Water Injection: Why did it take so long to enter service?

[Koopernic]

Musing on water injection I get the first dates of deliver to service squadrons for Water Injection in WW2 as being no earlier than 1944.
1 Me 109G6AM and Me 109G6ASM about February or March 1944.
2 P-47 Probably January 1944 though P-47 had provision for the 15 gallon tank about 1 year earlier. (long wait?)
3 Hellcat and Corsair. Sometime in 1944 as well, probably latter rather than sooner.
4 Fw 190D9 about November 1944. Ta 152 maybe Jan 1945.
5 Some sources credit the Fw 190A9 in late 1944. BMW801 is a bit cryptic as some sources say it was tried in Fw 190A5 in 1943 but caused micro cracks in the heads but at that time the Luftwaffe was upgrading its C3 fuel and the engines were injecting C3 as a rich mixture so why bother with MW50? Some sources speak of BMW801 on bombers getting it for takeoff.

Japanese aircraft I have no knowledge.

Hence 1944.

I had thought that water injection would be an aggressive modification but studying this curve of the Me 109G14AM I note that even above the sustainable full pressure altitude of 1.3 atmospheres the DB605A of 6600m/22000ft that the Water is providing about 3% (10mph/16kmh) more speed which must be 10% or 100 extra hp all without any additional boost and the negative effects of higher temperatures. The late service entry is even more puzzling regrading German aircraft given their lack of high octane fuels.

This is perhaps why the V-1710-9 on the P-51H received this. It would have given 150hp even without boost.

 

[gomwolf]

Basically, water injection is Anti-Detonant Injection. It means, it makes engine sucks more air and spray more fuels to cylinder and compress it higher pressure and water itself cannot be compressed. As a result, engine cylinder will be under far higher pressure when you using water injection. So, if you want use it, you need more solid engine cylinder. (My poor English kicking my ass...)

And... AFAIK, Germans use MW50 field installation package on Bf109 at 1943. In principle, it had to use recon only, but it was too good to use restrictively. So pilots demanded it for all aircraft. RLM approved it and they produce the aircraft with MW50 since 1944.

 

[yulzari]

In considering the workings of water injection in aero engines one must remember that they operate at extremely cold air temperatures at height so the liquid cannot be plain water. In auto engines water can be used neat as an antidetonant. Even then generous use can cause intake manifold freezing. In aero engines the water is mixed with alcohol, usually methanol which is an even better antidetonant and contributes to both the cooling of the charge and plays a role in combustion.

Methanol fuel pipes need to be flushed out to avoid it reacting with the pipe metal if left standing in the pipes . We used to flush with petrol. One feature of methanol is a relative indifference to mixture strength and many a crude use of methanol in supercharged motor racing only worked by virtue of pouring huge amounts of methanol into the intake to cool the engine of which only a part was burned. Older members may recall Hillborn-Travers fuel injection once popular in Offenhauser Midget racers. That simply linked the fuel supply to the throttle position. I recall one Hillclimb event that was delayed by several minor incidents so the final runs were in deep twilight. The invisible in daytime burning of alcohol was demonstrated by the afterburner appearance of the cruder practitioners as the exhaust gases were full of unconsumed methanol which was still burning as it came out of the exhausts in the gloom.

The overall point is that there is much more to water injection in a WW2 piston aero engine than just spraying some into the intake. IIRC the early BMW801 attempts found that suddenly spraying large amounts of cold water/methanol into the intakes of an already hot engine caused cylinder head cracking as the hot heads were suddenly sprayed with cold water/methanol. In aero piston engines it is not a simple quick fix. Very handy for those who are limited by the detonantion limits of their fuel supply (i.e. Axis). The Wallies worked on the problem chemically by brewing noxious ultra high octane fuels which coped with the extreme heat of ever higher intake pressure charging. One can remove the feline epiderdermis by a variety of techniques.........

