FW190-D9 Engine Boost

[CUJO_1970]

Please help me understand  - was the supercharger/boost/manifold pressure regulator really unable to develop full power to the Jumo 213 in winter? Why do German tests with D9 done in December/January reference full boost pressure if it is not possible?

 

Granted - I do not read technical German or even regular German, but I have never once seen this claim for the engine ever made or ever referenced by any pilot about it. Is this referenced in the original German POH or engine manual? I've not seen it in the English translations.

-------------------------

Wie aus der Auftragung ersichtlich; entsprach die Zuordnung bei den Flügen 1 und 2 am 12. und 14.10. nicht ganz den Soll werten. For the 1900 PS – power, the boost pressure below full throttle height, pLC = 1,75 ata, lay too high and for methanol operation with pLC 1,77, somewhat too low. Therefore, the speed increases of ΔVa = ~12 mph (20 km/h) (for 1900 HP) and ΔVb = ~20.5 mph (33 km/h) (for methanol) are to be used. After repeated adjustment, the maximum boost pressures lay in the correct range during flight 3 on 14.10, i.e. 1.7 ata for 1900 PS and 1.8 ata for methanol.

------------------------

Callum Douglas in his excellent new book goes into technical detail about all Jumo 213 systems but does not mention this being unable to achieve full boost.

My understanding is 1.7ata 1900PS was developed for early boosted Dora and then 1.8ata 2100PS when MW50 was used but this would be in any season, summer autumn or winter conditions. For example...

 

Flight tests in winter in Germany also reference full boost in winter months, for example in December and January reference specifically 1.8ata being normally reached with this engine during tests.

[AndytotheD]

As I recall, the engine control unit for the Dora uses a mass air flow sensor instead of manifold pressure to meter the air charge. Colder air is denser and so less pressure would be delivered for a given mass of air. It was my understanding that this was why engine power ratings (cruise/combat/emergency) for the Dora were given in revolutions alone as opposed to a combination of revs and boost pressure; boost pressure would vary wildly with temperature. 

[ZachariasX]

4 hours ago, CUJO_1970 said:

Please help me understand  - was the supercharger/boost/manifold pressure regulator really unable to develop full power to the Jumo 213 in winter? Why do German tests with D9 done in December/January reference full boost pressure if it is not possible?

You have an air mass regulator in The Jumo. Hence in cold weather you have more air mass (higehr air density in cold weather) for the same ata and the engine hence reduces ata to normalize the absolute amout amout of air to what it should be. The PN feature a table of ata attained depending on temperature. It's correct as it is in the game.

[Kurfurst]

If its a regulator, in all likelyhood it could be and was adjusted (caliberated) for winter months, so...

[Bremspropeller]

40 minutes ago, VO101Kurfurst said:

If its a regulator, in all likelyhood it could be and was adjusted (caliberated) for winter months, so...

 

Maybe, but we don't really know for sure.

 

Here's how you'd do it:

 

 

 

 

 

 

MAP / Temperature relationship and MAP-regulator adjustment.

(Note: It's a 1943 Engine Manual)

 

 

Fig. 61

Edited January 4, 2022 by Bremspropeller

[Kurfurst]

2 minutes ago, Bremspropeller said:

Maybe, but we don't really know for sure.

 

It is mere common sense that aircraft received in the cold months of late Autumn 1944 would be set up according to instructions in the manual (BTW, thanks for digging that up. ). Setting up, tuning new planes was routine at the units. 

 

It would appear to me that if the measurements in this thread are correct, our D-9s do not perform simply because they oddly use summer engine settings on winter maps..

[Bremspropeller]

16 minutes ago, VO101Kurfurst said:

It is mere common sense that aircraft received in the cold months of late Autumn 1944 would be set up according to instructions in the manual (BTW, thanks for digging that up. ). Setting up, tuning new planes was routine at the units. 

 

It would appear to me that if the measurements in this thread are correct, our D-9s do not perform simply because they oddly use summer engine settings on winter maps..

 

That's not what the instructions suggest. They're there for initial regulation of a newly installed (replaced) Motorbediengerät and the table's there for initial tuning at a given temperature. Read: If it's 20°C outside, the MAP should be 1,53 ata at 90° Power Lever Position. If it's -20°C, the same PLP should give you 1,41 ata. You're normally only using one value.

 

The MAP is supposed to be fine-tuned by the means described. It's not an instruction for additional performance-tuning. Note that if the ata rating is fiddled with, the Füllungsregler will give the input mamount of fuel (set by Power Lever Position) a less than optimal amount of air (richer or leaner mixture than desired), so the possible power-outcome of a "tuned" engine might actually be less than un-tuned.

 

Here's how the engine is controlled:

 

...and...

 

Edited January 4, 2022 by Bremspropeller

[CUJO_1970]

Thanks I don’t understand the German text, unfortunately. It’s hard enough trying to learn about German aircraft, lol…appreciate our German speakers here.

 

This failure to calibrate for winter temps does not make sense at all. It basically renders the Dora obsolescent on winter maps…especially when you have aircraft getting 20-25 mph speed buffs and now you are 59 mph slower than the competition? There is no way this is correct.

 

There is zero chance something like this would not be referenced or talked about by any pilot if the are suffering like this in cold temps…it is literally a deadly handicap. How could pilots reference getting away from or catching tempests and mustangs in this condition? It’s literally impossible.

 

also, why are German flight tests in winter showing full boost 1.8ata if they couldn’t adjust the regulator? Tests in December and January in Germany show 1.8ata.

