Jump to content

Top SL speed


Hummingbird
 Share

Recommended Posts

Hi,

 

Just tested the top SL speed of the DCS K4 which turned out to be 316 KTAS (585 km/h), which is is 10 km/h lower than the real life figure:

xwblZvB.jpg

 

 

Curiously I was able to achieve ~600 km/h @ SL a few months ago, not sure what changed, maybe it's related to the issue with the D-9.

Link to comment
Share on other sites

The problem is that this diagram IS NOT A REAL LIFE test result but just calculated estimation.

Ніщо так сильно не ранить мозок, як уламки скла від розбитих рожевих окулярів

There is nothing so hurtful for the brain as splinters of broken rose-coloured spectacles.

Ничто так сильно не ранит мозг, как осколки стекла от разбитых розовых очков (С) Me

Link to comment
Share on other sites

I tested the DCS K4 on current OB.

15°C, 760mm/hg

 

 

Interestingly MW50 still adds power when in reality it shouldn't above 7.4km.

 

 

I tested the speeds at some altitudes twice, so there are two colours per powersetting.

 

 

The worst discrepancy is about -2.8%, but c

 

 

 

It appears that this DB605DB's supercharger was built on a monday. Underperforming till FTH, overperforming above FTH

 

 

 

When time permits, I'll do a third run.

 

 

 

 

attachment.php?attachmentid=244785&stc=1&d=1597484830Fox

5026-27_DBSonder_MW_geschw.thumb.jpg.fc94c9abe1fb2b843a906601d76a2b7d.jpg


Edited by iFoxRomeo

Eagle Dynamics, ceterum censeo: dare nobis ME CCLXII!

Spoiler

PC Specs: Ryzen 9 5900X, EVGA RTX 3080 FTW3 ULTRA GAMING, 64GB RAM 3600Mhz, Samsung SSD 970 EVO Plus 1TB, HP Reverb (G1)

Link to comment
Share on other sites

MW50 still does add some power above FTH, just not nearly the same amount as below.

9./JG27

 

"If you can't hit anything, it's because you suck. If you get shot down, it's because you suck. You and me, we know we suck, and that makes it ok." - Worst person in all of DCS

 

"In the end, which will never come, we will all be satisifed... we must fight them on forum, we will fight them on reddit..." - Dunravin

Link to comment
Share on other sites

Yes MW 50 cools inlet air, adding some power because of this.

MW 50 alone add a little power across all alt, main power increase comes from ability to run engine at higher then normal ATA which is 1.78 or something like that, while MW 50 is engaged, this power gain persist until supercharger critical alt is hit then power will drop rapidly from this point, only little gain from cooling effect persist at high atl.


Edited by grafspee

Windows 11, I9 9900KF @5.0, Gigabyte Z390 Aorus PRO, Ram 32 GB G.skill, Palit Gamerock OC 3090, Hotas Warthog, Thrustmaster Pendular Pedals, OLED 48" 120Hz.

Link to comment
Share on other sites

MW50 still does add some power above FTH, just not nearly the same amount as below.

 

Yes MW 50 cools inlet air, adding some power because of this.

MW 50 alone add a little power across all alt, main power increase comes from ability to run engine at higher then normal ATA which is 1.78 or something like that, while MW 50 is engaged, this power gain persist until supercharger critical alt is hit then power will drop rapidly from this point, only little gain from cooling effect persist at high atl.

Why does the sheet say MW-Abschaltung at 7.4km, if it would be benefitial to keep MW at and/or above this altitude?

 

I saw a excerpt of a 109 manual that stated, switch off MW above 7.8km, as there is no gain, just useless burning of MW-Stoff. When I find it, I'll post it.

 

EDIT:

I found something for the DB 605 ASM in the G-14. There it is stated, that MW is useable up to 8.5km. For the DB 605 AM it is 6km. Above that altitude the MW system has to be switched off, as it has no benefit, but still burns the MW.

