This is sort of an odd thread because I am actually most interested in where/how you learned the answer to the following question:

What is the best ratio of ethylene glycol to water in coolant for a car which has overheating issues (and freeze protection and rust protection are not a concern).

In my forum reading here and elsewhere, the common answer is that less ethylene glycol is better because the specific heat of water is higher than EG. But I have read some older Society of Automotive Engineering papers which contradict this common wisdom. They say more EG makes a car less likely to overheat (up to 70%). So I am confused right now.

So ignoring all other issues (freeze protection, rust protection, water pump lubrication, etc.):

1) To avoid overheating, is less ethylene glycol better?
2) Where/How did you learn this (1st hand experience? the internet? engine handbooks, etc.)?

(for simplicity, can we assume that no other additives are being used in the cooling system (like Water Wetter or Purple Ice), because these would just complicate the question)

When time allows, I will try to post some of the observations from these old research papers. But for now, I would really appreciate your answers to those two questions. The more responses the better, as I am most curious about how people have learned about this issue. Thanks so much.

__________________
1968 220D, w115, /8, OM615, Automatic transmission.
My 1987 300TD wagon was sold and my 2003 W210 E320 wagon was totaled (sheds tear).

I have lived in the mideast and some tropical african countries where the lowest temperature in winter hits 90F.

usual coolant mix there is 70/30 or 60/40 - the higher content being water.

When in the bush after repairing a hose pipe which blew out and refilled with 100% water from a stream or rain channel. You will see that the engine runs extremely cool. So much so that old MB temperature gauges actually will move when the thermostat oscillates. The only downside is the corrosion, pitting and cavitation that will render the block and head deck useless along with the water pump housing and also the welch plugs in a few thousand miles.

__________________
2012 BMW X5 (Beef + Granite suspension model)

1995 E300D - The original humming machine (consumed by Flood 2017)
2000 E320 - The evolution (consumed by flood 2017)

It's fairly complicated and not so cut-and-dried as the mix ratio. The design of the cooling system plays a huge role in how well the coolant accepts and releases heat. Water has one of the highest specific heats of any liquid, meaning it can absorb a very large amount of heat from something with little temperature rise (this is why steam engines are so powerful).

Antifreeze/ethylene-glycol raises the boiling point and reduces the freezing point of water. The result is that an EG/H2O mix can operate at a higher temperature before it boils. Pressurize the system and you can run at an even higher temperature before the water boils. Raise the boiling point of the fluid you're working with and you can operate at lower pressures for a given temperature fluid.

What does this translate into for your engine? Localized boiling. If you have hot spots in the cooling channels in the engine, a water-only coolant can cause overheating despite having the highest specific heat due to localized boiling. Steam can't conduct heat like liquid water can.

The old adage of 50/50 water/antifreeze is there for ease of mixture, benefits of corrosion inhibitors, and a good compromise of boiling point and specific heat of the coolant. Hot-rodders often run 60/40 water/antifreeze for that extra bit of heat capacity.

It goes without saying, but if your cooling system is working as designed, you shouldn't be overheating.

My knowledge comes from old car service manuals and growing up around mechanics.

__________________
Current stable:
1995 E320 149K (Nancy)
1983 500SL 120K (SLoL)

Black Sheep:
1985 524TD 167K (TotalDumpster™)

Gone but not forgotten:
1986 300SDL (RIP)
1991 350SD
1991 560SEL
1990 560SEL
1986 500SEL Euro (Rusted to nothing at 47K!)

Focusing on how/where I learned the answer to the best EG/water ratio, I just read the bottle. 50/50 recommended for most areas, ratios up to 70/30 recommended for extreme conditions.

In my(admittedly limited compared to many here)experience, the ratio is pretty moot for cooling provided the cooling system is in sound mechanical condition. I've had no issues with straight water at 100+ deg temps. If you're relying on your radiator fluid mix to prevent boilover, you likely have other mechanical issues you should resolve, regardless of your mixture.

Much more important is freeze protection in the short term, and corrosion protection in the long term.

__________________
617 swapped Toyota Pickup, 22-24 MPG, 50k miles on swap

I use 2/3 water, 1/3 coolant. You need some coolant for rust prevention.

https://hellafunctional.com/?p=629

__________________
CENSORED due to not family friendly words

I totally agree w/ Diseasel300. I published my MS Thesis in the Int J of Heat & Mass Transfer and have engineer'ed in many industries (not automotive), so not a total idiot, but nobody is paying for my thoughts so value as any free comment.

