Analysis of real life 109E performance for CloD

[Timberwolf]

LoL Everyones talking about top speeds, and Alt speeds ... Most people is using a test from other people and countries. That would fuel up a 109 take it up and see.. One of the most major factors left out was the use of Synthetic fuel that the Germans used. And American testing on German Aircraft. The American fuel which had a higher octane And had a great affect on speeds even more with supercharged planes Yet no one has talked about it just lines from other websites ..and those don't have which fuel was used on the test

[TomcatViP]

Originally Posted by Timberwolf

LoL everyones talking about top speeds and Alt speeds ... Most people is test from other people and countries that would fuel up a 109 take it up and see.. One of the most major factor the use of Synthetic fuel that the germans used To American fuel which had a higher octaneand had a great affect on speeds even more with supercharged planes Yet no one has talked about it just lines from other websites ..and those don't have which fuel was used on the test

How right you are.

But perso I became allergic to all fuel's grade discussion with month arguing my best (with honesty) on the recently closed thread abt 100oct fuel

[Kurfürst]

First, thank you for putting time in an obviously well thought out and coherent posts. Its a rare thing to see something like that on this board.

And now the barrage

Originally Posted by camber

The disagreement can be summarised as:

Position 1 (Fast guaranteed 109s): Messerchmitt guaranteed 500kmh on the deck and 572kmh max TAS for the 109E. The E1 prototype duplicates these figures theoretically once the actual flight tests are corrected for a measured engine deficiency of around 50PS.

Position 2 (Slow low 109s): Actual flight tests of production 109Es, both by Messerchmitt and by foreign governments on captured 109Es, indicate lower performance (primarily at low alt). Speeds are around 475kmh IAS/TAS on the deck and 560 kmh max TAS.

I support position 2, for reasons as follows.

Here is all the test data summarised. I am appending the data links at the end of the post. Most of this data is on "competing" websites (http://kurfurst.org and http://www.wwiiaircraftperformance.org). Most of it is on both, and it is exactly the same data..... however the interpretation is different. I will give kmh and mph figures, plus boost psi and ata.

Postion 1 (Fast guaranteed 109s) requires explanations to why all the 109E actual flight tests show lower performance than "guaranteed".

Similarly Position two (Slow 109s based on trials with uncorrected/unknown conditions) requires explanations as to why ignore the most detailed actual flight tests which the manufacturer choose to guarantee.

A good start is assigning lower significance to tests on captured aircraft, as these may not operating at maxiumum efficiency (for various reasons). However this makes no real difference to the data set, as the flight tests (besides the E1 prototype) are all similar. The only exception is the Russian captured aircraft which appears to be a (low performance) outlier at low altitude, and can perhaps be discarded from consideration.

For test 1 (E1 prototype) the converted (theoretical) data appears reasonable. The engine power deficiency was measured on the test stand and used to provide a theoretical correction to actual flight data.

The Russian results are probably easy to understand if anyone looks at the climb curves. They show an abnormally steep fall of climb rate near SL, which is a clear indication something was wrong with near-SL power outputs of the specific plane.

For tests 2 and 3, the suggested reason for low performance is that the results are not corrected for nominal DB601A output, as was done for the E1 prototype (http://kurfurst.org/Performance_test...MP16feb39.html). However this begs the question, why were multiple production 109Es failing in tests to reach their guaranteed performance because of underperforming DB601s? This seems an unlikely scenario, and rather a bone of contention between Messerchmitt and Daimler Benz!

While Test 1 was made at 1,35ata and higher output, with known (measured) engine outputs Tests 2 and 3 were made with a 1,3ata - test 2/3 were flown at lower boost. This explains some of the difference but not all.

The important difference between Test 1 and Test 2/3 however is that while Test 1 is a performance test, and wanted to obtain accurate and absolute performance figures for the whole altitude spectrum, Test 2/3 were never intended to do the same.

