What's new
What's new

Finish on cylindrical grinder

ballen

Diamond
Joined
Sep 25, 2011
Location
Garbsen, Germany
A few months ago I bought a cylindrical grinder: a 1960s era Studer RHU-450. Another thread in this forum describes some of the work I have done to fix it up. Now I'm starting to learn how to use it. My experience so far is limited to two test pieces. One is a piece of 16mm (5/8") steel from the scrap drawer, and the second is a scrap of 12mm (1/2") diameter stainless. The first I did between centers, and the second with a collet.

The finish is not as good as I had expected, and I want to track down why. I'd be grateful for advice from the experts here. It's probably technique or the wheel, but I am not sure. I also have a J&S 540 surface grinder and can get mirror finishes from that without trying very hard.

First off, here is the wheel that came with the machine. This is 300 x 30 x 127mm (12" x 1 1/4" x 5"). I am not experienced at reading the codes, but I think 81A is aluminium oxide, 80 is the grain size, I don't know what the -4 means, G is the hardness (on the softer side) 13 is a very open bond. M is not a bond type that I could identify, perhaps the V means vitrified bond. I removed it from the hub, ring tested it, reinstalled, balanced, trued, and balanced again.

attachment.php


Wheel surface speed is about 28m/s = 92 feet per second = 5500 feet/minute. I am using the recommended pulleys for this wheel.

I am dressing the wheel using the tailstock mounted diamond point that came with the machine. It's about a half-carat made in Germany diamond, looks in good shape to my inexperienced eye.

I am grinding wet, using the same synthetic grinding coolant as on my surface grinder. I did clean out the coolant sump, which has "3 level" system of weirs to trap grit, but is not as extensive a filtering system as on the J&S 540. I don't have any kind of paper or other filter.

Here is the grind on the 16mm (5/8") steel rod. I spun this at various speeds from 50 to 500 rpm. It didn't make any real difference in the finish. I also tried various feed rates from inches per minute to inches per second. I had dressed the wheel with a fairly quick pass, intended to leave it open, the same that gives me good finishes on my J&S surface grinder.

attachment.php


Here is the grind on a 12mm (1/2") diameter bit of stainless (just the part to the right was sticking out from the collet). In this case the finish is better than before, because I experimented dressing the wheel with a very slow pass rather than a quick one as before. I mounted this in a collet in the work-head, and spun it around 400 rpm.

attachment.php


You can see this rod lying next to a piece of flat steel (under the dial indicator) which I have surface ground. The surface ground steel is a mirror surface, whereas the stainless has a more "frosted" appearance. If I run my fingernail along it, I can "feel" the texture.

Here are some possible reasons for this not-very-good finish:

(a) wrong wheel
(b) wrong dress
(c) wrong feed/speed

Potentially

(d) lack of a solid base, the machine is still on a pallet

(e) problems with the (solid-bearing) spindle

The spindle is perfectly smooth and quiet, and the runout is at the micron level (0.00004"). But I have not taken the spindle apart or gone through the procedure for setting the spindle end-play and runout, because I didn't think it was needed. When it runs it gets up to temperatures of about 37C = 100 Farenheit, pleasantly warm

I'd be grateful for advice about what to try or experiment with to get a better finish.

Cheers,
Bruce

PS: here is the wheel side of the coolant nozzle. I don't know why there is this little hole. Also, it seems that most of the coolant just bounces off the air around the wheel. But this looks like the stock nozzle shown in the manual...

attachment.php
 
I have limited experience with cylindrical grinding. I found to get really good results you need a decades experience then add some eye of newt to the coolant and say some magical incantation.

There are a lot of variables. Workhead speed, wheelhead speed, coolant cleanliness, table speed.

Not hard to get good results, hard to get those results you see the experienced people get.
 
From here that looks like a popcorn wheel. Popcorn is a slang expression for a very open wheel often made by adding walnut shells to the mix before baking the wheel. The walnut shells completely disappear at baking and leave the grain particles a distance apart. This allows much better cooling but also can produce a wind to make it seem blowing off, often the coolant is more directed to the work piece not to the
wheel.
Think the nozzle hole just a safety device so coolant pressure can not build.