 

Something is very wrong here.

Edited January 4, 2022 by CUJO_1970

[ZachariasX]

20 minutes ago, CUJO_1970 said:

This failure to calibrate for winter temps does not make sense at all. It basically renders the Dora obsolescent on winter maps…especially when you have aircraft getting 20-25 mph speed buffs and now you are 59 mph slower than the competition? There is no way this is correct.

It is correct because what you are seeing is the very calibration for witer temperatures. Would it give the same ata as on warm days, then the engine would run overly lean mixture with al, the consequences.

 

This is also why you regulate the Jumo according to rpm and not MAP.

[Bremspropeller]

24 minutes ago, CUJO_1970 said:

Thanks I don’t understand the German text, unfortunately. It’s hard enough trying to learn about German aircraft, lol…appreciate our German speakers here.

 

This failure to calibrate for winter temps does not make sense at all. It basically renders the Dora obsolescent on winter maps…especially when you have aircraft getting 20-25 mph speed buffs and now you are 59 mph slower than the competition? There is no way this is correct.

 

There is zero chance something like this would not be referenced or talked about by any pilot if the are suffering like this in cold temps…it is literally a deadly handicap. How could pilots reference getting away from or catching tempests and mustangs in this condition? It’s literally impossible.

 

also, why are German flight tests in winter showing full boost 1.8ata if they couldn’t adjust the regulator? Tests in December and January in Germany show 1.8ata.

 

Something is very wrong here.

 

The Jumo is regulated exactly the opposite way to a common engine, where you'd regulate MAP and then give the appropriate amount of fuel.

In the Jumo, an amount of fuel is regulated, and the mass-flow regulator then tries to give that fuel the right amount of air by throttling the supercharger inlet.

 

I do agree that the performance-differences are probably too great in relative terms.

[CUJO_1970]

Just now, Bremspropeller said:

 

The Jumo is regulated exactly the opposite way to a common engine, where you'd regulate MAP and then give the appropriate amount of fuel.

In the Jumo, an amount of fuel is regulated, and the mass-flow regulator then tries to give that fuel the right amount of air by throttling the supercharger inlet.

 

I do agree that the performance-differences are probably too great in relative terms.

 

Thanks for explanation Brems! I knew I could count on you lol.

 

May I ask what your thoughts are on German test dated in winter that reference 1,8ata for the Dora-9? I will try to post those later but you guys blessed to be fluent in German have likely already seen them.

[Bremspropeller]

15 minutes ago, CUJO_1970 said:

May I ask what your thoughts are on German test dated in winter that reference 1,8ata for the Dora-9?

 

Original report:

http://www.wwiiaircraftperformance.org/fw190/Fw_190_D-9_210002_FB_Nr2.pdf

http://www.wwiiaircraftperformance.org/fw190/Fw_190_D-9_210002_FB2_level-speed.jpg

 

This test has a clearer view of the achieved airspeeds during the test above:

http://www.wwiiaircraftperformance.org/fw190/Fw_190_D-9_210002_FB_Nr3.pdf

 

Maybe @JtD and @ZachariasX could give it a shot, too.

The way I read it, is like they tried to increase power by using the 1900PS mod and they initially mis-rigged the MAP-regulator about 0.05 ata too high and - with the MW50 injection - 0.03 too low. They way it's written, it suggests, they were using a baseline of 1.52 ata at 1700PS take-off power, which would be the normal MAP for around 15°C, which is not an uncommon day-temperature for mid October (12-14 October '44). So it's not really a winter-temperature adjustment.

 

 

[JtD]

As far as Junkers was concerned, the mass flow regulator was the way for the future. Not only was it a lot simpler than a boost control unit, but it also kept the power more constant below full throttle altitude, in fact it gained a little. Which meant you had maximum power available when you typically needed it the most - during take off. And you had it all year round, as opposed to boost regulated engines, which lost power in summer (where you'd rather have more than less).

 

Boost as such hardly ever is the technically limiting factor, and I don't know what really were the limiting factors with the Jumo213. I'm assuming the engineers at Junker did put in a good amount of thought and ended up with a solution that for their particular design was the best all around compromise.

[Real_NBD]

I do wonder if there is an excessive increase in aircraft performance from cold air in the game (apart from the D9).

 

Could ram air have something to do with it?

 

Testing at Rheinland, 300m alt. The P51B-7 can reach 81" at as low as 85% RPM in winter, giving us 696 kph. Same RPM gives us 74" in autumn, maxing at 644. This setup means that we should have no throttling losses in either case.

 

81" is 273,375 kPa, 74" is 249,750 kPa. Call it a 24 kPa difference.

 

Ram air should give us 0,5*rho*v^2 of pressure increase over ambient. For 696 kph it's 25,5 kPa, for 644 kph it's 19,6 kPa, so we should gain ~6 kPa from the speed and density increase.

I used 1,225 kg/m^3 and 1,367 kg/m^3 for autumn and winter air densities. 

 

Where is the additional 18 kPa or 5.3" of manifold pressure coming from? I know this is slightly off topic but the D9 is the victim of winter as it does not gain any engine performance..

Edited January 5, 2022 by SCG_NoBigDreams

[Cpt_Siddy]

 

3 hours ago, SCG_NoBigDreams said:

 

 

Where is the additional 18 kPa or 5.3" of manifold pressure coming from? I know this is slightly off topic but the D9 is the victim of winter as it does not gain any engine performance..

