Warping sheet metal - sandblasting

[AlanD]

Ok, I have read many times that sandblasting sheet metal causes it to warp, and I proved it because I did it. But I want to know why it warps.
I was always told that it warps because the heat generated in the metal from the sand blasting on it. This is where I loose belief.

I was using a small harbor freight blaster on a 55 degree day, so the metal is staying pretty cool, certainly not hot. Now on the same day I used a wire wheel on another piece of sheet metal and got it almost too hot to touch, but no warpage. I should get as much or more warping from the wire wheel, but it doesn't. Why not?

[Sparky]

I think it's due to a "shot peening" affect. The metal gets compressed on the blast side, so it warps. It's got nothing to do with heat.

[MikeK]

Media blasting particles blown with sufficient (too much) pressure will dent the top surface of the sheet. The moved metal is pushed in several directions, including sideways. The cumulative effect of the sideways compressive forces from all the impacts causes only the blasted side of the sheet to expand, bending it.

[mramc1]

I believe there is heat generated in the blasting process because you are abrading material under high pressure. I have blasted in low light (evening) conditions and the rust particles coming off have a dull glow almost like grinder sparks, but not that bright. There has to be some heat generated to cause the particles to glow. It's so concentrated that it doesn't have enough effect to raise the entire panel temperature significantly.

[Joe K]

Plus one to all of the above.

One of the marvels of my engineering experience is "pein straightening" which is the use of a rather dull impact tool placed against a shaft and "hammered" with a device like a chipping hammer - but making no discernable marks or hammerblows on the surface of the shaft.

Contrary to expectation, the tool is placed not on the "high" side as in effort to push the shaft into straightness, but rather on the "inside curve" of the bent shaft.

I saw a boiler feed pump shaft that was fully 1/4 inch out of straightness brought back into 0.000 by the use of such a tool.

Another advantage of pein straightening are it stress relieves the shaft as you go along - existing stress, such as caused by the impact or whatever that bent the shaft - are "erased" by the sonic shock wave that such straightening creates.

It also does not change the diametral dimension of the shaft.

As you do it, you can literally see the shaft "walking" towards you.

When the shaft is straight - one can be assured that no residual stress will remain that will cause the shaft to act "funky" under increase in temperature or normal stress of operation.

It's magic.

I marvel.

Joe K

[Bob Bidonde]

Sponsored Links (Register now to hide all advertisements)

Sand blasting, shot peening, hammering and the English wheel all cause a compressive residual stress on the surface of the metal being worked. This compressive stress causes the metal to curl up (AKA warp).

The compressive stress also has the effect of work hardening (embrittling) the material, more so in non-ferrous metals such a aluminum, copper, brass, etc., than in ferrous metals like steel.

The two effects, compressive residual stresses and hardening make it impossible to completely eliminate the curling / warping without special heat treatments.

[28ACoupe]

It should be noted here that just because a piece of sheetmetal is sandblasted, doesn't mean that it WILL warp. If the blaster is skilled and knows what they are doing, it is a very effective way to remove rust, paint, etc from metals.

I have had all my projects sand blasted for the last 15 years with no warpage to sound metal. However, I give very specific instructions to the person doing the blasting beforehand.

[AlanD]

I must not be skilled. Whats the secret?

And I appreciate the above posts and explanations. Peening makes sense.

[Tom Wesenberg]

To demonstrate what Mike said in #3, after I first built my big sandblast cabinet, I blasted an old 1" thin blade putty knife. The blade curled over at least an inch in just a few seconds of blasting. I held the nozzle close and had about 80# pressure. I flipped the knife over and straightened the blade the same way.

[28ACoupe]

AlanD - What I have the guys do is hold off the metal starting at about 2' at a 45 degree angle or more to the surface and come in closer slowly till the media just starts to remove the surface material. It takes a bit longer to do it this way, but it saves me countless hours of fixing the warping in the metal.

This method has worked every time.

[Kevin in NJ]

No it is not due to heat. The sparks you see are from individual pieces of sand breaking. You do know that quartz is squeezed with a snap to generate the electricity to light your grill?

To cause heat distortion you need to get up into the 500 degree area where metal turns blue (might be a bit off on the temp, but close enough). Expanding air takes heat out of the metal. What does air conditioning do? It compresses a liquid and lets it expand to remove heat. So you are cooling the metal in one respect. You are also compressing the air against the metal but let us not confuse things. I suggest you blast the panel and feel it immediately. If you got it hot enough for heat to warp it then you should burn your hand.

Sand blasting is best thought of a tiny hammers impacting the surface. The amount of energy the sand can impart into the metal depends on the velocity of the particle or the number of particles at any given time. Think one light hammer dropped on your head slowly is not so bad. One light hammer moving fast is going to hurt. Now what about a box of light hammers dropped slowly on your head, ouch that will hurt bad.

