What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Lowest Fuel Burn for Best TAS LOP

rv6ejguy

Well Known Member
We've had some people on the forum ask why we feel it's so beneficial to have user adjustable ignition timing. First thing to say is, maybe it isn't so important if you fly at medium to lower altitudes ROP. However, if you fly high and/or LOP we can see substantial benefits in advancing ignition timing under those conditions. The charts below may help you understand why:

flamespeed33_zpsetkdixdl.jpg


The chart above shows how flame speed within the combustion chamber varies with air/ fuel ratio (AFR). Note how much slower the flame speed is at lean mixtures. This means with fixed timing, peak cylinder pressure (PCP) will arrive too late when running LOP to get the most push on the piston, rod and crank for the fuel burn.

AFR33_zpsgjvshtnq.jpg


The chart above shows the relationship between AFR, power and BSFC. Peak EGT occurs around .065, best power at around .078. Note how power drops off rapidly on the lean side of peak but also how BSFC is best at around .055.

mapflame_zpsy4iawhpj.jpg


The chart above shows flame speed vs. manifold pressure. Again you can see at lower MAP, flame speed drops so we'd have to initiate the spark sooner to optimize the point of PCP. This again shows the compromise of fixed timing like mags have and why EIs give more speed and/or better economy in cruise, especially at higher altitudes where the MAP falls off.

If we can get all cylinders to peak nearly simultaneously and still light off the lean mixture at the right time to achieve PCP at the optimal point we can maximize the amount of energy we extract from each pound of fuel. Witness Dave Anders' RV4. 200 mph TAS on only 6.8 GPH. Granted, this is a clean aircraft compared to other RVs but if you work out the power this engine is likely making at the MAP and RPM he's running, the BSFC is probably better than .35 lbs/hp/hr.

Is this possible? We know the Wright R3350 turbo compound engines were around .375 running LOP. They were recovering energy back into crankshaft through 3 Power Recovery Turbines but at the same time, they had much lower compression ratios than what Dave is running ( 6.7 vs. around 10 for Dave presently) The higher CR boosts thermal efficiency substantially. We also know that the Continental 550 engines can achieve something around .38 with relatively low CR, fixed mag timing and mechanical injection.

It's not much of a stretch to see that with the higher CR, superior EFI fuel atomization and equal mixture distribution along with variable and optimized ignition timing, we can do better than the the older technology.

Toyota recently achieved an amazing 40% thermal efficiency on their spark ignition 2016 Prius engine. It uses similar thoughts- high CR, lean cruise mixture and advanced timing along with careful EGR feed.

It would be very interesting to put a strain gauge on Dave's crank to measure true torque and therefore hp so we could calculate his BSFC accurately.

Thanks to Liston for the charts.
 
Last edited:
Nicely done Ross. People should notice that the left side hook up in the BSFC occurs due to misfire, and with "better" combustion, going more lean can drive that down even more. This shows very clearly what the possibilities are for the SI engines. Even many diesels are not better than .350 as they compromise for ... well... lots of things.

Ross, I was quoted a number of .330 on a production Conti when I was there, and sadly the only thing I truly remember is the SFC number. Now I wish the engine and operation details were known, it was highly LOP , obviously.

This illustrates why the diesel has a hard time competing at cruise on the basis of economy. The diesel SFC curve/map is much better at part load conditions, like climb and idle and part power cruise, but up at wot, they are not that far apart. So, it becomes driven by fuel availability. The Navy says that gasoline costs $400/gal with all the infrastructure/logistics to get it on a ship. (RPV)

That prius engine uses valve lift/timing control to extend the expansion stroke longer than the intake stroke, yielding excellent conversion to work. It works for diesels too. Lousy throttle response, but not missed with hybrid. Was that 40% @ stoic so the 3 way catalyst still functioned? If so even more impressive.
 
Last edited:
Happen to know what is meant by "thermal plug"? Is the source material available?
 
That Prius engine uses valve lift/timing control to extend the expansion stroke longer than the intake stroke, yielding excellent conversion to work. It works for diesels too. Lousy throttle response, but not missed with hybrid. Was that 40% @ stoic so the 3 way catalyst still functioned? If so even more impressive.

Yes there are a few other technologies present in this engine that are not easily applied to a Lycoming.

My reading on modern catalysts seems to suggest they are able to process the bad stuff quite effectively well lean of Stoich which is where these types of engines spend a lot of their time cruising.

Some people blindly believe diesels are always better in regards to BSFC but in fact, the Otto cycle is theoretically more efficient than the Diesel cycle at the same compression ratio.
Otto cycle efficiency is 60.2%
Diesel cycle efficiency is 53.7%

Of course, diesels can tolerate higher CRs generally but the interesting trend is a raising of CRs on Otto engines and a dropping of CRs on Diesels of late. Many new diesels are down around 16-17 now and some production SI (Otto) ones are as high as 14 now- witness the Mazda SkyActive engines.