 

[rinkol]

Musing on water injection I get the first dates of deliver to service squadrons for Water Injection in WW2 as being no earlier than 1944.
1 Me 109G6AM and Me 109G6ASM about February or March 1944.
2 P-47 Probably January 1944 though P-47 had provision for the 15 gallon tank about 1 year earlier. (long wait?)
3 Hellcat and Corsair. Sometime in 1944 as well, probably latter rather than sooner.
4 Fw 190D9 about November 1944. Ta 152 maybe Jan 1945.
5 Some sources credit the Fw 190A9 in late 1944. BMW801 is a bit cryptic as some sources say it was tried in Fw 190A5 in 1943 but caused micro cracks in the heads but at that time the Luftwaffe was upgrading its C3 fuel and the engines were injecting C3 as a rich mixture so why bother with MW50? Some sources speak of BMW801 on bombers getting it for takeoff.

Japanese aircraft I have no knowledge.

Hence 1944.

I had thought that water injection would be an aggressive modification but studying this curve of the Me 109G14AM I note that even above the sustainable full pressure altitude of 1.3 atmospheres the DB605A of 6600m/22000ft that the Water is providing about 3% (10mph/16kmh) more speed which must be 10% or 100 extra hp all without any additional boost and the negative effects of higher temperatures. The late service entry is even more puzzling regrading German aircraft given their lack of high octane fuels.

This is perhaps why the V-1710-9 on the P-51H received this. It would have given 150hp even without boost.

View attachment 471965

Water injection will cool the air, thereby increasing its density at any given boost pressure. So some power gain could be expected even without an increase in boost pressure. Note that there would be an increase in fuel consumption on top of the water used.

I recall that the Bramo 323R engines generally specified for the Fw 200C are supposed to have had water injection. If this is correct, they might be the first example of the technology being used in service.

 

[Shortround6]

Water injection will cool the air, thereby increasing its density at any given boost pressure. So some power gain could be expected even without an increase in boost pressure. Note that there would be an increase in fuel consumption on top of the water used.

In actual practice (and it depends a bit on which engine) many engines actually used a de-enrichment circuit in the carburetor when the water injection was switched on. Some radial engines, in particular, had been using large amounts of fuel as an internal coolant, well over what was needed for combustion.
The R-2800-8W in the Corsair for example used 45 gallons an hour less gas when using war emergency power than when using take-off or full military power despite making 250 more hp (with aux blower in neutral).

 

[InlineRanger]

The limited development resources available in the US were focused on exploiting increases in fuel octane first. The benefits of water injection were known since the late 1800s, but it has some significant downsides for aircraft. Not only do you have to carry the water injection/storage equipment and the water itself, but also additional fuel to exploit the hp increase. High octane gives more horsepower at comparatively cooler temperatures and reduced fuel consumption. It was the lowest hanging fruit that everyone pursued first.

Water injection wasn't really utilized until thermal limits were reached. It was a way to generate more horsepower beyond the cooling capacity of an engine. Air cooled engines were a natural beneficiary. Liquid cooled engines tended to have a higher reserve cooling capacity. The P-51 didn't get water injection until the H model when 90" of boost was approved.

The German's were in a more unique situation due to their available fuel. High performance aviation fuels were produced through the hydrogenation of coal, which gave a product consisting mostly of straight chain alkanes having an octane rating of ~72. This fuel responded poorly to additions of tetraethyl lead due to the lack of branched chain alkanes, so the Germans were forced to boost the octane primarily by mixing in $$$ aromatic hydrocarbons (e.g. benzene, toluene). Their 87 octane fuel contained ~15% aromatics and their 95-97 octane contained ~40%. Aromatic hydrocarbons burn with a flame type that radiates more heat into the cylinder walls/pistons than branched/straight-chain alkanes. The result was that their engines ran hotter than allied models. Another issue they faced was that aromatics tend to dissociate at a faster rate than branched alkanes at elevated temperatures. So while their lean octane rating was high, their full rich octane stayed about the same. The net result was that Germans were willing to pay the weight penalty of water injection sooner because they were the first to run out of development potential.

 

[yulzari]

Tthe Germans were forced to boost the octane primarily by mixing in $$$ aromatic hydrocarbons (e.g. benzene, toluene).

Which is pretty much the route taken by Formula 1 engine manufacturers in the first turbo era exploiting the questionable definition of 'petrol' in the regulations. Water injection was not permitted so this is the way they went. It prompted rumours of the carcinogenic properties of some of the chemicals thrown into the brews. They reached @85% toluene by the end.

