Low spot on milled surface?

[zoltan]

I milled a face on a 2.5"x3.5"x1.5" block of 1018 steel and a 3/16"x3.5"x1.5" 1018 steel plate. The two milled faces will be fastened together. I used a 5/16" HSS end mill at 1,100 RPM, 7"/min feed, lubricated with Boelube paste.

When I checked fitment of the plate to the block I noticed a slight gap at one corner (~0.002"). Regardless of how I turned the plate the gap remained in the same place, so it appears the block has a slight low point at that corner. The affected area is approximately 1/8" deep by 1/4" long.

I plan to re-mill the surface of the block tonight, but I'd like to get some idea of what caused that low point so the same thing doesn't happen again. Any ideas?

 

[mcostello]

Are Your parallels in good shape? Vise bottom flat? Chips anywhere?

 

[digiex-chris]

Easier if I had the part on a surface plate to map out the highs and lows and see exactly what the thing is doing, but a few things come to mind, starting with what I think is the most likely.

Yeah, good point about the chips and parallels and vise. I try to clamp straight to the table when I need flatness and accuracy because my vise sucks. Confirm what you're supporting the part with is in good shape and is as accurate as you need. Add in "slipped in the vise" as another possibility. Don't forget to lock the axes that are not in motion. I get lots of odd movement on my old worn machine if I forget that.

1018 is cold rolled, giving it internal stresses. Very often, cutting will release some stresses, causing it to warp. It may have warped slightly during the cut, then cutting the opposite side made it warp unevenly back the other way. Your block isn't thin, and it depends on how much has taken off, but 0.002" is easily within range of a cold rolled stress relief warp. When doing this kind of thing, with cold rolled, try using a larger piece, and milling an equal amount off of all sides to get close to your final dimension. Then I'd attack this with a very light skim pass on all sides with a well trammed and very sharp fly cutter to try to flatten it without taking much off and introducing another warp. If you can tolerate taking another .008" on your part (0.003-4" per side), a very light pass might work for you.

Check the tram of your head, I'm wondering if it's cutting a slight scallop shape so when it's run on the edge of a part, with maybe only 1/2 or 1/4 of the diameter of the endmill involved in the cut, it's actually producing a slightly curved spot ending low? I think it's less likely this is the case since you didn't describe ridges, but when anything funny happens I always check tram.

milling machines also tend to suffer from table tilt when the table runs off to one side because of gravity pulling down the long end. I'm not positive that's what happened here, since I'd expect to see two corners not matching up, but it's definitely an effect that happens. It may be combining with out-of-tram effects to give a slanted surface. Stick a dial indicator on the table near where you hold the part and run the table to one side and see how much the short end lifts. Might be nothing, might be surprising. If so, it might be worth running through a gib setup proceedure and then trying to plan the work so the table stays as centered as possible.

 

[zoltan]

I'm pretty sure there weren't any chips between the block and vise (I always wipe before clamping). The vise is a screwless and I've found it to have good accuracy. I didn't have to use parallels. The tram on my mill isn't perfect, but it's as close as I can get it with shims. Currently it's out 0.0025" over 12" on X and out 0.0015" over 5" in Y.

When I try to fix it tonight I'll probably try clamping directly to the table. Fortunately it now has a couple holes in it which should make that possible.
Honestly, I've never had too much luck fly cutting steel on my mill, probably because of the low RPM needed even for my 1" fly cutter. It works great on aluminum though! If I don't use a fly cutter, would possibly a smaller end mill work well, since it takes off less material per pass?

 

[digiex-chris]

Don't worry to much about the traming or diameter of the endmill, I re-read your original post and it's not describing scallops or ridges, which would be well repeated across the part. Sounds to me (on remembering my reading comprehension skills heh) like it's very localized. something let go slightly? Endmill slipped in the holder a bit? Maybe table rock is more of a possibility? It stayed put until it was unbalanced enough to move?

 

[purpleknif]

Try going deeper with your z and then set the depth by coming up. This will put the pressure going up so your tool can't drop.

 

[toolznthings]

I believe you are fighting the stress in the cold rolled stock. Very hard to control on parts that are thin. Switch to hot rolled steel for better results. As mentioned, light cuts working both sides may help. Do not tap the part tight to the parallels until you are sure you are getting a flat surface. You get spring back when releasing the vise on the part.

 

[Swifty]

Can you put the vice in a different position on the mill and machine a bit off the block, this may show that the mill is the problem.

Paul.

 

[Ken I]

Cold rolled steel can seriously warp.

I machined a piece of 50x50mm x 800 to a required size of 50 x 49 - only a millimeter removed - part held in two hydraulic vices and when it was released it bent like a banana - about 12mm depth of bow over the 800mm length.

I had several of these to do and had to get the stock normalised / stress relieved before I could go further.