 

MW50 is 49.5% for cooling, 0.5% to prevent corrosion. But also 50% to "level up" the bad B4 fuel and thus increase its octane number to prevent detonations in the engine because of the high ATA. Hence, in this case, it doesn't make sense to use it above altitudes, where the supercharger can't supply more than combat power pressure, and not as I stated before, already at FTH.

 

Fox


Edited by iFoxRomeo

Eagle Dynamics, ceterum censeo: dare nobis ME CCLXII!

Spoiler

PC Specs: Ryzen 9 5900X, EVGA RTX 3080 FTW3 ULTRA GAMING, 64GB RAM 3600Mhz, Samsung SSD 970 EVO Plus 1TB, HP Reverb (G1)

Link to comment
Share on other sites

The problem is that this diagram IS NOT A REAL LIFE test result but just calculated estimation.

 

I find it hard to believe that the series prop curve is calculated Yo Yo, the K4 had afterall been in service for quite a while before this graph was written on January 19th 1945.

 

The Dünnblatt curve, sure, this is calculated, and also what the report concerns, i.e. projected performance gain with the new prop. Hence this had to be based on something solid first, which obviously would be the established performance (from test flights), with the series prop, which by then would've been long achieved as the K4 had been in active operational service since October 1944.

 

Finally MTT calculated performance is in general conservative, i.e. the real aircraft in general flew slightly faster than the calculated performance, hence why the engineers in the same report write "this performance will be reached for certain by well built series produced examples".


Edited by Hummingbird
Link to comment
Share on other sites

I tested the DCS K4 on current OB.

15°C, 760mm/hg

 

 

Interestingly MW50 still adds power when in reality it shouldn't above 7.4km.

 

 

I tested the speeds at some altitudes twice, so there are two colours per powersetting.

 

 

The worst discrepancy is about -2.8%, but c

 

 

 

It appears that this DB605DB's supercharger was built on a monday. Underperforming till FTH, overperforming above FTH

 

 

 

When time permits, I'll do a third run.

 

 

 

 

attachment.php?attachmentid=244785&stc=1&d=1597484830Fox

 

 

There is no use to compare calculations of WWII with the model.

Ніщо так сильно не ранить мозок, як уламки скла від розбитих рожевих окулярів

There is nothing so hurtful for the brain as splinters of broken rose-coloured spectacles.

Ничто так сильно не ранит мозг, как осколки стекла от разбитых розовых очков (С) Me

Link to comment
Share on other sites

I find it hard to believe that the series prop curve is calculated Yo Yo, the K4 had afterall been in service for quite a while before this graph was written on January 19th 1945.

 

The Dünnblatt curve, sure, this is calculated, and also what the report concerns, i.e. projected performance gain with the new prop. Hence this had to be based on something solid first, which obviously would be the established performance (from test flights), with the series prop, which by then would've been long achieved as the K4 had been in active operational service since October 1944.

 

Finally MTT calculated performance is in general conservative, i.e. the real aircraft in general flew slightly faster than the calculated performance, hence why the engineers in the same report write "this performance will be reached for certain by well built series produced examples".

 

You should not believe. Just inquire how flight tests report were arranged, and what is the source of the charts you are referring to.

Ніщо так сильно не ранить мозок, як уламки скла від розбитих рожевих окулярів

There is nothing so hurtful for the brain as splinters of broken rose-coloured spectacles.

Ничто так сильно не ранит мозг, как осколки стекла от разбитых розовых очков (С) Me

Link to comment
Share on other sites

You should not believe.

 

So you believe the series prop figure is calculated despite the fact that the aircraft had been in active frontline service for 4 months by that point ?

 

Just inquire how flight tests report were arranged, and what is the source of the charts you are referring to.

 

The source of the charts is Kurfurst, and he also provides MTT comparisons between calculated & flight test results for the 109 where the general trend is that flight tested performance is slightly higher.

Link to comment
Share on other sites

There is no use to compare calculations of WWII with the model.

Why not? And what do or would you use to compare your FM with?