So you can cause damage from too much velocity (air pressure) or low pressure and too much sand in the stream. So you need a pot that lets you control pressure and the amount of material in the stream. These cheap siphon blasters allow too much media in the stream and will warp a panel.

From experience of blasting several full cars including a thinner French metal and not warping the metal (even when trying) I can say it is easy to get great results. We have always run 40 PSI (from a Smiths A head conversion) through a pressure blaster and just cocked the media lever on the bottom till I can see media in the stream. It is slower going, but I know I have never ruined a panel.

So your solution is to experiment with pressures and quantity of media in the stream till you find what works for the media you are using.

The warp is caused by too much energy displacing the surface metal. In effect making once side longer then the other. Very hard to correct. Normal stretching with say a hammer is through the full thickness of the metal. So you shrink the whole thickness. But you can not shrink just the one side. You end up shrinking the full thickness so now the other side is short.

[wrndln]

I have sandblasted many, many model A parts and have never seen any war-page of the parts I blasted. I normally sandblast sheetmetal at an angle and use 80#'s or less with a small tip -3/32". In my opinion, it is the best way to clean metal of paint and rust and give an excellent surface for an epoxy primer, like DP 90. If one were to sandblast sheetmetal perpendicular to the metal with a big tip at 90# of more, there might be some warping. I plan to keep sandblasting sheetmetal and other parts as it works better for me than electrolysis or other methods of removing rust and paint I have tried. I use fresh air breathing system and coveralls, which makes the job tolerable. Just stating my experience.
Rusty Nelson

[pooch]

The metal is expanded not compressed on the top side being blasted.

The surface , if flat or concave, which is most of what auto panels are, will rise towards the nozzle.

A dead flat panel may move in with the air pressure, but it will want to raise initially, but it could oil can, but there are very few dead flat panels on a car, the hood louvres would be the ones to watch most.

The trick to not warping is to know just far you can raise the metal within its normal spring back structure.

Too much raising and it will stay warped, with the correct amount for efficiency and it not stay warped.

I used to use 95 PSI and a 5/16th nozzle and after 3 decades of blasting, I knew when a panel would warp and one that would not.

The worst I have ever found on the exterior of a car was a VW beetle door.

Because it is turned at edge only 90 degrees, and spot welded, they always warped just in from edge.

A usual door skin is folded right over and spotted, it has residual tension and strength to resist any warpage.

And never blast a concave large panel or over an internal rib.

You may get away with inside a Model A tub across the back, but go careful.

Concaves inside a fender or cowl is OK, as these are tight curves with plenty of resistance to raising in.

Inside door skins are very susceptible to staying warped in too.

[crumbaa]

WOW another timely topic. How about soda blasting rather that sand, oxide, or slag media? I am under the impression that soda is much more gentle and produces a smoother surface. Is soda less apt to cause warping of thinner materials?

[wrndln]

Tony,
I don't think getting a smoother surface is necessarily a good thing. Assuming you are going to prime the part that is sandblasted, a little rougher surface will give more "bite" for the primer. Soda blasting is probably OK, but I suspect it is more expensive and takes longer to perform. I have never sodablasted before, so I realy don't know much about it. I will continue to sandblast all my rusted sheetmetal panels.
Rusty Nelson

[700rpm]

Originally Posted by AlanD

Ok, I have read many times that sandblasting sheet metal causes it to warp, and I proved it because I did it.

You read the warnings and you proceeded anyway? Why?

[Ross/Kzoo]

Originally Posted by crumbaa

WOW another timely topic. How about soda blasting rather that sand, oxide, or slag media? I am under the impression that soda is much more gentle and produces a smoother surface. Is soda less apt to cause warping of thinner materials?

Before you go that route read the negative reviews before you proceed.

[karasmer]

How long does it take you guys to blast something for example a front fender? I have little experience except on a frame and it took forever.

[wrndln]

I have sandblasted five or six 28/29 front fenders with a small tip (3/32"), of course the tip slowly enlarges to about 1/8" then I replace it. Front fenders have a lot of surface area when you sandblast both sides. With a small tip it take a while - maybe 3 hours, give or take an hour depending how bad the rust is. If the rust is really bad it may take a little longer. I have sandblasted a front fender in 2 hours when there is mostly paint and just a little light rust. BTW, you need a large 2 stage compressor to sandblast well. You also need a a good moisture trap (or two of them) or the sand will clog up a lot - that is real aggravating. Hope this helps.
Rusty Nelson

[pooch]

Originally Posted by karasmer

How long does it take you guys to blast something for example a front fender? I have little experience except on a frame and it took forever.

It all depends on pressure/ size of nozzle/ type of media and skill of operator.

I used to do 3 fenders an hour.

About 10 minutes a side.