 

[Koopernic]

The limited development resources available in the US were focused on exploiting increases in fuel octane first. The benefits of water injection were known since the late 1800s, but it has some significant downsides for aircraft. Not only do you have to carry the water injection/storage equipment and the water itself, but also additional fuel to exploit the hp increase. High octane gives more horsepower at comparatively cooler temperatures and reduced fuel consumption. It was the lowest hanging fruit that everyone pursued first.

Water injection wasn't really utilized until thermal limits were reached. It was a way to generate more horsepower beyond the cooling capacity of an engine. Air cooled engines were a natural beneficiary. Liquid cooled engines tended to have a higher reserve cooling capacity. The P-51 didn't get water injection until the H model when 90" of boost was approved.

The German's were in a more unique situation due to their available fuel. High performance aviation fuels were produced through the hydrogenation of coal, which gave a product consisting mostly of straight chain alkanes having an octane rating of ~72. This fuel responded poorly to additions of tetraethyl lead due to the lack of branched chain alkanes, so the Germans were forced to boost the octane primarily by mixing in $$$ aromatic hydrocarbons (e.g. benzene, toluene). Their 87 octane fuel contained ~15% aromatics and their 95-97 octane contained ~40%. Aromatic hydrocarbons burn with a flame type that radiates more heat into the cylinder walls/pistons than branched/straight-chain alkanes. The result was that their engines ran hotter than allied models. Another issue they faced was that aromatics tend to dissociate at a faster rate than branched alkanes at elevated temperatures. So while their lean octane rating was high, their full rich octane stayed about the same. The net result was that Germans were willing to pay the weight penalty of water injection sooner because they were the first to run out of development potential.

You are clearly well informed on these matters and I've been lead to such concepts as "adiabatic flame temperature" and "laminar combustion velocity".

You're saying that components of the fuel the Germans were compelled to use radiated more heat (infrared?) that lead to higher peak temperatures on the engine?

Some fuels seem to burn faster despite having the same RON it seems. I know of some hot roders posting that they get more heat coming out of the coolant flow when fuels are changed and some fuel produce often higher temperatures. I'd assumed it was from a faster deflageration.

It's very much emphasized that fuel can not just be measured by its knock rating alone. Flash point, ease of vaporization, other burn characteristics all matter.

You've probably solved one puzzle for me. When the Technical Oil Missions came to Germany in 1945 to interrogate and report on the German synthetic fuel plants the US experts asked the Germans why they didn't use cracking. German petro chemists stated that cracking was not applicable to the Bergius Process hydrogenated material.

A big portion of the Synthetic iso-Octane the US used to create US 100 Octane came from iso-butylene gas that itself came from cracking (presumably thermal cracking) and that was latter derived from catalytic cracking. (Houdry Process, about 1938) that produced even more.

The octane rating that came out of the German hydrogenation plants, about 72 was about the same as could be obtained from high quality crude from British Contolled Borneo, Trinidad or the Caribbean ie about 75. The Germans had access to Romanian crude which was also in the 70-75 range. (Soviet Troop buildups near these oil fields were a big part in triggering Barbarossa). Appart from California crude most US crude was valued at less than 55 RON.

The German plants initially produced iso-butylene by dehydration(over alumina) of Butanol that had been obtained by Fischer Tropsch synthesis over chromium catalyst. The iso-butylene was also needed as a feed stock for synthetic rubber so it was limited in supply and little could be used for upgrading fuel.

I suspect that the Bergius Hydogenation plants were up to 55% efficient, the Fischer Tropsch 35%-40% though good for kerosene/diesel though the gasoline obtained was only 45 octane. I think they upgrade this with 30% methanol and TEL to be an aviation fuel used in trainers of about 77 ON.

I suspect Synthesis of iso octane by the Butanol route must have been only 25% efficient. Latter in the war they were able to use butane, butylene that came of the hydrogenation plants and bypass the Butanol step in many plants. In 1943 they added on alkylation plant that had been started in 1940 but couldn't commission any more.