Most of the aircrafts regulate manifold pressure based of thermodynamic parameters of compression heating. Where the mass flow of the gas is secondary concern to the pressure (and by proxy, the temperature) of the gas.

 

 

The D-9's engine seems to be metering the air based on optimal fuel air mixture. This seems to be a carry over from bomber system as this will increase range and engine reliability at the expense of peak performance where you are still not hitting against the pre detonation conditions. 

 

Considering that given the same amount of fuel, bore and stroke geometry, the final compression is pretty much where you are going to get your power from, more boost is always more output power. 

 

On 1/4/2022 at 11:38 AM, ZachariasX said:

You have an air mass regulator in The Jumo. Hence in cold weather you have more air mass (higehr air density in cold weather) for the same ata and the engine hence reduces ata to normalize the absolute amout amout of air to what it should be. The PN feature a table of ata attained depending on temperature. It's correct as it is in the game.

 

But thats not how it how ATA works. If you have X density of air at 0,9 ata at -10C, you dont need to correct it to +15C nominal ata value and lower the actual ATA unless your goal is to have stoichiometric air fuel mixture, which is idiotic thing to do in high performance fighter engine.

 

If you want your maximum performance from engine, mass flow is really only relevant factor for fuel metering. Your concern come from the absolute pressure because pressure is interlinked with temperature. And temperature is what restricts you in your final boost due to limitations on ignition and fuel delivery system. 

 

 

23 hours ago, VO101Kurfurst said:

 

 

It would appear to me that if the measurements in this thread are correct, our D-9s do not perform simply because they oddly use summer engine settings on winter maps..

 

 

I am also of this opinion. It seems idiotic to limit engine based on mass flow when you are nowhere near the limiting temperature rises due to under boosting. And my flying experience of D-9 at both summer and winter seems to confirms this. 

Edited January 5, 2022 by Cpt_Siddy

[ZachariasX]

38 minutes ago, Cpt_Siddy said:

But thats not how it how ATA works.

Ata doesn't "work", it is as simple pressure unit like psi and doesn't compensate for anything. 1 ata is one kilopond per square centimeter, as opposed to 1 psi being one pound per square inch. It is very much the same als inches Hg, as it also reflects a column of something, in case of ata it is 10 m water colum insead of a Mercury column. And that is all there is to it.

 

If the air is cold, you have more air at the same pressure. Although your same bucket of air is heavier at low temperatures, it still has the same pressure as a similar bucket in warmer temperature. The mass regulator compensates for just that. In the end, the engine is very agnostic of the pressure it is fed the air with. The amount of air is important however.

 

How the game deals with that is an entirely different matter.

[Cpt_Siddy]

26 minutes ago, ZachariasX said:

 

 

If the air is cold, you have more air at the same pressure. Although your same bucket of air is heavier at low temperatures, it still has the same pressure as a similar bucket in warmer temperature. 

 

 

PV=nRt

 

your pressure (P) remains equal in buckets (V) in changing t in relation to n, this is given. That said, engine cares about n only as far as how much fuel it consumes and additional energy release at the end of the compression stroke. 

 

The t component, however, is  important factor in ICE, because it will determine the final P achievable at the end of compression stroke before bad things start to happen. 

And the lions share of efficiency in ICE is dictated by its max P at the end of compression stroke. 

 

This is, to this day, have remained a common sense in any performance engine, where fuel efficiency is secondary concern.

 

If you have to optimize your engine for power, you care about n only in relation to how much energy it will release at the point of ignition. Rest is cramming as much initial P in to end of compression stroke to extract maximum efficiency from the release of that energy. 

 

The way many in here have described the Jumo engine only make sense if you are making fuel efficient prius, not a fighter. 

 

Now, pedantic among you might say that P is actually irrelevant if you can cool the air charge to absolute zero and cram a Bose-Einstein condensate in to the cylinder, and you are correct, but we live in real world where your ability to cool the charge air is limited and any real increase in n in your combustion chamber is more or less limited to the P. 

 

Edited January 5, 2022 by Cpt_Siddy

[Bremspropeller]

The way to "tune" the Jumo would be tuning the fuel-flow at given power-lever-angle. Not sure if the MBG would automaticly figure out the additional amount of air mass-flow, or if that would need additional tuning, but I'd hazard a guess it's smart enough.

 

Two things to keep in mind:

 

1. There's a switch from "lean" cruise settings to "rich" high power settings at a given PLA.

2. The Jumo was initially designed to use B4 fuel, so there's probably an upper bound in terms of temperature and pressure the engine is going to accept. And that point may not be too far off the normal working conditions.

 

FWIW the charging air temperature reduction between the 1900PS power-setting and the MW50 power-setting was measured at 35°C.

 

[ZachariasX]

1 hour ago, Cpt_Siddy said:

PV=nRt

 

your pressure (P) remains equal in buckets (V) in changing t in relation to n, this is given. That said, engine cares about n only as far as how much fuel it consumes and additional energy release at the end of the compression stroke. 

The engine cares for n. If you reduce t, you have to increase n for the whole equation to remain at an equilibrium. Now, the engine doesn't want that (it wants a given n). Hence to keep the equation in equilibrium, the engine can only reduce P, as swept volume (V) is given by the engine design. This is what you see in the manual charts. And that is how it works.

 

The engine doesn't care about how much fuel it uses, it just uses the amount required for a metered n amount of air. This way, your n air defines the power output at given rpm. It  is just consequential.

