carefully studying the geometry. After a bit of head scratching I realised that the geometry of the top part of the slide is almost perfect except that it's warped across the diagonal. If I lay it upside down on a surface plate (ground surfaces down) then it's about 10 microns = 0.0004" high on two opposite corners. If I tap the four corners with a plastic hammer you can hear it clearly. Now here's the thing. If I push those two opposite corners down flat on the surface plate, then the geometry of the dovetail ways is really good, within a couple of microns (0.0001"). But if I don't push those opposite corners down, then the geometry is off by that same amount.

In use, the pressure from the gib and the wheel head pushes this flat. So I think this only matters if I want to use that top half of the slide (which is the longer part) as a master for the bottom part (where the geometry is off by a lot: 50 microns = 0.002").

I was thinking about your demo of bending a gib straight, and what you have told us about 3-point support and parts getting a deform set into them from their own weight. So I was wondering, can I put the top slide upside down on my workbench with a strip of wood underneath the diagonal, and use a couple of woodworking clamps to apply force (say 50 pounds) to the high opposite corners, then let it sit like that for a few days? The top slide is cast iron. Would the pressure deform it back to the correct shape again?

Best wishes for the holidays and the new year!

Cheers,
Bruce

Hi Bruce,

You sure get anal on this...lol...I am assuming your talking about the wheel head slide? No picture, so I went back to the Studer thread and looked at it. I do recall during the class you didn't scrape something I had suggested, just not sure now what. Going back to basics you need to make the slide 4 corner hammer tap tight in the relaxed position and when you tighten down the slide so it doesn't change. I suspect when it was ground it warped or didn't grind parallel to no noise tap. That's why we scrape after it's ground to improve it. To improve on how accurate the grinder can grind. Remember its a machine tool and not a CMM or inspection machine where you need millionths and not tenths.

Also double check the straightness of the ways and be sure they are not high in the middle. Time to be that anal detective again. Take a surface plate and check the surface that the wheelhead base sets on and work up. As I am writing it is coming back...I think I suggested leaving the wheel spindle side of the slide high .0005" to compensate for the weight of the wheel, wheel guard and wear as that side where the wheel throws more grit on that side so it wears faster. How wide is the wheel? 1 to 1 1/2" and how far apart are those flats and dovetails? 10"? Think about the multiplication of the error. .0002" or even .002" divided by 10". .00002 or .0002" ? And the wheel gets dressed....right?

If I were you I would double check things with the hammer test and scrape it keeping in mind to leave the wheel side high, double check to be sure the middles are low a couple of passes or in your case .0002". Middle low keeping in mind the travel of the slide. Not the middle 40% if the side moves to far or into the 40% ( I think you know what I mean, even if the readers don't) One more thing is the oil film. On that machine with the scrape marks being aprox. .0002" deep you should have .0002 to .0005" oil film or shake above the high spots. So if you have say .00005" to .0001" twist how does that matter. Is your shop temperature controlled? Is your coolant temp. controlled? I would take all that into consideration and remember the swivel slide needs to be tightened down that can pull a small twist in. Scrape all the surfaces and be sure to check the hinge.

I doubt pulling the slide with a wood clamp will change a .002" twist. A thin gib compared to a casting that thick? But you have time to test and recheck it, go for it. I would tighten it down and worry about what that condition is and not so much when it is loose sitting on a plate independent of the assembled wheel head weight. Merry Christmas to you too. :-) Rich

Rich, thanks for the fast reply! I'll post some photos after the work is underway.

Richard King said:

I doubt pulling the slide with a wood clamp will change a .002" twist.

Click to expand...

The top of the cross slide is 0.0004" high on two opposite corners, but otherwise the geometry is good. That's what I would like to remove with clamps (if possible) and was asking about.

The bottom of the cross slide is off by 0.002". I'll have to scrape that in, as you said working my way up and keeping your other points in mind. On the left (wheel side) the front of the slide is low, and on the right the front of the slide is high. That's probably what twisted the top. Note that the grinding head is reasonably well balanced left to right because the weight of the grinding wheel on one side is compensated by pulleys, belt and belt guard on the other side.

Don't know what you have Bruce but I have done a lot of straightening. If straightening gibs only a very small amount My trick is to brace the part on top of two places, and set a stack under the part to straighten at the center of the desired bend place, then, pressure to a certain dead stop amount of bending. with removing the height of the stack to change so the height the bend/straightening is kept in control. I have removed a little as .015" per push on a very valuable non replaceable hard part.
I was taught this method by a journey blacksmith (Bobby Doda) and have even used it on hard enough to break parts.