Dave's over 2 point drop in CR may not be offset by the efficiency gains of the EFI at higher power settings and fuel flow but I suspect the EFI is much better than the Bendix style injection at the lower fuel flows. Dave always had a big GAMI spread under these conditions before.
 
Last edited:
Specific Range v Ignition Timing at high altitude

A lot of folks love their RV-8 because they have a high top speed. I like mine because it is economical going fast, which makes long distance trips really nice. After reading the recent post on the 20nm/gal club I went out to see how well my plane would do. I also have dual PMags and an EICommander so I took the opportunity to experiment with ignition timing. Based on previous tests the optimum cruise altitude for my RV-8 is 18,000'. This gives me the highest specific range and also the best cruise (which I define as the highest product of TAS and nm/gal, a modified Carson Speed of sorts). I can go above 18,000' but chose not to because its cold, the time of useful consciousness rapidly decreases and the risk of DCI increases.

For this test I used 17,500' PA which was as high as I could go in the airspace I was in at the time, where it was 0 C (ISA +20C), the plane was fairly light at 1500 lbs with a CG at 79.9". For all the test series the prop was set to 2,500 RPM and I started with the throttle fully open. I leaned to 25F LOP at WOT for all the runs except the 25.2 deg advance one as the engine would not run LOP with this setting so I did this at 25 ROP ( as lean as it would run in the interests of maximum efficiency). This I found interesting as my GAMI spread typically varies between 0 and 0.2 gph, so even with good mixture distribution the engine was rough on the lean side of peak with only 25 deg advance up high. For each series once leaned I just reduced throttle in approximately 1" MAP increments, with prop and mixture controls constant, for subsequent data points. I repeated the 35.0 degrees advance run (which is my current maximum timing setting) at 100F ROP just for a comparison.

The results are shown in the following graph. There is a big difference in economy between the 25 deg advance and higher advance curves but as this advance used 25 ROP vice 25 LOP it is hard to say how much is attributable to each factor.

Interestingly there is only 5% difference between the maximum SR from 30.8 to 39.2 degrees advance with the 25F LOP runs. This could be as much due to subtle difference in how far the mixture was leaned at the start of each test run as due to the difference in ignition timing. There was not a great deal of difference in CHT with any of the advance settings, but I only waited long enough at each point for the IAS/TAS to stabilize and not the full 5 minutes it takes for my CHT's to stabilize, but at any rate with all speeds, ignition advance & throttle settings they were from 300 - 330F. At 39.2 deg advance the engine was slightly less smooth than with 30.8, 35 or 36.4 degrees advance.

In winter I used 36.4 degrees as the maximum ignition advance and I could use the target EGT climb method without CHT going over 400F but now in summer I have to either use a slightly richer mixture or reduce advance to 35.0 degrees.

In my plane it seems that the biggest effects on SR and best cruise are the altitude, airspeed, mixture, prop RPM and ignition timing, in that order. To fly the furthest I need to go as high as possible up to 18,000', fly at 150 KTAS, using 2,500 RPM at 25F LOP with around 35 deg ignition advance. To get the highest product of TAS and SR I need to do as above but fly at Vh, i.e. WOT.

Ignition%20Timing%20v%20SR_zps7ara4psl.png
[/URL][/IMG]

Cheers

Nige
 
Last edited:
Happen to know what is meant by "thermal plug"? Is the source material available?

No idea. I was hoping you might! :) Bill?

Sorry for the response delay - these look like little set screws that are screwed into the parent material that can be analyzed after removal to determine the temperature of the material during operation. They are definitely old school, but much quicker to use than routing thermocouple wires up a connecting rod for piston temps. They only give peaks, so one has to be careful on how the test is run to generate the proper data. A real PITA, actually. One shot per install.

For the more definitive information - http://testing-engineers.com/images/TEMPLUGMAN.pdf

Edit - These plugs might have been used within the combustion chamber protruding from the surface like a pin fin to measure combustion temps, but since they only average them the absolute temps are lower. Just a guess, as they too high for piston/head/cylinder temperatures.
 
Last edited:
Interestingly there is only 5% difference between the maximum SR from 30.8 to 39.2 degrees advance with the 25F LOP runs...

Very advanced timing (39.2) to get the last of the 5% may not be a worthwhile trade. For example, peak cylinder pressure will be near TDC and high, which has has an effect on physical structure, blowby quantity, and ring pressure at the point in piston travel where lubrication is worst. That said, I'm a big believer in quantifying, and we just don't have that data. Plus, yea or nay depends on goals. If a guy wants to set an efficiency record, it's worth it to him.

Nice work Nigel, as always. Sure is wonderful to have a professional test pilots in the community.