 

1 hour ago, Bremspropeller said:

2. The Jumo was initially designed to use B4 fuel, so there's probably an upper bound in terms of temperature and pressure the engine is going to accept. And that point may not be too far off the normal working conditions.

The Jumo has a very good cooling system (despite the suboptimal three valve design) and is inherently a B4 fuel engine. It mainly gained more power by upping rpm, quiet like the Sabre where 150 octane didn't give the engine anything more (sleeve valves run inherently cooler than traditional poppet valves) and the upped rpm was that scaled the power of the engine. In both cases, remarkably so.

 

The design was based on C3 not being available, hence they made do with B4 and methanol injection for top end power.

Edited January 5, 2022 by ZachariasX

[JtD]

6 hours ago, SCG_NoBigDreams said:

Testing at Rheinland, 300m alt. The P51B-7 can reach 81" at as low as 85% RPM in winter, giving us 696 kph. Same RPM gives us 74" in autumn, maxing at 644. This setup means that we should have no throttling losses in either case.

 

81" is 273,375 kPa, 74" is 249,750 kPa. Call it a 24 kPa difference.

 

Ram air should give us 0,5*rho*v^2 of pressure increase over ambient. For 696 kph it's 25,5 kPa, for 644 kph it's 19,6 kPa, so we should gain ~6 kPa from the speed and density increase.

I used 1,225 kg/m^3 and 1,367 kg/m^3 for autumn and winter air densities. 

 

If you used indicated air speed for these speeds you already have the dynamic air pressure, including ram effect, measured, directly. In your case an 8 % difference, or a relation of 80" to 74". A good enough match.

 

If you use your approach you'll also have to add the static pressure, which can be different. And then you have to scale up the relation of the uncompressed air by the compression ratio of the compressor, which is roughly 2.5 here. Or in other words, your calculated 6kPa at the intake translate to 15kPa at the manifold and the remaing 9kPa could be 3-4kPa difference in ambient pressure.

 

----below this line is supposed to be a second post

 

2 hours ago, Cpt_Siddy said:

The t component, however, is  important factor in ICE, because it will determine the final P achievable at the end of compression stroke before bad things start to happen.

 

And T in terms of intake/charge/manifold temperature is typically higher at lower altitudes, so more often than not there's room for reserve to slightly increase boost as you go up, which is exactly what the Jumo213 with its constant mass flow regulator does. In other words, constant boost pressure regulators may waste more engine potential than constant mass flow regulators do.

 

Now I wish I could provide a lot of charts, but there isn't a lot around. One chart I found is for the BMW801, and I've got a book on the RR Merlin stating the same (both from theoretical approch and measured values in a table).

 

Edited January 5, 2022 by JtD

[Cpt_Siddy]

2 hours ago, JtD said:

 

 

 

 

And T in terms of intake/charge/manifold temperature is typically higher at lower altitudes, so more often than not there's room for reserve to slightly increase boost as you go up, which is exactly what the Jumo213 with its constant mass flow regulator does. In other words, constant boost pressure regulators may waste more engine potential than constant mass flow regulators do.

 

 

 

 

 

That is only true if the boost regulator compares the manifold pressure with ambient and not some absolute ambient independent way. 

I imagine that Germans understood this as their whole variable speed supercharger  kind of depends on this to work.

 

3 hours ago, ZachariasX said:

The engine cares for n. If you reduce t, you have to increase n for the whole equation to remain at an equilibrium. Now, the engine doesn't want that (it wants a given n). Hence to keep the equation in equilibrium, the engine can only reduce P, as swept volume (V) is given by the engine design. This is what you see in the manual charts. And that is how it works.

 

 

 

I get that, but my point was that you are wasting power when you don't take advantage if additional n  you can cram in when you have lower t in the winter. If the Jumo engine is misconfigured for summer temps, it will waste power. 

 

The whole point of winter vs summer discussion in this thread is that boost regulated engines dont need to take outside temperature in to account as they more or less assume compression dependent temperature rise. While misconfigured massflow system can under preform if it is not calibrated for right temperature range. 

Edited January 5, 2022 by Cpt_Siddy

[JtD]

But all control systems measure temperature, one adjusts the throttle, the other the injected fuel. Both can be misconfigured.

 

If your engine is rated at 1200hp 1.5ata@2000m@standard conditions, would you rather have an engine that produces 1100@SL@standard conditions, 1000hp@SL@summer, 1200hp@SL@winter, 1300hp@2000m@winter (but always has 1.5ata boost) or an engine that produces 1200hp under all conditions between SL and 2000m, no matter if summer or winter (but has boost vary between 1.4 and 1.6ata)?

 

Do you think you're wasting more engine potential with an engine with constant output than with one where maximum power output varies by 30%?

[Cpt_Siddy]

1 hour ago, JtD said:

But all control systems measure temperature, one adjusts the throttle, the other the injected fuel. Both can be misconfigured.

 

If your engine is rated at 1200hp 1.5ata@2000m@standard conditions, would you rather have an engine that produces 1100@SL@standard conditions, 1000hp@SL@summer, 1200hp@SL@winter, 1300hp@2000m@winter (but always has 1.5ata boost) or an engine that produces 1200hp under all conditions between SL and 2000m, no matter if summer or winter (but has boost vary between 1.4 and 1.6ata)?

 

 

The thing is, engines should produce more power in the winter, this is thermodynamics 101. I don't see no reason why D-9 should not do this as well. 

Mass flow metered engine is no magic, it operates exactly under same constrains as any other engine. 