The Germans did seem to produce a 96/125 to 96/130 C grade fuel from late 1942 or early 1943. British oil intelligence analysis of drop tanks and captured fuel and aircraft detected a measurable increase. This fuel was used in the Fw 190 in 1943 and allowed rich mixture injection and increased boost (1.42ATA to 1.58ATA) but not the Me 109 till about 1944 and then only some of the time.

I darent call it C3, it was probably C4.

 

[Koopernic]

Which is pretty much the route taken by Formula 1 engine manufacturers in the first turbo era exploiting the questionable definition of 'petrol' in the regulations. Water injection was not permitted so this is the way they went. It prompted rumours of the carcinogenic properties of some of the chemicals thrown into the brews. They reached @85% toluene by the end.

Because of the somewhat paranoid removal of MTBE oxygenates from motor car fuel over ground water contamination fears arising from poor storage we are seeing increasing quantities of BTEX used to increase octane rating instead. BTEX refers to the chemicals benzene, toluene, ethylbenzene and xylene. These kind of mixtures, previously known as Benzol, were used in the Schneider Trophy float plane racers along with acetone. Rolls Royce found that just using mainly methanol produced the most output.

Benzene is definitely carcinogenic and is nowadays limited to 1% in fuels though it used to be up to 20% of some fuels. Benzol was obtained from coal coking, production of town gas from coal or direct pyrolisis of coal (probably the main source of fuel for the Germans after the allied oil campaign)

I actually think the current drive to 98 RON and 100 RON automotive fuels is a marketing madness when the refinery costs and requirement for dubious anti knock additives is considered.

Benzene and probably some of the other BTEX compounds are likely endocrine disrupters that maybe giving girls early puberty but feminizing boys, possibly effecting sperm count. Explains a lot of modern society.

I'd say your likely to get more exposure to aromatics in a nail salon, brutal environments. Benzene used to be used in aftershave because it smelt nice!

I think Formula 1 now uses standard road fuel whereas Indianapolis racing is moving from methanol to ethanol. They're at 10% ethanol now.

 

[wuzak]

Quite simply, water injection allowed higher boost, which caused additional stress on engine components. Until the engines were strong enough to take the extra boost it wasn't much point.

 

[Koopernic]

In actual practice (and it depends a bit on which engine) many engines actually used a de-enrichment circuit in the carburetor when the water injection was switched on. Some radial engines, in particular, had been using large amounts of fuel as an internal coolant, well over what was needed for combustion.
The R-2800-8W in the Corsair for example used 45 gallons an hour less gas when using war emergency power than when using take-off or full military power despite making 250 more hp (with aux blower in neutral).

The question then is if you are de-enriching the mixture to obtain higher boost through water injection why bother with a complicated to produce 100/130 fuel with a rich mixture response when a 100 octane fuel would be as good.

Without the British leading the US down this path my guess is that the US would graduate to the higher power levels water injection offered far sooner.

The Fw 200C3 Condor entered service in 1941. It's BMW Bramo 323R2 engines offered 1200hp with 87 Octane plus Water injection over the 900-1000 of Earlier engines.

Rinkol I think was referring to an engine going from lean/stoichiometric to over boosting through water injection as opposed to one transferring from over boosting through rich mixture injection to over boosting through water injection.

It's interesting data you provided: 45 gallons an hour extra fuel consumption on the p47 for rich mixture over boosting is enormous and puts the cost of carrying water in context.

 

[Shortround6]

Please note the P-47 and Navy two stage engines were using all three methods to reach high boost.
1.High performance number fuel
2. Intercoolers.
3. water injection

Take one away (or reduce it's contribution) and you won't get the same result.
Trying to use water injection to make 100PN fuel act like 130PN fuel is going to require a lot of water.

I would note that the specific fuel consumption of the navy engines was over 0.70 at full power.

 

[Koopernic]

Please note the P-47 and Navy two stage engines were using all three methods to reach high boost.
1.High performance number fuel
2. Intercoolers.
3. water injection

Take one away (or reduce it's contribution) and you won't get the same result.
Trying to use water injection to make 100PN fuel act like 130PN fuel is going to require a lot of water.

I would note that the specific fuel consumption of the navy engines was over 0.70 at full power.