As mentioned, several aspects to over-heating. If you "boil over" (i.e. exceed pressure of the radiator cap spring), you will lose coolant and then gets much worse. 70% water/ethylene glycol has the highest boiling point of the mixture, but you can get much higher w/ pure glycol (~350 F). Evans Waterless is that plus a mixture of propylene glycol, and the later is sold as Sierra Coolant (up to 5% water, but can boil that off). Racers run pure water (w/ cooling additives), but perhaps more for track rules since glycol leaks are slippery.

If your T-stat is controlling (80 - 90 C, i.e. not wide open), then the specific heat of the coolant doesn't matter since "good enough". Of course, the heat rejection also depends on the flow-rate. Heat transfer involves several variables, which proves confusing to those who said, "I'll never need this" in high school math and science. But doesn't stop them from spouting their confused "common sense" on websites. One long passed around among hot-rodders is that a restriction disk in coolant path helps by "giving time for the coolant to cool off in the radiator, if it passes thru too fast it doesn't have time". Not kidding. You can find a web article where an engineer at Robertshaw traced this silly concept to 1930-40's cars w/ a weak radiator cap spring which would burp coolant with a beefed up water pump that gave more flow pressure, thus the restrictor reduced line pressure. But, can't kill that myth.

Of course, anti-corrosion is usually more important. In the 1990's, I couldn't convince relatives in Indonesia to use anti-freeze. They ran pure water and their radiators clogged w/ rust. At least, they were smart enough to use condensate from the home AC. No auto parts carried it and all said, "never freezes here". I noticed they use it now and no old vehicles are seen on the road.

I run Evans in 6 of my 7 vehicles, and both 300D's, since I value anti-corrosion most. I also retrofit w/ silicone hose when I can. Even new cars have cooling limits. I understand one place manufacturers test new designs is on the I-5 south grade going to LA (i.e. Grapevine) on a hot summer day (meaning >110 F to you easterners). My 300D's hit 104 C there on a hot day, so "good enough" and without corrosion should stay that way.

__________________
1984 & 1985 CA 300D's
1964 & 65 Mopar's - Valiant, Dart, Newport
1996 & 2002 Chrysler minivans

I design and manufacture high performance radiators. I recommend 50/50 regardless of climate. As has been pointed out, there are many factors you need to evaluate when designing a coolant. These include latent heat, density, viscosity, lubricity, surface tension, boiling point, and oh yeah, specific heat. I just want to point out some misconceptions that have been voiced here, elsewhere, all over the web, with regard to specific heat.

First, let's take another look at the chart that TJTS put up. I've attached an annotated version. The first question you should ask is what are the conditions of measurement? If specific heat is represented as a single number, it only has validity for one temperature and pressure. Specific heat isn't a point, it's a curve. For water, it doesn't matter much because it's always going to be close to 1. But glycol's specific heat increases significantly with increasing temperature. So while it may be .59 at room temperature, by the time you get to 212F, it's around .69.

The next question is what units are you looking at? Take a careful look at the chart: the dimension is BTU/lb/degreeF. The importance of this is that you're seeing a number that's reflects weight, not volume. You don't fill your radiator with 16lbs of coolant, you fill it with two gallons. Glycol is 12% heavier than water, so if you fill your radiator with it, you have more coolant by weight, and 12% more thermal mass is an important difference.

When you add it up, at higher temperatures a 50/50 mix will have about 90% of water's heat transfer capacity, and pure glycol about 80%. And that's without taking into account that glycol mixes can run much hotter than water. Not what you would expect from a casual glance at TJTS's chart. So you're not giving up much with 50/50. And you get more than a few benefits:

- Higher boiling point. Overheating is boil over, not just running hot. A glycol mix can raise boiling point by up to 20f.
- Less likelihood of cavitation in low pressure areas due to lower vapor pressure and lower surface tension. If you look at the pump involute of a car that's been running pure water, it usually has pockmarks. This isn't ordinary corrosion, it's metal blasted away by localized boiling.
- Pump lubrication
- Corrosion protection
- Freeze protection
- Viscosity compatible with system design
- Consistent with manufacturer's warranty if applicable.