Test 2/3 were about finding the relative difference in performance in various aerodynamic conditions, presumably for seeking out the viability for future improvements (guns installed not installed etc.). All they wanted to gain is relative speed difference on the same plane to get an idea how much drag penalty these items induce. They were never meant to be representative performance flight trials.

For test 5, the explanation for underperformance relates to the function of the DB601 supercharger.

Also for Test 2 and 3. One part that is not mentioned in the table that all flight trials (none of which we know the supercharger setup) show very close agreement, within a few km/h to Test 1's performance with the high altitude supercharger speed in operation.

The hydraulic supercharger coupling of the 109 was intensely interesting to the British and there is a couple of very good, detailed explanations of it's operation in the period press:

www.wwiiaircraftperformance.org/me109/DB-601A-Flight-7Nov40.pdf
http://www.flightglobal.com/pdfarchi...0-%200516.html

Both the DB601 and Merlin had the same problem...a supercharger powerful enough to allow high altitude performance would give so much overboost (unthrottled) at sea level to cause predetonation and engine damage. The Brits developed a boost controller that automatically throttled the supercharger inlet to avoid overboost. The Germans worked out an elegant hydraulic supercharger coupling that spun the supercharger impeller slower at sea level and faster at height.

There seem to be a misunderstanding. You seem to assume that the British used a boost controller, the Germans did not, and relied solely on the barometrically controlled hydraulic supercharger coupling and - did not have a throttle on the engine at all!

This is however a false assumption, as apart from controlling the supercharger, the DB engines had automatic boost controller as the RR engines did. In fact those engines that did not and the throttle was controlled manually were the exception.

The hydraulic coupling provided the apprx. amount of boost needed, usually a bit more just to make sure, but fine adjustment was made by the throttle on the DB, which limited boost to the nominal pressures.

The DB601 hydraulic coupling had two oil pumps which pumped oil into the engine to impeller junction. One pump operated continuously but insufficiently for efficient coupling, allowing slippage of about 30%. As altitude rose, a barometric control increased the action of the secondary oil pump, reducing slippage to the minimum possible, about 2%. The corresponding impeller ratios were 7.2:1 at sea level rising to 10.2:1 at FTH.

The continuously varied supercharger output accounts for the curved nature of the 109 speed vs alt curves (e.g test 4) Unfortunately some tests seem to omit data points in favor of straight lines, making it hard to assess whether the curves are characteristic of the hydraulic coupling or not (e.g test 5

There seem to be an assumption that the supercharger output on the DB engine was continously varied. However it is not true to the 1st s/c speed, ie. the Bodenlader speed or MS gear in British terminology, at which only the first, constant supply oil pump supplied oil to the hydraulic clutch, and was therefore a fixed speed unit.

The second variable oil supply pump was only engaged at around 1.5-2 km altitude, above which the s/c output was contiously varied indeed - but not before that.

This characteristic is clearly visible on DB power curves (straight power line between 0-2km).

The (fast guaranteed 109s) explanation for the test 5 underperformance is based on the hypothesis that the hydraulic supercharger setup can be altered in tests to act a two speed supercharger, and the Swiss were testing in high altitude configuration at all altitudes including at ground level. As a consequence the high altitude speed is about right but the sea level speed is low (http://kurfurst.org/Performance_test...109E_J347.html)

Correct. It should be noted that this explanation gives a rational explanation for the great difference between various measured values. We are not talking about 10-20 km/h but 40 km/h speed difference, which is rather tricky to explain with individual variation between planes.

Also it not only true for the Swiss testing (conditions completely unknown, but the lack of any curvature and good agreement between V15a Hoehenlader results also suggest this), but also to the two other German 109E flight tests.

This does not seem to be possible in light of the operation of the hydraulic coupling as explained in the references above.

You are simply wrong.. there is no technical difficulty in doing so.