I'm fundamentally questioning whether 100/130 fuel mixed lean will have any different response to plain 100 fuel mixed lean when used with water injection. In the plain air it doesn't make any difference lean by definition nor should it make any difference lean in the presence of water from water injection. Also, the presence of the alcohol antifreeze is going to push the mixture towards rich. Unless the gasoline is substantially de-enriched there are going to be uncombusted alcohols and/or fuels in the exhaust.

 

[Deleted member 68059]

So while their lean octane rating was high, their full rich octane stayed about the same

Can you clarify this - what are you saying stayed the same as what ? - Are you implying that later German aviation fuels had elevated lean mixture knock ratings; but rich ratings more or less similar to their basic service fuels in early 1940 ?

This fuel responded poorly to additions of tetraethyl lead due to the lack of branched chain alkanes, so the Germans were forced to boost the octane primarily by mixing in $$$ aromatic hydrocarbons

I should like to see an original German mixture-response curve that illustrates that please; and a definition of what % rise in anti-knock constitutes the description "poorly"

The result was their engines ran hotter than allied models

I would be very interested to see a table of specific heat rejection for a selection of German watercooled engines vs Allied water cooled engines that demonstrates that.

 

[Shortround6]

I'm fundamentally questioning whether 100/130 fuel mixed lean will have any different response to plain 100 fuel mixed lean when used with water injection. In the plain air it doesn't make any difference lean by definition nor should it make any difference lean in the presence of water from water injection. Also, the presence of the alcohol antifreeze is going to push the mixture towards rich. Unless the gasoline is substantially de-enriched there are going to be uncombusted alcohols and/or fuels in the exhaust.

Just notices this

When running rich there was plenty of uncombusted fuel in the exhaust. P-47s were notorious for black/smoky take-offs for instance.
Spitfires doing supercharger checks on the ground were sometimes noted as having raw petrol running out the exhaust pipes.
The effect of the alcohol in combustion is minimal. You need roughly twice the amount of alcohol per pound of air than gasoline.
On the Corsair when the water injection kicked in the fuel system cut 35-45 gallons an hour from the fuel flow. If the "water" was 50/50 alcohol the alcohol comes nowhere near replacing the the amount of gasoline being cut. A Corsair carried just over 10 gallons of "water" and this was good for about 7 minutes or about 1.4 gallons per minute or about 0.7 gallons of alcohol per minute which is just over 1/2 the amount of alcohol you need to replace the gasoline on a heat value (BTU/energy) basis.
A Corsair at sea level can make 2250hp using water injection but using 245 gallons per hour it is still burning (??) or better using 0.6533lbs of fuel per hp/hour which is certainly on the high side.

 

[Deleted member 68059]

As for fuel flow, in the 109 with a DB605ASM, at full chat its flowing 100 gal/h of C3, which when the MW50 is switched on is automatically increased to 150gal/h of C3 combined with 40 gal/h of MW50 into the impeller eye. Which is made use of by increase of the boost (automatically controlled) from 1.3 to 1.7 ata. Which takes you from 1250 to 1800bhp - in combination with revs increasing from 2570rpm to 2850rpm.

(Below in English is more or less...."Engine Specification Card - DB605 - Issued July 1944" - copyright goblins prohibit me from posting these in full )

View media item 25551

 

[Juha2]

Hello Calum
thanks for the info!
ImpGal I suppose because the DB 605 A charts says 480 ltr/h at Start- und Notleistung (1.42 ata).

Juha

 

[Deleted member 68059]

Glad the info was useful - however is 605ASM chart......for which 1.42 ata doesnt exist. You get 1.3 or 1.7 - and its 370 to 570 lts/hr with and without.

But yes I chose to convert the units because the preceding post had used american units.

 

[Shortround6]

For the Corsair the engine (with no change in RPM ) made 2000hp while using 290 US gallons per hour (using 6lbs per gallon) for a specific fuel consumption of 0.87 pounds per HP/hour which is not only pretty lousy but really points to the fact they were using an excess of fuel to act as an internal coolant. The increase in boost using water injection was minimal (52.5/54in to 57in) at sea level and pretty much inline with the increase in power.

This use of excess fuel (more than can be burned by the available air) dates back to at least WW I and the "air cooled" Renault V-8 engine.

 

[pbehn]

On these Aero engines are the intake and exhaust valves open at the same time?