RonKMiller

Well, like AdamC says, I guess if one is going to advance a thesis, it's time to start backing it up with a little science.

By the way, "thesis" is merely a proposition, not conclusive proof. To me it's meant to be a stimulating and interesting conversation backed up by some intelligent thought - and that's all this
thread was ever intended to be. Hopefully it will advance our knowledge of why things go bump in the night.

Without getting into a lengthy technical dissertation I'll try to distill this into simple terms that hopefully
everyone here can follow. I am certainly willing to go into more detail with anyone who can follow
fairly advanced chemistry, although that is not the point.

In order to explain electrochemical degradation (ECD) down to it's most simple terms, it is simply the end result of
electrochemically reactive dissimilar materials forming a galvanic reaction. It is usually and most often caused by metals - however
there is potential (literally and figuratively) for many different materials to cause the process to happen.

It would take an exceptionally well equipped laboratory, and a lot of time to identify all of the chemical processes that
happen over the lifetime of a combustion engine, however there is quite a bit of known information when it comes to modern
cooling systems. For the sake of simplicity we'll stick to glycol and water in a 50/50 concentration, since that is what the
vast majority of us use.

The PH of water is typically around 7.0 to 7.2. The PH of antifreeze (glycol) is 10.5. Mixed together this comes out
to around 8.7 or so. It is generally accepted that for cooling system longevity (and when I say cooling system this
includes ANY components that come into contact with this solution like O rings) this is far too acidic, and the recommended level
should be around 10. In order to slow ECD the system should be adjusted to make the solution more basic, thereby
extending component life. PH changes don't happen overnight in a closed loop system, but over a course of months.
Slowly but surely the PH of a cooling system's solution will become more acidic. In the water cooled aircraft engines
I have worked on (which are pretty far and few between) we always add a PH concentrate and try to stabilize
the solution at around 10.5. The whole exercise is done to eliminate cooling system failures. It should be done
on cars as well, but I will bet dollars to donuts the average shop or home schooled wrench would never take
the time to do this, and would probably dismiss this much effort as "over the top". I would venture that very
few race car teams take the time to do this since longevity over years use is not a priority for them! I would also venture
that very few FD3S owners change their coolant on a regular basis - which would have the same effect as keeping the
acidity down to a reasonable level.

Two of the metals that don't get along well with each other are aluminum and iron. They form an immediate dislike
when ever they come into contact. That's why you should never use aluminum rivets on a steel airplane wing, they will
literally decompose rapidly and the end result is the wing falls apart. It has happened many times to unsuspecting homebuilders...
When the metals can be isolated the disagreement can be dramatically slowed down. Unfortunately the solution in our cooling systems allows them to contact each other rather easily - the brawl starts over again. Rotary engines have massive
amounts of iron and aluminum in close proximity to each other. The probability is that this will accelerate
ECD vs. a standard (iron) piston engine. Metallurgists are very specific about coming up with alloys that "get along" with
each other, and in recent years we have seen huge advances in this field. That's why you see a lot more
aluminum components in car engines - not to mention they are much lighter which makes the whole enchilada better
from fuel economy to handling, etc. Unfortunately the alloys that are in current FD3S engines were designed 20 - and sometimes
even 30 years ago!

Now we can throw heat into the equation. Turbocharged rotaries run hot. Period. If the cooling system is neglected or has a weak
spot (hard to imagine on the FD3S...) the engine can operate at an elevated heat range for a long time before the owner is aware of the problem. Our stock temperature guages are not exactly accurate. Heat is a catalyst, it causes reactions to accelerate dramatically.

So now we have an acidic solution, a pronounced galvanic reaction and heat all working together to start ECD.

How does this whole situation affect O rings, polymers and rubber compounds in general? Not favorably.

O rings specifically can be manufactured from about 25 or so different compounds ranging from styrene-butadiene rubber to
high temperature nitrile rubber (which I "think" our O rings are made out of but have no proof) to ethylene-propylene-diene-rubber to
perfluorinated elastomer to silicone. Each has it's strengths and weaknesses depending on the intended application.

Nitrile rubber has good heat resistance, excellent resistance to aliphatic hydrocarbons (petroleum oil), yet only fair
resistance to acids and fair electrical properties. It is especially susceptible to ECD, and the high acid concentrations that
live in our cooling systems. Whatever the specific compound that our O rings ARE made out of they certainly fall
into what is known as the R group: an unsaturated hydrogen carbon ring. All the compounds in this group
exhibit very poor properties when it comes to acid resistance and electrical properties.

So back to my thesis: substitute silicone O rings (which have superior acid resistance and electrical properties)
and a PH stabilized solution like Evans and you SHOULD have O rings that last a lot longer since you are minimizing ECD.

But, like I said, it's only a thesis. If anyone has any better ideas or would like to blow holes in my
logic - FIRE AWAY!

