Making Steel Bench Top Flat and Clean

Grinder- cutting disc, cutting torch, chisel+ hammer and a lot of blood, sweat and tears.

How getting it flat depends on the surface: if you can remove the obstacles, is that enough?

Otherwise, consider straightening with heat, press or machining it, or both. My welding table is within 0.1 mm straight. (forged steel and machined)

Obviously the original builder/welder had no real clue what he was doing, otherwise your table wont like like it does. Hopefully not too many people got furniture off this table. However the welder started welding in the center and kept goind instead of first spotwelding and checking before going mad. ( Bet the he had the amperage very high as well.) The best way, and the best end result will be to loosen the top, by any means possible, and start again. Obviously this time it can be done correctly. The top DONT need to be welded fully. A number of 1" welds every 24" or so will be more than enough.

All true - but I think the OP bought a clamping jig built for a glue-lam operation.

That being the case, 30 mm over 2.4 m was/is a bit light-on without extensive ribbing and support. Add to that, the flexing due clamping forces (cracking stitches) and no doubt someone decided full welding was an answer.

Naturally, eventually, it sells cheap - and "looks like" a flat steel bench.

It is probably cheaper to sell it for scrap and start from scratch, with a proper "flat surface" design and construction, and maybe a more useful size than 8 feet square.

Or - as said - it is likely flexible enough to be jacked into being "flat enough".

Nothing has ever been built that can't be de-constructed. Follow dvmdsc's advice.

I'd add this. Do not pour epoxy or any other liquid gunk on it. Do it right. You'll never be able to keep the feathered edge from lifting and flaking.

You'll thank us in 10 years.

An off the wall suggestion-

cover the table top with two layers of high grade charcoal- like for a grille- light it and let it heat top to about 500F or so, then let it burn out. When cool, check top level.

The thought is that the charcoal will heat-normalize any issues that the welding triggered.

If it doesn't work, you are only out the cost of charcoal and lighter and where you are now.

As someone that can weld, but has never be "taught", therefore, I get to stick two pieces of metal together. Are you saying that by bringing the entire surface to a high temperature that it will soften and in effect "destress" the metal? Thanks, Jack

Yes- This approach is used frequently in large manufacturing sites- with a dedicated oven- to remove stress points and any potential distortion.

Sure way to prepare surface of steel plate is use hand scraper, a single-edged tool used for scarping metal from surface. Hand scraping is done where a surface need to be trued, to fit a mating part, need to retain oil and give decorative finish. This is highly skilled job requiring patience and tenacity.

Google search "hand scraping of surface table" should give a good lead.

Nice suggestion krishnan.ng, but I estimate to flatten the table would require the removal of about 25 kg of steel, that's major machining, not hand scraping.

Tony

Agreed!

After machining a perfect surface with good oil retaining surface is achieved by hand scraping the top layer.

Hi Guest - come and join us, we're really nice people

Depending on what you want to spend, I assume the top is higher in the centre.

1. Heating the top with a BIG oxy torch in the centre (it will use a lot of gas) then pouring copious amounts of cold water on it will shrink it slightly
2. Heating the welds to stress relieve them
3. Find a company with a large plough grinder

How about unloading a photo.

Good luck Tony

do you need to bronze bushed holes?

I would suggest put a new metal plate on top of the old one. (1/2" steel?)
It will be flat. You can weld the 4 or 5 points of closest contact on the edges..

If you need the holes, you will have to drill them through the new plate to match, but that can be done by accurate measurement and layout.

Chris

This is a heart breaker. What were the people who built that table thinking of when they went at it with the welder?

My approach is based on one of my favorite sayings: "If I can measure it I can make it". (I wish I had an attribution for that sentence. That's lost in my memory. I'm tempted to say Thomas Carlyle or one of his 19th century contemporaries)

Anyway, faced with this problem the first thing I'd do is figure out how to map and record what I actually had. This thing is 2-1/2 meters square. It's going to take more than a straight edge and a set of feeler gauges or machinist's levels. This is something I only have a passing familiarity with.

When machine shops need to certify their inspection surface plates they bring in specialists that use optical techniques to measure flatness down into the tenth range. Autocollimators or lasers or something like that. That may be expensive overkill for half a millimeter. Maybe a simple laser level?

