TheSamba.com :: View topic - Cautionary Tale: sand blasting / bead blasting on body panel

Thu Dec 27, 2012 2:52 pm

As we worked on "Kick Azz", I got pretty confident with our little sand blaster running glass bead media.

The chassis turned out great:

as well as suspension parts:

We even had the suspension parts plated with Zinc Chromate, and they look awesome:

I had heard many "the heat from sand blasting will warp parts" stories, so I was very careful to be gentle on any thin parts. But I was also very skeptical that "heat" was a problem. The compressed air expanding makes things cold . . . I had a hard time believing that the tiny amount of heat generated from the impacts could withstand the huge blast of expanding cold air.

So first up, I blasted a couple of fenders and the sunroof and had them plated:

My thinking was that the zinc plating would be a great base for the 2K primer to come.

Both of these fenders had a pretty dire need of body work (I had removed a good chunk of bondo), plus the fenders have relatively tight complex curves (curved in both direction, similar to a section of a sphere, rather than a simple curve that is in one direction only, similar to a section of a cylinder) that make them quite rigid and resistant to the warping effects of the bead blasting. So my thought was that I had been "successful" with blasting and plating the fenders.

The sunroof also looked like it had been worked on and when blasting it you could see some areas had more primer+paint layers than others. So when the sunroof ended up with some "oil can" after bead blasting, I just put that down to previous work, not my blasting. If you look at the photo, you can see lots of waviness.

Then I did a hood. It looked great, or so I thought at first glance.

Then I did a door. This door had huge amounts of filler on it (1/4" plus) which I removed using a wire brush wheel on the grinder, then I proceeded to blast the entire door. That door was a mess, and then I remembered that I have three bugs, and one of the other doors must be better. So I grabbed another door and blasted it just like I had done the others. It was pretty wrinkly and I started to suspect that my careful bead blasting was causing warping.

So I started yet another door. This one appeared to have no body work, and was very straight. I carefully blasted all the tight-radius curves, checked the panel. Still straight. Using 2x to 3x my normal distance (~18", giving time for the glass to slow down before hitting the metal), I very carefully blasted just enough to remove the paint. Every few seconds I would check for warping and I found none. So at this point I have a door that is blasted all around the edges and the entire outer side, and no sign of warping.

So far, so good.

I flipped the door over and started blasting the inside surface of the outer skin. I was still using my long-distance (~18") so the blasting should be very gentle. As I just started blasting the inside of the flat panel, I watched it warp up towards me maybe 1/4" to 1/2" (this made a dent from the outside). Now having absolute evidence pounded through my thick skull that even my careful blasting was warping the panels, I shut off the compressor and tried to understand what is going on.

I then re-checked the hood. The hood is not nearly as flat as the door panels, so the warping effect was much smaller than the the door panels, but there it was: running my hand down the hood gave a wavy ride. Ruined.

I found this thread:
http://www.hotrodders.com/forum/sandblast-warping-heat-myth-76938.html

which matches my experience almost perfectly.

Except that there is no explanation why the blasting from outside did not cause warpage, only the inside blasting ?

What I think is going on is that the metal is stressed from the factory press. A normal part keeps that stress and stays springy through its entire lifetime. I remember a neighbor kid sitting on the hood of our old 1972 1302 Super, and being horrified as it buckled in, then relieved when it miraculously sprung back to its original shape.

So when the beads blast into the surface of the metal, they re-arrange it, relieving stress (anneal?).

On the outside curve of the door, the outer layer of metal is in tension. Relieving that tension does not really do anything visible, because the door is not going to suddenly spring into more of a curve. It might reduce some or all of the factory panel's ability to spring back to its original shape.

On the inside of the door, the outer layer of the skin (towards the inside of the car) is under compression. When that compressive layer is stress-relieved, it is like taking the support bricks out from the underside of a stone arch bridge. The metal collapses towards the inside, since the compressive inner layer is now relaxed. So in this one case, I think the metal actually shrunk a tiny bit, not stretched.

Can damage from a blaster be fixed?

