What's new
Van's Air Force

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

Test stand detonation data??

Your numbers here:

670 * 2700 / 5252 = 344 HP at 2700
550 * 2500 / 5252 = 260 HP at 2500

Sorry, but this data is GROSSLY inconsistent with Lyc engine performance data seen by the rest of us.

Larry

Just because you don't believe it, doesn't mean it's not actual, measured data. Note the engine setup; significantly different than a 540C4B5. Parsing this might be fun, but note that it doesn't distract from the fact you can run 91AKI fuel in a 9:1 piston setup (which also is not OEM for the 540C4B5).

BTW, I determined my parameters, built the engine and am quite happy with the performance.
 
Last edited:
Please see my previous posts showing my dyno testing at LyCon, which includes both power and torque curves. ]

So, what cam did you put in there to make the torque curve so different from a stock Lyc? The stock cam loses 4% of HP from 2700 to 2500 and yours losses almost 25%. Sorry, but it seems to defy logic or at least mine. I have never seen torque drop off so rapidly near what should be peak torque.
 
Last edited:
Just because you don't believe it, doesn't mean it not actual, measured data. Note the engine setup; significantly different than a 540C4B5. Parsing this might be fun, but note that it doesn't distract from the fact you can run 91AKI fuel in a 9:1 piston setup (which also is not OEM for the 540C4B5).

BTW, I determined my parameters, built the engine and am quite happy with the performance.

I'm glad you're happy, but it is not the same as all others and therefore you shouldn't make recommendations to others that have a very different performance profile than yours. Your torque curve and engine set up may support 91, but because it is so different, that can't be applied to the rest of us.
 
Last edited:
I'm glad you're happy, but it is not the same as all others and therefore you shouldn't make recommendations to others that have a very different performance profile than yours. Your torque curve may support 91, but becuase it is so different, that can't be applied to the rest of us.

I disagree with your synopsis of my engine's torque curve, however that doesn't affect the ability of a 9:1 air cooled lycoming engine to run MOGAS.
 
So, what cam did you put in there to make the torque curve so different from a stock Lyc? The stock cam loses 4% of HP from 2700 to 2500 and yours losses almost 25%. Sorry, but it seems to defy logic or at least mine. I have never seen torque drop off so rapidly near what should be peak torque.

Feel free to take this offline and PM me. Happy to discuss Larry.
 
Last edited:
For my engine at full throttle (same manifold pressure set point), 2700 RPM will provide about 670 ft-lbs of torque to the prop. 2500 RPM will provide about 550 ft-lbs of torque. Those are measured data points and not theoretical.

Ron, perhaps a fixed pitch test club, i.e. the 2500 RPM torque was not full throttle?

If a constant speed prop and full throttle at 2500, it says some interesting things about the intake manifold and camshaft. Stock is tuned for cruise. It's not unusual to see just slightly more torque at 2400 than 2700.

Just for fun, I've attached a snip, data from the FAA dyno at Hughes. It's a standard 540K. Column on the right is horsepower. I've backed out torque for each RPM. Within this narrow RPM range, manifold pressure sets torque.
.
 

Attachments

  • 540K.jpg
    540K.jpg
    64.7 KB · Views: 114
Just for fun, I've attached a snip, data from the FAA dyno at Hughes. It's a standard 540K. Column on the right is horsepower. I've backed out torque for each RPM. Within this narrow RPM range, manifold pressure sets torque.
.

That's really interesting data Dan, specifically the influence of manifold pressure on torque. For reference, in the data I provided earlier, we measured 30.2" MP at the intakes ports during the power runs noted. It's the change in port manifold pressure that drives the increase in mass air flow (which is joined by the required increase in fuel flow), hence increase in torque. How fast that changes, or the ramp rate, drives how quickly torque builds or falls off.

A point of trivia for the casual observer, is that dynamometers measure the force applied through twist. The force the engine applies to the sensor produces a ft-lbs measurement, and then horsepower is calculated, vice the other way around (although the math is accurate in both directions).

It might be best to start a new thread and continue this topic of torque and the parameters affecting it.
 
Last edited:
So Ron, the 550 lbs-ft 2500 RPM torque was taken with a fixed pitch test club (i.e. at part throttle) or with a constant speed?
 
So Ron, the 550 lbs-ft 2500 RPM torque was taken with a fixed pitch test club (i.e. at part throttle) or with a constant speed?

Dan, it was with a Hartzell constant speed prop.

Here's a snap shot of a couple runs, one at 30º advance on 100LL and one at 24º running on 91 AKI. Note not only the ramp up in torque and rpm, but MAP, which was measured at the primer port on cylinder #6 for these two runs, both performed at full throttle.

We did something like 30 different runs with different combinations of prop, exhaust fuel and ignition settings. The best power was predictably found in the 100-125º ROP range.

Disregard the BSFC number; due to setting up with the dual ECU SDS, we didn't get around to calibrating that. Fuel flows were taken by differential meters and not the SDS fuel flow.
 