Manual operation of the variable oil pump (if possible) would have the following effects:

* If the variable oil pump is set to off, impeller slippage remains at 30%. Aircraft will perform normally at sea level, but boost will decay rapidly with altitude with consequent underperformance.

Yet this is not what is happening in real life. As a matter of fact the variable fuel pump (2nd speed) is completely off even in normal operation up to around 2 km, ie. impellor slippage is constantly at around 30% up to that altitude. Yet the boost does NOT fall, its constant 1.3ata.

In short, the DB 60x series supercharger's first speed is ALWAYS fixed ratio, yet as opposed what you speculate, that the engine boost would decrease above sea level, it doesn't. On the contrary, power curves show it increases, peaking out at about 2000m, where the variable oil pump is engaged by barometric control.

The reason is simple, when the slippage is at 30% (first s/c or 'Bodenlader') the supercharger compresses far more air (at around 1.6-1.7ata) than the necessary (1.3-1.45 ata), but the boost regulator only feds the engine with the necessary amount. This also means that at near SL the air is much more compressed, than a bit above, and as a result heats up more, decreasing output a bit.

The actual pressures can be seen on this graph on a 109G. See the difference between Ladedruck (boost) and Geblaesedruck (which is the pressure in the supercharger itself): http://kurfurst.org/Performance_test...les/blatt6.jpg

* If the variable oil pump is set to maximum, impeller slippage is held at 2%. The engine will be highly overboosted at sea level and will be unable to perform at all!

Nope, the engine won't be overboosted. The supercharger with minimum slip will develop a lot-lot more pressure at low levels, heat up the air a lot more in the process, decrease output greatly as a result.

What will simply happen is that the supercharger will develop at low altitudes a lot more pressure than needed, and before it enters the engine it will be throttled down to the nominal levels.

The whole 'not possible because of the use of a hydraulic coupling' is nonsense. The only thing the hydraulic coupling change is the gear ratio of the supercharger, between 7.2:1 and 10.2:1, just like a mechanical gear in a more simple two-speed supercharger would do, the only difference is that the hydraulic is capable of any intermediate interval, ie. any ratio between 7.2:1 and 10.2:1, the mechanical linkage is not.

This also contradicts the idea that the test 1 (E1 prototype) performance is from a DB601 with hydraulic supercharger coupling manually manipulated to act as a two speed system. The data explicitly refers to a two speed system, so in the prototype it appears that the supercharger is an earlier direct and not hydraulic type.

There is no such thing stated in the E1 prototype. The report Blatt 6 curve shows the supercharger operation in Bodenlader and Hoehenlader, which is always used by Germans to describe the low altitude (MS) and high altitude (FS) supercharger speeds/gears.

Furthermore Rechlin's much later trials with G-6 show exactly the same operation of high/low supercharger gears. Unless you want to tell me that hydraulic coupling was also missing on the DB 605A/109G-6...

http://kurfurst.org/Performance_test...G-6_DB605A.jpg

The fact that the E1 prototype achieved higher sea level performance but similar maximum performance suggests that the hydraulic coupling had an efficiency penalty at low altitude as compared to a direct coupling. But the advantage of the hydraulic supercharger coupling in reducing pilot load in combat was probably considered an overriding factor.

The whole of this thesis assumes that the E-1 prototype had more power at low altitude because it supposedly would have used a direct coupling (something that is nowhere stated in the report) which was supposedly] more efficient at low altitudes, and in the end gave more power to the propeller.

Problem is that all of the above runs aground since the report clearly states the aircraft doing 493 km/h with 951 PS power, at 1.31ata with the supposedly more powerful engine.*

For comparison the DB 601A-1 static output at SL was 990 PS at 1.3ata, and that of the 601Aa at 1.35ata was 1045 PS.

In short the tested aircraft's propeller had exactly the same power as serial production machines, regardless of what happened in the engine before that.

*The results were later corrected to 996 PS, the nominal output in flight.



I am not aware of other explanations of underperformance in the other actual 109E flight tests.