I do not think so. That original shape that was pressed into the metal at the factory has some layers of the panel under compression, some under tension. Of course a person could pound and bondo things back to near the original shape, but the metal under the patch will never regain its factory springiness and "memory" of its original shape.

Conclusion:

1) Careful as I might think I was, if there is enough energy in your media to effectively remove the paint, then there is probably enough energy to relieve the stress of the metal, which will likely botch the part forever.

2) I think it is possible to use fine glass beads where there is a sharp compound curve (on a door, this would be everywhere except the big flat panel).

3) Blasting on one side of a panel may have drastically different success than blasting on the other side.

4) This post is based on using glass beads. Other "soft", or "light" or "fine" blast media may have completely different results.

Trying to make myself feel better:

Even though I ruined: 3 doors, one hood and a sunroof, I still had not taken to blasting the main body of the car. If I had blasted the main body I would have completely ruined the it.

After ruining the driver's doors from the 74 and 75, I discovered that the doors from the 1973 are in fantastic shape: original paint, only a tiny spot of rust. They are in much better condition and will take very little work to make them perfect.

I think that zinc chromate dipping for body panels is a good possibility, but there would need to be a way to completely remove the paint without resorting to blasting.

Thu Dec 27, 2012 5:31 pm

If I had a nickel every time I heard someone say "the heat from a sandblaster will cause warping" I'd have a pretty little stack of nickels.
Picture this - one hundred pick hammers hammering away at an area an inch or two square. That's where the warping comes from. It's inevitable.
When the door skin warped while blasting, it's the exact same as if you had done some hammer and dolly work - striking sheetmetal with a dolly one one side and a hammer on the other will pull the metal towards the striking hammer. That's the basis of metal working.

I have to ask - why would you blast the inside of a door skin?

Thu Dec 27, 2012 5:58 pm

baxsie wrote: My thinking was that the zinc plating would be a great base for the 2K primer to come..

think again :lol:

take a look at dry ice blasting

Thu Dec 27, 2012 7:22 pm

CanadianBug wrote: I have to ask - why would you blast the inside of a door skin?

My idea (now recognized as "failed idea") was to remove all the paint from the door and do the Zinc Chromate plating to the sheet metal before paint.

My reasoning was that zinc on the bare metal is a good thing, zinc is being removed from paints for environmental reasons, 2K primer is widly accepted as great, but does not have zinc. Solution: 2k over zinc chromate :)

skills@eurocarsplus wrote: . . think again :lol: . . .

Am I missing something? Is there a reason that the two component epoxy primer not bond to the zinc chromate?

Thu Dec 27, 2012 8:45 pm

When blasting a panel it is not the heat but the impact of the beads that warps it. Each hit causes a slight stretch & with many hits it amplifies it until it warps. One way to minimize this is to hold the nozzle at an angle to the surface being blasted. This will make the impacts less severe & reduce the chance of warpage.

Thu Dec 27, 2012 10:42 pm

baxsie wrote: Can damage from a blaster be fixed?

I do not think so. That original shape that was pressed into the metal at the factory has some layers of the panel under compression, some under tension. Of course a person could pound and bondo things back to near the original shape, but the metal under the patch will never regain its factory springiness and "memory" of its original shape. Sure it can!...I fixed some pretty heavy blasting damage along with hail and various other damage on a customers bug. :D

Fri Dec 28, 2012 6:19 am

Right after blasting would have been a great time to put the epoxy down. That (in my opinion) is the best adhesion known to man.

Have you ever painted over cad? You can peel the paint off by the sheet. I honestly think you will have a epic failure like no one has ever seen.

You might get lucky, but there is no way I would ever in a million years do it that way

Fri Dec 28, 2012 9:16 am

skills@eurocarsplus wrote: Right after blasting would have been a great time to put the epoxy down. That (in my opinion) is the best adhesion known to man.

Have you ever painted over cad? You can peel the paint off by the sheet. I honestly think you will have a epic failure like no one has ever seen.

You might get lucky, but there is no way I would ever in a million years do it that way

I'm with skills on this one, you should have done some homework first. Sorry dude, epic fail.

