Another surface grinder conversion thread.

kuraki said:
That looks like wheel runout to me. If you slow down the table speed (not crossfeed) does the pattern change? Not sure if that's what you meant from what you've tried. I could see it being related to the serrated wheel too, have you tried lower belt tension? Could the belt be conforming to the serrations enough that it's actually presenting a faceted wheel to the work surface?
Click to expand...

I put a dial indicator on my spindle once upon a time and got almost zero run-out, so I really don't think it's the bearings. I also dressed my contact wheel, so theoretically that should be ok as well.
You might be on to something with the belt tension, though I recently increased it and it seemed to get a little finer, especially with spark out.

This is kind of just one of those things I've been living with, as a little bit of hand sanding or cleaning up on a flat platen will take the pattern out pretty quick, but I would like to get to the bottom of it eventually.

I'll have to play around with it some more when I get time, and see if I can get some sample blocks to compare with different feeds, speeds, tensions, etc...
 
knife to a gunfight said:
I put a dial indicator on my spindle once upon a time and got almost zero run-out, so I really don't think it's the bearings. I also dressed my contact wheel, so theoretically that should be ok as well.
You might be on to something with the belt tension, though I recently increased it and it seemed to get a little finer, especially with spark out.

This is kind of just one of those things I've been living with, as a little bit of hand sanding or cleaning up on a flat platen will take the pattern out pretty quick, but I would like to get to the bottom of it eventually.

I'll have to play around with it some more when I get time, and see if I can get some sample blocks to compare with different feeds, speeds, tensions, etc...
Click to expand...

How did you dress the wheel? Something like a lathe bit on the magnet?

I was trying to come up with a realistic way for you to simulate a solid wheel with your serrated one by covering it or something but nothing I can think of would stay round or be something you could dress.

One thing I haven't really heard anyone talk about in regard to surface grinding and conversions, that I was taught and have experienced first hand, is that a dressed rock wheel is only dressed until the spindle is shut off. Once the spindle is stopped, you dress the wheel again or you will get similar patterns to what you have above. Some grinders are worse than others. But things settle enough when stopped to throw the concentricity of the wheel out .000050 or .0001 and that's enough to cause visible deviations in your work.

If I were you, I would get set up to dress your wheel, then run the spindle motor for a few minutes to let the bearings warm up, no belt, then dress a thousanth or 2 off of the wheel. Stop the motor and put your belt on and start it again quickly before the bearings cool off, and see if that's improved anything.

The other thing is your dressing the contact wheel without the belt tension on it...that's going to change things too.

If you have a 2" wheel, can you split a belt to 1" and track it over to one side, start the grinder, dress the wheel where the belt isn't tracking, then track the belt onto the freshly dressed portion?
 
kuraki said:
How did you dress the wheel? Something like a lathe bit on the magnet?

I was trying to come up with a realistic way for you to simulate a solid wheel with your serrated one by covering it or something but nothing I can think of would stay round or be something you could dress.

One thing I haven't really heard anyone talk about in regard to surface grinding and conversions, that I was taught and have experienced first hand, is that a dressed rock wheel is only dressed until the spindle is shut off. Once the spindle is stopped, you dress the wheel again or you will get similar patterns to what you have above. Some grinders are worse than others. But things settle enough when stopped to through the concentricity of the wheel out .000050 or .0001 and that's enough to cause visible deviations in your work.

If I were you, I would get set up to dress your wheel, then run the spindle motor for a few minutes to let the bearings warm up, no belt, then dress a thousanth or 2 off of the wheel. Stop the motor and put your belt on and start it again quickly before the bearings cool off, and see if that's improved anything.

The other thing is your dressing the contact wheel without the belt tension on it...that's going to change things too.

If you have a 2" wheel, can you split a belt to 1" and track it over to one side, start the grinder, dress the wheel where the belt isn't tracking, then track the belt onto the freshly dressed portion?
Click to expand...

Some good thoughts. I probably didn't let my machine warm up enough, and as for dressing, I just put a flat file on my chuck and ran the table in and out a few times (y axis). Probably not the absolute best way.
I have a 1" wheel that I split belts for.


Sent from my iPhone using Tapatalk
 
I have not been getting that chatter pattern with my wheel. I've seen it before with a stone, fixed mostly by dressing it.
My first thought was to dress the contact wheel, too. A little rough sandpaper glued to some scrap bar, mag chucked down- but you've dressed it.
It is belt splice? Do you get this on worn-out belts and fresh belts alike? How about if you change your power feed speed?

I still think it's either spindle or wheel runout. The pattern is so regular. But, is your belt slapping around any? Could that be introducing a little vibration into the wheel? Or, perhaps your contact wheel is concentric and the spindle has minimal runout, but the wheel itself needs to be balanced? I could see that being the problem.
 
I've been using the diamond dresser on my gator belts. I tore them in half which was easier than i thought it would be.
It's important (and I'm very much a beginner) to not only dress the belt, but also to dress the chuck with the wheel or belt you're using.
You should be able to eliminate alot of that chatter.
 
Grinding chucks....sucks.

Part of our 6 month PM on our little Mitsui grinders at HTI was to regrind the chuck. We would grind 5 blocks, 1 in each corner and 1 in the center and then air gage them on a surface plate. I can't remember what we considered acceptable but I think is was only .0002" or so deviation across all 5 blocks.

