Looking for TAS vs OAT data

Some people are interested in keeping a close track of their aircraft's performance in level flight, so they can evaluate the effect of small modifications. This requires some way of dealing with the fact that performance varies with temperature, and the ambient temperature varies quite a bit over the course of a year.

Conventional wisdom suggests that on days with non-standard temperature, the test altitude should be adjusted to achieve a target density altitude. This approach would work very well if the engine power available was a strict function of density altitude. If this were true, we should get the same TAS, no matter how the temperature varied, as long as we kept the density altitude the same.

However, the temperature correction that Lycoming publishes on their power charts suggests that the engine power is not a strict function of density altitude. If we have a warmer than standard day, and we lower the test pressure altitude to achieve our target density altitude, the power available will be higher than if we tested at that density altitude on a standard temperature day. Thus the TAS will be higher than it would be on a standard temperature day. All the above assumes that Lycoming knows what they are talking about with their temperature correction.

I would like to gather some test data to determine if Lycoming's temperature correction fits real world data. I need several people with aircraft with constant speed props who are prepared to do TAS tests on several different flights on days of different temperatures at altitude. No changes that could affect drag or power should be made between flights. Once I have enough data, I will analyze it and publish my conclusions here. I have an idea for a better way to correct TAS for non-standard temperatures, but I need to get some data from several aircraft to see if my concept will work.

Interested parties should adopt the following test protocol:

1. Choose a pressure altitude that you will use for all these tests. It should be high enough to offer smooth air, but low enough that we get as wide a variation as possible in air temperature. Use your local knowledge to pick the pressure altitude - it doesn't matter to me what altitude you use, as long as you tell me what it is, and use the same one for every test. The test series may go on for several months, to help get a range of temperatures.

2. Use the same weight and CG for all tests, as close as possible.

3. Set the altimeter to 29.92, so you are reading pressure altitude. Fly all tests at the same pressure altitude.

4. Use the same rpm for all tests, and full throttle.

5. Use the same leaning technique for all tests.

6. Test only in smooth air.

7. Allow the aircraft to accelerate to its max speed. Be patient, as this will take several minutes. Record the IAS, OAT, rpm, MP, pressure altitude and fuel flow (if you have an indicator - this isn't needed for my data analysis, but it provides a quality control check on the leaning technique).

8. Measure the TAS by one of the following methods:

a. Record GPS ground speed and track on four runs in a box pattern. Use the NTPS spreadsheet to calculate the TAS. Report the TAS and the standard deviation from the NTPS spreadsheet.

Or,

b. Do flight testing to establish your airspeed system errors (see my Determing Static System Error page for details on how to do this). Then on each flight test, record the IAS, correct for your airspeed system errors to get CAS, then use OAT and pressure altitude to convert CAS to TAS.

Send me the data from all your flights where the data is of good quality. I want to get IAS, CAS (if you used method b above), TAS, standard deviation (if you used the NTPS method), OAT, pressure altitude, rpm, MP and fuel flow (if your aircraft has a fuel flow indicator). Also tell me what model engine and propeller you have. My e-mail address is khorton01 AT rogers DOT com.

It may also be interesting on the same flights to do another TAS test at the density altitude that is equal to the target pressure altitude. This would allow the consistency of results from the conventional constant density altitude method to be compared to my proposed constant pressure altitude + analytical correction method. If anyone is interested in doing this, contact me and I'll provide a table of OAT vs pressure altitude to give your target density altitude.

Kevin, great idea. Another variable is the dewpoint/temperature combo. Hot, humid summertime air contains a lot of water by percentage and reduces power. A dewpoint of 20C means about 2% water in the air by mole.

AlexPeterson said:

Kevin, great idea. Another variable is the dewpoint/temperature combo. Hot, humid summertime air contains a lot of water by percentage and reduces power. A dewpoint of 20C means about 2% water in the air by mole.

Click to expand...

Interesting comment Alex. I've always looked at this just from a density point of view, but I need to ponder this some more. Just looking at it from a density point of view, it is quite a small effect. For example, at sea level, standard day (59 deg F), going from completely dry air to very moist air with a dew point of 58 deg F only increases the density altitude to about 210 ft. The reduction in density is only about 0.6%.

But, looking at it from a mole perspective, if there is that much less O2 to burn, that would be more significant. Man, I wish we were having this conversation back when I was in university - I was a lot smarter back then . I've got to dig back into my old text books to help put your comment in perspective.