[CountZero]

ah they just need to make D9 faster 30-40kmh on winter maps and no one would complain, now its a joke when you see some one who take d9 n winter maps, its like flying e7 on kuban front lol

[CUJO_1970]

12 hours ago, CountZero said:

ah they just need to make D9 faster 30-40kmh on winter maps and no one would complain, now its a joke when you see some one who take d9 n winter maps, its like flying e7 on kuban front lol

 

I actually figured this out flight testing for top speeds - I could not believe it when the fully boosted D-9 was nearly 96 kph slower than 81" Mustang on the deck.

 

This how the sim is currently interpreting things on winter maps.

[JtD]

13 hours ago, Cpt_Siddy said:

The thing is, engines should produce more power in the winter, this is thermodynamics 101.

 

I disagree. The thing is that supercharged aeroengines are more complex than thermodynamics 101. And unless you dig into the particulars of a specific engine, you'll never know how this guideline may be applied to that specific engine.

[ZachariasX]

7 hours ago, CUJO_1970 said:

This how the sim is currently interpreting things on winter maps.

I did some quick testing as well and on rated settings I get the 81'' Mustang 67 km/h faster. You can game the game and get another 15 km/h. So I guess we are talnking about the same. (My speeds are all IAS at 400 m)

 

What is of note is that in cold weather/winter, the Mustang can hold 81'' at some 2750 rpm on the deck. The D9 reduces boost from ~1.81 ata (summer) to ~1.68 ata in winter and in consequence gets about the same speed, 593 summer to 606 km/h winter.

 

If I look at the considerable power creep of the Mustang at 61'' (summer 571, autumn 593, winter 611) and at 81'' (s 612, a 631, w 673), then I am tempted to say that the increased airframe drag of the denser atmosphere in winter is not propperly modelled (or in an odd way) in the FM.

 

I say this because (if we care to look at things outside the scope of this game) for instance the Spit IX (that cannot get a speed boost by lowering revs) gets an IAS of 521 km/h (@400 m) in stock config and 542 km/h in winter and 521 and 542 in summer. It is almost as if a ~7% power increase from dense (winter) over less dense (summer) gets almost translated linearly into airspeed, whereas the 20% power increase from going from 130 to 150 octane is rightfully factored 1.2^(1/3) = 1.06 -> 6% addition.

 

Adding 20% power by going from 130 to 150 oct gives me 21 km/ in summer, some 30% short of what is expected (I would expect going from 521 to 552, but I get 542) but ok.

Adding only 7% power by going from summer to winter in stock config, then I get 47 km/h added speed! In fact, I would expect maybe a 2% speed increase! (521 summer, 531 winter, but winter in the sim is 568!)

 

In this sim, denser air translates exponentially in speed gain over what raising boost does. I get what corresponds to a net ~25% more engine power by going from summer while the actual gross power addition is a mere ~7%.

 

But the bottom line is that it makes me think what we are seeing is a general trait of the simulation design (for reasons), and less a particularity of a respective aircraft. It all seems to come down to the denser air that gives more power gives not more drag than thinner air.

 

But as far as the game is concerned, I don't really think it is a huge problem for themoment. We'll always game the game and what I wrote above I consider hypothetical unless confirmed by others.

 

Edited January 6, 2022 by ZachariasX

[Cpt_Siddy]

5 hours ago, JtD said:

 

I disagree. The thing is that supercharged aeroengines are more complex than thermodynamics 101. And unless you dig into the particulars of a specific engine, you'll never know how this guideline may be applied to that specific engine.

 

 

Duuuuude, did you just say that the law of thermodynamics can be broken. Are you perhaps one of those "overunity engine" people who think that jet engines can produce free power? 

[HR_Zunzun]

9 hours ago, Cpt_Siddy said:

 

 

Duuuuude, did you just say that the law of thermodynamics can be broken. Are you perhaps one of those "overunity engine" people who think that jet engines can produce free power? 

 

He said that is more complex thermodinamix, not that is against it. Please leave the "ad hominum" for other forums....

[ZachariasX]

I tried to be a bit more systematic now about the power differences between winter and summer maps. I see this as mainly an academic example, just because... well, you know I like this kind of activity. It is by no means something I see as an immediate basis to pester Jason with or to throw it in AnP's face making him spit his breakfast ceral across the table. If there is something at odds, it certainly would require a more thorough description that I can offer for now.

 

TL;DR: Dense winter air gives some 8% more airspeed that would reflect some 25% more engine power. Of this 8% more airspeed, ~2% increase in airspeed is probably due to the leaning of the mixture, as all carbs do not compensate fuel flow for air density, they only seem to do so for air pressure. As I consider it not likely that the sim makes such a blunder as inflating the engine power, it must be because of other factors, MAYBE one being that drag in denser air is not suitably modelled.

 

Winter/summer speeds [km/h]
	Summer	Winter	Speed increase	Nominal power increase required for observed winter speed	Theoretical winter speed  (expected, scaled by 1.02)		comment
Yak-1 ser.69	496	544	48 (10%)	32%	507		30% rads
Yak-1b s.127	512	559	47 (9%)	30%	524		30% rads
Bf-109K4	596	641	45 (8%)	24%	610		Auto rads!
Bf-110G2	488	523	35 (7%)	23%	499		37% rads
Fw190A5	552	593	41 (7%)	24%	565		rads closed
P-51B 61''	571	611	40 (7%)	23%	584		Auto rads!
Spit9 +18lbs	521	568	47 (9%)	30%	533		Auto rads!
N28	192	211	19 (10%)	33%	196
Camel	181	200	19 (10%)	35%	185		Mixture s: 83%, w:90% -> 8% difference in air,available!