Hey, any chance you'll repeat the test at say, 2200 RPM? I'm sure you saw Andy Turner's RV-10 data.

http://www.vansairforce.com/community/showpost.php?p=1103408&postcount=11
 
This is a fascinating thread ... I'm learning about things I've often pondered. Thanks to all for sharing their test results and knowledge! :)
 
Yes there are a few other technologies present in this engine that are not easily applied to a Lycoming.

My reading on modern catalysts seems to suggest they are able to process the bad stuff quite effectively well lean of Stoich which is where these types of engines spend a lot of their time cruising.

Some people blindly believe diesels are always better in regards to BSFC but in fact, the Otto cycle is theoretically more efficient than the Diesel cycle at the same compression ratio.
Otto cycle efficiency is 60.2%
Diesel cycle efficiency is 53.7%

Of course, diesels can tolerate higher CRs generally but the interesting trend is a raising of CRs on Otto engines and a dropping of CRs on Diesels of late. Many new diesels are down around 16-17 now and some production SI (Otto) ones are as high as 14 now- witness the Mazda SkyActive engines.

Dave's over 2 point drop in CR may not be offset by the efficiency gains of the EFI at higher power settings and fuel flow but I suspect the EFI is much better than the Bendix style in section at the lower fuel flows. Dave always had a big GAMI spread under these conditions before.

Warning: TMI

All the autos use a 3-way catalyst formulation, but not the exact same one. Most today have it all incorporated in one wash coated substrate. 3-way - NOx, CO, HC. The fuel system perturbates the A/F between lean and rich. Pretty fast too 10-30 hz. The CO/HC oxidation process needs oxygen for it's chemistry to work, NOx catalyst chemistry (a reduction process) can not have oxygen present or the reaction preference will just oxidize more. So - there is an oxygen capture (adsorption) element in the washout that allows the CO/HC actions to take place, then the rich side reduces the NOx. The smell from a car going up a hill is from the nickel being used to make the catalyst cheaper. There are better formulations that other companies use. The very low NOx standards will keep autos in the primarily stoic region for a while.

The Prius uses a modified Atkinson cycle substituting valve event control over linkages. It takes what is typically a part load condition (throttled) and eliminates the throttle losses by going near WOT using valve events to control power. It has very bad load acceptance and response, so the integration of the battery/electric system and the split-torque transmission allow the transients to be all handled by the electrical side allowing the engine to run very efficiently. I have an article from Popular Mechanics from 1968 talking about GM and Toyota hybrid systems. Toyota continued to learn, GM did not. Thus the constant "ICU" of the US auto industry. The toyota hybrid system is brilliant example of deep system integration. Not an add on like all the others.
 
Always nice to see some actual flight test data like this, thanks Nigel.

What induction setup are you running?

Any idea why the engine is rough only running slightly LOP with 25 degrees of total timing? This is not something I've frequently heard of.
 
Thanks Bill for your ever deep well of knowledge!

Reportedly some of the European auto OEMs (and possibly others) have been running far LOP (around 16-17 AFR) at light load for a decade now to boost cruising fuel economy. I suspect they can do this because emissions are usually spec'd in grams per mile and the engine isn't doing much mass flow in cruise at 15-20hp so it still meets the spec. In harder parts of the test cycle, I'm guessing they're running stoich as much as possible to keep the cats happy.

BMW had Valvetronic a decade ago now where there is no throttle plate at all. The valves are opened to control power. They found 10% fuel savings reportedly with that technology at low load.
 
Last edited:
Thanks Bill...thermal plugs calibrated to steel hardness. Whodathunkit? ;)



Can you ask Liston where they came from?

Mr. Liston is likely long gone. Looks like he's referencing a lot of work from Taylor, Hersey, Wood, Purdue labs, Weigard, Eberhardt, Hottel.

Much data was complied from single cylinder CFR engines.

That book was written in 1953 but it continues to amaze me what these folks could do and learn using very dated technology like thermal plugs and moving core/magnet transducers.

They had a very good understanding of the chemical processes going on in the combustion process and had high speed photography processes to capture the images of the combustion process inside the chamber.

Some very smart people doing aero engine development in the 1940s.
 
Last edited:
More ignition testing

Hi Dan,

I was planning on repeating the test at 2k, 6k & 12k with variations in LOP & ROP. My engine is happiest at 2500 RPM so I wasn't planning on varying that for the first cut but could always be persuaded.

Ross,

My engine was built up from a Superior kit and has their 'cold air induction' intake and a airflow performance fuel injection servo with a K&N filter under a Sam James cowl. Not sure why the engine would not run smoothly LOP with 25 degrees advance. I tried leaning a couple of times thinking that I might have been a bit fast the first time, but to no avail. Each time the GAMI spread was 0.1 gph, but it felt like an inflight mag check once 25 LOP. I really cant say what caused it to be a little rough but with 30+ degrees advance it was normal. Given the airspeed, throttle, MAP & RPM were the same as other test points I don't think it is airflow. The GAMI spread was similar to other test points so I don't think it is fuel distribution either but I don't have any data to support that feeling. Normally the engine is smooth 0 - 50F LOP and starts to get less smooth as you get to 70F LOP.