Kevin Horton said:

Man, I wish we were having this conversation back when I was in university - I was a lot smarter back then .

Click to expand...

You're right about that! I'm lucky to even remember that I have books to go look at !

Books?

Is there a good technical book on reciprocating engine performance available these days? I can tell you quite precisely how to correct for humidity (dew point) for turbine engines, but I'm guessing we don't get the mass flow benefit in our Lycs!

If we fly in high dew point wx do we have more horse power or less horse power available. There were once engines that used water injection to increase the power. So if we are filling the cylinders up with water does that cause them to run cooler, does it raise the compression level? Fodder for thought.

Norman CYYJ said:

If we fly in high dew point wx do we have more horse power or less horse power available. There were once engines that used water injection to increase the power. So if we are filling the cylinders up with water does that cause them to run cooler, does it raise the compression level? Fodder for thought.

Click to expand...

If the air is humid, it is less dense than dry air, so the engine makes less power. There may also be an additional effect of water effectively displacing O2 for combustion, but I need to ponder this a bit.

If we inject water, this is additional mass, over and above what the engine would suck in all by itself, so the total mass flow is higher than it would be with dry air. A turbine engine can make a bit more power because of the higher mass flow. I'm not sure if this helps a piston engine or not. The water evaporates, and this cools the air temperature, which helps on both piston and turbine engines.

keen9a said:

Is there a good technical book on reciprocating engine performance available these days? I can tell you quite precisely how to correct for humidity (dew point) for turbine engines, but I'm guessing we don't get the mass flow benefit in our Lycs!

Click to expand...

Ben - I've got some reference material on piston engine performance at work. I've got a couple of small reference texts from test pilot school, but they are in French. One of our flight test engineers has a reference from Pratt and Whitney (or maybe it was Wright). I'll take a look next time I have some spare time in the office (not until next week, probably).

As far as the effect of humidiy on turbine engine performance, all the info I have shows that dry air is more dense than humid air (assuming the humidity is not due to water being injected into the engine). I would have expected humid air to produce less thrust. You seem to disagree. I'm interested in figuring out the source of the discrepancy and learning from it. But, as turbine engine RVs are quite rare, maybe we should take this discussion off the forum - contact me at [email protected] - I'd love to learn why higher humidity leads to more thrust in turbine engines.

For piston engines, Richard Shelquist has some info on an "official" SAE way to correct automotive engine dyno power for temperature, pressure and humidity. I'll see if I can track down a copy of SAE reference.SAE J1349 to see if it provides any info on how their corrections were developed. For what its worth, the official SAE correction for automobile dyno data shows that higher humidity results in lower power (in other words, if the air is humid, the SAE correction says you need to multiply the dyno result by a number greater than "1" to correct it to what it would be if the air was less humid). This suggests very strongly that piston engines make less power as the air becomes more humid.

The SAE correction seems to show that the adverse effect of humidity on piston engine power is more than 3 times as great as would be expected just based on the reduction in density. So, there is definitely more to this story than I suspected. Maybe it is Alex's moles at work

Cold Temperature Altimetry

Kevin,
When we flew out of Hong Kong (Tropics) to Alaska (Arctic) we had no idea about "Cold Temperature Altimetery" until it was bought to our attention by the Canaks.
Subsequently, a section was inserted in the manual for corrections to the Minima under very significant IAS minus temperatures.

Do you want me to dig up the data and send it to you.

Pete.

PS. Carl Morgan flew over from NZ last week and we did some flight testing, engine out, and recorded it to Memeory Stick and then Carl did wizard stuff on my Mac with it and Decoded the data and we plotted it with Excel.
We learnt a lot, mostly on how I need to fly to get more reilable and easily identifieable data.
Carl has gone home and I am going to fly some more engine out descents and we are going to endevour to produce a Polar for MY RV-7. Firstly I shall go through your Pitot Static calibration.
I shall keep you posted.

Pete.

Kevin, I was only contemplating the displaced O2, not the density change. N2 weighs 28 amu's, whereas H2O weighs 18. The density change for a glob of humid air would really be small, but a 2% reduction in O2 is a 2% reduction in power. This would be noticable at higher speeds.