Think better to test grinder on other than SS perhaps a hard steel part.….and 5/8 x 6 1/2” is a stout part and so should run OK with not having a steady. Photo looks longer than that. Might not be a bad idea to add an under steady and back..just a nylon/brass/carbide/ceramic support so the part can not bend down or away from grinding pressure.
SS is such a bugger to grind you may be getting some chatter.

Good test is to mount a part then with one, two, three finger pressure see how much deflection to a tenths indicator with part and wheel parked. testing at head, mid point and tail end..

We would often just switch to a lukewarm water tap water last few passes to rid the finish from picked up particles for super finishing.
80 grit the dressed wheel particles perhaps .010” or so.

I don’t like the abrasive contact to your wheel mount. If possible would add a brass ie SS shim between wheel and mount.

Some times good to dead-center grind the lats tweak..but that may change your straight or taper amount ir one center is not dead true. Dead center grinding rules out center run-out as a problem cause.
 
Tail and head end centers needs to be near zero run - out if live because fall way and fall into wheel can cause crushing wheel and make wheel out of true round..Grinding in place can make head end near dead true..then regrind if taken out and replaced.

Tail live has to be ground in its own bearing...not taken apart and ground out if its own bearing.
 
Last edited:
RC99: one of the things that I like a lot about PM is that there are some very experienced people here who also know how to teach beginners. I am sure that their advice will cut the learning curve down.

Buck: thanks a lot for your comments.

- Yes, the wheel has a very open grind. I didn't realise that this makes the "wind" worse. I am hoping to get a couple of more hubs and set them up with two more grinding wheels. What would you suggest to complement the wheel I show above?

- I'll try some hardened steel next

- The 5/8" part is just over 150mm = 6" long, between dead centers. I'm quite sure it does not need a steady. There is no vibration and the finish is the same along the whole length.

- I'll get in the habit of doing your 1,2,3 finger deflection check

- I can see switching to tap water for the final passes to get a great finish. But I am still pretty far from that now!

- Abrasive contact to wheel hub. There are blotters on both sides of the wheel, but they were sticking to the hub. On the other hand there is a strip of paper on part of the ID of the wheel. I could keep that, remove it, or replace it with some 0.1mm = 0.004" or 0.12mm = 0.005" shim stock. What do you suggest?

- With the 5/8" test piece I did not use a live center, just dead centers at both ends. Question: how deep a center hole should I drill in my parts? What OD on the center hole is about right?

- I have not yet had to regrind my centers, but will heed your advice when I do

Richard: thanks for helping out (again).

I hope you are right about the pallet being the problem. If so it's an "easy" fix. Unfortunately the grinder is not parked underneath my chain hoist and the pallet is the same size as the machine, which weighs 1.8 metric tons (4000 lbs). So to lift it off I need to fabricate a couple of 2m long I-beam or U-beam lift bars to bolt on either side. I can use these to lift it with car jacks. Anyway, making this rails and taking down the machine has just moved up my priority list, hopefully will get it done in a week or two.

Cheers,
Bruce
 
Center lap mounted points are good for truing up centers..Some times a center will not be exactly round and that can affect finish. Center Lap Mounted Points | Norton Abrasives United States of America & Canada

Sides blotters should be perhaps .015/,025 or so (.5mm) or thicker..Make your own is fine and they may be greased/or not. Petroleum jelly males a fine grease for such. OD hub protector SS or brass shim. Blotters keep the high abrasive grain form putting a pressure that could crack a wheel..and keep your hubs from getting abraded.
You can glue side blotters them on with Elmer's school or white glue..water soluble so a little soak takes the off... and a little grease at the outside
 
Hi Buck,

Thanks. I've ordered a Norton Center Lap Mounted Point. The side blotters I have are the stock cardboard ones that came with the wheel. I will reglue them with white PVA glue as you suggest. I will add some 0.12mm = 0.005" strip as a hub OD protector.

I'll put a bit of grease on the side blotters as you suggest. Is this to keep them from sticking to the wheel flanges in the future?

Cheers,
Bruce
 
Hi Richard,

Look in the base for round holes with covers. Many machines have 4 - 2 on each side so you can slide in a rod of steel and either use a fork lift or jacks to lift the machine

Unfortunately the machine does not have those. It does have these "slots" on the back and the sides, but not on the front. They extend inwards about 1 inch (25mm) and I suppose that they could be used for "toe jacks". But I haven't such jacks.

attachment.php


So for the moment I don't have a better solution than my two lifting bars. But I am open to suggestions!