Cheers
Nigel
 
Hi Dan,

I was planning on repeating the test at 2k, 6k & 12k with variations in LOP & ROP. My engine is happiest at 2500 RPM so I wasn't planning on varying that for the first cut but could always be persuaded.

Ross,

My engine was built up from a Superior kit and has their 'cold air induction' intake and a airflow performance fuel injection servo with a K&N filter under a Sam James cowl. Not sure why the engine would not run smoothly LOP with 25 degrees advance. I tried leaning a couple of times thinking that I might have been a bit fast the first time, but to no avail. Each time the GAMI spread was 0.1 gph, but it felt like an inflight mag check once 25 LOP. I really cant say what caused it to be a little rough but with 30+ degrees advance it was normal. Given the airspeed, throttle, MAP & RPM were the same as other test points I don't think it is airflow. The GAMI spread was similar to other test points so I don't think it is fuel distribution either but I don't have any data to support that feeling. Normally the engine is smooth 0 - 50F LOP and starts to get less smooth as you get to 70F LOP.

Cheers
Nigel

Yes, does not seem like fuel as all points to ignition. Anyone else experience similar roughness when running a similar configuration?
 
Leaning Techniques with EFI

I'm adding to this thread in response to the BSFC discussion in the recent CD155 thread.

For those not up to speed on the latest possibilities and techniques being used to make your fuel $ go further with your EFI Lycoming, I'll present a bit of information here.

Below is a screen shot of Rusty Crawford's 7A in cruise LOP with EFI.

rustypanel_zpsmzqvfhcq.jpg


His plane is powered by a stock compression PV 360 Lycoming. You'll notice the TAS is 172 knots on 7.7 gph (46.2 pph). There are no serious engine or airframe tweaks here, so this is representative of what can routinely be achieved with a well built airframe, relatively stock engine with stock CR and modern engine controls using good technique.

Several of our clients run lean at idle (about 15 AFR) and use the prop control to a large degree as the power lever in flight. They will pull the rpms way back after the initial climb WOT until they are under 75% power and then will pull back the mixture LOP to cool the heads in the climb to cruising altitude.

Leaning can be done either manually or with the SDS LOP switch which can simultaneously lean and add the programmed amount of ignition advance to optimize the point of peak cylinder pressure for the slow burning LOP mixture.

Almost all flight is done WOT to minimize pumping losses and many users fly at 11,000 to 15,000 feet, especially if the tail winds are good as in this instance above.

With the EFI/ EI, we're also able to sharply advance timing on the ground for lowest fuel burn during idle and taxi operations. The engine will run smoothly even leaned out because we can trim each cylinder for low AFR spread between cylinders and the hot EI spark.

We even save fuel during warmup because the EFI vaporizes fuel better than carbs or mechanical injection at very low airflow rates and we have superior mixture distribution as well. Warmup enrichment is controlled by the ECU.

We can leave the engine lean in the decent as well, the ECU adjusts for changes in MAP and barometric pressure automatically.

Here's a video of our O-200 test engine running EFI/EI.

https://vimeo.com/221821595

We're able to idle down to 350-400 rpm. The idle you see here at about 600-700 rpm is at around stoich (14.5 to 15 AFR) on 87 octane mogas. We can do similar things with the Lycoming engines.

Despite what others may believe, the old Lycoming can turn in some pretty impressive numbers and we've come a long ways with modern glass engine monitors and electronic engine controls from the days when I used to fly Warriors and Tigers ROP most of the time and lean by feel with carbs and mags doing the fuel and spark chores.
 
Last edited:
For those not up to speed on the latest possibilities and techniques being used to make your fuel $ go further with your EFI Lycoming, I'll present a bit of information here.

Below is a screen shot of Rusty Crawford's 7A in cruise LOP with EFI.

rustypanel_zpsmzqvfhcq.jpg


His plane is powered by a stock compression PV 360 Lycoming. You'll notice the TAS is 172 knots on 7.7 gph (46.2 pph). There are no serious engine or airframe tweaks here, so this is representative of what can routinely be achieved with a well built airframe, relatively stock engine with stock CR and modern engine controls using good technique.

Strictly speaking, "modern engine controls" are not required to match the above. And let's remember a speed-density EFI is a long way from cutting-edge modern. I'm not criticizing, just keeping things real.

Here's 178 on 7.8, same altitude, using mags with fixed timing and an AFP FM200. I've posted the photo previously. I attribute the speed difference to airframe differences, not engine efficiency.