Another thing to note here is that the quality control guys make a big deal over the way a granite surface plate is supported and where to put the support points to get the least distortion. You definitely don't bolt these things down.

Whatever, get the map and then go to work. The thing that comes to my mind here is that the "UB perimeter beam" is supporting the distortion of the steel plate. A question come up at this point. Was the steel plate ever any better than 2.5 mm flat before the welding? Like this thing was probably hot rolled steel, which may be that way right out of the mill. You could Blanchard grind it to get much better flatness; but do Blanchard grinders come that big? Time was that this job could only be done on a big pit planer.

Anyhow, rambling here, one way to kill some of the ridgedity of the shallow box structure formed by the perimeter beam is to cut away each corner. This could be done with a portable bandsaw or a jury rigged big abrasive cutoff wheel to make slots the full depth of the beam. Then measure the results. Taking due note of what you should have already learned about where to support the plate while measuring.

If the minor surgery on the perimeter beam didn't fix anything at least you are on the way toward being able to measure whatever you make of this problem. Perhaps the only answer is grinding and scraping. Hopefully the availability of a measuring method will enable a minimizing of the grunt work.

If nothing else the answer may marking an orthogonal grid on the plate and using predetermined shim thicknesses per your final flatness map at appropriate grid points under any work piece setup. So you have a box of custom thickness metal shims each marked with it's grid coordinates. Use then when a high precision setup is called for.

Ed Weldon

Comprehensive response, thanks.

I've got some big sanding belts (1150 wide by about 3 m long) in various grits. I was planning on cutting them up for linishing belts, but it crossed my mind to glue some to some large sheets of laminated glass and rig it up to some sort of powered system that would spin it on the table top. Seems like a lot of trouble to go to, but using a smaller abrasive pad seems fraught with likelihood of making the situation worse. Would probably first get the top clean of glue etc to avoid clogging the sandpaper before I achieved any material removal.

As stated elsewhere, I might just decide its fine the way it is.

cleaning off the mill scale, won't be a problem

surface grinding or machining could be

you never did say what your goal is? [how good is good enough]

Send the whole table to a machine shop that has a horizontal grinding mill with a table big enough to handle it, they can grind/machine it true and flat in a fraction of the time.

this is the only one that put any brain in answering this question about a flat table - a Blanchard segment grinder will do the job very fast within .001 flip it grind it again - put levling feet under it find earth leval and be happy - if you can not do this you do not need it flat!

knowledge is the only thing we have that works every time

Withers Tool, Die and Mfg Co., 1238 Veterans Memorial Hwy SW, Mableton, GA 30126

http://www.witherstool.com/contacts.html

This outfit has a 144" dia Blanchard grinder. Looks like an answer if you can afford it and can find a shop reasonably close with that size grinder. Could take some setup effort to level it on the table given the welded pieces underneath.

Ed Weldon

thanks for the insult. as the OP hasn't been back, and hasn't provided enough information... anybody's suggestion is valid, without a better understanding of the situation.

I believe my suggestion of adding a new table top to be faster and cheaper than yours.

obviously you think rudeness works pretty well too.

Chris

Thanks for the hint Chris. Should have read through them all. My tin foil hat is in service, should have known you were in the aether. The best way to do it in my opinion, whats his problem?

Just doing a KISS check ... I understand you stated 'flatness'; but, I am just checking to see if you meant level and not flatness. IS THE FLOOR LEVEL? Check the floor/foundation to see if it may be the culprit ... if it's not level, the table may be twisting. If the table is as heavy as I suspect, it may not teeter ... it may just warp the top.

Is the variation linear across the top? If so, shim it on the bottom. --- I bought an unleveled table before ... it was like that by design, since it was more or less in a permanent place in a garage where the floor was sloped. Oddly, the table was at an angle in the garage and ran high at front right to low in the rear left. The guy whom I bought it from, said he took a grinder to the legs until he was satisfied with the level. Apparently, he thought screw on leveler legs were an eye soar.

Otherwise, turn it upside down on something flat, run a rotisserie across the legs ... and roast a pig on it ...

IE ... do what energygod said ... charcoal ... stress relief ... you might even want to drill some holes in the beams at the corners to aid the stress relief.

