[AlexPeterson]

I went for a short hop today, and collected a little data. I fixed rpm, and set MAP to various settings. I was not scientific about mixture, but basically just went lean until a noticeable airspeed drop, then nudged it back in. The idea was to start this process, not to rigidly claim this data is correct.

The engine is a O360, FI/EI. The 4500' was pulled out of thin air, pun intended.

Here are the results:



Although this is somewhat noisy data, it does appear that peak mileage is in the 110 to 120 KTAS range. Collecting the data is a little tough in the lower speed areas, as it takes a bit longer to stabilize things. In fact, I'm not sure the speed was completely stabilized at the lower speeds.

The next step for this is to do somewhat the same thing, but gather data from a bit ROP to LOP for each MAP setting. Further expansion will be to do this at various altitudes.

 

[llavalle]

Interesting data.. I'm waiting to see more.

 

[petehowell]

GEEK

Yes, you heard me, GEEK. Just the way I like you........

 

[Kevin Horton]

If the SFC and prop efficiency were constant, a plot of fuel flow times TAS vs TAS^4 at any given weight and density altitude should make a straight line. SFC and prop efficiency will vary somewhat, and the weight and density altitude will not be exactly the same for all test points, but we can still use this type of plot to help us deal with noise in the data.

Here is a plot of FF times TAS vs TAS^4, from Alex's data:


A linear fit has an equation of the form Y = a + b*x.

In this case, we get:

FF * TAS = a + b * TAS^4

The linear fit for Alex's data is FF * TAS = 262.6 + 0.000001522 * TAS^4.

If we divide both sides by TAS, we get:

FF = 262.6 / TAS + 0.000001522 * TAS^3

A plot looks like:



FF / TAS = 262.6 / TAS^2 + 0.000001522 * TAS^2

Mileage is TAS / FF, so:

Mileage = 1 / (262.6 / TAS^2 + 0.000001522 * TAS^2)

That plot looks like:

 

[airguy]

Now THAT was straight up Geek - and I thank you for it.

 

[Jamie]

Kevin took the Geek title a long time ago when he did his POH in Latex.

 

[Steve Sampson]

Alex, I was interested to see your data. As you said, early days, but I think you will be able to get better mpg than that.

I have been working from a different direction, and trying to find the regime - MP/prop/fuel - where I can cruise around 160 mph (139Knots) most efficiently.

I'm an O-320, P-mags, MT, 2 blade -4. Its turning out that at 3000' running 2050rpm and 22.5" I am getting about 24.5 Nm/USG. Because I have been doing everything down low I have not bothered to correct from CAS to TAS, and my purpose was to find an efficient regime.

I have graphed my data and there is an example in a recent post of a recent flight. Before you click over, the mpg is the grey line up near the top. It is displayed in statute miles per litre times 100. So the 700 tick is 23Nm/USG and the 800 mark is 26.3.

In the algorithm that calculates that curve, if the recorded height is dropping the graph cuts out. There is also some smoothing. It does not start displaying until the height has been steady for a while or is increasing

The -4 is probably a little more efficient than the -6, but I would not have thought the engine size was much of an issue. The P-mags might be helping to run very lean. This graph is at just under 1500' though normally I do the test at 3000'

I might give it a shot cruising down at 115Knots (132mph) for comparion though its not where I want to operate. I think my engine would freeze up!

 

[Kahuna]

Here is another reference to FF vs TAS from actual inflight measurements. I would expect this to be linear.

 

[Kevin Horton]

Here is another reference to FF vs TAS from actual inflight measurements. I would expect this to be linear.

Interesting. At high speed, the power required is dominated by the profile drag, which varies as the square of the speed. Thrust power [1] required is equal to drag times speed, so thrust power required varies as the cube of the speed. If the engine's specific fuel consumption [2] and prop efficiency were constant, the fuel flow would vary linearly with thrust power, which means it would vary as the cube of the speed. If your data shows a linear variation of fuel flow with speed, it would suggest that on your aircraft the prop efficiency changes significantly with speed, and/or the SFC is much, much better at high power than at low power.

[1] Thrust power is the power produced by the prop as it acts on the air - i.e. it is engine power times propeller efficiency.

[2] Specific fuel consumption (SFC) is the ratio of fuel flow to power.

 

[Radomir]

I like Kahuna's numbers better.. his charts don't have LSA speeds on 'em..

 

[AlexPeterson]

Kevin, wow, thanks. It ought to really look nice once I get curves at each power setting varying mixture. Then repeat for altitudes.

Steve, it seems your data is quite similar to mine. Your 28 mpg at 160 mph converts to a little over 24 n.m./gallon at 139 knots, which almost exactly fits my data. This assumes your miles are statute miles.

 

[Steve Sampson]

Units....perhaps Napoleon was right!

Alex, Checking back through I dont think units is the problem, but then I bet the Hubble guys probably said that first time around!

The graph I linked to was in statute miles per litre.

Because Kevin had done such a nice job of the graphing I took his third graph and read off your optimum was 25 Nm/USG at about 115Knt. I then looked at your fuel consumption at the speed I was working, 139Knt, on Kevin's graph and that indicated your performance as about 23.2 Nm/USG. At that speed I was getting about 24.5Nm/USG which equates to about 28sm/usg.

That is about a 5% improvement. I think we both thought our aircraft could do a little better. I was ignoring the small adjustment for TAS and just using a CAS.

I will be really interested to see your data progress.

I find it really hard to get good data. Just look how that grey line on my graph wobbles up and down!