Since then (summer 2014) I've installed a pair of electronic ignitions with fully programmable dual advance maps. The dramatic change has been in operation aspects like hot starting, and 75 to 100 LOP loafing when low and slow, not cruise performance.

Unquestionably an EFI will distribute more evenly at very low fuel flows, where the Bendix/AFP style system is distributing based on the flow divider rather than the nozzle size, and and SDS EFI injector trim can be further tweaked in flight, a major advance. I've also been very impressed with EFI starting ease, which constant flow can never match. I suspect LOP smoothness at altitude is better with the EFI system. I'm sure they idle better. All might be good reasons to choose EFI. However, there is no "magic 100 MPG carburetor"; I would not assume a significant improvement to fuel mileage merely by delivering fuel to the port with a pintle injector rather than a nozzle.

Might we do better with airframe drag reduction? It's nearly free, requires little or no maintenance, and no supporting electrical system. I recall really good numbers being reported by Tom Martin, and without question, Dave Anders got a lot more from drag work than any engine accessory change.
 
I generally agree Dan. On the CD155 thread we even saw impressive numbers with a carbed 320 9A running LOP with a touch of carb heat.

My point on this thread was to show how many of our customers are running for best efficiency using EFI and where they are seeing additional gains over traditional engine controls during the entire sortie (startup, idle, taxi, decent, high altitude/low power). A few percent here and there can add up to another bit saved when you refill your tanks.

The EFI gives people the ability to fly high at very low power settings efficiently as Dave Anders has demonstrated (3.7 gph@ 17,500 feet). As you say, this is hard to do with mechanical injection. EFI gives the best of all worlds. Dave changed out his AFP system because it didn't do a good job for his cruise mission at low power settings. It worked well at higher power settings he said.

While we don't yet have the full FADEC capability of the Lycoming IE2 system for Experimentals, the gains available past what we have now when properly programmed and managed would be minimal IMO. You're splitting hairs for maybe another 2-3% decrease in sortie fuel burn. You would reduce pilot interaction and workload though which would be welcome and useful for many. That won't come at a reasonable price until we have readily available unleaded avgas though.

Speed/density systems with baro correction work just fine in this relatively steady state application and the addition of closed loop feedback would make it perform just like a MAF system. I might be missing your reason for saying this. Over my many years in the EFI and auto repair business, I've seen far more MAF failures than MAP failures. We feel MAP is simply more reliable just as Hall Effect sensors tend to be more reliable than magnetic pickups in the long term.

Studies have been done on fuel droplet size with regards to hp years ago and it's a large area of research today by all the OEMs. Almost certainly low pressure mechanical FI simple nozzles don't break up the fuel to optimal size or shape. One of my friends with a couple SCCA championships behind him found some extra hp on the dyno by increasing FP 20 more psi. Before DI started coming on the scene in autos, the OEMs were raising FP into the 60-65 psi range, finding both slightly better emissions and slightly more power plus higher resistance to vapor lock with the returnless systems generally in use then. Port fuel injector design also changed quite a bit in the last 20 years to obtain better atomization at a given fuel pressure compared to the original pintle design. The gains are small but measurable. Better atomization provides less stratification of the charge as it enters the cylinder. This may aid mixing, especially at low MAP which ensures more linear combustion rates. Now, the majority of auto engines are using DI but spray studies are just as important to improve performance, emissions. I wouldn't assume a fixed orifice mechanical injector is just as good as more sophisticated and highly developed/ tested EFI injectors.

An RV8 has less drag than a 7A and you have some aero tweaks as well over Rusty's 7A. Dave Anders likely has less drag than you and turns in better numbers than your -8 up high- some of which is due to the EFI. I agree, some effort in the drag reduction area is time well spent.
 
Last edited:
While these are good fuel flow numbers for that TAS, is it no better than my data with an IO-360 (180hp) with AFP injection, balanced injector nozzles and dual pMags.

Good options on the market for people to decide what they want. I for one am not a fan of the current "Single Lever" type trend for engine control. Others will disagree and vote with their wallets.

Carl
 
Carl Froehlich;1185283 Good options on the market for people to decide what they want. I for one am not a fan of the current "Single Lever" type trend for engine control. Others will disagree and vote with their wallets. Carl[/QUOTE said:
I fly for fun. Playing with the knobs is part of the fun for me.

That said, this and other similar threads and discussions have helped me understand what is going on, and what isn't, in many aspects of opreratng my power plant. Knowing why I hit the wall with my limited technology is beneficial.
My cruise numbers with standard mags, servo, and balanced injectors are not that far off either, certainly less efficient, but only marginally for most of my flying.
Hot starts have never been an issue but Inknow some struggle.
Good stuff.
 
While these are good fuel flow numbers for that TAS, is it no better than my data with an IO-360 (180hp) with AFP injection, balanced injector nozzles and dual pMags.