Fri Dec 28, 2012 10:04 am

I want to see some PROCESS pics on the hail damage. That is very impressive work. That would have been a ragtop graft for sure.

Fri Dec 28, 2012 12:05 pm

type241 wrote: I want to see some PROCESS pics on the hail damage. That is very impressive work. That would have been a ragtop graft for sure.

I agree, the work looks amazing.

Fri Dec 28, 2012 12:27 pm

543fold wrote: baxsie wrote: Can damage from a blaster be fixed? Sure it can!...I fixed some pretty heavy blasting damage along with hail and various other damage on a customers bug. :D
Wow, that body work looks fantastic! Do you want to come to Spokane for a week this summer and work on Kick Azz?

Your pictures obviously show that this kind of damage can be repaired (and quite well, I might add). My argument is that once the original stress that gives the part its original shape, springiness and "memory" is lost, and the part's metal now has different characteristics. The only way to test it would be to take a part that had been sandblasted heavily then expertly worked back into shape like 543fold did to those panels, and then (destructively) compare the re-worked part's memory and springiness against a factory panel.

Ian, skills@eurocarsplus:
Please understand: I am not trying to come off like an expert . . . I have no formal knowledge or training and very little experience in this area. I am trying to share my embarrassing mistake so others may proceed with caution. I do have some technical training in another field (electronics), and I personally have a very great need to understand what is going on. Perhaps my biggest failing is not being able to take things at face value and just practice a technique, I need to understand (or at least think I understand) what is going on underneath the technique. This often leads to me learning by hard experience what I do not understand and therefore will not take as granted.

skills@eurocarsplus wrote: Right after blasting would have been a great time to put the epoxy down. That (in my opinion) is the best adhesion known to man.
For the chassis, this was and still is the plan. We are waiting on the seats to arrive, so we can fabricate and weld in those brackets, then the complete chassis will have a very light blast to remove the tiny bit of surface rust that has formed in the few months since the original blasting, then it will get wrapped up and taken to the paint shop for DP90 primer.

skills@eurocarsplus wrote: . . . epic failure like no one has ever seen . . .
No disagreement here. Ruining all those parts with the bead blaster has been a serious slap in my ego face. I think blasting the body panels is an epic fail already.

My new plan, starting with the beautiful doors from the 1973 parts car is to use very minimal bead blasting only at the tiny rusty spot in the weatherstrip track at the bottom of the door. The the rest of the paint prep will be done using sanding or scuffing. Lesson learned.

I will need to deal with the miscellaneous suspension parts that are already zinc chromate plated. As skills points out, there may be some challenges getting paint to adhere to them. These threads back him up:
http://www.finishing.com/2400-2599/2444.shtml
http://www.finishing.com/150/46.shtml

On the other hand, these pages indicates that yellow zinc chromate is a good base for paint:
http://www.infinitechfinishing.com/Articles/Article7.htm
http://www.allegiscorp.com/docs/materialFinishGuide.pdf

I guess you can find backup for any argument on the intarwebs.

My plan was to paint the zinc chromate plated pieces with the KBS RustSeal system (de-greaser, etcher, then paint). I guess I will use that approach on a disposable part then destructively test the finish adhesion.

I know that a lot of replacement automotive parts are shipped in yellow-zinc chromate, if you needed to paint one of those, how would you approach it?

Fri Dec 28, 2012 12:32 pm

type241 wrote: I want to see some PROCESS pics on the hail damage. That is very impressive work. That would have been a ragtop graft for sure.