 

It is Rheinland winter/summer, 400 m altitude and 340 L fuel where possible.

One thing we know about the air density at equal altitude is, that in winter, the air is ~7% denser. This is what the Fw190D9's mass flow carburettor compensates for. Hence it drops ata some 7% in winter to make for a similar air mass at 100% power, producing the same amount of power. Funny enough, these 7% also manifest directly in the mixture setting of the Camel (~8%). It has a sensitive mixture curve and a nice dial for exact reading. (Not so the N.28). It is clear that the way those pipe carbs are modelled, air density directly scales with required mixture ratio, e.g. at half the air density, you retard the mixture by 50%, at 7% more air density, you advance the throttle by ~8%. (That is factually correct enough for my purposes.)

Thus, in all aircraft with a mixture regulated according to only MAP, in principle we have 7% more more power available then at the same MAP in winter vs. summer. This however (as detailed above in the thread) should only make for 1.07^(1/3)=1.02, a 2% increase in airspeed. What we see is an increase in airspeed that corresponds to a 10 to 15 fold the power difference bewteen winter and summer.

 

Now, I do not believe that the simulation does this kind of blunder in terms of engine power output, as overall the airplanes perform remarkably well according to specs. As said, the only factor I see in this is the drag of the airplane when different Reynolds numbers are applied, and thus something that is "maps" related and not aircraft related.

 

When I look at the equation behind all that, this one:

 

D=1/2ρV^2*S*C_D (D:drag, ρ: density, S=surface area, C_D:drag coefficient)

 

Then I have a drag function that in effect should scale to the cube, same as the power requirement. For practical putposes, I consider the multiple of drag coefficient and S to be similar in alike aircraft (P-51, Bf-109, Fw-190), as the aircraft are somewhat similarly fast at same power (at least exact enough for my purposes), hence I have only ρ that scales my cube funtion of the airspeed. And that speed increase is almost the same as the power increase gained through denser air, between 7% - 10%. If I held ρ constant, would that be enough to impair the drag? Or would there be a more suitable way to compute drag? I have to leave this open for now. But It is the only dirction in which I can speculate.

 

It is of note that both the biplanes are on the top end of the speed gain. In both cases I can perfect the mixture ratio manually. In case of the Yak... that is one sleek aircraft. I hope I am being reasonable here, maybe @Holtzauge can lend a wooden eye?

 

As far as the mixture mechanics go as implemented in the game, I can speculate that the fuel flow is entirely directed by the MAP. This has consequences. It means that the aircraft only run on specified mixture ratio in reference atmosphere, and I would guess these are aour autumn maps. In summer, the aircraft run too rich, in winter, they run too lean.

 

I took the Yak-1 s.69 to test this:

 

In the world of indestructible engines that we inhabit here, the power output is best at a "best power" air-fuel mixure ratio. I can find out by cranking the mixture lever, by leaning or by enriching, if I am rich or lean of peak. At the peak, the aircraft is fastest. (In the real world going there at full power would blow the engine clear off the airframe, but we indestructibles can do it.)

 

Now the Yak: In summer, it is a tad faster at 50% mixture (499) than at full rich mixture (496), meaning that in summer with a less dense athmosphere, the mixture is calibrated such that full rich is actually just marginally on the rich side, while 50% mixture puts it at the peak (or very near of it). Knowing what I mused about above, then I would expect that any increase in air density (such as winter) would lean the mixture, as I get more air per fuel. As said, the fuel flow metering in the engine is assumed agnostic to the repective weather. What I get when I take the Yak up in winter, is that the aircraft is now considerably faster at 100% mixture (544) then at 50% mixture (531).

 

So we have two things that add speed to the aircraft, one is the increase in air mass that increases power, the other is the leaning of the mixture that in rich mixture settings lead to higher power output (if you don't care for predetonation etc.). I am somewhat certain that this leaning in mixture is the cause for the Fw190D9 getting a slight speed bump (593 -> 606) in winter, that is some 2%. Or, about 1/4 of the total speed gain in aircraft that have no air mass regulated carb. 2% in speed gain would  be about 7% in added engine power. Depending on the power curve used, this could actually make somewhat sense, but is definitely on the high side. Especially sice C3 fuel has very, very good rich-rich burn qualities.

 

Having diagnosed what might be an issue in competitive multiplayer, I'm not sure (if that all were applicable and true) what to make of it. First of all, I have no detailed info on how for instance the Bendix carb deals with changing density altitude, but I would expect the mixture effect described above. (In fact, controlling the mixture was a reason why in the P-51B, the mixture lever was restricted to "AUTO LEAN" and I hope they fix that.) But fixing the rest? the more Ilook at it, the less I'd want to touch the innards of the sim to mess with all of that and make all the planes fly in funny ways. And the rework all of them. But who knows.

 

Edited January 7, 2022 by ZachariasX

[CUJO_1970]

Thanks, wow!

 

One quick reminder: D-9 is running B4+MW50.

[ZachariasX]

18 minutes ago, CUJO_1970 said:

Thanks, wow!

 

One quick reminder: D-9 is running B4+MW50.

Good point. Yet both on Methanol or "just B4", the mixture will be richer in summer than in winter. It is actually a downside of the MW50 vs "just water" injection because it runs the engine richer, hence it is less power efficient. (Why MW50 is still a good idea for the specific Jumo213 is a different issue.) This is why I suspect that the mentioned  leaning effect of winter still applies for the 190D9. With the P-47 it might be different, as the water injection might set it back to best power. Then again, if it is calibrated for autumn, then this would make "best power" is too lean in winter and too rich in summer.

