I have been doing body work on cars and vettes for 30 years and never have used epoxy under filler. I never had anyproblem with rust coming thru from under filler. Why is it made out to be a must do thing here when it really iosn't necessary.

I agree, body filler has been put on top of bare metal for over 50 years, with no problems, if it is done correctly. I have done it for over 35 years myself, and still do sometimes.
But with the newer Epoxy primers and 2k primers, it is just a little insurance just in case you did not do it exactly correct....sooooooo if you have the chance to put the Epoxy on first, then why not....just a little more piece of mind.

I think one of the reasons you see second restorations on some cars, is the lack of epoxy under the filler and paint.

Not always the case I know, but down the road if you want to sell and you have pictures or proof of the epoxy on bare metal, its a good selling point.

Rob

To take it to the extreme. Heat enables higher humidity. Curing filler gets hot. Left exposed, you may not see it blatantly, but it may be there to some degree depending on humidity exposure at that time. I have also appied filler over bare metal decades ago. It was standard practice. But panels I have had around since then and taken down showed evidence of rust creeping under the feathered edges. Bare metal, filler, variprime, lacquer primer then topcoat. Yea I know it's lacquer and porous but nonetheless. No failure but it would seem only a matter of time. Some say if properly applied and no surface rust is apparant it's OK to then simply epoxy over the filler with no pre epoxy process. I now put epoxy prior to filler. That way there is always a barrier, especially on bike frames, they take forever to smooth welds and finish. I don't epoxy over and over between coats of cheese, that would be ridiculous; only initially. Look what heat and humidity did to the car exposed to fire in another post. An extreme example yes but indicative of corrosion possiblities.
I had a customer call me the morning following removing the lead from a 40 Chevy. The adjacent areas were all surface rusted.
As far as adhesion of filler goes. Read this for entertainment but I choose no sides in this person's tests. Just something for people to add to the debate issue. http://autobodystore.com/filler_&_epoxy.shtml

in my 30 years in the same body shop [GM dealer] that had customer satisfaction ahead of painters pay [we got paid flat rate per job and did our own warranty free] i learned to make paint STICK, when epoxy first came out [DP-40] it was agod send, but whether it was the newer steel or the VOC something changed. it wasnt working as before [i have a 77 chev i blasted severe rust 24 years ago not a single blister now] 5 years ago i did my S-10 same procedure guess what rust blisters. epoxy USED to stick like s**t but i would not use it unless some sort acid was applied first. having said that this is coming from canada maybe your VOC or whatever havent changed. all im saying is something is different, and i have seen it.

Is that comparing the old DP epoxy with the newer DPLF epoxy?
The DPLF is a completely different animal then the DP and not even remotely close to DP.

Ive had problems with rust coming back under DPLF and being able to wipe off lacquer thinner and now I will NOT touch the stuff. IMO DPLF is a sub-par product.

Since then Ive used Diamont's epoxy with great success along with Valspar's VP epoxy, which to me is comparable to the old PPG DP. Now I use a product from SPI or Southern Polyurethanes.