Truing up a disc grinder - where can I get this machined?

[scholar]

I recently acquired (from the nicest UKW member) a rather shabby 20” disc sander (actually marketed as a disc grinder for metal, but anway).

I had thought it just needed a good clean up, but I am finding it needs quite a bit of fettling - I think these problems go back to its original manufacture and became ignored with various parts getting broken and loose! Anyway, the problem I could use some help on is that the disc is not running true - that is, the sanding face wobbles in and out as the disc turns.

I have had the dial indicator on this and, with the disc properly mounted, measure a discrepancy from the low point to the high point (at the outer edge of the face) of approx 0.65mm. I don’t actually have a reliable way of measuring the disc itself, but I am pretty sure that the culprit is the central mounting flange that fits on the motor spindle - the flange is approx 71/2” wide and shows a discrepancy of approx 0.22mm from low point to high point. In both cases, the low point is opposite the high point and the high and low points identified on the flange line up with those on the larger disc - the disc is just under 3 times the diameter of the flange and so the measured discrepancy appears consistent.

I have looked at the spindle - there is no apparent play and whilst there is a small flutter on the dial indicator, this is not consistent and so I think that is not the problem. I am thinking that if I could get the face of the flange re-machined to be perpendicular to the hole for the spindle, this should pretty well true everything up. By the way, the disc only fits on the flange in one position, so rotating the relative positioning is not an option - it wouldn’t make any difference anyway if the main disc itself is true, but I cannot experiment to check this.

Please does anybody have any idea how i could get the flange machined??

Some photos:

Cheers

 

[TRITON]

Are you sure it actually needs it ? .65mm is not a lot, and it isnt a precision machine,rather just a sander, for sanding to a marked line.

 

[Rorschach]

I suspect you could shim out an error that small quite easily.

 

[TheTiddles]

It looks like it’s just a grub screw keeping it on? If so that pretty poor, hopefully there’s a key in there too.
Edit- yes I see the key now
Either way, take the flange off and have it skimmed

 

[Inspector]

A machinist would have to make a temporary shaft to mount the flange onto for turning. They would then face it flat. Then mount the disc to it and see if it now runs true. If not they would have to either face the back flat or face the sandpaper side on the lathe with the temporary shaft and flange together. Then it would be as good as you can get. It would likely be expensive because of the size of the lathe needed and the time needed to do the setups and work. Before doing that do as Rorschack suggested and shim it as best you can and call it good enough. The sandpaper you buy won't be a perfectly even thickness anyway.

Pete

Curious if you have put a straight edge across the disc to see if it is flat and not dished? Or placed it on a flat surface and checked flatness with a feeler gauge?

 

[scholar]

Hi All

The wobble is quite noticeable and so really does need to be corrected.

There are two grub screws actually, but they just keep the flange from moving forwards - as you say, there is a (substantial) key - I think one of the grub screws bears on the key.

I have had the flange off - my real question is where/how to get it skimmed?

Cheers

 

[Inspector]

Did a quick search and found a possible.
https://stratford-upon-avon.cylex-uk.co.uk/company/g-jones-12760023.html
Pete

 

[scholar]

A machinist would have to make a temporary shaft to mount the flange onto for turning. They would then face it flat. Then mount the disc to it and see if it now runs true. If not they would have to either face the back flat or face the sandpaper side on the lathe with the temporary shaft and flange together. Then it would be as good as you can get. It would likely be expensive because of the size of the lathe needed and the time needed to do the setups and work. Before doing that do as Rorschack suggested and shim it as best you can and call it good enough. The sandpaper you buy won't be a perfectly even thickness anyway.

Pete

Curious if you have put a straight edge across the disc to see if it is flat and not dished? Or placed it on a flat surface and checked flatness with a feeler gauge?

The disc is pretty flat across the face when mounted and the variation on the dial indicator moved consistently as the disc was spun.

I have got some shim stock and so will give it a go with that - the low point nearly coincides with one of the three mounting holes which helps.

Thank you

Cheers

 

[scholar]

Did a quick search and found a possible.
https://stratford-upon-avon.cylex-uk.co.uk/company/g-jones-12760023.html
Pete

Thanks! That’s cycling distance

 

[okeydokey]

I don't have a dial indicator and for my fiddling with bits and bobs never needed one so the following is from a not engineering background but hope it might help or set a train of thought in a direction.
I'm sure you have considered this but when the flange slides onto the shaft is the end of the flange where it beds against the shaft really clean with no build up of crud on either the flange or where it beds against as that could - depending on how close the tolerance between shaft and flange cause the small wobble.
Another thought - would tightening a grub screw pull the flange one way or another again causing the wobble?
Does it make a difference if you tighten the grub screws gradually and alternately to help keep the flange fitting square on?
Failing any of these a shim or two one each side of the fixing bolts to the sanding disc might be the easiest solution.
regards

 

[TRITON]

Had a thought, probably not correct, but here goes.
Wouldnt you need to machine both surfaces for a super flat fit ?. If you macchine just the flange, then refit it it could make the wobble worse, given that the main disc you are attaching it to hasnt had any material taken from it. It would have to rely on the main disc being precisely flat for it to mate directly to the flange

 

[Rorschach]

Had a thought, probably not correct, but here goes.
Wouldnt you need to machine both surfaces for a super flat fit ?. If you macchine just the flange, then refit it it could make the wobble worse, given that the main disc you are attaching it to hasnt had any material taken from it. It would have to rely on the main disc being precisely flat for it to mate directly to the flange

Yes or no, the disc could be flat and all the error could be coming from the flange, the disc could also have errors that either cancel or compound the error in the flange. Skimming the flange could solve the problem, or make it worse. This is why I wouldn't machine anything until I had tried shimming the flange/disc interface. There is no shame is using shims, precision machine tools are often shimmed to remove errors or wear. The great thing about shims is that if you make the problem worse, you remove the shim and try again, if you try and machine the problem away you can never put that material back. (ok technically you can through welding etc, but it's bloody hard work compared to removing a shim.)

 

[scholar]

Yes or no, the disc could be flat and all the error could be coming from the flange, the disc could also have errors that either cancel or compound the error in the flange. Skimming the flange could solve the problem, or make it worse. This is why I wouldn't machine anything until I had tried shimming the flange/disc interface. There is no shame is using shims, precision machine tools are often shimmed to remove errors or wear. The great thing about shims is that if you make the problem worse, you remove the shim and try again, if you try and machine the problem away you can never put that material back. (ok technically you can through welding etc, but it's bloody hard work compared to removing a shim.)

Thanks for the further comments:

The fit on the spindle is tight over the full length of the flange boss and keyway (takes a bearing puller to remove it). I don’t think the grub screws would adjust the angle at all.

I am optimistic that the main disc is true enough - this is based on my assessment that the measured discrepancy on the face of the flange is consistent with the measured discrepancy on the face of the disc when fitted to the flange (as mentioned earlier).

I am going to try some shims between flange and disc and report back.

Cheers

 

[clogs]

I wouldnt bother with shims...just something else to get loose.....
also no good taking it off to get machine as they NEVER go back the same anyway.....
I would just glue some decent abrasive paper to a heavy block and offer the paper to the disc on the machine with it running.....
just taking off or reducing the high spots.....
if u face machine a disc that big the disc will however lightly the tool touches the disc will push it backwards.......

I bought a home made version of ur's.....it to had a wobble and the above is what I did to reduce the wobble.....
worked for me...

and as an add on......when using a big faceplate on a metal lathe its quite normal to lick the face off to get it running true....
because it always goes back different.....

 

[Alpha-Dave]

Just out of curiosity; have you thoroughly cleaned the faceplate and checked for burrs around the edge and mounting holes? There is definitely some worn paint around the edge and surface discolouration. A burr will definitely throw the faceplate out of true.

I would be tempted to shim with aluminium from a coke can at ~100 microns thick you will be in the ball-park with 2 layers. The high appears to be opposite a mounting hole, so cutting the shim into a washer could stop them from shifting.

 

[Phill05]

"The fit on the spindle is tight over the full length of the flange boss and keyway (takes a bearing puller to remove it). I don’t think the grub screws would adjust the angle at all.

I am optimistic that the main disc is true enough - this is based on my assessment that the measured discrepancy on the face of the flange is consistent with the measured discrepancy on the face of the disc when fitted to the flange (as mentioned earlier)."

If you have had it pulled off with a bearing puller it might have introduced a slight twist in the flange boss, I would be tempted to face it up in situ on it;s own shaft, that way it would be true to your machine then refit the disk.

 

[scholar]

Shims rock!

After a bit of trial and error with some 0.2/0.1/0.05 mm shims, I got a pretty good result using just a 0.2 mm shim at the fixing hole nearest the previous low point. here is the result - 0.1mm variation which I will live with for the time being (feels much better at the sanding face now).

 

[Stevekane]

I guess your sticking PSA sanding discs onto it,,will it matter if its 20thou out prior to this? My much smaller sanding disc made up from an old bench grinder had a horrible wobble on its diy mdf platter but responded perfectly to a couple of gentle taps with a lump of wood and remains quite true,,,well as long as I dont go near it with a dial gauge!