Cheers,
Bruce

PS: the photo above does NOT show the current configuration. I picked it because it shows the slots. Currently the machine is sitting on a pallet which is the same size as the base.
 
I am not that particulary anal about oil
One exeption however A accurate plain bearing
Before operating it I would change the oil
And use a oil that has the same specs as in the manual
I suspect your oil should be thinner as water
you say the spindle runs smooth
Perhaps there is the problem
A spindle like that should after shut down coast to a stop quit rapidly
Especially the last bit
If you turn it by hand you must feel some decent resistance ecpecialy at start up
Also if you measure endplay while it is filled with oil you are fooling yourself
Measured endplay depends on the oil it is filled with

Always check your oil on a accurate plain bearing


Peter
 
Last edited:
You said fairly quick dress. Slow down your dress. Show a picture of the diamond, for a wheel that size 1/2 carat is a pretty small diamond. Your coolant nozzle shape looks poor to me. I like one that fits the wheel much closer, like it was adjusted and then pushed into the wheel so the side away from the wheel is sharp.

If you grind wet, dress wet.

How does the workpiece act when you stop your infeed? Does it spark out immediately, or does it spark for several passes?
 
Peter from Holland is bang on about the spindle oil. Grinding machines are super sensitive to having the correct oil.

I once spent a week pulling my hair out on a large roll grinder that was giving awful finishes. Look up a " three penny bit " on the Internet and you'll know what I mean.

I tried everything without success and work was piling up waiting for this machine. It being the only big roll grinder this company had. I done everything I could think of and I didn't know what to do next when the operator just casually mentioned he'd changed the spindle oil the previous week. When I asked him what he'd used it became apparent he'd used the wrong oil.

Although the oil he'd put in was really thin the correct oil ( Velvus 3 ) was like water.

When I changed the oil back " happy days were here again " and you could have combed your hair in the finish we were getting !

Regards Tyrone.
 
From here the finish looks better at the drive end..is that the case or just the photo?

Less open wheel and a bit harder will give a better finish but will have more tendency to burn, warm and wheel pressure the part. Driver dog shown is not a balance dog but Ok perhaps turning 300 RPM and less. Fast turning the part can make the part become a wheel crusher and so dress the wheel with the part..not good doing this.

This vendor suggesting a lot of k wheels I have also used J and Ls ( still 46-k/L are my most used wheel for general grinding)
Grinding wheels for cylindrical and roll grinding

Changing wheels can be troublesome so at times I have chosen a wheel for a common surface need and then abrasive cloth finished for a higher or rare surface finish need. Hold cloth (with not catching fingers or thumb)and travel only one way like cutting a thread, Don't go back and forth. Best with wheel not running...*Think about how to hold cloth as many have caught a finger doing this...Go from big end to small if a micron off)

A bit of grease or the like on a dead center can be good.
I have put a rubber band from dog to a machine feature to keep it tight. even with a small cut of nylon between dog and driver with the rubber band.
Fresh dress is not always the best finish, but a little load makes finish better. often a very soft wheel can not load but will wear to expose new fresh grits so not getting better. This is very good to avoid burning and heat and maintaining existing finish, if good enough for spec..
Have loaded wheel with a candle (wax) for last few passes.

Hand surface finish test gauge is handy,
Surface Roughness Gage Tester Set GE Machinist Tool Box Find Milling Finish | eBay

Along with Peters good post about oil is a warm up time for plane bearings.
 
Did you check endplay on the workhead
Thats another possibillity
Before messing around with feeds and speeds and stones you must be sure the machine is up to par
Endplay on the plain bearing can be adjusted but many times the taper on the spindleshaft where it runs on needs a regrind and lapp
Also looking at the drawings for the grinding head you see a "oilfilter with Jet-workhead"
So that indicates the machine has a oilfilter to clean and a pump to check
This is a old machine so it is realistic to assume the grinding head and/or workpiece head have wear

Peter
 
Hi All,

Thanks for all of the good ideas. I will start trying them this weekend.

- Richard, I am working to get what is needed to bring the machine down off the pallet.