(But you better back up your ideas with some solid science or you'll be labled as a whack job like Mr. You Know Who, and no, out of respect for the rest of the
gentlemen forum members I will not engage in any additional derogatory tit for tat or name calling.)

paw140

Very nice write-up! These are the kind of discussions I like to participate in Silicone o-rings may be a good idea, but I've seen specs that say silicone doesn't hold up well in constant oil/fuel environments, although it has *excellent* thermal stability. I'd lean towards the fluoronated elastomers, like Viton. Anyone ever heard of anyone using Viton?

paw140

Mr. Stock did bring up a good point, though, about the fact that the aluminum and iron housings in a rotary engine are in contact, and thus the o-ring is not completing a circuit between two dissimilar metals. Do you think ECD would be an issue in that situation?

(I know next to nothing about electro-chemistry - I'm an organic guy)

RonKMiller

It really would not matter if the housings are in contact or not - the fact that the O ring is exposed to the fluid is all that matters - it is of prime importance. It is etched by the mildly acidic solution over time, and that's what causes the microcracks and eventual catastrophic failure.

Viton would probably be an excellent choice, but I am not aware of any production Viton O rings that you could buy off the shelf. It would probably require a custom order and be wildly expensive.
There may be mechanical issues involved like hardness, stretching or compression problems that would eliminate it as a candidate.

Fluorosilicone and Silicone both exhibit desirable properties in a static setting:
Good to Excellent resistance to acid, electrical properties, chemical resistance, heat, and "set" or crushing. Curiously though, they only rank fair in water and steam resistance (hence the idea of
a non-aqueous solution like Evans) and only fair to good to continuous exposure to mineral oil, which really is not an issue in this application. I don't know how they react to burned or unburned hydrocarbons but would guess they do well due to the excellent overall chemical resistance rating. Maybe with your experience in organic chemistry you could investigate this further.

Probably a teflon encapsulated O ring would be the best overall choice - and they are readily available in many different metric sizes. It would take a really experienced chemical engineer to determine the "ultimate" compound for our O rings. One thing is for sure: there are many off the shelf choices now that did not exist when the engine was designed.

It will be curious to see what the Renesis engine uses.

johnchabin

iTrader: (2)

That's what I'm thinking. If it's a different o-ring with the proper material propeties (and the right size), then this will probably be the cheapest option as well.

Svelte_7

iTrader: (4)

Viton is readily available in many different sizes and durometers. Anywhere from NW10 flanges to 10 foot in circumfrence vacuum chambers.

Check out:

http://www.aceseal.com/products2.html

Kalrez is another oring of interest. It's upper continuouly operating temperature is around 550F and has excellent chemical resistant. It's not as commercially available as other types of orings but here's a write up for those interested:

http://www.marcorubber.com/material_chart.htm

RonKMiller

Cool stuff, Kalrez is amazing. Way overkill for our apps, but amazing....

stik6shift00

whoever was trying to argue with him

pomanferrari

iTrader: (1)

Mr. Stock, you have my respect as the ultimate mechanic: you replace and repair cardiovascular systems for a living (yeah a heart surgeon for those non-mechanic out there) but you're wrong about the rotary not having a circuit: Disconnect the battery, take a voltmeter, stick the ground probe onto the block and the positive probe into the coolant without touching the positive probe to any metal part. You can see at least 0.8 Volt with fresh coolant and 1.5 volt with old coolant. So there is ionic migration between metals giving us excess electrons and therefore electromotive force or voltage. Thus, there is an electric circuit in our rotary or any engine with dissimlar metals.

The Information Minister has some good points but I don't think just ECD is the sole cause: it's a combination of things.

I'm pulling the motor right now and will be running Evans coolant to see if the motor will last longer.

GP1200R

I don't think you could go wrong by spending a little extra cash on viton o-rings, or possibly teflon....I thought teflon o-rings were avaliable for the FD?

RonKMiller

...I could not agree more, and that's exactly what I said:

"So now we have an acidic solution, a pronounced galvanic reaction and heat all working together to start ECD."

ECD is the end result of several processes, it is not a sole cause. But it is the culprit. IMHO.

johnchabin

iTrader: (2)

Complete o-ring kit:

http://rotaryaviation.com/o-ring_kits.htm

Some Viton, some teflon, some stock...

Anyone with the o-ring failure (but not due to overheating) ever just replace the o-rings, then reassemble? How long has/did this fix last?

RonKMiller

I think the Viton is a new addition.. and they have the best prices I've ever seen for complete kits.

Decisions, decisions........

RonKMiller

Ahhh, but the outer jacket water O ring is still OEM.

They have absolutely the best prices anywhere on the complete kit. This guy is building some awesome engines.

Svelte_7

iTrader: (4)

Kalrez dammit!

I suppose teflon encapsulated viton would suffice...

rex u.k

hello all,i just registered with the club forum from England u.k,very informative and quite emotional too(o-ring thread)

RonKMiller

Welcome! Yeah, us Yanks are pretty passionate about our engines....Cheers.

Mr. Stock

It's about time.

Wow, does this mean that you can be wrong?