As to the Messerchmitt guaranteed performance, I am somewhat intrigued as it seems serial production 109s with hydraulic supercharger couplings did not meet it (at least at sea level).

1, All Bf 109s with DB engines had hydraulic supercharger couplings

2, The guarantee was for 1.35ata performance, +/- 5% for 500 km/h. That means that anything between 475 km/h and 525/h is OK for acceptance, for an Emil with DB 601Aa and operating at 1.35ata.

Compared to that, it would the French got around 480 km/h at 1.3ata from a crashed aircraft and the Germans iirc 474 km/h for another test machine, again at 1.3ata. In both cases the nominal output of the engines was at least 55 PS less than that of the Aa engine.

I would say serial production machines stack up really well to the specs.

So based on the above a reasonable (serial 109E with hydraulic supercharger) performance is based on the actual flight tests (minus the prototype):

475kmh IAS/TAS at SL 1.35ata 2400rpm
560 kmh TAS at 5000m 1.35ata 2400rpm top speed

And this is based on which test or official data...?

1.35ata is 5 min combat limit. It is rather open to debate what effects sustained 1.35ata should have in the sim, what performance 1.45ata should give and whether it should work at all at height (historically it is a takeoff only boost after all).

Uhm, so by what connection how does 1.35 ata performance come into the picture? You basically show threww German tests (at uncorrected power) at 1.3ata, which only show performance for SL, and suddenly 1.3ata performance is now the same as 1.35ata performance?

How are uncorrected 1.3ata tests are in connection with 1.35ata performance?

[camber]

Dear Kurfurst and others

Thank you for the response. That was exactly the critique I was looking for.

Originally Posted by Kurfürst

While Test 1 was made at 1,35ata and higher output, with known (measured) engine outputs Tests 2 and 3 were made with a 1,3ata - test 2/3 were flown at lower boost. This explains some of the difference but not all.

The important difference between Test 1 and Test 2/3 however is that while Test 1 is a performance test, and wanted to obtain accurate and absolute performance figures for the whole altitude spectrum, Test 2/3 were never intended to do the same.

Test 2/3 were about finding the relative difference in performance in various aerodynamic conditions, presumably for seeking out the viability for future improvements (guns installed not installed etc.). All they wanted to gain is relative speed difference on the same plane to get an idea how much drag penalty these items induce. They were never meant to be representative performance flight trials.

Also for Test 2 and 3. One part that is not mentioned in the table that all flight trials (none of which we know the supercharger setup) show very close agreement, within a few km/h to Test 1's performance with the high altitude supercharger speed in operation.?

I agree with you that Tests 2 and 3 are comparative (interestingly the changes made never affected performance). I consider them useful for consideration as they are among a limited range of official Messerschmitt 109E flight tests I am aware of, although they are certainly non-optimal (comparative, no height curve, sea level speed probably extrapolated etc.). More German speed tests would be be very useful, a few more and we could remove these ones if desired.

I would say however that provided the rated conditions are known, the "control" speed (i.e without configuration changes, not that this changed anything) should be considered in the data set in light of stated boost (1.30ata).

But I can't agree that (lack of) engine power correction is a good explanation of low altitude underperformance in multiple tests (i.e. other than test 1). It suggests that 109E underperformance due to DB601 defficiency is typical and might be a factor in actual production aircraft (unlikely). In this case the 109s underperform but it is a DB rather than Me problem.

Originally Posted by Kurfürst

There seem to be a misunderstanding. You seem to assume that the British used a boost controller, the Germans did not, and relied solely on the barometrically controlled hydraulic supercharger coupling and - did not have a throttle on the engine at all!

This is however a false assumption, as apart from controlling the supercharger, the DB engines had automatic boost controller as the RR engines did. In fact those engines that did not and the throttle was controlled manually were the exception.

The hydraulic coupling provided the apprx. amount of boost needed, usually a bit more just to make sure, but fine adjustment was made by the throttle on the DB, which limited boost to the nominal pressures.