Picture would help.

If the dogs are going to be used to aid in squareness ... don't forget to check the dog angle to the table, and take that into consideration when used. Once you get the top level you may be able to drill the dogs to 'square' them. Just trying to be thorough ... they will likely be just fine.

That must weigh a tonne or more.

I use a Makita Belt sander with a 40 grit belt to shine up my welding bench and remove the spak. The brass will easily give up with that but plug the threads first with brass threaded rod and cut a deep screwdriver slot in the top of the rod so as to remove them l8r.

Epoxy won't work.

What do you want to do with said bench in the future?

2.5 in 2400, or 2500 is fairly flat.

First question: Do you really need more flat than that?

If you do, then you need to establish if the table has been machined or ground previously. If it was machined, the underside should show 'roughing' evidence.

If not, then 'breaking the skin' will cause large movement - a true "can of worms"

All in all - I'd be inclined to try jacking, wave a belt sander over it for the glue and rust - and decide it's flat.

Other wise I'd say you need to cut off the top and have both sides ground - and the top of the frame/legs machined flat as well, then bolt it back together.

That's assuming the frame it came with, is actually correctly designed to support/keep 1.4 tons of table top "flat" over 8 feet in both directions.

Well, I'm slightly overwhelmed. I've been away from an internet connection, and now I find a treasure trove of good advice waiting for me, possibly the best being 'decide its flat'!

Yes, the top is a piece of 30 mm thick steel with mill scale still attached. It is also plumbed for compressed air and vacuum, plus a double GPO at each corner. The table was installed in an architect designed workshop where the legs went straight through the wooden floor (36 mm ironbark planks) and were welded to the UB subframe. Quite the struggle with a nine inch grinder under the floor to release it, and then some tricky moves, using a one tonne electric hoist while operating a bobcat with forks fitted. Not recommended, but got it on the truck and then a jerry built gantry has got it onto the gravel at my place where the workshop is yet to be built.

Probably won't do anything about it until I've got some more important infrastructure sorted, but at this stage I'm leaning toward cutting the welds under it and see if that helps, and then going at it with a floor sander or belt sander.

Tried to attach a photo but it doesn't seem to work. Thanks again to all for the input.

This is the Apple appearance but PC is much the same

For photos - easiest is first drag them onto your desk top.

In "reply" mode;

Click on the camera button (green icon on task-bar above) - it opens this;

select "browse" button. It opens this;

Select 'desktop', (if not already selected)

Find / click on the image e,g.

At foot, click "open"

It then reverts to menu 1 above; click "submit"

It may think for a while if the resolution needs adjusting to CR4 limits.

Once the image appears, in the reply window, (click on it and handles appear for resizing) proceed as normal to 'preview' and 'submit'.

If an image does not show up in preview - you have a file type conflict.

CR4 "camera" accepts most image file types like .jpg, .png. (there is a list somewhere) but not some fringe dwellers.

If the camera you have and the CR4 camera don't get on - use 'save as' .jpg (to desktop) and go back to step one.

Good luck - nice you have registered - welcome to the fun house.

Thanks Guru, I'll give it another go this time with smaller images (pulled out of iphoto as if for email)...

Brilliant. (Note workshop under construction).

If I'd built it, I would have moved the perimeter in another 100 mm to make more room for clamping and blind tapped some holes to bolt it, not welded the top at all. One more pic from the workshop it came out of:

Reply to 'what do you want to build on it?', many things, impossible to predict. Currently building the trailer you can see the back corner of in above photo. It will be a little fold out food kitchen for operating at markets and festivals. Using some nice honeycomb aluminum composite board and good looking big radius corner extrusions. Table will be good for restraining the sink/splashback/benchtop/draining rack unit that I'll weld up. (Trailer is the alternative income plan since moving far from capital cities and regular client base)

I've built quite a bit of circus equipment - free standing acrobat poles, aerial spheres (curved st. st. tubing with 4130 clam joints), free standing tight rope, 360º swing, customised rigging equipment, a two person office that can roll around the stage while people sit at their desks and various other weird and sometimes wonderful contraptions.