With the ability to really fine tune fuel flow as a function of egt, we notice lower fuel flows in hot, humid days vs winter by a large amount (at the same MAP/RPM), owing to both density and humidity, density being the larger player. For example, at 80F vs 20F, the ratio of molecules per volume would be (460+20)/(460+80)= 89%

Regarding what we've forgotten from the engineering college days - the only way I keep from crying is to notice that I must have learned something over the years that employers value even more.

Power v. Altitude

Convention has it that piston engine power varies with the ratio of seal level air density to density at altitude (commonly designated sigma, the density ratio) raised to the power of 1.2 . For turbines to the power of 1.0, i.e in direct proportion to the density ratio.

Hope this helps reduce reinventing the round thing.

fodrv7 said:

Kevin,
When we flew out of Hong Kong (Tropics) to Alaska (Arctic) we had no idea about "Cold Temperature Altimetery" until it was bought to our attention by the Canaks.
Subsequently, a section was inserted in the manual for corrections to the Minima under very significant IAS minus temperatures.

Do you want me to dig up the data and send it to you.

Click to expand...

Hi Pete - I know all about cold temperature altimetry, so I don't need the data. But thanks anyway. It doesn't have too much relationship to the TAS we would get at a given OAT, unless you know something I don't.

garnt.piper said:

Convention has it that piston engine power varies with the ratio of seal level air density to density at altitude (commonly designated sigma, the density ratio) raised to the power of 1.2 .

Click to expand...

Unfortunately, this does not produce the same power vs temperature relationship as the Lycoming power charts. If we only change temperature, and leave the pressure the same, the Lycoming power chart says the power changes by the square root of the density - i.e. we use a power of 0.5, not 1.2. I don't claim to understand why this should be so, but I have to start from the assumption that Lycoming should know how the power produced by their engines varies as you change the conditions. Once I get some data to analyze, hopefully we can figure out which is correct - Lycoming, or one of the rules of thumb, or some new rule of thumb.

i doubt it.

Kevin,
I doubt there is anything I know about Performance that you don't know.

What i was thinking about of course related to Altimter error; not TAS.

And of course i meant to write "significant ISA minus temperatures", not "significant IAS minus temperatures".
Pete.

Kevin, great thread!

I have been testing my RV8 using the 4-legged GPS method presented on the National Test Pilot School website for the last 4 years. I always do the test at the same density altitude. The temperature (and the resulting pressure altitude vs density altitude) absolutely does make a difference. I get about 2 kts faster in the winter than in the summer.

I will look through the data (I have a giant Excel spreadsheet) and try to pick some examples to send you. It is unfortunatly complicated by the numerous mods I have made, trying to make the airplane go faster.

It is also complicated by my living in the mountain west, where smooth air without verticle movement is hard to find.

John

Water injection

As I understand the use of water injection in big supercharged reciprocating engines, it is used as an anti-detonation measure. They ran pretty high MAPs (2 atmospheres) achieved with big super- or turbochargers, so the induction air was pretty hot. This despite inter- or aftercoolers. So injecting water reduced the air charge temperature and prevented detonation.

This effectively increased the power, because without water injection, they would've had to reduce the boost to avoid detonation. Less boost -> less oxygen -> less fuel -> less power.

That does not alter the fact that a cubic metre of moist air has less oxygen than the same volume of dry air.

Nuisance said:

I have been testing my RV8 using the 4-legged GPS method presented on the National Test Pilot School website for the last 4 years. I always do the test at the same density altitude. The temperature (and the resulting pressure altitude vs density altitude) absolutely does make a difference. I get about 2 kts faster in the winter than in the summer.

Click to expand...

I'd love to see that data John, if we can somehow find points at a similar configuration.

My first cut at trying to estimate the effect of air temperature on power and TAS suggested that the TAS should be a bit higher in warm air and lower pressure altitude (i.e. summer) than it would be in cold air and higher pressure altitude (i.e. winter). This differs from what you report, so I obviously have some more head scratching to do, or maybe I made a calculation error (first one this year, I swear), or Lycoming's temperature correction is out to lunch, or you've remembered it backwards.

My calculations did have estimates for ram air recovery and air box pressure losses. I doubt the effects of any errors here would be enough to turn my predictions completely around, but I'll need to look into this some more. Do you have any idea on how much MP pressure loss there is in your air induction (filter + snorkel), and how much ram pressure recovery you see? For example, how much lower is the MP during the take off roll than it is with the engine stopped on the ramp? If you did a high speed pass down the runway, quite low, how much higher would the MP be than it would be during take-off? And, what would the speed and altitude be during that high speed pass? Do you have a stock air induction system, or have you changed it?