But flatness of the chuck (and it's parallelism to the ways) isn't really going to effect surface finish like that, unless there's a severe problem supporting the work.

Salem may be on to something with the taped joint and removing grit above it. That belt bump will effectively gouge the material that's backed up by the magnet rather than your hand. Putting a belt on and dressing it with a diamond like you would dress a rock wheel might be another good experiment. You'll be able to tell right away how much higher that tape joint is projecting the grit.
 
I appreciate all the suggestions/discussion. I hope I'm not derailing Salem's thread too much. :foot:

The pictures I showed above are pretty old, and I've since improved the finish from what you see there, BUT it is still showing that same pattern, albeit very faint. I do always knock the grit off of the splice with a dia-pad, though I think I should try dressing with a diamond point as well. I'm definitely going to look at tension on the belt, and adjusting everything after the machine has warmed for a good while (especially since my shop is hovering about 40-45 degrees Fahrenheit these days).

Another thing I think I'm going to do is see about speeding up my VFD that's powering the spindle motor. Right now, it's running at 60hz, but I'd like to bump it up if I can to see if it affects the pattern one way or the other.
 
Well I've only ground one chuck so that's the extent of my experience. Maybe that doesn't help, but my finish seemed to improve after I ground the chuck then redressed my wheel or belt.
 
Knife to a Gunfight, not derailing at all- indeed this thread is becoming everything I hoped it would be; a virtual surface grinding symposium! Thanks for the input everyone, let's keep the dialogue going!
 
Thanks Salem for a much needed thread. I would like to know what HP your SG is so I might get a better understanding to apply your numbers to my machine, which is only 1/2 hp. (actually 1hp, my bad).
 
Last edited:
Hey, no worries man! I have a lot to learn on these too, definitely the conversation benefits me as well.

My machine has 3 hp. 3 phase at the spindle.
 
Salem, Count,

Thanks for the information. I've definitely learned a few things from this thread and once I move I'm going to convert mine to belts. Or maybe sell it and buy one with power feeds and then convert to belts :D

-Clint
 
HAZAAH!!!! I think you guys nailed it!

When I got home from work tonight, I turned my machine on and started cleaning up the shop for about an hour, to let everything warm up. I then chucked up a small file and very lightly dressed the contact wheel. After grinding with a dull 120 grit, I could already see marked improvement. A fresh 220 really put a nice finish on.

I took a couple quick clips:
[video=youtube;sO2IE9LrE2o]https://www.youtube.com/watch?v=sO2IE9LrE2o[/video]

Easily the best finish I've gotten out of this machine since I got it, although it is just a piece of 1/4" mild steel. This was also done with maximum step-over, and a moderate table speed. I think I could get an even better finish if I change a couple minor things, but letting the machine warm up first was definitely the key to success here.

Thanks to all who made suggestions and got me thinking about this old problem again!
 
Killer! That last finish was looking good indeed. Nice to work out a long-standing issue like that.

Here's a vid I took tonight, of my grinder taking off .010" depth with .100 or so stepover. No problem at all, and when I felt the workpiece, only a little bit warm to the touch. After aggressively taking about 30 thou total off each side, I took it off the chuck and it was flat and true all over to within a thou, and very uniform with no chatter or even ramped edges.

That's exactly what I need it to do. Not precision finish grinding folder parts (most of the time) but hogging and dimensioning pieces for billet work. This was a 52100 bearing, forged out to 2" wide by about .180 with scale, pressed flat, annealed, and taken to the grinder to clean up and match in thickness to the .120" thick 304L sides for san mai. Works great!

In the vid you can see that I need to build a spark deflector and a dust collection scheme for it. When I ran it with coolant, it didn't throw dust and sparks...

[youtube]PlxXGRFPIkU[/youtube]
 
Last edited:
Salem, will your grinder do a continuous cross feed? Sometimes I find that to be even faster without creating a lot of additional heat. I would imagine it might be particularly useful with a belt conversion since you don't get the same corner breakdown (therefore lead in radius) as with a rock wheel, by getting a greater portion of the belt abrasive in play.
 
My grinder, being from 1946 and mechanically auto-fed, will not continuous crossfeed... although that's an interesting note!

Something I found elsewhere, that helped me in my grinder mod, that I wanted to include here- a quote from Hank Knickmeyer:
"If possible move the idler assembly off center, usually to the left. This allows the belt a longer time on the contact wheel before contacting the metal and allows the belt to "settle down". Makes for better accuracy and longer belt life.
The above information came from Dan at Bader Corp. They did some real testing to come with it. Good people."
 
Salem Straub said:
My grinder, being from 1946 and mechanically auto-fed, will not continuous crossfeed... although that's an interesting note!

Something I found elsewhere, that helped me in my grinder mod, that I wanted to include here- a quote from Hank Knickmeyer:
"If possible move the idler assembly off center, usually to the left. This allows the belt a longer time on the contact wheel before contacting the metal and allows the belt to "settle down". Makes for better accuracy and longer belt life.
The above information came from Dan at Bader Corp. They did some real testing to come with it. Good people."
Click to expand...

VERY interesting. I suppose it makes sense too. I've been thinking about redoing my "frame" anyhow. To make belt changes a bit easier. May give this a try.
 
You must log in or register to reply here.
Back
Top