Yes, I did make that (false) assumption. In my defense, I will say that I was seduced late at night by the romance of a boost control system engineered by hydraulic slippage of the supercharger by German engineering geniuses. I will look further into the total system but point conceded as the system you describe makes more sense.

This certainly knocks over one of the legs of my argument (but not all ). With separate boost control and provided ability to alter the variable oil pump operation, the coupling could indeed act similarly to a two geared system. However, it is still a hypothesis that the Swiss 109 was tested in high altitude mode at all altitudes (decreasing low alt performance). An alternative hypothesis is that the speed/height curve has been simplified as a straight line, similarly to the original 109E specification. Without actual data points we can't tell.

This unfortunately gets us back to the original problem of a somewhat large range of possible low altitude 109 performance, with reasonable grounds (in my opinion) to argue both ends of the spectrum:

* A guaranteed technical specification
* A prototype that meets the spec provided engine power is corrected
* Multiple flight tests (with boost variability) that are specification "passes" for max speed but around the pass/fail level at sea level speed. Some plausible but unproveable hypotheses are presented for underperformance in these real tests.

Originally Posted by Kurfürst

The guarantee was for 1.35ata performance, +/- 5% for 500 km/h. That means that anything between 475 km/h and 525/h is OK for acceptance, for an Emil with DB 601Aa and operating at 1.35ata.

Compared to that, it would the French got around 480 km/h at 1.3ata from a crashed aircraft and the Germans iirc 474 km/h for another test machine, again at 1.3ata. In both cases the nominal output of the engines was at least 55 PS less than that of the Aa engine.

Actually better, I get ~480kmh at 1.26ata from my pressure conversion for the French test. I'm not sure what the relevance is to the Aa export motor (comparison to Test 5?)

Originally Posted by Kurfürst

And this is based on which test or official data...?

Uhm, so by what connection how does 1.35 ata performance come into the picture? You basically show threww German tests (at uncorrected power) at 1.3ata, which only show performance for SL, and suddenly 1.3ata performance is now the same as 1.35ata performance?

How are uncorrected 1.3ata tests are in connection with 1.35ata performance?

Well, I admit that was rather sloppy, but we do have to deal with a set of tests with varying boost levels . It was a rough estimation of performance from the flight tests excluding the prototype and Russian test, to answer my original question : "if you had to derive a single set of figures, what would they be?". A better method would be to attempt to correct all speeds for a standard boost (1.35ata) in some manner, then average. Otherwise one can only say (based on flight tests minus tests 1 and 7)

SL speed 465-480kmh at 1.25-1.3ata
Top speed 547-565kmh at 1.25-1.3ata

With a theoretical correction to 1.35 ata, I would estimate we are looking at a low specification "pass" at SL and close to bang on spec at height.

Originally Posted by robtek

I've seen such a acceptance diagram from Meserschmitt 109's, but i don't know which series it was, anyway the general deviation was about 10 km/h below the guaranteed value, iirc. A few were above, the majority slightly below and five had to be refurbished because they've failed.
Found it: http://www.kurfurst.org/Performance_...catter_web.jpg

Looking at these tests for the 109G (Thank you), it appears (although the plot is a bit hard to make out) that acceptance testing was based on max speed only. The failed 109Gs are underperforming at height, and it doesn't appear that SL performance is being considered. The pattern of the true max speed average falling a little under the spec average seems rather plausible as well.

Based on this all, I would suggest that "typical" series 109Es
* Performed close (maybe a little below) to their max speed average spec at 5000m (572 kmh TAS)
* Performed close (but above) their minimum "pass" rate at sea level (475kmh IAS/TAS)

Based on the 109G test data I would hypothesise that slight deviations under sea level minimum spec might be possible in individual "passed" aircraft (provided they passed the max speed test at height).