 

 

Fox

Eagle Dynamics, ceterum censeo: dare nobis ME CCLXII!

Spoiler

PC Specs: Ryzen 9 5900X, EVGA RTX 3080 FTW3 ULTRA GAMING, 64GB RAM 3600Mhz, Samsung SSD 970 EVO Plus 1TB, HP Reverb (G1)

Link to comment
Share on other sites

Why not? And what do or would you use to compare your FM with?

 

 

Fox

 

It's a bit complicated engineering matter... First of all, as it was shown many times, the beautiful straight lines plotted at the grapfs are result of real measuring. Any measuring, especially flight test measurings has its accuracy. For example, the test that seems to be very simple - max speed - requires at least THREE independant measuring: the speed itself whatever method is used - optical or IAS measuring, temperature and pressure to reduce the result to ISA. All of them has it's own error and they cumulate in the result. During the test the wind varies affecting the optical method. The known errors of each measurements were not less 1% due to all factors.

 

Even for the certain aircraft the results of two flight tests can be a bit different.

 

That's why the actual test results look like this one, so speaking about "2.8%" is absolutely senseless in this regard because this percentage is inside the confidence interval.

 

Anyway, the test reports is the best source of reference, but the best case is if a number of similar tests of the same aircraft are available. For sure they won't coincide, but if the curve for FM is somewhere inside the bunch of curves it can be considered as accurate.

 

Theoretical studies can contain much more errors because they usually did not consider a lot of minor factors, for example, radiator drag/thrust and engine jet thrust (it was directly shown for 190D9 calculations). The common method was to use prop efficiency curves obtained in WT tests and airframe drag obtained with different methods.

In addtion to the fact that these measurements had their own errors, the subsequent calculation had their own simlifications, and sometimes real life added its own surprises like hot air from 190D9 radiator to the engine inlet.

 

So, the point is that finding proper reference is not very simple engineering task. We use all methods to have reliable combined reference using all methods: WT tests for airframe and props, flight tests reports and add CFD studies if we are not confident in some cases.

L-39_00106.thumb.jpg.cde281f8b325ef0278d2c7fbee822d46.jpg

Ніщо так сильно не ранить мозок, як уламки скла від розбитих рожевих окулярів

There is nothing so hurtful for the brain as splinters of broken rose-coloured spectacles.

Ничто так сильно не ранит мозг, как осколки стекла от разбитых розовых очков (С) Me

Link to comment
Share on other sites

Just out of curiousity, what PS figure is used, 1800 or 1850 PS? Asking as the German chart shows results with 1850 PS.

 

50PS probably within measure error :)

Windows 11, I9 9900KF @5.0, Gigabyte Z390 Aorus PRO, Ram 32 GB G.skill, Palit Gamerock OC 3090, Hotas Warthog, Thrustmaster Pendular Pedals, OLED 48" 120Hz.

Link to comment
Share on other sites

It's a bit complicated engineering matter... First of all, as it was shown many times, the beautiful straight lines plotted at the grapfs are result of real measuring. Any measuring, especially flight test measurings has its accuracy. For example, the test that seems to be very simple - max speed - requires at least THREE independant measuring: the speed itself whatever method is used - optical or IAS measuring, temperature and pressure to reduce the result to ISA. All of them has it's own error and they cumulate in the result. During the test the wind varies affecting the optical method. The known errors of each measurements were not less 1% due to all factors.

 

Even for the certain aircraft the results of two flight tests can be a bit different.

 

That's why the actual test results look like this one, so speaking about "2.8%" is absolutely senseless in this regard because this percentage is inside the confidence interval.

 

Anyway, the test reports is the best source of reference, but the best case is if a number of similar tests of the same aircraft are available. For sure they won't coincide, but if the curve for FM is somewhere inside the bunch of curves it can be considered as accurate.

 

Theoretical studies can contain much more errors because they usually did not consider a lot of minor factors, for example, radiator drag/thrust and engine jet thrust (it was directly shown for 190D9 calculations). The common method was to use prop efficiency curves obtained in WT tests and airframe drag obtained with different methods.

