French coolant for the HS 12Y?

[elbmc1969]

Does anyone know what coolant the Hispano-Suiza 12Y was designed for? Did it change in the later models? Was it changed in service for the M.S.406, say?

I'm wondering if the M.S.406 had such an awkward radiator because it was using pure water or something else with a low boiling point. I was just reading that the early Merlins were designed for pure water, which caused problems with over-large radiators.

Thanks!

 

[waroff]

D520: Aérodiol/water, 50/50
MS 406: Ethane-Diol(ethyl-glycol)/water 40/60
Hispano Suiza prescribe, 10 to 60% glycerin, ethyl glycol, or ethyl alcool

 

[elbmc1969]

D520: Aérodiol/water, 50/50

Which is a 12Z. (No idea what I was thinking there ...

Thanks for the info! Unfortunately, it leaves me searching for the reasons that the M.S.406 had such a terrible radiator.

 

[elbmc1969]

From Liquide de refroidissement pour le HS 12Y-3 (M.S.406)

"Selon la notice technique du moteur (1940) : Les liquides utilisés sont l'eau ou certains autres liquides spéciaux. Nous voilà bien avancés..."

Liquids used are water or some other special liquids. Not "and" or "mixed with," but straight water as a possibility. This leads my back to the idea that the radiator was designed for pure water and an operating range of 60-90 C (as a guess), which would explain a lot about the radiator design.

 

[tomo pauk]

...
I was just reading that the early Merlins were designed for pure water, which caused problems with over-large radiators.

Is there a proof for any of those claims?

 

[elbmc1969]

D520: Aérodiol/water, 50/50
MS 406: Ethane-Diol(ethyl-glycol)/water 40/60
Hispano Suiza prescribe, 10 to 60% glycerin, ethyl glycol, or ethyl alcool

What is "Aérodiol"? Where can I find information about it?

 

[elbmc1969]

Is there a proof for any of those claims?

It's just what I was reading about the development. I can't say that it's authoritative.

 

[tomo pauk]

It's just what I was reading about the development. I can't say that it's authoritative.

Per manual for Fairey Battle, Merlin I engine, coolant listed is a glycol mixture.

 

[elbmc1969]

Per manual for Fairey Battle, Merlin I engine, coolant listed is a glycol mixture.

I meant earlier development, pre-deployment. The source said that the Brits got a supply of ethylene glycol from the U.S. and switched to a much more compact radiator during development. Don't know if that's accurate or not.

 

[tomo pauk]

I meant earlier development, pre-deployment. The source said that the Brits got a supply of ethylene glycol from the U.S. and switched to a much more compact radiator during development. Don't know if that's accurate or not.

Earliest Merlin prototypes (3 types) were to combine water and steam cooling, combining both radiator and condensers. The engines proved very unreliable, and switched to 100% glycol with Merlin E.
Merlin F entered service as Merlin I on the Battle.

 

[waroff]

Aérodiol is probably a french commercial name of ethylene glycol.
Liquide de refroidissement pour le HS 12Y-3 (M.S.406)

hispano suiza 12Hb of NiD 62 use pure water (distilled water or filtered rain water)

 

[ThomasP]

I believe Aerodiol was a french commercial brand name for ethanediol, just with some additives making it more suitable for use in aero engines.
Ethanediol is in turn a contraction of ethane-1,2-diol.
Ethane-1,2-diol is in turn the IUPAC standard nomenclature for ethylene-glycol (C2H6O2).

The French automobile and aircraft industries were well aware of the cooling properties of the various blends of coolants - but like the US, UK, and Russia in the late 1930's - they had not yet settled on the later more-or-less standard universal mix of 70/30 (water/ethylene-glycol) mix for aero engines.

The US in 1939(?) had specified a 2/98 mix for all liquid cooled aero engines, including the V-1710 (this was a major part of the problem with getting rid of heat in the turbocharged V-1710 systems used in the early P-38s, why the non-turbo V-1710s had problems developing higher power ratings early war, and why the V-1650-1 engine charts for the P-40F&L were never rated at more than 1300 HP). AFAIK The US did not switch to a 70/30 mix until the adoption of the Merlin 60/V-1650-3 engines in the P-51B Mustang.