Good options on the market for people to decide what they want. I for one am not a fan of the current "Single Lever" type trend for engine control. Others will disagree and vote with their wallets.

Carl

The thing is the EFI will do this at "normal" power settings but also at very low power settings. AFP won't do that. Dave tried it with balanced nozzles which is why he switched. He also has no more hot start issues, smooth idle and smooth running. There are multiple reasons why people are leaving the mechanical injection side and moving to EFI these days. Mechanical FI works well enough for most people most of the time but lots of our customers had it before, didn't like it for various reasons (usually hot start and lumpy idle concerns) and would never go back to it after using the EFI for a while. Likewise, some folks get the willies from EFI and prefer a mechanical system. To each their own.
 
Last edited:
The EFI gives people the ability to fly high at very low power settings efficiently as Dave Anders has demonstrated (3.7 gph@ 17,500 feet). As you say, this is hard to do with mechanical injection.

Perhaps "easy starting and great idle" would be a more effective sales message than great fuel burn at 17K...seriously.

As for increased cruise efficiency at settings in common use, do you recall a magazine article in which I reported hard dyno numbers? At the time, a few readers acted like I had taken a dump on the Vatican steps, but lo, behold the data, Bendix RSA-10 and Slicks on the left, Paisley's SDS-based system on the right:



Using corrected HP and fuel temperature, that's a BSFC of .492 for Bendix/mags, vs .455 for EFI/EI. Put another way, the EFI/EI burned roughly 9% less fuel for 2% less power. This particular run was subjective, hunting the best balance of power and economy, just as the thread requests. Had we enriched slightly to match HP, I think the EFI/EI would have still turned in a better BSFC. So, the data suggests slightly better fuel economy is a reasonable expectation. I merely pointed out that it's not the only path...and anyway, every choice has its compromises.

Speed/density systems with baro correction work just fine in this relatively steady state application and the addition of closed loop feedback would make it perform just like a MAF system. I might be missing your reason for saying this.

I'll clarify. I am not knocking speed-density. Simplicity is highly desirable.

However, calling it modern is a stretch. At the fundamental level, we're talking about the Bosch system installed on a 1968 Type 3 Volkswagen...fifty years ago. What we have here, compared to that 1960's system, are modern semiconductors and board fabrication, i.e. increased electronics reliability, plus improvements in the injectors themselves. All good of course, but incremental improvements to an old system, not cutting edge bragging stuff.

Yellowed page from a 1971 how-to manual:



Back in 1968, a fellow could go to the hardware store and buy a pretty good claw hammer. Today a fellow can buy what is arguably a better hammer, with a comfortable rubber and fiberglass handle. It may be more pleasant to use, and harder to break, but it is still just a hammer.

SDS builds an excellent hammer. Here your friend is trying to tell you to stick with excellent, and leave hyperbole to the other guy.

Studies have been done on fuel droplet size with regards to hp years ago and it's a large area of research today by all the OEMs. Almost certainly low pressure mechanical FI simple nozzles don't break up the fuel to optimal size or shape.

Agree. Can you list titles/authors for your favorites? I'd really like to read them.
 
Last edited:
Perhaps "easy starting and great idle" would be a more effective sales message than great fuel burn at 17K...seriously.

As for increased cruise efficiency at settings in common use, do you recall a magazine article in which I reported hard dyno numbers?...

As a SDS customer I'm not envisioning flying around at 17k either. The illustration does resonate with me as an example of stable EGT at very low fuel flows - and the EGT divergence happens with my Bendix systems much lower than that. As soon as I get under about 8.0 GPH in the -8, the nozzle pressure is just too low to be an effective metering scheme and my previously well grouped EGT's start to run amok. This fuel flow occurs at most of my normal cruise altitudes. This divergence is a LIMFAC to going higher. Careful swapping of parts can improve this, but it remains a characteristic of mechanical fuel injection, and let's face it - it's a bunch of work. To recap, I think the "sales pitch" here is "stable fuel delivery at any fuel flow".

To your other point (BSFC), I'd use caution when comparing Robert's implementation of EFI to Ross'. Yes, Ross' computer is the basis for EFII, but I'm skeptical that the "upstream and backwards" placement of the injectors that Robert went with is optimum. Without a side by side shootout we are not going to be sure, but empirical evidence suggests that the direct port injection offered by SDS will have an edge. Therefore, the slight edge over Bendix you attributed to EFII above is likely to be incrimentally improved again with the SDS direct port configuration. I will grant you that there is not enough performance headroom to make radical improvements with the Lycoming engine configuration, but a little here and there adds up.
 
Perhaps "easy starting and great idle" would be a more effective sales message than great fuel burn at 17K...seriously.

