DIY surface grinder

[Sonic EscapeParticipant @sonicescape38234]

For some time I started thinking about what the construction of a surface grinder would involve. I’m not sure what are all the critical aspects of this machine. I suppose rigidity is not as critical like on a milling machine. From what I read one important feature is the ability to move vertically with high accuracy.

To move the magnetic chuck horizontally I’m thinking to use HGR20 linear guides. The chuck weights around 20kg. But there are two problems for which I have no solution yet. How to build a perfectly flat surface where to mount the guides? I could play maybe with an indicator and shims but that sounds like a last resort option.

And second, I understood that welding is out of question when long term stability is desired. The structure will twist because of the tension caused by heating. I have a wild idea of clamping the two guides on a surface plate and then turn the plate upside down and encase the guides in a concrete slab. But again, not the first choice. Any better suggestions are welcomed.

 

[John HinkleyParticipant @johnhinkley26699]

Have a look at

Home

There you will find plans for (amongst many other things) a surface grinder.  They will not cost you much but will save you a lot of time and hassle.

John

 

[JezParticipant @jez]

Physics Anonymous on YouTube built a surface grinder – quite interesting how they went about it.

 

[Martin KyteParticipant @martinkyte99762]

Swing grinders are easier. MEW had a series if I remember rightly.

regards Martin

[Andrew JohnstonParticipant @andrewjohnston13878]

The OPs thinking is not really correct. While a grinder does not have to resist the cutting forces that milling machine does any vibration will completely trash the finish. Hence grinding machines tend to be very heavily built with material choice designed to minimise vibration, both internally and externally.

Two guides are not a good idea, they are over-contrained. There should be only one feature guiding the table. On both my industrial grinders (surface and cylindrical) the table rests on one inverted V and a flat. The V does the guiding and the flat simply helps to support the weight. Both tables are very heavy but just rest on the ways; they can be lifted straight off if needs be. The designs also make provision for a continuous film of oil on the ways.

The down feed needs to be in small increments, but a screw and wheel is fine, provided the mechanism is loaded in one direction. On a surface grinder the weight of the head and spindle will be sufficient. On my cylindrical grinder the infeed fine pitch screw has large weight on the end to ensure the mechanism is loaded in one direction. The manual claims that a cut of 25 millionths of an inch can be applied, but I have no means of verifying that.

Andrew

[Martin CargillParticipant @martincargill50290]

Build your grinder then use it to make its own table flat before fitting the rails?

Martin

[Nigel Graham 2Participant @nigelgraham2]

Not casting the guides directly in concrete but you do have the germ of a method that is used where both rigidity and vibration-damping are critical.

Some very large machine-tool beds and columns are steel fabrications filled with concrete, and I believe the Peatol lathe’s hollow bed extrusion is similarly filled, if not with concrete then with something similar.

[Nigel Graham 2]

On Nigel Graham 2 Said:

Not casting the guides directly in concrete but you do have the germ of a method that is used where both rigidity and vibration-damping are critical.

Some very large machine-tool beds and columns are steel fabrications filled with concrete, and I believe the Peatol lathe’s hollow bed extrusion is similarly filled, if not with concrete then with something similar.

I was thinking on something like this. I can connect each guide to a large steel bar using long hex nuts. In a way that only the last few mm of the nuts remain outside the concrete. And until the concrete hardens I could use a granite plate to keep the guides perfectly aligned.

I saw the Physics Anonymous surface grinder a few months ago. Their video among other sparked my interest in this type of machine. But I’m not a big fan of their method. Why wasting effort with shims on something that is unstable in time?

http://www.machineryplans.com is an interesting site. I didn’t knew about it. But their SG plans require cast iron dovetails, so you don’t need a surface grinder to make them? 🙂

[HopperParticipant @hopper]

If you already have a milling machine I reckon you would be better off to make a grinding head attachment to fit that existing machine. Then a simple lever mechanism or perhaps rack and pinion etc to move the table back and forth at grinder speed rather than milling speed.

[DiogenesParticipant @diogenes]

Concrete and steel assembled in that way isn’t dimensionally stable (go look at a motorway bridge) and you have no means of correcting errors after assembly.

You need a separate ‘carrier’ between the concrete and the rails.

You’d be better off restoring an old grinder.

Where is Joe Noci these days?

[Diogenes]

On Diogenes Said:

Concrete and steel assembled in that way isn’t dimensionally stable (go look at a motorway bridge) and you have no means of correcting errors after assembly.

You need a separate ‘carrier’ between the concrete and the rails.

You’d be better off restoring an old grinder.

Where is Joe Noci these days?

It isn’t dimensionally stable because is shrink while it hardens? Or later?

My milling machine has a horizontal spindle and I also considered to use it as a surface grinder with some modifications. If I would really need one that would be my first choice. But is more fun to build one from scratch.

I would like to restore an old grinder. I’m frequently checking the local market. But for some reason they are very expensive. They cost much more that I paid for the lathe. I don’t understand why. I think they are used to refurbish car engines. Here there is a huge market for old vehicles.

[Martin Kyte]

On Martin Kyte Said:

Swing grinders are easier. MEW had a series if I remember rightly.

regards Martin

Hi Martin Kyte, yes there was a Swing Grinder construction article in MEW Nos.45 &46 (Oct/Nov 1997) and the one in the article was shown at the Harrogate exhibition.

Regards Nick.

[Alan JacksonParticipant @alanjackson47790]

I wrote an article in MEW No 294 (June 2020) about a surface grinder I made from the botton half of a Dore Westbury milling machine.