In addtion to the fact that these measurements had their own errors, the subsequent calculation had their own simlifications, and sometimes real life added its own surprises like hot air from 190D9 radiator to the engine inlet.

 

So, the point is that finding proper reference is not very simple engineering task. We use all methods to have reliable combined reference using all methods: WT tests for airframe and props, flight tests reports and add CFD studies if we are not confident in some cases.

 

Statistically speaking, the "confidence" interval MUST be bellow 0.5% or even 0.05% according to the newest tendencies in the field. 2.8% is too HIGH to yield statistically coherent results.

 

It would be very interesting to know what input data you used in CFD modelling and what were the results..

Specs:

Asus Z97 PRO Gamer, i7 4790K@4.6GHz, 4x8GB Kingston @2400MHz 11-13-14-32, Titan X, Creative X-Fi, 128+2x250GB SSDs, VPC T50 Throttle + G940, MFG Crosswinds, TrackIR 5 w/ pro clip, JetSeat, Win10 Pro 64-bit, Oculus Rift, 27"@1920x1080

 

Settings:

2.1.x - Textures:High Terrain:High Civ.Traffic:Off Water:High VisRan:Low Heatblur:High Shadows:High Res:1920x1080 RoC:1024 MSAA:4x AF:16x HDR:OFF DefS: ON GCI: ON DoF:Off Lens: OFF C/G:390m Trees:1500m R:max Gamma: 1.5

 

Link to comment
Share on other sites

Statistically speaking, the "confidence" interval MUST be bellow 0.5% or even 0.05% according to the newest tendencies in the field. 2.8% is too HIGH to yield statistically coherent results.

What? That makes no sense whatsoever.

Link to comment
Share on other sites

Curiously I was able to achieve ~600 km/h @ SL a few months ago, not sure what changed, maybe it's related to the issue with the D-9.

Hi, look at radflaps position. They could be in a more closed position.

What? That makes no sense whatsoever.

It makes, when you have 2.8% less maximum speed for one, and 2.8% more for another. You absoltely gonna feel that in combat. And, one more thing: acft we fly are identical. In real life acft are different, so you wont see the difference. +/- 3% if your talk about hundreds of acft is realy negligible error.

[sIGPIC][/sIGPIC]

Реальные хотелки к ЛО3 по Су-25 в основном...

ASRock PG9, i-5 9600KF, MSI 2080Ti, 32GB 3466

Link to comment
Share on other sites

Statistically speaking, the "confidence" interval MUST be bellow 0.5% or even 0.05% according to the newest tendencies in the field. 2.8% is too HIGH to yield statistically coherent results.

 

It would be very interesting to know what input data you used in CFD modelling and what were the results..

 

What?? 0.5% or even 0.05% ?? During WWII? Absolutely fantastic.

 

The most accurate photo-registration method of SL speed measuring gave circa 1.5% accuracy. The speed at altitude measurements usually conducted using IAS measurement, then corrected to PE, compressiblity and actual atmosphere parameters. It was not a single measurement - it was a chain of measurements, each having it's own error. Then, the correction to ISA conditions has it's own systematic simplifications adding additional error.

 

Take a look at the IAS gauge - what accuracy of reading you can wait from this scale? 0.05%.

 

Moreover, it is impossible to have engine power absolutely the same even for the same engine. And even if the tests for -10C and +25C were corrected regarding air drag and prop and engine power, the radiators were in very different position during both tests to maintain normal engine temperature.

 

And after all - we are talking not about differences between actual tests and DCS but between the THEORETICAL ESTIMATIONS and DCS tests.

Ніщо так сильно не ранить мозок, як уламки скла від розбитих рожевих окулярів

There is nothing so hurtful for the brain as splinters of broken rose-coloured spectacles.

Ничто так сильно не ранит мозг, как осколки стекла от разбитых розовых очков (С) Me

Link to comment
Share on other sites

What? That makes no sense whatsoever.