The UK was just starting to switch over from pure ethylene-glycol to the 70/30 mix in 1940 with the adoption of the Merlin X and XII engines. The earlier Merlin II-VIII powered aircraft had to continue using the earlier pure ethylene-glycol unless/until their cooling systems were modified for pressurized cooling.

I am not sure when the Russians switched to ~70/30-60/40 but from what I have seen of their engine charts and manuals I think it would have to have started in 1942.

 

[elbmc1969]

The UK was just starting to switch over from pure ethylene-glycol to the 70/30 mix in 1940 with the adoption of the Merlin X and XII engines. The earlier Merlin II-VIII powered aircraft had to continue using the earlier pure ethylene-glycol unless/until their cooling systems were modified for pressurized cooling.

I'm ... very surprised by that.

 

[ThomasP]

I too was very surprised when I read about the evolution of coolant in aero engines just prior to the war. I first ran across references to the pure ethylene-glycol use in a few pre-war and early-war engine/aircraft manuals. Then I read about some of the early-war combat where it turns out that pure ethylene-glycol is flammable, and some of the early-BoB aircraft still used pure ethylene-glycol as the systems had not all been modified.

Apparently pure ethylene-glycol can absorb a more heat than liquid water but cannot absorb or get rid of the heat anywhere as quickly as water. The pre-war US and UK designs relied on the volume of ehtylene-glycol being pumped through the engine to take the heat away. But that type of cooling has two serious problems:

1. The overall volume is limited by the size of the system including radiator, hoses, engine channels.
2. The volume of any coolant passing through the engine per unit of time limits the amount of heat than be safely generated by the engine without overheating and pre-detonating (either in the cylinder or the intake manifold).

The only way to increase the engine HP beyond a certain point (at least where cooling systems are concerned, every thing else being equal) is to do one or more of three things:

1. Increase the overall surface area of the radiator and engine coolant channels (the larger coolant channels would require redesign of the engine, the increased surface area of the radiator would add drag, and the increased volume would add weight and drag).
2. Increase the volume of coolant passing through the engine coolant channels, hoses, and radiator. This is limited by the size of the coolant channels (the smaller the channels and hoses, the more pressure required to pass a given volume of coolant per unit time) and the strength of the pump (which has to be powerful enough to push the required volume of coolant per unit of time).
3. Change the coolant type to one that absorbs and releases heat faster/easier.

The UK solution to this problem was a gradual step-by-step evolution of the engine and cooling system.

The first step was to change to 70/30 water/ethylene-glycol.

Then engineers at Rolls-Royce increased the size of the engine coolant channels in the subsequent Merlin X and XII, allowing greater engine BHP (1150 to 1200 BHP) while still using 87 octane petrol, yet maintaining safe engine operating temperatures using higher boost pressures.

The adoption of the 70/30 mix took place about the same time as the official adoption of (and access to) 100 octane petrol.

The 100 octane petrol allowed higher boosts without detonation in the cylinder, and the 70/30 mix allowed the higher boost without pre-detonation in the intake manifold and cylinder. Together, these changes allowed the Merlin III and VIII to generate about 1300 BHP vs the earlier 1035 to 1085 BHP. The changes in BHP for the Merlin X and XII were similar.

The next generation Merlin XX and 45 had were designed to take full advantage of the 70/30 mix and 100 octane. The changes included increased engine coolant channel and hose sizes, larger area radiators, and significant mechanical strengthening for higher power. These changes, when combined with a redesigned supercharger which generated less heat in the incoming air/fuel charge, allowed even higher power outputs of ~1400 BHP, with low altitude versions like the Merlin 25 producing ~1600 BHP.

 

[K5083]

pure ethylene glycol is flammable, isn't it? Pure glycol was a cause of aircraft losses in the Battle of France, IIRC.

 

[elbmc1969]

Apparently pure ethylene-glycol can absorb a more heat than liquid water

Ethylene glycol has a lower specific heat capacity than water. My father and I looked this up in the CRC handbook when I was curious about why the He70 would use pure E-G. I've double-checked. E-G has a slightly higher density, which makes up for this a bit. However, if the E-G system is being run well above water boiling (at 120C), the density difference is very marginal. Of course, at 120C the E-G is carrying a lot of heat, helping to make up the difference between it and pure water at 90C.

but cannot absorb or get rid of the heat anywhere as quickly as water.