What is happening is that the surfaces are under compressive stress - remove the stressed layer from one side and the compressed steel on the opposite side force the material to bow towards the destressed side - secure clamping helps not at all - it simply bends on being released from the clamps.

I had something equally dramatic happen after grinding a keyway down one side of an induction hardened shaft.

The laws of physics are not always on your side.

2c worth.

Regards

 

[zoltan]

After working on it some more last night I'm almost positive it's cold rolled stress I'm dealing with. Before I milled the face of the block I checked the opposite face and it was flat, when I checked it again last night it was slightly warped. So I milled 0.008" off that face as well, along with the previously milled face. However, the block has warped slightly yet again. It's only out about 0.001" (maybe a hair more), but it's definitely there. How could I stress relieve cold rolled steel so it doesn't warp like this?

I guess the other question is, do I keep taking light cuts to try and get the two faces flat, or is that a fool's errand? I still have about 1mm I can remove.

Also, thank you for all the help everyone has given me here.

 

[goldstar31]

Rough machine it oversize. Then heat it up until it is red hot and leave it in the embers over night to cool- and then machine it to size.

Regards

Norman

Normally, I'd use a magnetic table but remember that it will need packing because it will be warped. I've always used shim stock- to avoid more distortion. If there isn't a magnetic table/vice, I'd probably squeeze car bodge/filler between the clamps and the job. One way or another , it must be flat underneath.

 

[digiex-chris]

if it needs to be flatter than that (flat is relative), imo it's best to either stress relieve it (heat it to about 1000-1200f, soak for a bit, and let it cool in still air) or start with a different type of steel. It'll be a game of chasing the warp one way or the other otherwise. Hot rolled exhibits much less tendancy to do this, and is the same cost or cheaper.

 

[Ken I]

Zoltan,
Chasing the error, as you say, is a fools errand - it might work but more likely it won't - particularly if you want to be fussy about flatness.

The stresses extend 0.8mm to 1.5mm into the plate - so you almost certainly don't have enough material to play with.

Heat it red hot (hanging or supporting to help prevent warpage - depending on size) and allow to cool slowly - if you do it in a furnace simply turn it off and let it cool down. If you do it in a fire bury it in ash rather than embers to avoid carburising (or well ashed down embers - whatever).

Cold rolled is great used as finished size / drilled full of through holes etc. but machining off the skin is a no no - use hot rolled.

Regards,

 

[MachineTom]

Think of this, you are gripping the plate on rolled or sawcut sides, when you tighten the vise, the plate may be distorting from the pressure of being clamped.

The "block" is 1.5" thick, get some parallels and first square the gripping edges, then space the block up so only 3/8-1/2" are gripped. Be sure the tap down the work as you tighten the vise. A 5/16l EN is not going to cause these issues. BTW if your tram is not good don't waste time with a fly cutters, or larger EN.

 

[zoltan]

Quick question, if cold rolled steel is going to warp, does it do so immediately or can it happen over several hours?

 

[zoltan]

Zoltan,
Chasing the error, as you say, is a fools errand - it might work but more likely it won't - particularly if you want to be fussy about flatness.

The stresses extend 0.8mm to 1.5mm into the plate - so you almost certainly don't have enough material to play with.

Heat it red hot (hanging or supporting to help prevent warpage - depending on size) and allow to cool slowly - if you do it in a furnace simply turn it off and let it cool down. If you do it in a fire bury it in ash rather than embers to avoid carburising (or well ashed down embers - whatever).

Cold rolled is great used as finished size / drilled full of through holes etc. but machining off the skin is a no no - use hot rolled.

Regards,

Unfortunately, I don't have the equipment to heat it up to 1200* and hold it there for an hour. I may just scrape this piece and try again, this time not machining any faces.

If I have a long bar of cold rolled steel and I need to cut off a piece, do I risk warping if the cut is at 90* to the exterior faces? What if the cut is at 45*?

 

[digiex-chris]

the change is immediate in cold rolled steel from what I've experienced. Cast iron can change with time if it's not seasoned well but cold rolled is immediate. You can see it very obviously by hacksawing a slit into a cold rolled ring shape (hole drilled through rod). Sometimes it'll collapse and pinch the blade, and sometimes it'll spring open as you're sawing it. Sometimes your vise grabs on to the part well enough that it won't warp until the clamping pressure that is holding it in shape is released.

 

[digiex-chris]

a 90 degree cut to the rolled surface tends to be ok. It's when you remove some of the rolled surface and leave the opposite side that you get into trouble.

 

[purpleknif]

Know anybody with a surface grinder ?

 

[zoltan]

a 90 degree cut to the rolled surface tends to be ok. It's when you remove some of the rolled surface and leave the opposite side that you get into trouble.

So if you do an angled cut you'll generally be asking for trouble?