 

Thanks for mentioning, my eyebrows raised, was afraid to say anything because statistics usually get ugly. All in the eye of the beholder as they say.

 

But yeah, those interval numbers made me sit up and take notice.

Pointy end hurt! Fire burn!!
JTF-191 25th Draggins - Hawg Main. Black Shark 2, A10C, A10CII, F-16, F/A-18, F-86, Mig-15, Mig-19, Mig-21, P-51, F-15, Su-27, Su-33, Mig-29, FW-190 Dora, Anton, BF 109, Mossie, Normandy, Caucasus, NTTR, Persian Gulf, Channel, Syria, Marianas, WWII Assets, CA. (WWII backer picked aircraft ME-262, P-47D).

Link to comment
Share on other sites

Thanks for mentioning, my eyebrows raised, was afraid to say anything because statistics usually get ugly. All in the eye of the beholder as they say.

 

But yeah, those interval numbers made me sit up and take notice.

 

FYI..

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4111019/

Specs:

Asus Z97 PRO Gamer, i7 4790K@4.6GHz, 4x8GB Kingston @2400MHz 11-13-14-32, Titan X, Creative X-Fi, 128+2x250GB SSDs, VPC T50 Throttle + G940, MFG Crosswinds, TrackIR 5 w/ pro clip, JetSeat, Win10 Pro 64-bit, Oculus Rift, 27"@1920x1080

 

Settings:

2.1.x - Textures:High Terrain:High Civ.Traffic:Off Water:High VisRan:Low Heatblur:High Shadows:High Res:1920x1080 RoC:1024 MSAA:4x AF:16x HDR:OFF DefS: ON GCI: ON DoF:Off Lens: OFF C/G:390m Trees:1500m R:max Gamma: 1.5

 

Link to comment
Share on other sites

It makes, when you have 2.8% less maximum speed for one, and 2.8% more for another. You absoltely gonna feel that in combat. And, one more thing: acft we fly are identical. In real life acft are different, so you wont see the difference. +/- 3% if your talk about hundreds of acft is realy negligible error.

I'm not arguing with that.

 

The "statistical reasoning" doesn't make sense, though.

Link to comment
Share on other sites

And after all - we are talking not about differences between actual tests and DCS but between the THEORETICAL ESTIMATIONS and DCS tests.

 

Again, where's the proof that the curves for the series prop are calculated? The aircraft had been flying in combat for months at the time the graph was drawn, hence only the Dünnblatt & K6 curves can be ascertained as calculated.

 

But if our model is flying with 1800 PS instead of 1850 PS, then maybe that's what is causing the slightly lower performance.

Link to comment
Share on other sites

Again, where's the proof that the curves for the series prop are calculated?

 

 

Because of the place the document was issued.

 

Oberammergau - is a small town deep in mountains area without any airfield near it. And, finally, it is the place where RESEARCH AND DESIGN department of Messerschmidt were placed.

The same type of documents have "Projektburo" as a name of the origin.

 

Test flight documents usually were issued in Augsburg and looks very different having actual experimental points along the curves, boost or temperatures, etc. Having this points plotted over generalised lines is the compulsory style of experimental reports.

322569419_DB601Eamp605AGer08.thumb.png.c97601c410feece296547f1204d1734b.png

Ніщо так сильно не ранить мозок, як уламки скла від розбитих рожевих окулярів

There is nothing so hurtful for the brain as splinters of broken rose-coloured spectacles.

Ничто так сильно не ранит мозг, как осколки стекла от разбитых розовых очков (С) Me

Link to comment
Share on other sites

How does that in any way prove that the series prop curve was calculated? It's pretty obvious that the R&D department would have access to flight test results for an aircraft that has been in active service for over 3 months, and would use this known performance to compare with and calculate the performance of the Dünnblatt. Otherwise the blunt statement that "this performance will be reached with well built series production aircraft for certain" probably wouldn't be made either.

Link to comment
Share on other sites

  • 2 weeks later...
 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...