Surface grinder gibs

[Steve355Participant @steve355]

Hi

around a year ago, I extensively refurbished an old Eagle surface grinder, nearly scraped it to death, poor thing. I put it back together, but since then I haven’t really used it for any precision work, I’ve actually been using it for tapering wood plane blades. But now I’m at the point where I have a few jobs to do that would be much easier if I had a at least semi-accurate, surface grinder.

By the time I finished the scraping, I had got the total error in the main table down to less than 0.001” everywhere apart from a few extremities. I’ve verified this many times on surface plate.  The only thing I didn’t manage to do was scrape the dovetails.

The problem I’m having, I think is with the gibs. at the far reaches of the travel, maybe 3/4 of the way left and right, due to the weight of the table overhanging the knee, it causes an error. Clearly this should be not happening if the gibs are correctly tightened, that should remove any lateral or vertical movement of the table. However, I’m struggling to adjust the Gibbs in such a way that there is no droop at the end of travel, and they are not so tight that I can’t operate the left right motion of the table.

Have some videos of this, but it doesn’t really clearly show the problem in action.

Today I made some new grub screws for the table, and I have some new locking nuts for them. in the hope that this would allow me to solve the problem. But sadly not, it’s either so tight it won’t move, or the table droops by gravity at either end.

Perhaps I need some shims? Or something. I really don’t know.

any advice gratefully received 🙂

 

 

[peak4Participant @peak4]

I suspect you probably need to scrape the dovetails too.
Both sides of the saddle, but only the underside of the table opposite the gib.
If it’s anything like my Herbert, I needed to then make a new thicker gib strip as well.
Personally I’m no expert in this stuff, having only ever scraped in one machine in my life, during the first lockdown.
In the end, it came out to a couple of tenths over 8″ by the time I’d done, so not perfect, but as good as I was going to get.
I didn’t start photographing anything, until after I’d done the underside of the table, but there’s a record in my Flickr album HERE or on Facebook HERE.
Most of the photos are really trying to illustrate how I learned to set up for the measurements.

I did make an angled flat plate out of a lump of cast iron, to the same 55° as the dovetails.
Seen being used here for a slightly different purpose

And the setup for measuring parallelism of the dovetails, albeit on the knee here, rather than the table

I’ve a feeling these images might go missing again, as after submitting, they were viewable, but I couldn’t click on them to show as a larger photo.
Right click & open in new tab came up with 404 error image not found.

I’ll try inserting again one last time

Getting brassed off with this now. 🙁

Bill

[peak4Participant @peak4]

Sorry Steve, The first and second missing images were of my setup for measuring the internal dovetails, after I’d re-machined the tapered gib key on the knee; you might not have the same problem if you’re on parallel gibs; easier to just look in the album, rather than this site

The last missing image was of a screenshot showing how the first image had gone missing, but even that’s vanished now. Just to prove they were there, for a while at least, assuming this screenshot doesn’t go AWOL too.
Ho Hum 🙁

Bill

[Steve355Participant @steve355]

Hi Bill

yes, I remember reading your blog about this when I did it and gaining a lot of ideas and inspiration from it.

When I did mine, I ordered a camelback with a dovetail from the usual UK supplier…. It looks great quality but it turned up 6 months late, long after I’d put the grinder back together. I could use it but it needs scraping itself first and I’d guess that scraping the camelback then scraping the dovetails is likely to take me a couple of weeks at least, and at this point I have other priorities. Its sitting on the shelf waiting for me to feel like scraping again, which may be a while!

I think what might be happening is that the gib is designed by application of lateral pressure from the set screws, to apply some lateral and some vertical pressure to the dovetail. Due to wear and scraping (and I think I may not have been the first to refurbish this grinder) it is providing more lateral pressure than vertical pressure, which is why it is locking up before steadying the table in the vertical direction. That’s my theory anyway.

When I first disassembled the table, a bunch of shim “piano” wires fell out. I assumed that if I scraped the ways flat, they would no longer be needed. I now think they may be needed more. I just don’t know where to put them, or what size to get. Or even if I’m correct.

There’s loads online about shimming Bridgeport tapered gibs, but nothing about standard “parallelogram” style gibs.

Who knows!

 

[Pete RimmerParticipant @peterimmer30576]

If you have the facility to measure how parallel the dovetailed ways are then that should be your first port of call. Otherwise, you could make the best of your setup by carefully setting the moving gib at the limits of travel then in the middle.

[Steve355Participant @steve355]

Well, yesterday I measured the saddle dovetails (I think correct terminology, the part that sits on the knee and supports the table). I did this using a 0.0001” DTI and also a ground piece of round stock in one dovetail. I got mostly 0.0005 max deflection, perhaps increasing to 0.001 at one end. So that’s good news.

That means it looks like it’s the table dovetail. This is covered in oil grooves, and is opposite the gib, so is far more difficult to measure as there isnt an obvious reference surface. But l will have a go today.

[Steve355Participant @steve355]

One thing of interest is the procedure for setting surface grinder Gibbs. Obviously the table needs to be loosen enough to be easy to move, but tight enough to be stable. I guess the balance is more critical than a typical milling machine.

I think my table has seven set screws. I’m not sure if the order that they are tightened and re-tightened in, makes any difference. Perhaps there is a trick to this.

[Pete RimmerParticipant @peterimmer30576]

Well the table is quite easy to evaluate since it has the flat ways directly on the bottom. You can do it with a decent sized surface plate. Mine was convex on the underside slightly (about a thou high in the middle as I recall with the plate up-turned) so I had to scrape it from the middle outwards after filing off the wear ridges. Once you have the flat ways straight you can cobble up a simple gauge to check the dovetail spacing but you must make those bottom ways flat first.

[Steve355]

On Steve355 Said:

Well, yesterday I measured the saddle dovetails (I think correct terminology, the part that sits on the knee and supports the table). I did this using a 0.0001” DTI and also a ground piece of round stock in one dovetail. I got mostly 0.0005 max deflection, perhaps increasing to 0.001 at one end. So that’s good news.

That means it looks like it’s the table dovetail. This is covered in oil grooves, and is opposite the gib, so is far more difficult to measure as there isnt an obvious reference surface. But l will have a go today.

I’m certainly no expert on this, but how is your machine built?
Is it like this, or is it effectively built the opposite way with this side of the dovetails on the table.
I’m guessing it’s like this, by the looks of your gib key adjusting screws

Photo from Tony’s site
If like this, is one of the dovetail sides removable as above, or is it all one piece?

How did you measure it using a clock?
Maybe use two bits of ground round bar opposite each other and a micrometer, but you also need to ensure the dovetails are straight first.

Also, are you quite sure it’s the table rocking on the saddle, and not the saddle rocking on the knee as the table moves, and alters the centre of gravity?
Maybe even the knee rocking on the column, or a combination of both.

Bill

[Ady1Participant @ady1]

If I was going to get serious abut gibs and flatness I would look at using a laser which can be used to inspect across the entire surface

[Steve355Participant @steve355]

Hi Bill

To try to answer your questions….

Yes it is like that but without the removable dovetail. I measured it with a setup as below –

And as I said I got about 0.0005 error, perhaps a little more at the extremity.

I am also sure it’s the table at fault, not the knee or the column. I have made new grub screws and tightened them right up. I can feel the table moving and I can correct the sag at either end by giving it a little pull.

So on to the table, on one side it has the gib and on the other a oil-groove infested dovetail, so it’s very hard to measure.

https://youtu.be/Oa8-4LA7Qx8?si=DH9R44LV66NWUkrq

i could try to use the “apron” surface to measure it, but that’s still tricky and I’m not confident that I would get good results.

like this…

[peak4Participant @peak4]

Unfortunately your last image has gone awol, like several of mine, and the youtube link doesn’t work for me.
That’s the same sort of method I was using to check for wear on mine, but I also tried with a micrometer and two round bars just to confirm, repeating for several different diameters; Initially I used two round bars on one side and a long parallel, with a third round bar on the other, still measuring across with a micrometer, just to make sure the dovetail was straight as well as parallel if you see what I mean.
i.e. to ensure it’s not a perfect banana shape.