From the CloD point of view we are perhaps over-focussed on the sea level performance. We spend a lot of time chasing and being chased at sea level in 1v1 battles, and when we "test" the aircraft we skim the waves in a way no actual test pilot would care to do on a routine basis

camber

[IvanK]

Some more info.

I have a Russian (What I think) is an Engineering and Performance apprasial on the BF109E3 done in 1941. Its 117 pages and very detailed but being written in Cyrillic is hard going. Evident in the document the E3 has Auto Prop pitch.



Buried in the document for what its worth is this graph:



Looks like two sets of TAS v Altitude lines for different conditions and two sets of Time to height lines "t".

Be nice if a Russian speaker could repost the graph with the legends to the lines and the statement at the bottom in English.

[ZaltysZ]

This is level speed (TAS vs Altitude) and climbing (Climb time vs Altitude) chart. Units: m, km/h, min. I will look at this book later for translating the legends, because these are abbreviations in chart.

EDIT #1: curves with bubbles - "Manufacturer data", curves with crosses - "Data of Research institute (НИИ - Научно-исследовательский институт)"
EDIT #2: Tests were done at 2400RPM/1.35ATA. This is true for climbing test only until 5 minute mark, after which lower power setting was used.

[camber]

IvanK,

Nice document. That is clearly the full appraisal that goes with Test 7 (Russian captured E3) in the original thread post. The curve matches pretty much exactly.

The Russians clearly took the Messerchmitt guaranteed 109 performance seriously.. they have plotted it alongside. If I was shopping for a secondhand 109E for the local pylon races I think I would give that one a miss, it really seems a bit of a lemon.

Cheers, camber

[5./JG27.Farber]

In what conditions were all these aircraft captured? I remember reading somewhere the 109 needed a service after 10 flying hours.

[JG4_Helofly]

Is it still a discussion about correcting performance in the game or do we fight the real BOB again?
If it's still about the game, then I doubt that taking some perf. tests from captured aircrafts will get us very far. It's pointless to discuss about taking such data for the FM. Why? Simply because it would be impossible to estimate the average performance of several thousands of planes at a given time in a given battle. Especially when you don't have a representativ number of tests.
How can it be justified to change the top speed of the 109 based on a few testflights from foreign countrys?
And if we would go down that road, how about low production quality in late war scenarios? Or other factors who would affect performance?

IMO the only way to have acceptable performance data is to take the theoretical values which should be reached by production aircrafts under normal conditions. Everything else is just BS.

[camber]

Originally Posted by JG4_Helofly

Is it still a discussion about correcting performance in the game or do we fight the real BOB again?

Well actually neither, it was a discussion about whether the actual 109E performance possibly deviated from (or within) the Messerchmitt specs, based on German and captured aircraft tests. Of course then there is a further argument about whether it would be best for which data to be in CloD. Currently the sea level 109 is too slow for either side of the argument.

Originally Posted by JG4_Helofly

If it's still about the game, then I doubt that taking some perf. tests from captured aircrafts will get us very far. It's pointless to discuss about taking such data for the FM. Why? Simply because it would be impossible to estimate the average performance of several thousands of planes at a given time in a given battle. Especially when you don't have a representativ number of tests.

I disagree. It is never pointless to examine actual flight tests provided their limitations are considered...but you are correct captured aircraft tests should be treated with extra suspicion, as in the original post.

Originally Posted by JG4_Helofly

How can it be justified to change the top speed of the 109 based on a few testflights from foreign countrys?
And if we would go down that road, how about low production quality in late war scenarios? Or other factors who would affect performance?

IMO the only way to have acceptable performance data is to take the theoretical values which should be reached by production aircrafts under normal conditions. Everything else is just BS.

That is a valid position for argument. However I do not think that means alternative arguments are BS. I would also say the Messershmitt specs are +/- 25 kmh at sea level, even adjustments within these bounds may change your online experience with such closely matched aircraft as Spits and 109s.

camber