[JtD]

Zach, thanks for all the testing and the analysis. ?

 

I do wonder - are all your speeds in IAS? If so, keep in mind that power isn't IAS³, it is IAS²*TAS. The drag does go up square with IAS, but power is drag times actual distance per time. And without knowing the particulars of the map, actual speed increase is going to be quite a bit smaller than IAS indicates (no pun intended). If your 7% delta density are right, it will eat up about half the speed increases (in TAS) you listed and power requirements will be 4-5% lower than what you gave. Not an overall game changer, but somewhat less shocking.

 

Also for consideration would be prop efficiencies, which as we know are quite low at high revs high speed. Giving the engine more power will give the prop some more to bite - higher pitch - which can increase prop efficiency noticeably. I've been wondering how much this could and would account for in practice.

 

And then I think that most engine will give an overproportional power increase, because some losses like cooling and oil pump are dependent on revs only and don't increase when gross power output goes up (unlike supercharger). Same is true for some of the friction inside the engine. Small effects, probably, not enough to close the gap, but maybe makes it a little smaller.

 

I wanted to test some things myself today, but I got distracted. Maybe later, or tomorrow.

[ZachariasX]

30 minutes ago, JtD said:

I do wonder - are all your speeds in IAS?

Yes. For sake of simplicity, I was just interested in the relative quantities of the airspeeds. My rationale was that the mistake I am making with this is well below the mistakes introduced by my assumptions. At same altitude and weather, IAS should scale acceptably linear. But I see you point:

 

Here in more detail (using this), the errors I am making when comparing e.g. 500 and 550 km/h respectively and I am just interested in the factor of the speed differential between winter and summer:

 

500 km/h IAS -> 511 km/h TAS @ 15°C, @500 m

550 km/h IAS -> 562 km/h TAS

+10%               -> +9.98%   relative speed differential

 

500 km/h IAS -> 493 km/h TAS @ -5°C, @500 m

550 km/h IAS -> 542 km/h TAS

+10%               -> +9.93%   relative speed differential

 

500 km/h IAS -> 520 km/h TAS @ 25°C, @500 m

550 km/h IAS -> 572 km/h TAS

+10%               -> +10%   relative speed differential

 

And the error I am getting is usually below the 1/100th. I am sure I made other, much larger errors in my reasoning that that. But you are certainly right. I just didn't care to be too exact at this point. It was a shot from the hip.

 

It was my intention to visualize a trend. This is why I discarded anything like prop efficiencies, as I simply treat it equal in summer as in winter. I have never seen anyone put on a "winter prop" as they are supposed to put on winter tyres in my part of the world.

 

Still, I am curious what is really beneath that stone. All I have made just gave me indication to expect something under that stone.

Edited January 7, 2022 by ZachariasX

[JtD]

Took a Yak-9 for a spin to play a bit with airscrew settings in order to investigate my above theory. If it exists in game, it's lower than my accuracy of measurement, so forget about it. As for the Yak-9 in the tested conditions

 

760mm HG; Summer: 25°C, Winter: -15°C

Summer IAS: 498km/h, Winter IAS: 536km/h

Summer TAS: 506km/h, Winter TAS: 507km/h.

 

Funny, first one I picked matches my gut feeling that TAS shouldn't change, with extra power and extra drag cancelling each other out. It also means about 16% extra power in about 16% denser air, which again makes sense.

 

You can see that I do testing for a living because the first random sample I picked is the one where everything adds up. ?

[ZachariasX]

14 hours ago, JtD said:

You can see that I do testing for a living because the first random sample I picked is the one where everything adds up. ?

You. Are. Right. I guessed wrong. For the simple fact that I took a wrong temp offset initially to assess whether it was worth the bother to beat all the data through the calculator. You took -15°C and I changed the spreadsheet accordingly:

 

Winter/summer speeds [km/h] TAS
	Summer	Winter	Speed increase %	Speed increase	Nominal Power increase required	Theoretical speed increase (expected)		comment
Yak-1 ser.69	516	530	3%	14	8%	528		30% rads
Yak-1b s.127	532	545	2%	13	8%	544		30% rads
Bf-109K4	620	624	1%	4	2%	634		Auto rads!
Bf-110G2	508	518	2%	10	6%	520		37% rads
Fw190A5	574	578	1%	4	2%	587		rads closed
P-51B 61''	594	610	3%	16	8%	608		Auto rads!
Spit9 +18lbs	542	554	2%	12	7%	554		Auto rads!
N28	200	206	3%	6	9%	205
Camel	189	195	3%	6	10%	193		Mixture s: 83%, w:90% -> 7% difference in air,available!

 

It does fit. I must admit, that makes me happy. Because it directly shows the effect of the mixture ratio on cold air. It is about as exact as it could get for that.

 

It is notable how well planes like the Camel perform when you can adjust the mixture to optimal ration in winter. As for the rest, power gain competes with deviation from ideal mixture minus any kind of drag value that keeps the plane in line under common condition.

 

That makes me much happier than the finding above. And that was the main reason why I wouldn't want to belive it unless reproduced. But that's cool. Actually, very cool. It actually makes the simulation very, very good as such. I wasn't prepared to find the season having an effect on the mixture ratio initially, as as it should have. I was looking for an error but I found someone made their homework. Kudos!