I'm not aware of that and I can't think of any reason why it would be so.

Of course, running E-G coolant at 120C makes the radiator a lot more efficient. The result was that you could have a much smaller radiator, which meant much lower drag.

 

[tomo pauk]

I believe Aerodiol was a french commercial brand name for ethanediol, just with some additives making it more suitable for use in aero engines.
Ethanediol is in turn a contraction of ethane-1,2-diol.
Ethane-1,2-diol is in turn the IUPAC standard nomenclature for ethylene-glycol (C2H6O2).

The French automobile and aircraft industries were well aware of the cooling properties of the various blends of coolants - but like the US, UK, and Russia in the late 1930's - they had not yet settled on the later more-or-less standard universal mix of 70/30 (water/ethylene-glycol) mix for aero engines.

The US in 1939(?) had specified a 2/98 mix for all liquid cooled aero engines, including the V-1710 (this was a major part of the problem with getting rid of heat in the turbocharged V-1710 systems used in the early P-38s, why the non-turbo V-1710s had problems developing higher power ratings early war, and why the V-1650-1 engine charts for the P-40F&L were never rated at more than 1300 HP). AFAIK The US did not switch to a 70/30 mix until the adoption of the Merlin 60/V-1650-3 engines in the P-51B Mustang.

The problems the early turbo V-1710s had was probably that carb temperature was too high (= will lead into pre-detonation) due to insufficient intercooling at higher altitudes. Early 1-stage V-1710s were well capable to be run at high power levels, 1500-1700 HP.
The power table for the V-1650-1 might be in error with regard to war emergency power, listing for both gears 1300 HP WER on same rpm and boost.

The UK was just starting to switch over from pure ethylene-glycol to the 70/30 mix in 1940 with the adoption of the Merlin X and XII engines. The earlier Merlin II-VIII powered aircraft had to continue using the earlier pure ethylene-glycol unless/until their cooling systems were modified for pressurized cooling.

edit: Seems like Merlin III was able to run over-boost with glycol cooling:
table

 

[tomo pauk]

...
These changes, when combined with a redesigned supercharger which generated less heat in the incoming air/fuel charge, allowed even higher power outputs of ~1400 BHP, with low altitude versions like the Merlin 25 producing ~1600 BHP.

Merlin 25 was not a low-altitude engine.
1600 HP was possible, among other things, due to availablity of 130 grade fuel that allowed for greater boost.

 

[ThomasP]

In response to Elbmc1969's post#16

I am not a chemist (I have worked in manufacturing most of my career, either as a manufacturing engineer or systems engineer, but my first job out of college was as an automotive engineer for Continental and Chrysler on the alternative diesel power-plant for the M1 Abrams MBT) so I do not know if I am the best person to explain what I tried to say in post#14, but I will try.

Pure ethylene-glycol has a much higher boiling point than water (197.6ºC vs 100ºC) and the ethylene-glycol cooling system may be run at any temperature up to close to the boiling point of ethylene-glycol. For this discussion's sake lets say 180ºC vs 90ºC. If we assume the same volume of fluid is being pumped through the engine channels then the heat content absorbed by the two different systems would then be:

90 x 4.184 = 376.6 for water

vs

180 x 2.200 = 396.0 for ethylene-glycol

But only if the respective fluids can absorb the heat at a high enough rate. The rate at which the fluids will absorb heat is limited by 3 factors:

1. The area over which the heat is transferred to the fluid (aka the surface area of the engine channels).

2. The amount of heat the fluid can absorb per molecule (aka the specific heat capacity).

3. The thermal conductance of the fluid (aka rate of molecule-to-molecule heat transfer, or how quickly the fluid absorbs the heat).

Of course, the amount of heat removed must be enough to keep the engine within a safe operating temperature range. For this discussion's sake lets say less than 90ºC.

First let's assume that the engine has just enough channel surface area for a water based cooling system. If we replaced the water with pure ethylene-glycol, taking only no.1&2 above into account, we would have to approximately double the rate of ethylene-glycol passing through the engine.