Attached Thumbnails

     

Without question, I've never seen so many INTELLIGENT and well informed responses to a question on this site.

No s*** guys, well damn done.

Our subject car (87 300TD) is in CA, never freezes, and he driving a cooling system POS that really struggles with the AC on in temps above 80º. Been there done that.

Shortsguy1, glean from what has been shared, you need an optimized mixture for CA that will maximize the latent heat carrying capabilities in our 100º+ heat, especially if you go inland from SLO to PR to sip some fine pinot.

__________________
Mark
========
THE WHITE FLEET
2016 GLE300d 4-MATIC 38K BROWN!
2012 S350 Bluetec==94k WHITE

2007 ML320 CDI==166K WHITE (FOR SALE)

Under new management:
2005 E320 CDI--140K--WHITE
1995 E300-Diesel-133.5K--THE CAR IS BLUE
1986 300SL--97.5K (European) AND WHITE. Back in Europe!
1991 190E 2.3-73K California Perfect.--WHITE
1995 E320-Wagon-159K--WHITE (recently scrapped)
1987 300D Turbo-213K--WHITE
1987 190D 2.5 Turbo-288K--WHITE

I'm all too familiar with the 87 om603 cooling system limitations. Swap in the smaller water pump pulley from the 3.5, consider electric cooling fans.

__________________
CENSORED due to not family friendly words

I have been using 50/50 for about 50 years. My dad told me that was best.

__________________
[SIGPIC] Diesel loving autocrossing grandpa Architect. 08 Dodge 3/4 ton with Cummins & six speed; I have had about 35 benzes. I have a 39 Studebaker Coupe Express pickup in which I have had installed a 617 turbo and a five speed manual.[SIGPIC]

..I also have a 427 Cobra replica with an aluminum chassis.

What technique did dad recommend for achieving the desired ratio?

__________________
When cryptography is outlawed, bayl bhgynjf jvyy unir cevinpl.

Same same from my father-- If I remember correctly, there was a graph on the back of the prestone jug, much like the graphs in the earlier posts in this thread that provided a optimum temp for ratio. I think 50/50 was the best comprise for us Minnesota types that need lowest possible temp protection against highest possible boil over protections, and it always worked!

__________________
Mark
========
THE WHITE FLEET
2016 GLE300d 4-MATIC 38K BROWN!
2012 S350 Bluetec==94k WHITE

2007 ML320 CDI==166K WHITE (FOR SALE)

Under new management:
2005 E320 CDI--140K--WHITE
1995 E300-Diesel-133.5K--THE CAR IS BLUE
1986 300SL--97.5K (European) AND WHITE. Back in Europe!
1991 190E 2.3-73K California Perfect.--WHITE
1995 E320-Wagon-159K--WHITE (recently scrapped)
1987 300D Turbo-213K--WHITE
1987 190D 2.5 Turbo-288K--WHITE

His dad's dad probably recommended the the same 50/50.

__________________
85 300D turbo pristine w 157k when purchased 161K now
83 300 D turbo 297K runs great. SOLD!
83 240D 4 spd manual- parted out then junked

My son's Dad did some mixing and checking yesterday.

Wonderful innovation - the Zerex and MB jugs have a clear strip on side plus graduations to show how much is in jug! You fill 100% glycol up to the 2qt mark, then fill distilled water up to the 4qt mark. Amazing eh!

Actually, I was playing with my new Refractometer. Works like a charm:

1. 50/50 Zerex premix - Right on -40C
2. 100% MB coolant - no reading
3. above diluted 50/50 - about -42C
4. Coolant drained from my 350Sl when I switched to Zerex and new rad -70C (shop must have done that when they installed water pump for me!)
5. Some -100C prop glycol mix that had been used to flush my boat engine block - now -32C (Will use for house drain winterization when we go away in winter)
6. E320 coolant - -45C (probably because I once topped up with 100% MB coolant)

Neat new toy! Something like this one.

__________________
Graham
85 300D,72 350SL, 98 E320, Outback 2.5

Knowing what's in the jug is one thing. The challenge is knowing what's left behind in the cooling system.

It wasn't until I bought a refractometer that I realized that there is really no other option if you want your coolant/water ratio to be correct.

__________________
When cryptography is outlawed, bayl bhgynjf jvyy unir cevinpl.