As for increased cruise efficiency at settings in common use, do you recall a magazine article in which I reported hard dyno numbers? At the time, a few readers acted like I had taken a dump on the Vatican steps, but lo, behold the data, Bendix RSA-10 and Slicks on the left, Paisley's SDS-based system on the right:

I'll clarify. I am not knocking speed-density. Simplicity is highly desirable.

However, calling it modern is a stretch. At the fundamental level, we're talking about the Bosch system installed on a 1968 Type 3 Volkswagen...fifty years ago. What we have here, compared to that 1960's system, are modern semiconductors and board fabrication, i.e. increased electronics reliability, plus improvements in the injectors themselves. All good of course, but incremental improvements to an old system, not cutting edge bragging stuff.

Agree. Can you list titles/authors for your favorites? I'd really like to read them.

We have a fair number of people who fly regularly above 12,500 on O2 who don't like making fuel stops on long trips. With the lower drag up high in the 14-17K range, efficient running at the low available MAP is a consideration for them. The EFI allows that option to get 150-160ish KTAS on 6 to 7 gph. This may not be a consideration for others.

As Michael said, I wouldn't compare EFII with SDS, quite a few differences in mechanical items such as injectors, placement, orientation, throttle body size, pump module, regulator, plumbing plus several other features such as the fuel trim, internal FF output, dual access programmer, data logging which they don't offer. There are more new features in the pipeline.

I was never clear on this test if fuel return was being sent back to the pump inlet. If so, we don't recommend this.

Barrett has run a couple 540s now with SDS, you could check with Alan to see what he saw for hp on those. It think they at least matched the traditional offerings. I didn't get the final number on the last one.

I still don't see your point about the Speed Density thing. Honda and GM were using this into the 90s with solid state transducers replacing the moving coil transducers from the D Jet era. It's just another way to meter the fuel and with closed loop feedback as all cars have had since the early '80s, works just fine since the O2 corrects for any changes due to engine wear at part throttle. The D Jet ECU was analog, not digital- big difference in capability there. MAP sensors are more reliable than MAF and easier to mount, lighter and not sensitive to inlet conditions. These are all reasons to use Speed/ Density for load sensing. Very tough emission standards are why most OEMs use MAF now.

Injector spray studies? Google is your friend. Tons of studies listed, take your pick. This is a major area of R&D with the OEMs using laser techniques and HS imaging along with CFD studies which started back in the early '80s.

Yeah a hammer still drives nails no matter the vintage but the analogy may be a bit simplistic in this case. I'd compare it more to the glass panel revolution. Yes you can still fly with a VOR and a six pack but nobody does any more with new build RVs. Many people want the same sort of advantages that EFI/EI technology offers over the vintage hammers. Bendix FI and mags predate even D Jetronic Bosch systems.
 
Last edited:
I still don't see your point about the Speed Density thing.

Merely that it is a long way from being The Latest Cool Thing. Nothing more.

Fact is, EFI has been around nearly as long as the Bendix RSA, for which Elmer Haase was still being granted patents in the mid-60's. Both have been in continual development since then. Heck, you've personally been building speed-density EFI almost a quarter century, old bean ;)
 
I believe Bendix held the original EFI patents which date back to the 50s for aircraft use. Used in the late 50s on a handful of American cars. Bosch saw the potential and the rest is history.

Port EFI is passe now in the automotive world though. Mostly replaced by electronic, direct cylinder injection these days.
 
...Port EFI is passe now in the automotive world though. Mostly replaced by electronic, direct cylinder injection these days.

Maybe. At least Ford and Toyota have introduced engines which have both port and direct injection on the same engine. Ford just recently and Toyota for a while now.
 
Maybe. At least Ford and Toyota have introduced engines which have both port and direct injection on the same engine. Ford just recently and Toyota for a while now.

Correct. Toyota/Lexus did this from the start on their DI engines due to intake valves carboning up with the DI as there was no fuel behind the valve to wash it clean. This was a serious problem for most manufacturers who switched to DI without enough testing (Audi, BMW in particular). It seems, it's mostly been solved now- or so they tell us. Lexus doesn't think so apparently...
 
We know the Wright R3350 turbo compound engines were around .375 running LOP. ...We also know that the Continental 550 engines can achieve something around .38 with relatively low CR, fixed mag timing and mechanical injection.

What makes the Cont 550 so efficient? Can a Lyc 540 do the same? I stumbled across this video that has a review of the Questair Venture from 1992 (first 14 minutes of the video). The narrator talks of BSFC of 0.368 for the Cont 550 turning 2500 RPM. I presume that was still running fixed mag and mechanical injection.

https://www.youtube.com/watch?v=0WU8hlERnJw

Paul Klusman
 
I've seen .37-.38 mentioned for the IO-550 running mags and LOP but don't know where the data came from. Working some numbers back and forth, I believe Dave Anders is seeing better than that with his high CR pistons, tuned exhaust, EFI and EI on his Lyc IO-360- literally in light aero diesel territory which is .34-.35 at cruise according to data from Austro.

