Why do I keep destroying aluminum radiators???

[FLYNAVY53]

My daily driver is a 1972 C20 Suburban with an iron block 6.0L LS swap. I've been running Entropy's aluminum, LS swap radiator since I built the truck 10 years ago. My third radiator, which is just under 2 years old, started leaking from the core again today.....I'm assuming due to electrolysis and the aluminum welds failing.

I'm at a loss at this point. The radiator sits on rubber isolators but it has a grounding strap from the aluminum body to the frame of the truck, as well as a sacrificial zinc anode that is grounded directly to the battery per Entropy's instructions. The motor is grounded to the frame with a 1 inch braided grounding strap and the battery is grounded to the engine block with a 0 AWG cable.

Any thoughts or insight would be greatly appreciated. I've never had a problem like this before running an aluminum radiator.

Thanks!

[79 Camaro]

Sorry for your issues.
I've done a number of LS swaps using mostly generic LS aluminum radiators and never had an issue. No grounding or or zinc anodes.
Are you running a 50/50 anti freeze or straight water?
I do mount them with out exception in rubber mounts that capture the cooling tubes top and bottom not the tanks on either side. Pic of my 68.
My 98 Wrangler LS swap with an aluminum radiator has 14 years on it and zero issues.

[FLYNAVY53]

Impeccably clean install! Well done!

Yes, mine is mounted exactly like yours with the factory mounts and rubber isolators over the cooling tubes portion.

FWIW, I just dropped a probe from my multi meter into the coolant tank in the radiator, and grounded the other probe to the battery. It's reading a consistent .2 Volts. Not sure if that is enough to cause the aluminum welds to fail over time, but its definitely there and I dont have the first clue where to begin chasing it.

[FLYNAVY53]

Just curious, what kind of coolant do you run?? One of the guys on the C10 forum pointed out that it was probably my orange Dex-Cool that was eating the radiators. I had no idea that stuff was such a problem, but a quick google search yields no shortage of cases where Dex-Cool destroyed all sorts of aluminum components.

[andrewb70]

I have always used the Dex-Cool (orange stuff) and can't say that I have had any issues...

For your next radiator, I would skip the aftermarket stuff and see if a radiator from a new truck would fit your Suburban. Here is a 2002 Chevrolet truck radiator:

https://www.rockauto.com/en/moreinfo...t=2172&jsn=670

The dimensions are nearly identical to the 72 Suburban.

Andrew

[FLYNAVY53]

I have always used the Dex-Cool (orange stuff) and can't say that I have had any issues...

For your next radiator, I would skip the aftermarket stuff and see if a radiator from a new truck would fit your Suburban. Here is a 2002 Chevrolet truck radiator:

https://www.rockauto.com/en/moreinfo...t=2172&jsn=670

The dimensions are nearly identical to the 72 Suburban.

Andrew

Thanks Andrew….for 100 bucks probably worth a shot. Any thoughts on the .2 volts Im seeing across the coolant? Normal or something that needs trouble shooting?

Thanks!

[andrewb70]

Thanks Andrew….for 100 bucks probably worth a shot. Any thoughts on the .2 volts Im seeing across the coolant? Normal or something that needs trouble shooting?

Thanks!

Distilled water is not conductive. If your coolant is conductive, I suspect that is part of your problem. I always just use the 50/50 pre-mix.

Andrew

[raustinss]

make sure you use distilled water as well

[FLYNAVY53]

Yea, ive always used the 50/50 dexcool mix

[FLYNAVY53]

Did some more investigating tonight. With the probe in the coolant, the voltage reading is anywhere between .2 and .25 with the ground probe placed on the battery, frame, or engine block.

I did the same test on my 2002 Toyota 4Runner and my wife's 2016 4Runner and got .18 volts in both of those.

So if a minute amount of voltage through the coolant is "normal", why do I keep destroying made in the USA radiators?

[dontlifttoshift]

Does the core fail by the upper radiator hose? Is the top plate welded to both tanks?

[FLYNAVY53]

Does the core fail by the upper radiator hose? Is the top plate welded to both tanks?

All 3 times it has failed at either side where the core meets the tanks.

[dontlifttoshift]

Is the top plate welded to both tanks?

[FLYNAVY53]

yes, top plate is welded to both tanks

[dontlifttoshift]

This is all anecdotal and I don't really have any evidence.

I really think that is the problem with nearly all aftermarket radiators. The core tubes are swedged into the header and brazed so there is not a lot holding that together but the real problem is in a wide crossflow the tubes expand at a different rate than the top and bottom plates so it can't help but break the tubes loose from the header. It is almost always at the top near the upper radiator hose......where it is hottest. It's just a theory but I have seen these problems from CR, Autorad, and others that weld the top and bottom plates on.

DeWitts doesn't weld those top and bottom plates and I have not had a failure with them. Most cheap universal radiators do not weld those either and they tend to hold up surprisingly well. I have yet to see an OE radiator that ties both tanks together with anything other than the core and they also seem to live long lives.

[FLYNAVY53]

Thanks Donny, that is very helpful and in keeping with what Andrew said about using a factory radiator and what others have said about some of the cheaper options. I ordered the OEM, later model radiator that Andrew suggested so we'll see how that goes. If I cant come up with a viable mounting solution for that I'll explore some of the cheaper "direct fit" options.....or the DeWitts. I have a DeWitts in my '64 Corvette and never had an issue, although I'm running the old school green coolant in that car.

[marolf101x]

Though, like Donny, I don't have any real scientific data to prove this, but I have to agree with his assessment.

The failure is likely due more to thermal expansion and retraction than electrolysis.
OE tanks live as they have nothing tying both tanks together. . .just the core epoxied into the plastic end tanks.
As the core expands there's nothing stopping it from "growing".

Alternatively, an aftermarket radiator that has the tanks tied together is trying to keep the core from expanding, which results in cracks, typically at the edge of the weld.

If you look closely many of the more expensive radiator manufacturers have "expansion joints" in the upper and lower plates.

I'd assume if it's electrolysis (which is the fluid acting as an acid and "eating" the aluminum) the metal would be discolored, as it this photo:

(image acquired from this site via a quick Google search: https://ve-labs.net/electrolysis-101)

If the aluminum around the crack/failure is clean I'd say it's metal fatigue caused by thermal expansion.

I've seen this a lot on radiators mounted solidly (say a 1st gen Camaro). The mounts crack at the weld to the tank.
Look at how new vehicles mount their radiators. . .tons of room to "move".

[FLYNAVY53]

Much appreciated, you guys have me convinced....thanks very much!

[79 Camaro]

I will go against the grain a bit. I always run the green coolant 50/50 mix. On my Wrangler every two-three years I drain the radiator and replace with green 50/50 coolant.

[FLYNAVY53]

I dont think thats going against the grain.....nobody is debating that green coolant works fine. The question is whether or not Dexcool is causing my radiators to fail, or if it's electrolysis. And the consensus seems to be neither and that the culprit is in fact thermal expansion causing cracks because my radiator top cap is welded to the end tanks. Zooming in on the photo of your red C10 you posted earlier, you can see that the top cap on the radiator you are running is not welded to the end tanks which makes sense since you indicated you havent had any issues.