following from the recent thread on this topic, it seems there are some fundamental misunderstandings out there that need to be cleared up.
the first thing to understand is where corrosion comes from in the first place. very simply, it comes down to electrolysis. remember when you made a battery in school with a strip of zinc, a strip of copper and a lemon? well, that works because of the difference in electrode potentials between the two metals, and the presence of an accommodating electrolyte.
in car cooling systems, you have different metals used, iron, aluminum, copper, etc., and you have an electrolyte, the fluid in the cooling system. [there is more than that, but i'm simplifying for illustration]. so, that's going to lead to corrosion!!! how do you stop it?
basic methods:
- remove the electrode potential difference as much as possible - use an aluminum radiator with an aluminum block for example.
- passivate the materials as much as possible - slows thing down.
- use a non-electrolytic coolant fluid.
used together, all these work well and cooling systems can last many years with no obvious or at least, minimal degradation.
but what are the practical realities?
a. people tend to introduce electrolytes into their cooling system - the use of tap water being the prime example. not great, but it's life. and cars don't last forever.
b. use of the above can in fact cause some passivation. for example, buildup of calcium carbonate in a cooling system can slow down corrosion rates since it interferes with electron flow. but it also interferes with heat transfer, a strongly negative and unwanted side-effect. again, not great, but it happens. indeed, as a passivation strategy, silicates were use as a corrosion inhibitor in cheap antifreeze for this reason - it passivated the system by coating it. [but it also coated and ruined pump seals]
c. when, inevitably, a cooling system treated as above becomes too inefficient and fails and a radiator is replaced, the new radiator can fail rapidly afterwards. why?
this last seems to be the big problem that's confusing, even to experienced and otherwise very knowledgeable vehicle techs, and it seems it's being misattributed to use of de-ionized water as antifreeze dilutant.
how can this be? because there is essentially no difference between distilled and de-ionized, and certainly not for this application.
let's go back to what we know from the above:
electrolysis. electrode potentials for copper, iron and aluminum are ranked in that order. iron is more active than copper, but aluminum is more active than iron.
if you have an iron engine block, and a copper rad, even if you use a coolant full of electrolytes, the dissolution takes place primarily in the most active component, the iron. and with some considerable thickness of iron to eat through, you're really not going to notice any problems most of the time.
so why did the new radiator fail?
most new radiators are aluminum. so, as we learned above, now it will be the one that corrodes, not the iron. and, this aluminum is /real/ thin.
but we just used de-ionized water - didn't that cause the problem? nope
- there's no difference between that and distilled. not true de-ionized anyway. some products are sold as producing "de-ionized" water, but they're mis-described, and are merely water softeners, not de-ionizers. [and their product is highly corrosive].
remember that this is a repair of an existing system? well, that engine is full of years of corrosion product. you didn't care about it before, but now, unless your new antifreeze contains sufficient concentration and efficacy of corrosion inhibitors, all those products are going to re-equilibrate back into the coolant and become an electrolyte and provide the means for the corrosion to start. did we use a chemical de-scaler or coolant flush as part of the replacement? then magnify this effect even further because those chemicals are very aggressive and very hard to completely remove.
bottom line: if we want to avoid surprises like this, we need to understand the principles of what's happening.
i. replace like with like wherever possible. your system reached something close to an equilibrium as it stood before. if you change that, and complete electron flow reversal like swapping a copper radiator for aluminum on an iron engine block will do that, is about the worst thing you can do.
ii. use the best quality antifreeze with a decent corrosion inhibitor package. don't use "filtered" or recycled crap.
iii. consider very carefully before using a chemical flush of the system. no matter how you try, chemicals will remain on the metal surfaces and come back out into the new coolant fluid to act as electrolyte and facilitate corrosion. in extreme cases, it may be better to use them than have a system full of scale that's overheating, but if doing so, observe #i above. personally, i recommend leaving flush chemicals alone in aluminum systems unless you have no other choice. [use of decent coolant/dilutant will usually avoid all need for this though.]
iv. use high quality replacement parts! internal passivation and corrosion resistance varies. cheap stuff is cheap for a reason!
v. understand what's going on. don't misattribute a failure to the wrong cause [for this application, there is no difference between distilled and true de-ionized]. you'll spend a bunch of money and you'll have the same problem coming back again and again.
vi. don't mistake the difference between de-ionized and softened water.