What's new
What's new

eliminating waviness altogether on small surface grinders

metalmagpie

Titanium
Joined
May 22, 2006
Location
Seattle
I have a 718 surface grinder now for a few weeks. I haven't been happy with the grinds I've been getting. I have tried several approaches and I am writing to summarize what I had to do to eliminate the waviness. I'll provide more details and pics later, but for now here's a picture of my final finish, considerably enlarged. There are longitudinal lines but almost zero waviness.

finishGettingThere.jpg


The aluminum heat sink is there to allow my phone's camera to focus.

Here is what I needed to do to eliminate the waviness:

level the machine
lubricate the ways correctly
tighten the wheel adapter flange nut really tight
do a precision balance on the grinding wheel
dress the wheel
use a single point diamond that's still sharp
dress the face, MUST be deep enough
dress the sides of the wheel
last dressing pass no infeed at all
break wheel edges after dressing with dressing stone
touch wooden stick to wheel to get 'hangers'

The finish is also affected by the longitudinal traverse speed, crossfeed amount, and depth of cut. However, I could not prove to myself that these affect the waviness.

Similarly, the choice of wheel makes a difference in finish but I didn't see any correlation with waviness. I switched back and forth between a 46H and 100K wheel. Quite different, but again didn't touch the waviness.

Finally, I had heard that 3 phase power from a rotary phase converter is not as good as the power provided by a VFD. I have a machine with a VFD and an auxiliary receptacle, so I was able to plug my grinder into the VFD or use the power from my RPC at will. I didn't see much difference at all between the two.

I will have more to say later about how I did the wheel balancing. My results showed that balancing the wheel helped to eliminate waviness (aka chatter, cogging etc.).

metalmagpie
 
Last edited:
What in the world is with the lines? Are those your crossfeed lines? Looks terrible.

This was the first corner of the part I ground. I did it right after dressing the wheel but I forgot to break the edge of the wheel with a stone. The result was those lines. The lines aren't the point, though. No chatter. No ripples. No waves. That's the point.

I owned a nice K.O. Lee 612 once and never could make it grind without ripples. Eventually I sold it because I just couldn't stand to look at the grinds it made. My current grinder is also a K.O. Lee but a bit bigger at 718. When the ripples showed up on this machine I about had a conniption. But I'm older and more experienced now, plus more postings have been made on the Internet that I could research. So I took a few days and patiently tried all kinds of stuff. Eventually I got rid of the ripples. I figure the lines will be easy compared to those.

metalmagpie
 
Thank you for posting, I have never been happy with the surface finish I have been able to achieve on a surface grinder. Being self taught has its limitations.
 
Balancing the wheel is probably going to get you 90% of the way there. That is certainly a novel method... Not sure I'd rather do that than a static balance. I can test a static balance.
 
I think I've only seen one page where somebody tried to build that type of balancer. They weren't entirely successful, but I know they can work because they're used commercially for various purposes. My only comment on construction is about contact area and what holds it in place. The faces that clamp the wheel must be identical to avoid damage to the wheel. Does it press fit to the wheel arbor, or is it in contact with the wheel? I made a ring that accepts weights that's certainly sub-optimal, as you mention. I was going to make a new nut with a lip that could accept adhesive wheel weight pieces, but never got around to it. If the balance is good you should actually be getting better results than in your photos. I'm guessing your wheel is a bit overdressed, as a finishing pass with no feed isn't usually a good idea. Material matters. I get my worst results on mild steel, but that's what I grind the most of.
 
I think I've only seen one page where somebody tried to build that type of balancer. They weren't entirely successful, but I know they can work because they're used commercially for various purposes. My only comment on construction is about contact area and what holds it in place. The faces that clamp the wheel must be identical to avoid damage to the wheel. Does it press fit to the wheel arbor, or is it in contact with the wheel? I made a ring that accepts weights that's certainly sub-optimal, as you mention. I was going to make a new nut with a lip that could accept adhesive wheel weight pieces, but never got around to it. If the balance is good you should actually be getting better results than in your photos. I'm guessing your wheel is a bit overdressed, as a finishing pass with no feed isn't usually a good idea. Material matters. I get my worst results on mild steel, but that's what I grind the most of.

The back flange of my wheel adapter is only a small amount thicker than one of my autobalancer halves. So one side is pressed onto the adapter and the other side is pressed on also but is mainly held with the 3 pressed in dowel pins. Any contact with the grinding wheel is incidental.

I have seen your balancing setup, Conrad. Very cool.

My experience with small SGs is that the least little thing can have a profound effect on the result. I would have done a few grinds and got a much better finish but my hydraulic unit is giving me grief so I can only work for a few minutes. I'm certain I can get the straight lines out. It was the ripples that were killing me.

A balanced wheel is no assurance of good finish. It just eliminates one of the causes of bad finish.

metalmagpie
 
Balancing the wheel will just get rid of most of the jaggedy waviness... Assuming the spindle bearings are in good condition. You'll never completely eliminate it.The actual surface finish is obviously dependent on a lot of factors.
 
In my tool&die days, I never had to balance a wheel, and never got results that bad (unless the wheel was completely f'ed up). Maybe spindle bearings, maybe table not smooth motion, maybe chuck not flat and parallel..??

:confused::confused:
 
Here's a little page I wrote attempting to describe how I balance a surface grinding wheel.

metalmagpie

http://nwnative.us/Grant/shop articles/autobalancer

the way I understand function of this device is that it is not balancing anything, but instead it doesn't allow the system (spindle column) to go into its natural oscillating frequency by constantly shifting the imbalance phase in relation to spindle column resonance frequency

think of trying to push a swing, if your pushes (small amplitude) are in regular intervals all the time no matter the frequency, just that the frequency is the same all the time, eventually you will get the swing swinging at its natural frequency, the swing amplitude may not be great, but it will be steady

now 2nd case, you keep bumping the swing at irregular intervals, more or less chaotic frequency, there is little chance of getting the swing to swing at all, sometimes your bumps will combine, other times they will cancel out the energy transferred into the swing previously - eventually this still may "settle" and you might get the swing to swing somewhat, but in case of the surface grinder, you are taking some of this "bumping energy" accumulation out of the "swing" each time the wheel is cutting the work piece further disrupting the resonance which leads to smooth cutting action

at the end it accomplished the same result as perfectly balanced wheel on a perfectly rotating spindle nose (no runout), but I wouldn't call that effect - balancing

I may be wrong about this, but that is how I understand this "dynamic balancing", something similar is used with stepper motors to smooth out their motion and overcome some specific rpm ranges where they may lose steps
 
Same as Mike1974. Never balanced a wheel for a 6x12 SG. Balancing might have improved the finish but unbalanced wheel finish was plenty good/accurate enough most times. Any component needing a better surface finish in dies/molds got polished, ending with diamond paste. The lines in posted photo look, to me, like a loose grain in the wheel. Waves are either a wheel not completely dressed or a bad spindle. Sometimes a loose grain is hard to dislodge with wood/shim and requires another coarse dressing pass. I used either a .005 shim or an unmentionable method to feel the wheel before grinding. How much material are you taking per pass? Are you doing a finish grind or spark out of a couple of .0001's? Tried using a different material for test grinding? No offense meant but the finish looks like that when the wheel is loaded while grinding aluminum.
 
Last edited:
here is my amateur effort trying to solve this same issue few days ago (I had no way of balancing the wheel alone at the time), one pic shows the hub on my grinder with adjustable weights, the other pic is before and after of "balancing by feel" :D

balancing by feel - I dressed the wheel, then moved the weights to one side, all together, turn spindle on and off when it reaches full rpm, with my resting on the wheel guard, feel for vibration, there were quite a few rpm ranges the vibration could be clearly felt (like washing machine spinning up irregular load, there are few rpm ranges where it resonates), I kept moving the weights around till I could basically feel no vibration at all during spin up, full rpm and spin down, the resulting grinding result is in the second picture, full disclosure, the "wavy" part was also due to me slowing down the feed of the table too much trying to get more even finish, the better looking part had much quicker table feed, step over was the same in both cases

now the fun fact - spindle nose was probably crashed at some time, because it shows steady run out of 0,030mm (by steady I mean it is in the same spot in every rotation of the spindle, nearly same amount on narrow and fat end of the taper), which IMHO is a lot, the spindle sounds very quiet, no signs of bad bearing sounds (there are 3 pairs of angular contact ball bearings in this spindle), but for the price I payed for this 1000kg 150x400mm hydraulic grinder, I can't complain at all - 120EUR hammer price at an auction + 400EUR transport costs...