Thread here: http://www.thesamba.com/vw/forum/viewtopic.php?p=6345076

Fri Dec 28, 2012 3:17 pm

baxsie wrote:
skills@eurocarsplus wrote: Right after blasting would have been a great time to put the epoxy down. That (in my opinion) is the best adhesion known to man.
For the chassis, this was and still is the plan. We are waiting on the seats to arrive, so we can fabricate and weld in those brackets, then the complete chassis will have a very light blast to remove the tiny bit of surface rust that has formed in the few months since the original blasting, then it will get wrapped up and taken to the paint shop for DP90 primer.

you are walking backwards. the minute that pan was done, it should have been epoxied. when your stuff comes in, grind/weld the touch it up with epoxy.

now, you will re blast the parts you already blasted. and shame on the plating company for ripping you off. all of that plating needs to be sanded or blasted for paint to properly stick to it.

i may have a photo of some pan hardware i did an experiment with. after a light scuff and prime, i can take my fingernail and pull the paint right off. i will see if i have the parts and snap a pic

please, don't think i am pissing on your cheerio's just trying to save you from yourself 8)

Fri Dec 28, 2012 9:08 pm

skills@eurocarsplus wrote: . . . i am pissing on your cheerio's . . .
Damn. I thought they tasted funky! :) I realize you do not have malice, and hope to learn from your experience. Thanks for your continued input.

skills@eurocarsplus wrote: . . . i did an experiment with . . .
I have been working up an experiment of my own. Since I planned to use KBS RustSeal over the zinc plating on the suspension parts, and I already have material on hand, I will set up the following samples:
1) Bare metal (no plating) paint removed by wire brush (KBS acid wash "Rust Blast" used)
2) Bare metal (no plating) paint removed by bead blasting (KBS acid wash "Rust Blast" used)
3) Zinc Chromate as received from the plating shop (NO special wash)
4) Zinc Chromate cleaned (KBS acid wash "Rust Blast" used)My thinking is to use the paint to lay some strip of nylon cloth tape to the metal, then use a spring scale to see how hard the tape is to remove, pulling normal to the surface.

skills@eurocarsplus wrote: . . . the minute that pan was done, it should have been epoxied . . .
I realize I have the cart before the horse here. I have just have resigned myself to going over the pan one more time with fine glass beads. It will not take too long. Then it will be right to the paint shop for 2K.

skills@eurocarsplus wrote: . . . shame on the plating company . . .
I went in with eyes open. I am treating this whole mess as an experiment. If I learn something it will all be worth it. Somehow I had it in my head that Zinc Chromate would not give a paint adhesion problem, and I loved the idea of laying a real coat of metallic zinc on the part. As far as cost, they charge $60 per batch, and all those parts fit handily in two batches, so the expense is not so great.

My plan now it to experiment with finding a paint that will adhere to the Zinc Chromate. If I can find something I am home free. If not, I can re-blast the zinc chromate parts back to bare steel and use 2K.

Fri Dec 28, 2012 9:42 pm

you might try sanding the ZC and then hitting it with a product called Bulldog Adhesion Promoter. it sticks to chrome, and other plating. it smells worse that a dogfart, but seems to work. you can buy a rattle can to test, if it works, buy a quart and spray i on

Sat Dec 29, 2012 10:49 am

i'm glad you don't see any malice, as there isn't any. i have made almost every mistake in the book, and where it comes to saving time and money, i like to share what i know.

i read thru your thread, and no matter what or how it turns out, you are spending time with your kid, and that is priceless

Sat Dec 29, 2012 11:16 am

Master Series silver and/or black over the ZC would be fine for the suspension parts.

Sat Dec 29, 2012 12:46 pm

The folks over at KBS think that their RustSeal (POR-15 similar) will stick fine to Zinc Chromate:

http://techsupport.kbs-coatings.com/threads/using-...#post-1231

Quote: You can do the KBS 3 Step System on a Zinc Chromate Finish and get excellent adhesion. That is the best way to coat a zinc chromate finish and it is fairly common application.

I still think a test would be in order.

Sat Dec 29, 2012 1:00 pm

Mike Fisher wrote: Master Series silver and/or black over the ZC would be fine for the suspension parts.

Thanks for the suggestion.

I just took a look at the Master Series site. It appears to be in the same class of moisture-cure urethane rust products that also includes POR-15 and KBS RustSeal.

I'm sure each brand has its good points and failings.

Since I have the KBS system already, I'll give that a try first.

Mon Dec 31, 2012 7:12 am

Well the parts sure look cool with that plating on them. Just have them do the shell and the whole car will match, funky gold :lol: