Scraping Non-Sliding Machine Connection Points?

What is the conventional wisdom regarding scraping/flattening non sliding machine connections? Example the bottom of my lathe bed where the bed is bolted to the support leg castings? Both surfaces are in good OEM shape but I do notice some high spot lines when hitting them with stones(you can see the high points of the original grinding/planning passes. My curiosity is in particular the low spots around the actual bolt holes where the connection point is. Part of me says maybe I should try to increase the contact surface and reduce any stress induced during bolt torqueing. The other part of me says I'm being paranoid and over analyzing things (Thanks Robin Renzetti!)Thoughts?

I'm no pro but I would set up the bearing so that there was NO rocking and the bearing area's were next to bolts in supported places like underor next to a wall or gusset ect

Depends on the machine design.

If the lathe is old and was made with legs at each end of the bed like a saw horse, attachment fastener torque will have little effect on the bed's linearity or alignment so long as the gap to be closed is not eccessive..

If the lathe is a more modern design where the bed sits on pedistals, little improvement will be gained by scraping the faying surfaces on the tailstock end (or intermediate pedistals if equipped) but the headstck end is usually rigid box shaped casting and has a much larger face in contact with the bottom of the bed. If a simple dress with file and stone to remove burrs doesn't eliminate gaps, scraping to close the joint all around is an opition but with limits. I suggest 0.004 as the max gap to disregard (this is a moving target where much depends on the size and rigidity of the pedistal casting and the bed.) Over 0.004" gap chould probably be closed by drawing up the bolts. Scraping is certainly the gold standard but this is an assembly joint where shim washers fitted to fill the gap at each bolt hole is a valid fix.

Heavy duty lathes where the bed is attached to a rigid one-piece base casting poses a dilemma. Supposedly the factory fits the all-but-finished bed to the base where the joints are interrogated with a 0.001" (0.025 mm) feeler and any discrepancies are remedied before the way surfaces are scraped/finish ground. If your lathe has a gap in the headstock pedestal joint, it must be determined if the bed linearity is improved or not by closing the gap. This can lead to quite a program of checking and fitting.

I've been bitten by this snake re-grinding the bed of an 18" Okuma lathe whose bed was rigidly bolted and pinned to a sturdy cast iron base making a very stiff assembly. The customer sent only the bed without the base and the bed did not sit flat on the grinder table - there were wedge shaped gaps on the base connection faces. The decision from the white hats was to shim to neutral stress and grind it. We didn't get a complaint so the assumption was the bosses made a good call but I was never happy with their decision. Shoulda fetched in the base just to be sure.

It is a Monarch 12CK lathe so I would say it is the pedestal type. The pedestals have mounting bosses that contact the underside of the lathe bed so the contact area is reduced to those 4 points.

At least scrape the paint off the mounts then shim if needed


Sent from my iPhone using Tapatalk

You will want 80% to 90% contact for a lathe under head stock or column to base on a machining center, 99% on a steam chest or turbine housing.
File the burrs off and then scrape the surfaces the regular way as you scrape ways or plates, match fitting / scraping the 2 surfaces together bluing up and then stone the heck out of it with mineral spirits to get the high percentage of contact.

bmarsh said:

It is a Monarch 12CK lathe so I would say it is the pedestal type. The pedestals have mounting bosses that contact the underside of the lathe bed so the contact area is reduced to those 4 points.

View attachment 281516

Click to expand...

on a Monarch 10EE the contact surfaces between bed and base casting are scraped, so I would imagine that applies to a CK as well.

Im going to blue up a small surface plate on the pedestal and see what it looks like. Especially now that I may have more free time available LOL

Might be worth you time to research "liquid shim", which amounts to release-coating one side of the interface and bond prepping the other before buttering the bond-preped side with an epoxy paste.

We've probably liquid-shimmed acres and acres of interfaces in my 40+ years in the missile-and-space maintenance and manufacturing. It's a very effective technique.