- Peter, I did purchase 5 liters of Mobil Velocite No 3, which is the recommended oil. I have not yet replaced the old spindle oil. But now that you got me to think, it does seem too thick. The spindle does have some resistance when I turn it by hand, but when the oil is drained I will measure the endplay and compare to specifications. The manual says, "when the bearing is correctly adjusted, the bearing play amounts to 5 microns". Later, in the procedure to adjust this, one can see that this corresponds to an axial play of 50 microns (so the taper in the solid bearing must be 1:10). So I will remove the current oil, measure the endplay, put in the correct oil, and measure the endplay again.

- Gbent, I did slow down the dress, and it made the grind smoother, but not as good as I expect to get. I will post a picture of the diamond and the specs. I will also experiment with the coolant nozzle. I studied the manual and saw that there are three different types of nozzles normally delivered with the machine. Since I do not have those other nozzles, I could easily fit one of those flexible blue "hoses" made from segments and attach an orange plastic nozzle. You are right that I dressed dry, next time I will dress wet. To answer your question, when I stopped the infeed the work sparked out right away, I only saw sparks on the initial pass and nothing afterwards.

- Tyrone, thanks for telling your story, that is good motivation for me to change the oil. I just hope that the PO did not put in a thicker oil to hide spindle bearing problems, but just out of ignorance or haste. (PS: I didn't get the three penny bit, write to me privately if needed!)

- Buck, the finish was the same all along, not better at the drive end. You are right, the dog is unbalanced. I need to either make a set of balanced dogs or buy a set. I could draw these up and have them water-jetted from aluminium plate. Have you got a spare set to sell me (up to 200mm diameter) or can you recommend an inexpensive source? Thanks for the link to information about grinding wheels. Two wheels that I can get locally at reasonable cost are Flexovit WA 60 KVL and/or WA 46 KVL, in 300 x 25 x 127mm (12" x 1" x 5"). To get good finishes I would think that the 60 is a better choice, I can use my existing blue wheel to take off more material if needed because it is so open. What do you think? Alternatives would be a Norton 38A60K8V 300 x 32 x 127. A friend of mine has told me that moly grease is good on dead centers, I will try this. I will also try letting the spindle warm up more, with the right oil inside. The set of surface gauges that you liked to on ebay seems OK for milling but too rough for grinding. Or have I missed something?

Peter - there is no endplay on the workhead, I had it locked with no play and was using a dead center at both ends. I did clean the oil filter on the wheelhead but did not open up the wheel head housing to check on the functioning of the pump. I will try changing the oil and getting the machine on the ground to see what effects those have, and will measure the endplay as you suggest. I won't take the wheel spindle apart to look at the bearing and taper until I have eliminated other possible causes and still don't get a mirror finish.

Thanks again!

Bruce
 
Fantastic thread and great comments all around.

Endless threads and examples all over have shown us how various machines that have slop, bad bearings, backlash, wood pallet mounts, etc etc etc have been making excellent parts for years or decades .. when suitably set up and used ..
in effect usually pre-loading the kinematics so that good results appear at the workpiece.

Thus when people suggest a better machine mount, better oil, better tool, better parameters for the tool in cutting or dressing etc., perhaps better bearing fits etc. (not to go into spindle stuff in this case) they are all! correct, in general.
But the previous owner probably went to a different oil because it gave them better results in practice.
Often, vastly better results were got with unusual oils on non-typical materials (plastics, sintered metal) - I had examples of such as a machine tool distributor.

The tir numbers and mass/type-of-machine suggest to me the machine can/will likely produce good work with minimal changes.
The suggestions of new oil and new wheel and new dressings seem to me to be excellent and sensible ..
and cheap, easy, reversible, fast.

If this produces variable results all over the place, from fantastic finish to poor, one or 2 experienced old-timers on such machines might help in a few hours - for a controlled cost.
 
Slow turning a Part is often best (<300or so) so your dog is Ok…60K a good choice. [ QT: moly grease is good on dead centers] I just use Vaseline because it is clean and handy to and about the shop., [surface gauges that you liked] I did not look close just showing example of hand surface gauge…agree one more for grinding best.
Good to keep an eye on auctions and pick up a few wheels.. so you don't feel bad about dressing angles on wheels.

Seems nobody mentioned turn the diamond because we all know you know that, but if any nube grinders are watching I should mention that. A fresh facet edge dresses better.
 








 
Back
Top