 

-----EDIT: this chart is wrong, see below----------------

Let's look at the Fw190D9:

Summer

Winter

Speed  increase %

Speed increase

Nominal Power increase required

Theoretical speed increase (expected)

190D9

578

591

2%

13

7%

591

 

There, we get a net 7% increase in power output by leaning the mixture. Still ok'ish I think for our purposes.

--------------------

 

In practise, I would have my doubts if we would see such variations in mixture ratio, as ground crews probably would calibrate the mixture automatation regularly. If that should be represented in the game and how is an entirely different question though.

 

 

But that lets me sleep much better. Thanks man.

 

Edited January 8, 2022 by ZachariasX

[Roland_HUNter]

Hmmm and what about the over boosted Hurricane?
 

 

What did I make differently?

Spit IX 18 Boost Summer:

 

Spit IX 18 Boost Winter:

 

Spit IX 24 Boost Winter:

[ZachariasX]

So... Knowing a bit more, I redid the flights and corrected the IAS using the calculator for Atmospheric Temp:297.9 K |24.75 °C in summer and Atmospheric Temp:258.4 K |-14.75 °C in winter.

 

Lo and Behold it fits perfectly. Extremely perfectly so.

 

I was an oaf taking 0°C for winter as ballpark for corrections. While that makes sense on a Rhineland map, it should certainly not be the default for the Russian Steppe in winter.

 

Winter/summer speeds [km/h], TAS
	Summer	Winter	Speed increase %
Yak-1 ser.69	519	529	2%
Yak-1b s.127	535	544	2%
Bf-109K4	623	624	0%
Bf-110G2	510	509	0%
Fw190A5	577	577	0%
P-51B 61''	597	595	0%
Spit9 +18lbs	545	553	1%
N28	201	205	2%
Camel	190	195	3%
190D9	620	591	-5%
Hurri +14	471	471	0%

 

But that's really cool. When I can help it with the mixture, I can get the planes a tad faster in winter. But marginally so. It really depends on how the mixture ratio is calibrated. The Fw190D9 just pays the price for reatining rated performance in denser athmosphere. And it pays the price in a way that fits an remarkably exact way the discarded power gain by reducing ata while benefitting from leaner mixture.

 

8 hours ago, Roland_HUNter said:

What did I make differently?

Actually nothing, I get identical speeds as you do on the HUD. Them converted to TAS gets the values in the chart. That changes the picture.

 

8 hours ago, Roland_HUNter said:

Hmmm and what about the over boosted Hurricane?

It is a funny plane. In winter, you press the throttle into +14 (when the mod is installed) without pressing the "boost" button. Past +10, boost gauge jumps to +14. In autumn and summer, the boost gauge stops at +12 with throtte alone. Switching to SC2 at 500 m produces +18 boost in summer and +23 boost in winter. Same as in the A20, you actually get some "timer-time" instead of a KABOOM you'd get in the real aircraft. Hence, despite the higher power bleed of running the SC in high gear, you get the plane to  ~480 km/h TAS in Summer (+18) and 498 km/h in winter (+23). It goes without saying that these boost regimes are outside of the reasonable range.

 

Winter summer... things are as they should be, exactly. so.

 

Edited January 8, 2022 by ZachariasX

[JtD]

On 1/6/2022 at 11:33 PM, HR_Zunzun said:

He said that is more complex thermodinamix, not that is against it.

 

After the excursion into effects of cold air on performance, I'd like to get back to the particulars of the Jumo213.

 

One is the mass flow regulation we have already mentioned, and the next one would be an intake vane type throttle. As you may know, a throttle typically is just a plate in a pipe, when it's vertical, the pipe is closed, when it's horizontal, the pipe is open and all positions in between regulate the flow rate. Thermodynamically, this is very inefficient, because you basically create turbulence in order to reduce flow, and this is just wasted energy (lower dynamic pressure without lowered temperatures).

With the Jumo213, the Junkers engineers used a "Dralldrossel" - best translation seems to be a variable pitch intake vane throttle. This trottle is installed at the supercharger, and directs the air that hits the supercharger.

For full supercharging performance, the vanes are basically "fully open", meaning the intake airflow hits the supercharger wheel (impeller) perpendicularly. This requires the most power for the supercharger, which has to accellerate the air the most, and therefore produces the highest compression ratios.

If you need less supercharging performance, the vanes are angled and direct the air into the rotation of the supercharger, which then requires less power, and also produces less compression.

The nice thing about this type of throttle is that thermodynamically it is efficient, because directing the airflow redudes the power requirements for the supercharger. Or in other words, you don't create turbulence, but directly reduce pressure and air temperature. This is another reason for the constant power output over altitude - less wasted work in the supercharger.

On the downside this throttle is much more complex than the ordinary types. It is comparable to the hydraulic clutch of the DB60X series, though, and compared to that, not much more complex. Where the DB60X takes a hydraulic clutch to reduce the speed of the impeller, thus reduce air speed difference and thus compression, Jumo vectores the intake air speed and thereby reduces air speed difference and thus compression.

 

Picture:

You have the impeller on the left of both figures, it turns in the way the "Drehrichtung" arrow is pointed. On the right you have the variable pitch intake vanes.

Left full throttle/highest compression - Right reduced throttle/reduced compression/less supercharger waste

[Bremspropeller]

It's very similar and basicly the same principle (massflow control) as the Variable Inlet Guide Vane on the J79 jet engine.

 

 

Edited January 8, 2022 by Bremspropeller