Alan

[Nigel Graham 2Participant @nigelgraham2]

Converting parts of a milling-machine is one thing but I would never make a grinding attachment for a regular, complete mill.

How can you guarantee no abrasive will find its way into the three sets of ways and screws?

If wanting to build a surface-grinder (or cylindrical or tool grinder) it’s best to build it completely, designed for that purpose. It also means not needing to convert the table to rack or lever feed.

Some small, very simple horizontal mills, such as the Denbigh ‘H’ 1 to 3 series, made for batch-production tasks by semi-skilled labour, had lever-action long feeds, so might be suitable for conversion… if you can find one. Though it’s likely to be so old and battered it won’t be necessarily be any more accurate than as a mill, without thorough refurbishing.

(For similar reasons I am reluctant to use a grinding attachment on a lathe, but it may be feasible to make a cylindrical-grinding or surface-grinding attachment for a T&C grinder.)

[jaCK HobsonParticipant @jackhobson50760]

Accurate grinders are often heavier than milling machines.

If all you want is surface finish or to remove material then a belt grinder and a platten. I got a friend who uses huge surface grinders just for the auto-reciprocation and bolts on a belt grinder for material removal rates.

If you want the 1/10 accuracy – I think ‘massive’ might really help.

[Sonic EscapeParticipant @sonicescape38234]

I decided to do some experiments with different concrete recipes. I bought a bag of good quality cement and 3 bags of sand with different granulation. And a bottle of concrete plasticizer. The plan is to make a number of identical concrete bars with different recipes and to measure how well they dampen the vibrations. I’m going to use an oscilloscope to measure the ringing when the concrete is hit by a steel ball. I’m thinking on a setup that will have good repeatability.

I found also an interesting self leveling mortar that has compressive strength of 40MPa. Normally this stuff is between 6 to 20MPa. Since it is simpler to use than concrete, I bought a bag to give it a try. Here is the first sample:

The thickness is around 25mm. I found something even more promising from Sika. Unfortunately it was out of stock. It is an anchoring grout that has a compression strength of 80MPa. And it has low viscosity and it doesn’t shrink. Also it is much more expensive. Mapei has a similar product. Now I’m waiting for the first bar to cure…

One interesting thing I noticed while reading the different datasheets is that the concrete need up to 30 days to reach the maximum strength. After 2-3 days the strength can be as low as 15%. This will complicate the measurements.

 

[Sonic EscapeParticipant @sonicescape38234]

One more thing, I found this interesting diagram:

On the Y axis is the inherent damping factor in a material when it is dynamically loaded. On X axis is the strength. So concrete is better than cast iron. Better even than stone!

[mgnbukParticipant @mgnbuk]

Physics Anonymous on YouTube built a surface grinder

Not very well, though & they seemed to loose interest in it. Given the way that they kept welding bits on to the framework as they went along, I wonder how it maintained any sembelance of alignment over time.

On the special purpose machines made by one of my employers the frames were usually steel fabrications, but these were always sent away to be stress relieved before machining. Had PA done this with their frame, then it would have had a better chance of alignments staying as they expected over time. Even better would have been to encased a stress relieved space frame in a concrete block . German lathe builder Boehringer made slant bed CNC lathes this way – the steelwork had all the fixings & support structures for linear bearings, ballscrew support beraings, servo motor mountings etc. in the correct geometic positions relative to each other & this frame was encased in an epoxy concrete mix to give the finished product the look of a casting but far stiffer & better damped. IIRC they once exhibited at a machine tool exhibition with a machine hung from a crane producing accurate parts.

Nigel B.

[Grindstone CowboyParticipant @grindstonecowboy]

A couple of years ago, there was a chap posting on here (sorry, forget his name) who was making a concrete-filled/encased CNC machine. I think he was in Australia. Might be worth looking at?

Rob

[Frank GorseParticipant @frankgorse]

The chap with the epoxy concrete grinder project was called John Macnamara I think. Haven’t seen him on here for some time.

[Grindstone CowboyParticipant @grindstonecowboy]

That was the the name, thanks. Unfortunately, as you say, seems to have disappeared, along with a couple of my other favourite posters, Joseph Noci and Luker. However, Luker seems to be doing some good Youtube videos now.

Rob

[DiogenesParticipant @diogenes]

Don’t know how relevant this might be, but maybe some interesting background reading;

http://www.lathes.co.uk/multimachine/multimachine.pdf

 

 

[Sonic EscapeParticipant @sonicescape38234]

After 4 days I can stick a screwdriver in that mortar sample. I think that the 40MPa compression strength from the datasheet is a lie. Next I’ll try some recipes for regular concrete.

I saw a lot of new magnetic chucks on Ebay from Germany. I can’t decide what size to choose. They are between 6″ to 18″ in length. I think I will use the grinder for small parts. So maybe a 10″ chuck would be enough.

I’m not sure about the linear guides. There are so many types. Would a HGR20 be fine? What brands are better?

 

[Diogenes]

On Diogenes Said:

Don’t know how relevant this might be, but maybe some interesting background reading;

http://www.lathes.co.uk/multimachine/multimachine.pdf

 

 

Thank you for the document. It is very interesting. It mentioned how the concrete shrinks and how to avoid this.

[Colin BarronParticipant @colinbarron94178]

Have a look on face book, a manual feed Eagle for £500 with a magnetic chuck or a Jones and shipman 540 for £1500 which is much better but heavier.  Spend the time you would have spent on rebuilding a machine that will be rigid and give a decent level of precision.

[Author]

Posts