I did not see Richard's three point method but he really knows his stuff..better than me I suspect in most subjects.

Qt [clamps to apply force]...I would say no to that..... Still a .002 in a thin gib that a two finger push might push flat should not be any problem IMHO.

Buck

The way I straighten gib's. Scroll to minute 5:30 Richard King scraping class Norway August 214 - YouTube

Richard King said:

The way I straighten gib's. Scroll to minute 5:30 Richard King scraping class Norway August 214 - YouTube

Click to expand...

OK, I'll try this method on the top of the cross slide. Meaning, I'll support the top along the diagonal and then pull down the two high corners with some woodworking clamps, measuring the deflection, and monitoring my progress. If you think that's a mistake, please tell me!

Cash, the cross slide is about 20" long, 8" wide and 3/4" thick. I only need to remove about 0.0004" of twist. Do you think that's realistic? I won't employ clamps that are strong enough to break the part.

4 tenths twist I would scrape out plain and simple. Let's say you could pull the twist out of a cast iron part you 1. run the risk of breaking the part no matter how gentle you think you're being and 2. never know if it's going to settle back into the twist over time once again. If it means that both sides have to be scraped then so be it.

To avoid breaking a part is why I use a stack under a straightening job..It can only go exactly that far..If it does not straighten a tad then I remove a bit of the stack and push again. Mostly done on a straightening press but have done same on a table with a clamp.

I used to straighten hard spline broaches that were about 6 feet long. I used a carbide point held in a small hammed gun to expand an area opposite the high place.. some had multiple highs and lows so needing to back over a few times..Ye the pecks were in the gullets so were no harm to the broaches.

*For .0004 I think peter nailed it.

michiganbuck said:

For .0004 I think peter nailed it.

Click to expand...

It seems like a terrible waste of time to me. In normal use the pressure from the wheelhead and the gib remove this twist. (I can remove the twist with the pressure from two thumbs on opposite corners of the slide. The grinding head, wheel, motor, guards etc that normally sit on top weigh at least 150 lbs). If I scrape out the twist then I need to redo the geometry of two ground top surfaces, two plane lower surfaces, and one of the two dovetail flats (the other is behind a gib so gets adjusted when I do the gib itself).

I could likely straighten a .0004 out part with my technique being sure not to opposite bend it too far with the push… Likely I would stop at .0002 or so and call that good enough..Or consider a two thumbs push of perhaps 8-10 pounds push not a huge problem to allow the .0004 to be there . But I understand you are German and so can't bear the thought of error...sad to say I am that way also. “Oops” is a sad thing to say for a machine part.
Plus going too far and getting a twist the other direction is possible with not employing a control.

Agree Richard watches the indicator to know the extent of the push so with doing that he has control also. When a part gets to a certain place stress can often be relived with not breaking the part..but jut a little past that point "OOps".

OT: I once read in a government testing lab they discovered a grinding burn can produce a huge compressional surface stress, I think it was 200,000 tons.. I will have to search my books and see if I can fond that report.

OT: When are you next going to be in the USA. I have some clay surface grinder wheels for a very good surface finish and I could post one/couple/few to a US address, free because they are surplus to my needs.

You're are starting to remind me of Sheldon on Big Bang Theory. Check to make sure the middle isn't high. It may only need stoning or one or 2 scrapes. .0004" in 10" and you can push it straight with a finger...Have you rung it? Put it on 3 points with the single point off center of the twist on the low side. You could even pile up some weights on the high .0004" side. I can't believe I am even suggesting that as .0004" is nothing especially when it is tightened down. Better use a torque wrench too, and never grind on the machine, the strain of wheel push away might warp it. Oh and never lean on the machine or lay anything like a wrench on it as it may move .000001"

michiganbuck said:

*For .0004 I think peter nailed it.

Click to expand...

Yes. Seriously.

Just too small an increment to chase with counter-bending and expect it to be stable.

Presuming it IS "stable", as-had? Better to leave the balance of stresses right where they are and do but a MINOR touch-up to the critical surface(s) so they simply co-exist in harmony rather than "fight" each other.

Richard King said:

You're are starting to remind me of Sheldon on Big Bang Theory. Check to make sure the middle isn't high. It may only need stoning or one or 2 scrapes. .0004" in 10" and you can push it straight with a finger...Have you rung it? Put it on 3 points with the single point off center of the twist on the low side. You could even pile up some weights on the high .0004" side. I can't believe I am even suggesting that as .0004" is nothing especially when it is tightened down. Better use a torque wrench too, and never grind on the machine, the strain of wheel push away might warp it. Oh and never lean on the machine or lay anything like a wrench on it as it may move .000001"

Click to expand...