If we now add in the effect of no.3 above (a thermal conductivity for water of .609 vs .258 for ethylene-glycol) we see that for a given flow rate the cooling is still limited by how quickly the heat is absorbed. The rate of absorption of ethylene=glycol vs water would be:

.258 / .609 = .42%

This means that it takes 2.38 times as long for ethylene-glycol to absorb its specific heat capacity of 2.200 at it does for water to absorb its 4.184. (this is partly because the smaller water atoms move much faster for a given amount of heat and therefor bounce off each other more often, and partly because longer chain molecules like ethylene-glycol take longer to heat up than smaller short chain molecules like water).

Therefor we would have to further increase the volume of coolant passing through the engine channels by a factor of 2.38.

1 / .42 = 2.38

What all the above means for the poor engineer in charge of figuring out how to remove heat from the engine is that if you wish to use pure ethylene-glycol as a coolant, you will have to either:

1. Increase the volume of fluid being pumped through by a factor of (4.184 / 2.200 =) 1.902.

or

2. Increase the surface area of the engine coolant channels by a factor of (4.184 / 2.200 =) 1.902.

or both no.1&2, and possibly

3. Increase the surface area of the engine coolant channels some more to take into account that the coolant may be flowing through the channels fast enough that the ethylene-glycol does not have enough time to absorb as much heat as is required. (To calculate whether no.3 would be required I would need very detailed drawings of the engine cooling channels, including surface areas and discrete element lengths, and specifications for the pump.)

Now, after all the above figuring and designing and hair-pulling for the poor engineer assigned the task described above, he (or possibly some other poor engineer) will have to do it all over again in reverse when it comes to the radiator.

Elbmc1969, you are correct I think, that if you can run the engine at a higher temperature with a given coolant you can use a smaller radiator (ie if less heat has to be absorbed then less heat has to discharged from the cooling system through the radiator). If the water cooled system we used above could be safely run at 120ºC rather that 90ºC the radiator could have a 25% smaller surface area and still do the job.

However, if we switch to pure ethylene-glycol, the radiator would have to be redesigned with the exact same parameters in mind as we used in for the engine, just substitute radiator channels/surface area for engine cooling channels.

Assuming the radiator is designed for the same engine using a 90ºC operating temperature, one or more of the following changes will be required:

1. Increase the volume of fluid being pumped through by a factor of (4.184 / 2.200 =) 1.902.

or

2. Increase the surface area of the radiator cooling tubes/fins by a factor of (4.184 / 2.200 =) 1.902.

or both no.1&2, and possibly

3. Increase the surface area of the radiator cooling tubes/fins some more to take into account that the coolant may be flowing through the tubes fast enough that the ethylene-glycol does not have enough time to discharge as much heat as is required. (To calculate whether no.3 would be required I would need very detailed drawings of the radiator cooling tubes/fins, including surface areas and discrete element lengths, and specifications for the pump.)

Sorry this is so long winded but I do not know how to explain what is actually a very complex problem in a compact manner. Hope this helps clarify what I said in post#14.

 

[ThomasP]

In response to tomo pauk post#18,

UK 100 octane was what the US called 100/130 grade.


In response to tomo pauk post#17,

The early P-38 models used 2/98 mix in engine (until they redesigned the engine radiator systems to use 70/30 mix in the P-38x model). The reason the air/fuel charge entering the carburetor/intake manifold/cylinders was too hot was that they did not build enough extra heat dissipation ability into the intercooler and engine cooling systems. As I am sure you know, the P-38 was designed as a complete package, and the engine and turbosupercharger were part of the package. The original cooling systems were designed for 1000/1150 BHP sustained/maximum and no more. The early turbosupercharger system was at/beyond its limits at 1150 BHP. When the P-38 became active in the pacific the problems were less apparent because most missions were at relatively low altitude where the systems did not have to work so hard.

I may be wrong about what I am about to say, but I think the higher ratings for the 1-stage V-1710s did not occur until the switch was made to 70/30 coolant mix, and the US switched from their early-war 125 grade fuel to 130 grade fuel. This took place at about the same time as the new models of the P-38 entered service also.

edited for accuracy of description of intercooler system