That is probably correct since it reads coincidentally like the following:

"For best heat removal and metal protection 50% water and 50% antifreeze is desired. The proper mix is determined by checking the coolant protection of -35 degrees which equals a 50/50 mix. The problem comes when the PH of the water is 7.0 to 7.2 and the antifreeze PH is 10.5. A 50/50 mix of water and antifreeze would have a PH factor of 8.75. This level is too acidic for today's cooling systems. The recommended PH level should be between 9.8 and 10.5. This greatly reduces the acid content that together with electrochemical reaction of dissimilar materials causes electrochemical degradation. "

And that is taken straight from http://www.advancedautomotiveservice...arcaretips.cfm

Can you say "plagiarized", boys and girls? [SARCASM]Or perhaps you are the original author of that?[/SARCASM]

I wouldn't say aluminum and iron "form an immediate dislike of each other". I would say they like each other. As a matter of fact, they like each other so much, aluminum will donate electrons to oxidized iron(rust). In doing so aluminum becomes oxidized and corrode in the process.

How do you know that the alloys that are used in the current FD3S engine has not changed in the last 20 to 30 years? Unless you know something as a fact, I would not state it in a public forum. I would like to where you get this information from.

Mr. Stock

So which is your thesis again?

Is it the galvanic cell theory leading to electrodegradation of the O-ring?

Or is it the acid attacking the O-ring?

Mr. Stock

I did not say that the rotary engine has no circuit.

The electrochemical degradation theory of rubber hoses require a circuit between two metallic parts with an electrolyte solution. The circuit is completed by the electrolyte solution and the rubber hose.

In a rotary engine, the parts of different metallic composition are in contact with each other. Sure the O-rings are present and the coolant is there too. However, the O-rings do not complete a circuit because the path of least resistance of electrons are through the metalliuc parts thmseleves. The electrodegradation will occur where the different parts and the electrolyte solution(read coolant) are present.
The O-ring is an innocent bystander.

Here is an interesting factoid:
Pre 1986 rotary engines had O-ring grooves in the rotor housing.
1986 and post 1986 rotary engines had O-ring grooves in the side housing.
Why do you think that is?
My guess is that because the rotor housing which is made of aluminum saw more degredation(corrosion) than iron side housings.

DaedelGT

"For best heat removal and metal protection 50% water and 50% antifreeze is desired. The proper mix is determined by checking the coolant protection of -35 degrees which equals a 50/50 mix. The problem comes when the PH of the water is 7.0 to 7.2 and the antifreeze PH is 10.5. A 50/50 mix of water and antifreeze would have a PH factor of 8.75. This level is too acidic for today's cooling systems. The recommended PH level should be between 9.8 and 10.5. This greatly reduces the acid content that together with electrochemical reaction of dissimilar materials causes electrochemical degradation. "

I'm so very very confused. You guys are talking about acids in the pH scale above 7. Those are BASES, not acids. The closer to 7 you are, the less basic/acidic you are. If you were to follow those directions and have the recommended pH of around 10 your coolant would become more harmful than if it were around 8.75. Hell, most detergents are around 10 in pH. Make sense to anyone?

EricM

It's all relative Daedel, relatively speaking even 8.7 is already too acidic.

pomanferrari

iTrader: (1)

Not to hijack the thread but my engine finally gave up the ghost due to o-ring failure on Saturday.

These are the symptoms in my case:
1. coolant turning black with gas smell;
2. sucking down coolant to the quantity of 1 gallon (in addition to that in the system) over 15 miles (I have a Volvo S70 AST which allows me to run with extra coolant).

I'll check with Ray to see if they'll allow me to split the block so that I can take some photos of the o-ring. May be I can use a magnifiying lenses to check for "microcracks"?

Mr. Stock

You know, the more I read, the more I am convinced that you have no idea what you are talking about. I don't even think you know what "culprit" means.

"Culprit" according to Merriam-Webster dictionary -
Main Entry: cul·prit
Pronunciation: 'k&l-pr&t, -"prit
Function: noun
Etymology: Anglo-French cul. (abbreviation of culpable guilty) + prest, prit ready (i.e., to prove it), from Latin praestus -- more at PRESTO
Date: 1678
1 : one accused of or charged with a crime
2 : one guilty of a crime or a fault
3 : the source or cause of a problem

or if you do know what "culprit" means and meant to fool the rest of us by using such a [SARCASM]difficult[/SARCASM] word, I have to pronounce you the king of Gobbledygook and Doublespeak

rpm_pwr

Exactly where I get lost on all this too. I don't claim to know chemistry but I can't see how 8.5 is "too acidic" ???

pomanferrari -> Sorry to hear about your motor. Did you do anything that you think might have caused this? eg Run hot, low coolant, leaks, air pockets, restriction, old coolant etc etc?

I just dont get all this blown o-ring thing. I've blown hoses (x3), AST's (x2) cracked a radiator (x2) and even run the car at racedays in 40degree C heat. But never have I blown an o-ring. But then I'm always changing coolant because I'm always breaking / cracking something I also noticed that I rarely trip the thermofans, even on track days!

-pete