Many folks don't realize that the Otto cycle is theoretically more efficient than the Diesel cycle. Once you get the CRs up close, SI engines can match or nearly match diesels at WOT. Toyota achieved 40% thermal efficiency back in 2016 on some of their SI auto engines and is aiming for 45% by 2020.

In the WAM diesel thread a few years back, stock O-235 RV-9s were matching TAS and FF with the WAM powered one (lbs. of fuel/hr., not gallons). You could do better still with EI and EFI on them.
 
Philosophical question: So let?s say we put all the trick bits onto a Lycoming or Continental: supercharger, tuned exhaust, EFI and EI, custom cam, special oil, ported/polished induction, cold air intake, and we manage to get some kind of 0.34 BSFC out of it. Further we put it on a very efficient airframe with high-aspect ratio wings (i.e. Questair Venture). Put on a trick prop with custom cowl and careful attention to cooling airflow. Put the people on oxygen, fly very high, run LOP and I think we could hope to get 26-28 MPG out of if if with 200+ kts cruise. Perhaps 30-32 MPG with a tailwind.

Is there any kind of market for this?

I recall the NASA ?Green Flight Challenge? from 2011 where some of the aircraft achieved an equivalent 190 MPG with a super-optimized glider airframe/electric motor puttering along at 100 MPH. Impressive but so many years later not too many examples of that sort of aircraft flying today. At least not as traveling machines. As a practical matter the RV series are super-popular with their versatility, even if they only get around 20 MPG on a good day. I?m sure there is some room for absolute performance improvement over the RV4/6/7/8? but is it worth it?

What do you folks think?

Paul Klusman
 
What i think?

RV15, thats what I think. High speed high aspect ratio high wing with super efficient cruise. Big flaps for those who want a bush plane (ala Doug).
 
Philosophical question: So let’s say we put all the trick bits onto a Lycoming or Continental: supercharger, tuned exhaust, EFI and EI, custom cam, special oil, ported/polished induction, cold air intake, and we manage to get some kind of 0.34 BSFC out of it. Further we put it on a very efficient airframe with high-aspect ratio wings (i.e. Questair Venture). Put on a trick prop with custom cowl and careful attention to cooling airflow. Put the people on oxygen, fly very high, run LOP and I think we could hope to get 26-28 MPG out of if if with 200+ kts cruise. Perhaps 30-32 MPG with a tailwind.

Is there any kind of market for this?

I recall the NASA “Green Flight Challenge” from 2011 where some of the aircraft achieved an equivalent 190 MPG with a super-optimized glider airframe/electric motor puttering along at 100 MPH. Impressive but so many years later not too many examples of that sort of aircraft flying today. At least not as traveling machines. As a practical matter the RV series are super-popular with their versatility, even if they only get around 20 MPG on a good day. I’m sure there is some room for absolute performance improvement over the RV4/6/7/8… but is it worth it?

What do you folks think?

Paul Klusman

Dave Anders has already done much better than 30MPG (45.1 mpg at 167 mph) with all the electronic bits fitted to a 10.5 CR Lycoming on a modified RV-4. No doubt you could do even better with a clean sheet design.
 
Last edited:
Philosophical question: So let?s say we put all the trick bits onto a Lycoming or Continental: supercharger, tuned exhaust, EFI and EI, custom cam, special oil, ported/polished induction, cold air intake, and we manage to get some kind of 0.34 BSFC out of it...

...What do you folks think?...

I think this is very much why it has taken me 26+ months to do a "simple" engine overhaul on my Rocket.

...and EFI/EI, a new cowl, cooling system, induction system, etc, etc.

Yes, this is what interests me in E-AB - "mission optimization".
 
Ross;
I noticed the last couple of RV-10 videos you are running up high (16-20k) WOT but setting the prop RPM to about 2400. Reading some other publications some recommend lower RPM for best economy.

Would it cost too much speed to lower the RPM? I am just trying to resolve different information that I have read.
 
This rpm seems to be the smooth, sweet spot for Les' -10 but lower rpm should reduce engine frictional losses somewhat with the effect to reduce power some more too. Speed will certainly be less and fuel burn will be too. It's what the trade off will be that's important which is why I mentioned the savings at the end of the video at the SAME TAS. MPG is likely to be best near best L/D speed but we'll be going pretty slow there and most folks didn't build their planes to go somewhere slowly.

You'll have another variable with propeller blade angle and its effect on efficiency. This could be higher or lower.

I know Dave Anders sometimes flies at much lower rpm for best economy. This could be good if you catch a strong tailwind up high.

Maybe worth another test flight to examine the effects of lower rpm. :)
 
Back
Top