You have the opposite profiles to me, the dovetail section on your saddle, is under the table on mine.
I guess the same method still applies; I scraped the table top flat, though far from pretty cosmetically.
I then scraped the front of the table, vertical and flat.
That way, I could stand the table on the front edge on the surface plate, and use a similar sliding clock to make sure the dovetail internal profile is flat.

You could then use a stack of blocks & feeler gauges etc to ensure the other rear dovetail is flat and parallel to the front.
One thing that does spring to mind, is have you blued the dovetails to each other on the saddle  and the underside of the table?

As an exaggerated example, lets say you have 55° dovetails, but because of differential wear one end is 54° and the other 56°, the centre line with a clock will still show straight and parallel.
A similar issue on the opposing part might cause a tightening at the end, whilst being loose in the centre.

Just a thought of something different to measure.

Bill

[Ady1]

On Ady1 Said:

If I was going to get serious abut gibs and flatness I would look at using a laser which can be used to inspect across the entire surface

Lasers aren’t very appropriate. The dot is diffused so that it doesn’t blind people so it’s way bigger than even a fairly large wear error. You might be able to identify that there IS an error but if it’s a dip between two peaks it won’t even see it, certainly not provide any kind of measurement of wear.

[Steve355Participant @steve355]

Let’s try again with the pic and video

 

https://youtu.be/Oa8-4LA7Qx8?si=fpAgaED8TaRSQhvz

 

[Michael GilliganParticipant @michaelgilligan61133]

Pic andVideo link are both there  [in both Posts at the moment!]

… but only the latest Video link works

MichaelG.

[peak4Participant @peak4]

The last photo with the DTI will show whether the dovetail is parallel to the front of the table, but not necessarily show any high spots, other than along the line travelled by the stylus.
I made a couple of straight edges with a bevel; the one with the cream painted end is 55° and scraped flat; it was a slice of cast iron off a bus bar clamp/fishplate about 12″ long.
Even if it’s not exactly 55°, it doesn’t matter, as it was also used on the mating dovetail, which was further blued to this one.
This happens to be the column, but the overall principle remains

I’m not claiming to be any expert here, it’s just how I went about the task

See also for a close up

Bill

[Steve355]

On Steve355 Said:

One thing of interest is the procedure for setting surface grinder Gibbs. …

I think my table has seven set screws. I’m not sure if the order that they are tightened and re-tightened in, makes any difference. Perhaps there is a trick to this.

No one else reacted to so I’ll have a go!   The background is:

1. Gibs removed from machine
2. Piano wire shims came out with it, and assumed not needed
3. New set screws made.

Removing the gibs without noting how they were installed opens the door to re-assembly error  – things like the gibs being inserted upside down, or the right way up but with dimples not aligned with the set-screws.  In the worst case 3 out of 4 ways of fitting a gib are wrong.

I don’t understand what’s meant by “piano wire shims”.   A picture would help. They may be a clue.

Are the new set-screws correct, that is shaped to accurately fit any dimples on the gib?.   Dimples are often used to stop gibs moving sideways, and paired to mate with a set-screw.  If one or more screws and dimples are misaligned, or the screw end is misshapen, the gib may bend and bind causing the symptoms described.

With luck the problem is adjustment. Adjusting a gib correctly can be fiddly, though maybe my method is wrong!  I fit the gib and tighten all the screws gently by hand just enough to bring the gib and slide into contact.   Then I move the table end to end looking for binding.  If the dovetail and gib are in good condition and correctly assembled, then any binding is due to one or more screws being unevenly tight.  Back each screw off one by one, retest for binding every time, and re-tighten any as you go where loosening made no difference.   May be necessary to repeat the process several times.

Once the table moves without binding, tighten the screws with a screw-driver just enough to remove slop.   In my experience the final adjustments are delicate and it’s easy to over-tighten the gib at this stage.  Simply tightening the lock-nut can be enough to spoil the fit, so use the correct size screwdriver to stop the set-screw turning at all when the nut is tightened.

Sometimes I adjust my gibs in one go and think I’ve cracked it at last.  Not so far!  Mostly takes me 3 or 4 attempts to get it right.    Gib adjustment is one of those jobs where persistence and bad language get me there eventually.  I’m self-taught, could be doing it wrong, and am clumsy…

Dave

 

 

[Steve355Participant @steve355]

Dave – thanks for that. I will  answer briefly then have another go tomorrow.