Also would have been great to have the table for some 'luges' I built recently for access workers to travel through tunnels in a hydro scheme. They needed to pass through a 400mm hatch and assemble inside. I welded them up out of 50mm st. st. CHS and went with a close tolerance telescoping fit for assembly. Correcting the weld distortion to achieve fit was a challenge, but worked well in the end.

I've also done a few shipping container modifications (slide out bathroom in one, swing up sides, windows, etc.) and some vehicle body modifications, billboard sites, guard rail systems, sculpture pedestals and armatures, etc. I'll have a go at building anything that doesn't involve too much repetition and ideally calls for some imagination. Since moving to a beautiful bush block I've had to give up on 3 phase power, so no longer have the lathe or mill. Also don't have the same access to clients I had in the city, so other sources of income required but will always play in the workshop.

Don't mean to bang on, but must say that I'm pretty impressed by the quality of the responses to this query and hope that this reply is read as addressed to all. Any notes on forum etiquette welcome. Great to have found this resource. As my partner said, these people should be solving the world's problems. I pointed out that they were, one detail at a time.

Great stuff Mate. Welcome to CR4. Same here, if you cut out "the big smoke" you loose some but make up for it in other ways.

I can imagine you are not a friend of sitting in front of a computer but if you find the time have a look at something we have been discussing at length concerning shipping containers. This thread will also give you a good idea about the etiquette around here.

http://cr4.globalspec.com/thread/49244?frmtrk=cr4sd#newcomments

Could be of interest to you. Maybe you have some new ideas.

Have one on me, Ky.

Assuming eventually it gets bolted to a concrete slab, keep an eye out for one of those short Acrow props, to use as a jack under "where the center droop is most".

It's quite possible, when you get to banging about, some of the stress will relive. A adjustable 'center leg' is then handy - also movable is handy as it will always turn out to be in the way of something at some time by "just that much".

Miker -- Once you get that table flat enough for you it'll be perfect for welding up two rectangular frames that must mate close enough to seal a weather gasket with one latch and two hinges.

You will want to work on a way to measure flatness. If the table is flat over large areas but local areas that are too far out of the plane it will be harder to get good measurements.

Your work will be only as good as your straightness, flatness, length and angle standards. Without this you'll find yourself turning down jobs that you might otherwise take on and make money from. The big fab shops have the capabilities. But they have a lot more overhead than you do.

Keep an eye out for a discarded blade of a large (like 1.5 or 2 meter) metal shear. At some point they get ground down as far as they will go. They are good hardened and ground steel and the mounting surfaces will still be pretty good. Cultivate a relationship with a large machine shop operation that may have a large granite surface plate. That can be a resource for checking whatever candidate for a straightness standard you come up with. A coordinate measuring machine would do a better job; but time on those things is costly and your relationship with the shop guys might not bear that kind of strain.

Ed Weldon

what are you going to build with it?

G'day miker

We'd love to see photos from the table braking. Some here are still busy braking a bathtub. If you stay for a while you'll know what this is all about.

There is more than one way to straighten out a table. Good luck, Ky.

I was just thinking to myself, and I haven't seen anyone else broach this issue: Are you absolutely positively certain that your steel table top is "not flat"? I ask you (the OP) in what manner did you measure the table top? Did you measure it from the floor or some other reference point or plane, and are you absolutely certain that the floor on which the table stands is absolutely positively dead flat and is there a way to ascertain such a claim?

My reasoning is that IF the floor on which the table stands is used as a "BENCHMARK" and it is found not to be dead flat (or within extremely small tolerances across the plane), then the errors in the unevenness of the floor will be "carried up" the legs as the table was measured (or from the other plane of reference).

The only way I know to measure the levelness of a surface with any great deal of accuracy is to used a "LEVELED" rotating LASER whose project beam plane is parallel (or very nearly so) to the surface of of the table's surface plane, or you utilize one hell of a very accurate straight-edge that must be absolutely fixed @ each end to eliminate any sort of movement, etc.

Sorry to be a party pooper or rain on the party, but all sorts of postings to solve the "uneven" steel table top make be for naught if it is found that the original measurements are found to be "SUBJECT" of error.....