20200512_215631.jpg before-n-after.jpg
 
the way I understand function of this device is that it is not balancing anything, but instead it doesn't allow the system (spindle column) to go into its natural oscillating frequency by constantly shifting the imbalance phase in relation to spindle column resonance frequency

think of trying to push a swing, if your pushes (small amplitude) are in regular intervals all the time no matter the frequency, just that the frequency is the same all the time, eventually you will get the swing swinging at its natural frequency, the swing amplitude may not be great, but it will be steady

now 2nd case, you keep bumping the swing at irregular intervals, more or less chaotic frequency, there is little chance of getting the swing to swing at all, sometimes your bumps will combine, other times they will cancel out the energy transferred into the swing previously - eventually this still may "settle" and you might get the swing to swing somewhat, but in case of the surface grinder, you are taking some of this "bumping energy" accumulation out of the "swing" each time the wheel is cutting the work piece further disrupting the resonance which leads to smooth cutting action

at the end it accomplished the same result as perfectly balanced wheel on a perfectly rotating spindle nose (no runout), but I wouldn't call that effect - balancing

I may be wrong about this, but that is how I understand this "dynamic balancing", something similar is used with stepper motors to smooth out their motion and overcome some specific rpm ranges where they may lose steps

I believe the bearing balls actually balance the system. Take a look at the following video link. It's short and quite conclusive. My design is not a new concept as it has been used for years to balance truck wheels and even the tubs in clothes dryers. Link:

YouTube

metalmagpie
 
you are like a dog chaseing its tail

amusing but not productive

you need to solve your prolems stead of trying to find a work around

start over from the basics:is your dirt the best you can make it? is your concrete the

best you can make it? is your machine level and bearing eaven on all feet? are the

ways in the best shape that you are capable of getting them?

solve all the prolems that you know of includeing your hydraulic prolem

quit trying to grind til you solve your equipment prolems

if you are incapable of solveing equipment prolems get someone that can or work on

them til you figure out how
 
How are you holding it down? Is the bottom surface flat? If not try setting it on 3 machinist jacks and indicate it in the best you can, then block it in. Or set a Kurt Vise on the magnet and mag it down and clamp the aluminum in the vise after indicating it in the best you can. Do you have paper washers on each side of the wheel? Did you dress the wheel rough? With wheel spinning, I would spray some Dykem lay out fluid on the wheel and dress it until the fluid is gone. This will balance the wheel. Last dress pass (across one way)fast cranking should be .002" one time and stop. NEVER dress back and forth several passes as this with burnish the wheel. Are you using coolant? On on aluminum I used STP mixture with mineral spirits and brush it on before the grind. Wheel should be a 46 or courser grit as it will load up fast.
 
How are you holding it down? Is the bottom surface flat? If not try setting it on 3 machinist jacks and indicate it in the best you can, then block it in. Or set a Kurt Vise on the magnet and mag it down and clamp the aluminum in the vise after indicating it in the best you can. Do you have paper washers on each side of the wheel? Did you dress the wheel rough? With wheel spinning, I would spray some Dykem lay out fluid on the wheel and dress it until the fluid is gone. This will balance the wheel. Last dress pass (across one way)fast cranking should be .002" one time and stop. NEVER dress back and forth several passes as this with burnish the wheel. Are you using coolant? On on aluminum I used STP mixture with mineral spirits and brush it on before the grind. Wheel should be a 46 or coarser grit as it will load up fast.

Thank you for responding, Rich.

I have been alternately grinding the top and bottom of this part for a week or so now. The bottom is flat within a tenth as shown on my surface plate. I am holding it down with the mag chuck. Part and chuck stoned beforehand. Blotters on both sides of the wheel. I dressed the face of the wheel .001" deep several times. Your idea about Dykem sounds great but maybe messy. I inked up a wheel face with magic marker and dressed until all the black was gone. Same, only a little different.

I will take your dressing advice to heart and make my last pass a fast .002" deep.

I don't have my coolant going yet.

I had tried a 46 wheel but had moved to a 100 grit wheel just to see if things got better. I will go back, but I'm not grinding another speck with this machine until I make some progress with the hydraulic unit. The table is coming off today.

metalmagpie
 
you are like a dog chasing its tail amusing but not productive

I'm glad you're amused and I agree I'm not really productive and I appreciate your posting.

you need to solve your problems stead of trying to find a work around

Can't argue with that.

start over from the basics: is your dirt the best you can make it? is your concrete the best you can make it? is your machine level and bearing even on all feet? are the ways in the best shape that you are capable of getting them?
solve all the problems that you know of including your hydraulic problem. quit trying to grind til you solve your equipment problems.

The concrete floor is relatively flat and strong enough to hold a knee mill and a 14" lathe. I'm not aware of any problems underneath it. If I have to I'll saw the concrete and pour a foundation. But I don't see any problem with it yet. My machine has 3 feet, and I leveled it. I'll be looking at the ways soon. And I will certainly address the hydraulic problem.

if you are incapable of solveing equipment problems get someone that can or work on them til you figure out how

I'm a little thin on people that can help. But I do work on things until I figure them out. I also post what's going on because I find many comments to be helpful along the path to figuring things out.

metalmagpie
 








 
Back
Top