Set screws – good point, the grinding of the new ones may not be sharp enough to fit the dimples. I will check.

orientation of gib – I think this is right, due to the lock nut being different from the set screws, meaning there’s only one way it will go back, but I am not sure.

piano wire shims – I really don’t know. They fell out. These were several pieces of wire of different thicknesses, each piece about an inch long, that were obviously inserted somewhere, I’d guess on the dovetail side of the table. I am guessing but somehow I think these were used to compensate for wear, but I don’t know how or where, or how their sizes were worked out. By the time I worked out what they were, they had all fallen out. Perhaps 6 of them. I found some in the car when I drove the grinder back to London from Wales! Then more as I disassembled the machine.

Steve

 

 

[Steve355Participant @steve355]

And very embarrassingly… the gib was on the wrong way round.

I will reground the set screws and re test tomorrow.

if anyone has any thoughts on the piano wire shims ?

[Steve355]

On Steve355 Said:

…………….

if anyone has any thoughts on the piano wire shims ?

Is there any wear/flattening on them.
The only thoughts I have, since you confirmed there isn’t a removable dovetail.

Completely unrelated, and someone had lost them in the T slots, hence finding some loose after unloading.
If there is any flattening, did someone drop them into the oil grooves, to compensate for wear on one last job before they sold the machine.
No reason for them to be related to the gib, unless it was to push it downwards.

Bill

[Steve355Participant @steve355]

I’m not sure if there is any wear on the shims, Bill, to be honest I am not sure if I still have any of them.

Interesting idea that they may have been in the oil grooves, I agree I can’t think of anywhere else they may have been able to stay put.

I think my next step is to try it out with the gib in the correct orientation and see how it behaves.

Id love to finish the dovetail scraping, and one day I will, but I need at least couple of clear weeks to focus on it, which isn’t likely to happen in the short term.

[DiogenesParticipant @diogenes]

Bill’s comment made me think – I’ve seen wires used to ‘space’ a gib strip to stop it’s toe dragging along in the gutter at the bottom of the dovetail, (and/or it’s head riding in the angle at the top) – if this makes sense..

..a bodge alternative to ‘pinning’ them so that contact remains central to the shears and they are prevented from slipping or being forced up or down under screw pressure..

 

[Steve355Participant @steve355]

Well, I have reinstalled the table with the gibs the right way around this time, and reground the set screws to a nice point. The difference is night and day – its behaving properly now, and I can adjust the gibs as one would expect.

How embarrassing! I guess its  all part of gaining experience.

rule 1 – don’t be an idiot

rule,2 – see rule 1

i haven’t put a dti on it yet to see if it’s behaving itself at a precision level, but I’ll do that and see if any kind of shimming is necessary. Given that i scraped the main bearing surfaces flat, I would hope it’s ok. We will see.

[peak4]

On peak4 Said:

On Steve355 Said:

…………….

if anyone has any thoughts on the piano wire shims ?

Is there any wear/flattening on them.
The only thoughts I have, since you confirmed there isn’t a removable dovetail.

Completely unrelated, and someone had lost them in the T slots, hence finding some loose after unloading.
If there is any flattening, did someone drop them into the oil grooves, to compensate for wear on one last job before they sold the machine.
No reason for them to be related to the gib, unless it was to push it downwards.

Bill

Coming late to this one … with a wild guess that might prompt some thought:

Vincent motorcycles [and others, I presume] used wires inside drilled oil-passages, as a very effective way of tweaking the cross-sectional area and thus the flow.

Is there a possibility that these wires serve a similar purpose ?

MichaelG.

.

Edit: __ even managed to find an illustration after all these years !

.

[Howard LewisParticipant @howardlewis46836]

Nit picking, but since minimal distances are being sought and measured, I would try to re assess the measuring set up to maximise rigidity.

Can the scribing block be used the other way round (So that the pins are on an outer machined surface), and with the clock clamped further down the stem?

You are looking for deviations of less than 0.0005″ with a set up that appears to be more flexible than it could be.

Stiffening things up might improve results!

Howard

[Author]

Posts