Richard was that aimed at me...sorry if I got out if line
Buck

No...to Bruce...he was asking me...I asked him if he rung it...put it on 3 points...etc....lol

Richard King said:

I asked him if he rung it...put it on 3 points...etc....lol

Click to expand...

Yup, did all those things. But can't get the twist out, so I will scrape it.

I've decided that I might as well "do it right" which means starting at the bottom and working up. In this case "the bottom" is the lower (fixed at 90 degree) cross slide which has one flat and one V. A year ago when I put oil pockets into the long slideways, you commented that I ought to do the same for that lower fixed cross slide also. So I've pulled it off and will do those first, then work my way upwards correcting the geometry step by step.

The 90-degree cross slide geometry is still spot-on, so all that is needed there is new oil pockets. Richard, I can do those the way that I did the long slideways, which is to square cut them in a 1/8" checkerboard. Or (thanks to your class) I can half-moon them by hand. I just practiced the half-moons on a bit of cast iron and it works a charm. The half-mooning is quite a bit faster. Should I do that? Or should I stick to the square-cut oil pockets?

[EDIT: I think what you said in the class was half-moon pockets in the upper part and square-cut pockets in the bottom part. That makes sense, so unless you tell me otherwise, that's what I'll do here.]

Best wishes for the holidays and the new year!

Time to pull up the undies and YOU decide. I told you in class and on the DVD about exposed to air flaking. Also compare the depth, You tell me. I am glad you have been practicing as You know what I always say...:-) Practice makes perfect....and you are a perfectionist.
Fröhliche Weihnachten :-)

Richard King said:

I told you in class and on the DVD about exposed to air flaking. Also compare the depth, You tell me.

Click to expand...

I'm still not sure, which is why I am asking.

On the one hand, you've said that it's better to half moon on the top, so that the pockets don't collect grit. That makes sense to me. But then you've also said to half moon the shorter part, so that the half moons are not exposed to the air where they can collect grit. That also makes sense. The problem is that in this situation, these two guidelines run contrary to each other.

In the case of this grinder, the longer part of the fixed cross slide is the top part, and the shorter part of the fixed cross slide is the base. So if I want to half moon the part that is not exposed, that means half moons on the bottom. On the other hand if I want to half moon the upper part, that means half moons on exposed part.

Since this is an area (near the bottom of the grind wheel) that is near coolant splash and grit, it would be good to get it right. What would you do?

PS: I did search and found this post where you half-mooned both surfaces of the long ways on a Landis cylindrical grinder which has some parallels to my situation. You ground the blade differently in the two cases to get different depths. You did shallower depth 0.0005" half-moons on the lower exposed surface and normal 0.002" half-moons on the upper hidden surface. But that's a different situation from mine because the exposed surface is the bottom one. In my case the exposed surface is on top.

If I were doing it, I would scrape using the checkerboard pattern scraping I taught you, match fitting and lowering the middle keeping in mind the travel of the ways. If the travel exceeds the 40% we usually lower then I may only relieve 25 or 30%. Also you remember that hand scraping generally is a depth of .0002". On grinders we do not 1/2 moon the exposed surfaces the regular way. Meaning the blade sharpened at neg 5 degrees blade tip rake we use on mills and lathes, etc. Regular rake average depth is .002". Many times when I 1/2 moon machines I take into consideration the travel limits and what's exposed and on Mills, lathes, etc the depth of the 1/2 moon is .002"


Like I said you will need to measure travel and decide what area's to 1/2 moon so over travel isn't flaked the regular way. On the thread I did on the Landis grinder I did not 1/2moon the exposed side the regular way. I ground the blade tip to I believe 10 degrees neg rake so the depth of 1/2 moon was around .0002" like hand (or power) scraping. I believe I said in that thread my back was hurting and I didn't want to be bent over so long square cutting scraping it even when using a BIAX or especially by hand. That machine was like your Studer, it ground well but was loosing the scrape marks and was about to get stick slip and galling.

I talked it over with the machines owner (Minnesota Grinding) about 1/2 mooning the exposed top of base surface to the same depth of handscraping as he knows I never flake the exposed surface. So I 1/2 mooned the exposed bottom side with the neg 10 degree rake and 1/2 mooned the unexposed top or bottom of table the regular 5 deg. rake blade. I hope this helps you make your decision. Fröhliche Weihnachten (Merry Christmas.)