Isn't anyone else seeing this, or am I off track to some degree? My reasoning is that one must establish a unmitigated baseline or benchmark before proceeding to conduct any follow-through measurements, especially if they're conducted down to the gnat's ass whiskers in millimeters, right?

Just a thought.....

please have a pleasant day guys!

I propose that you go back and recheck

Like I said in #11 with to many words rambling if you like (yawn). If you can measure it you can make it.

CaptMoosie is right. First you need to get the correct measuring method.

Ed Weldon

I don't see the laser as a positive instrument for that kind of job. A regular straight channel will do the measuring better. Our crews use this to plaster walls, ready for high gloss paint and 0.5 mm you see very well with light under a certain angle. Like sundown. We use Aluminum for the weight. With sharp edges. With some movable light, you can easily work to 0.2 mm. For more precision, as what they use to scrape up the bed of a lathe, you can use a sharp edge precision lineal. But these are expensive if you want a long one. I don't see that happen with a laser on surfaces full with grinding dust. Just my opinion. With respect.

Everyone so far seems to have assumed your top is a solid thick piece of steel 30mm thick. My guess is, your top is a thinner piece of steel that is flanged down at the edges to form a 30mm flange or lip. Please clarify.

I repeat: Is the top a solid 30mm thick or more like 5 mm thick with edges flanged down 30mm??? If the top is solid steel, the weight would be about 3100 pounds, over a ton and a half. Another question: How flat is flat? Flat like a surface layout plate or just within a few 10ths of a mm? You haven't given us the complete picture.

Ronseto - You and I are pretty much on the same page here. But you bring out an excellent point. We're all assuming the plate is solid steel.

I'm thinking this guest measured the condition with something like a carpenter's level and I scared him into clamming up. He does have a tough nut to crack if it is the full thickness. If he would just open up and tell us more including where he's located we might be able to give him some usable information especially how to find professional help at a reasonable price.

I have another idea. If the table is sort of squarishly concave upward due primarily to the cooling shrinkage of the welds underneath then similar welds on top might reverse the effect. My instinct would be to put them 3-5 centimeters inboard (maybe a bit more) in a pre-ground deep V groove. Then grind it flush afterward. How deep the groove and how much amperage I'd leave up to a pro welder to advise on.

Does this make any sense?

Ed

Hi Ronseto, wasn't sure quite where to post an answer so I ended up replying to the most relevant post as replying to all would have been an exercise in repetition, so you will find some pics, my response etc in the midst of discussion. Yes, table really does have a 30 mm thick solid top and you're on the money with the weight estimate. It was quite an effort to relocate with inadequate lifting equipment and some resorting to jacks and stacks of blocks plus dodgy work methods.

Much as I'd love an 8 foot square surface plate, I acknowledge that this is impractical. Most likely I'll live with it as is, though I might go to the trouble of cutting those damn welds one day as they have clearly caused a bit of a trough. As for cleaning up the surface (it has intermittent mill scale remaining and plenty of old furniture glue, bit of rust, etc.), I suppose the big power wire brush and patience.

I've polished steel with the belt sander in the past but I'm a little spooked by talk of removing the mill scale opening a can of worms - need to find out more about this. Also need to decide on the best means of preventing rust once the top is clean; just a spray of silicone lubricant? rub down with machine oil regularly? penetrol,?? At the moment table is out in the open air. I've just downed tools and tarped it because of rain. It will live under the roof I will be building as soon as I get time. I don't care about getting ground for welding through the table as I prefer to go direct to the work when welding, so I could paint it, but prefer the bare steel if I can adequately protect it.

Is this just for cosmetic reasons?

What will you be using the table for?

I would take a thinner sheet and just weld it on top. Depending on usage that should last for a while, even if you just hate dints, variations of flatness.

You could place a lead or zinc filler and just shave it off, if you want to fill the space. Didn't read all the other comments so not sure if this is done and over with yet. Good luck, Ky.

Hey Miker,

welcome to our little goat rodeo

CR$ is an incredibly educational & entertaining time waster

As promised Link to the Pub enter at you own peril...

on the table deviation, is the problem worth fixing?

Sounds like your well aware of how to make big things to exacting specifications

Since it still has the mill scale

It will be real easy to make the table worse

maybe use it for a cook top & hope for the best :D