Out of curiousity today, I went up to our technical library and pulled out the Operator's Manual for piston engines from several different manufacturers to see how they handled temperature and humidity corrections. None of them mentioned a humidity correction at all. Every manufacturer had a different way to present a temperature correction. Lycoming has a correction that is based on the square root of the standard temperature over the test temperature. Continental says to add or subtract 1% power for every 6 deg F difference from standard on most engines. But some of their engines have a 1% correction for every 10 deg F. Rotax has a power correction that is strictly based on density altitude. The big Pratt radials just say to add a certain amount of additional MP for temperature changes (they are supercharged, so you don't have full throttle at low altitude - you set the throttle to get a desired MP).

The laws of physics are the same no matter who built the engine. So it is a bit puzzling to see very different approaches to correcting for non standard temperature. I don't see how all the approaches can be correct. We need some test data to help clarify things.

Over lunch I did a search for old NACA reports on the effect of humidity on piston engine power. I found one from 1933 that seems to shed some light on the issue - The Effect of Humidity on Engine Power at Altitude. I need to study it some more, but my initial impression is that it confirms Alex Peterson's suggestion that the effect on power is completely explained by the displacement of O2 by H2O. I could build a spreadsheet that does this correction, but it probably wouldn't be very useful, as we don't really know what the humidity levels are at altitude, unless we happen to be close to one of the sites where weather balloons are sent up.

What GRT does

I asked GRT what they do. Temp is used. This is the formula to which they referred me.
http://williams.best.vwh.net/avform.htm#Mach
I only offer this to round out the discussion.

hevansrv7a said:

I asked GRT what they do. Temp is used. This is the formula to which they referred me.
http://williams.best.vwh.net/avform.htm#Mach
I only offer this to round out the discussion.

Click to expand...

hevansrv7a - I'm having difficulty putting your comment in context, so I'm not sure what you are trying to tell us. Which of the formula on that page did GRT refer you to? Which part of the discussion should we look at when we consider that formula?

Answering Kevin

I'm sorry, Kevin, I'm over my head here. I asked them if they used temp. to figure TAS and they said yes. I asked for the formula and that is the link they gave me. I did notice it had other stuff in it too. I just wanted to point out that at least one "black box" was incorporating temp diff in the calculation and to reveal the formula if I could. Since many of us use GRT, I thought this might be helpful. I hope this is helpful, but I'm only a conduit, not an independent source of anything useful - at least in this case.

hevansrv7a said:

I'm sorry, Kevin, I'm over my head here. I asked them if they used temp. to figure TAS and they said yes. I asked for the formula and that is the link they gave me.

Click to expand...

OK - I understand now. No harm, no foul. Thanks for clearing it up.

GRT takes the IAS, then uses temperature and altitude to calculate TAS. I think they also offer a possible correction for errors in the airspeed or OAT system.

I am trying to do something different. I want to take a full throttle TAS, on a day with non standard OAT, and try to predict what the full throttle TAS would be if the OAT was standard.

Sorry, Kevin, got busy there and I didn't get back to it 'till tonight.

As I look through my data (27 runs with the variables recorded) I am afraid I didn't take enough data to help your cause. I see I usually recorded the OAT on the ground, which doesn't mean much since about half the time we have an inversion going. I would just put OAT and altimeter setting into the E6B function in the GPS and then fly at my 9,000 ft density altitude (except during mid-summer, when that would put me too close to the ground, then I would do 11.000 d.a. Meanwhile, I am madly making modifications, and it is hard to find much of a temperature swing without a modification (i.e. 6 months without a mod...perish the thought). 4 different props doesn't help of course.

I was more interested in making two runs at the same temp with a small mod to compare and test my theories.

I follow what you are saying re: a higher pressure at the same density altitude in the summer...I couldn't find anything to really disprove that. I'm not sure where I got the idea that I was going faster in the winter, because reviewing the data doesn't support that idea.

I will do what I can to get you some data to work with...maybe in the spring here I can get the temperature swings needed...but, with the race season starting, I must make my mods.

I did see I have gained about 15 kts. since I first flew the airplane in the fall of 2002, so I don't feel like a total idiot.

BTW, my record recently is 1 minute, 55 seconds time-to-climb brake release to 10K ft.

Ok, Ok, I start at 7650.

John