Attachments

  • Dyno Run.png
    Dyno Run.png
    492.7 KB · Views: 109
  • Dyno Run 24º.png
    Dyno Run 24º.png
    201.3 KB · Views: 103
Last edited:
Some folks commenting here are hung up on comparing Ron's results to a stock 540. This isn't stock. In particular the intake manifold and throttle body are significantly different with tapered runners and a larger volume plenum and throttle body with less restriction than typical servos. It also has EFI which offers different droplet size than mechanical injectors.

Add more compression, perhaps different cam timing and ported heads and it can behave quite differently compared to a stock Lycoming.

The folks at Sky Dynamics and Ly-Con have proven track records at the very top of the pile.

The combination that Ron has here really appears to work. The torque curve shows little sign of flattening out at 2700. This combo is breathing very well so VE is high and that makes lots of power.

Assuming the measured torque is accurate here, this is a most impressive PV 540.
 
For my engine at full throttle (same manifold pressure set point), 2700 RPM will provide about 670 ft-lbs of torque to the prop. 2500 RPM will provide about 550 ft-lbs of torque. Those are measured data points and not theoretical.

I can't help wondering if there was a recording error and your torque at 2500 RPM was actually 650 ft-lbs.

In this RPM range, torque curves are usually pretty flat.
 
Dan, it was with a Hartzell constant speed prop.
Here's a snap shot of a couple runs, one at 30º advance on 100LL and one at 24º running on 91 AKI. Note not only the ramp up in torque and rpm, but MAP, which was measured at the primer port on cylinder #6 for these two runs, both performed at full throttle.

Neither run indicates full throttle manifold pressure at 2500. Your 550 lbs-ft claim is at part throttle. Show us one with 30.2" and 2500.

Some folks commenting here are hung up on comparing Ron's results to a stock 540.

No, we're hung up Ron's claim of large torque reduction via a small RPM reduction.
 
For reference, in the data I provided earlier, we measured 30.2" MP at the intakes ports during the power runs noted. It's the change in port manifold pressure that drives the increase in mass air flow (which is joined by the required increase in fuel flow), hence increase in torque. How fast that changes, or the ramp rate, drives how quickly torque builds or falls off.

MAP must be measured near the throttle plate in order to be an indicator of power. Yes, increasing volume and pressure at the port will help the cylinder flow more air. However, it cannot flow more air than what was let into the plenum via the throttle body. Pressure at the port doesn't speak to total airflow and therefore not a power indicator only MAP taken at the TB or plenum can do that.

Most intake chambers suffer flaws that prevent the cylinder from pulling all of the air offered by the plenum and reducing these flaws will change the pressure and therefore volume. But you still can't pull more air than exists in the plenum and that is restricted by the Throttle plate. Therefore MAP measured in the plenum, not the intake port chamber, is the indicator of power or air fuel offered.

I agree with Dan that your 2500 based torque numbers are part throttle and cannot be compared to the 2700 numbers when talking full throttle performance. I suspect your 2500 WOT numbers are MUCH higher. This is why we are struggling to believe your torque numbers at 2500.
'
The data in you chart shows torque still rising at the highest RPM, so the torque peak must be above that. I suspect you have an aftermaket camshaft or someone re-profiled it.

Larry
 
Last edited:
Neither run indicates full throttle manifold pressure at 2500. Your 550 lbs-ft claim is at part throttle. Show us one with 30.2" and 2500.



No, we're hung up Ron's claim of large torque reduction via a small RPM reduction.

I did not make any claims regarding "torque reductions", however, I have provided data from several test stand runs showing the engine at full throttle, which do show the torque the engine was making as it ramped through 2500 RPM. I welcome you refer to those data runs and make any conclusions you choose to. As Ross noted, the peak torque for my engine setup is likely well above 2700 RPM, but we were limited by the propeller and did not explore that further.

I get that you and the other opiners here are not satisfied my engine doesn't comply with Lycoming's tech sheet and that the torque curve doesn't fit your model, however, you're largely ignoring the actual thread topic.
 
Last edited:
I get that you and the other opiners here are not satisfied my engine doesn't comply with Lycoming's tech sheet and that the torque curve doesn't fit your model, however, you're largely ignoring the actual thread topic.

The thread topic is detonation potential in a stock 540 with 9:1 pistons and you offered data to support the OP's question about avoiding detonation. Your engine is far from stock in many regards. Therefore, IMHO, your detonation data is not helpfull to those of us with stock engines. I could offer that I have a small block ford that runs 11.5:1 compression and doesn't detonate on 91 fuel. However, that also wouldn't be helpfull.

Dynamic CR is directly related to detonation potential; Static CR is not. Dynamic CR includes the cams intake closing angle and rod length. how much the air gets compressed cannot be computed without knowing exactly when the intake valve closes. It seems clear you do not have a stock cam and therefore, your detonation experience will likely not apply to the rest of us that have one.

FYI, I have NO dissatisfaction that your engine is different. To the contrary, I am impressed with the performance. Only making the point that the differences make your data not useful to Carl and the others. It could lead them to think their experience will be the same and that can't be expected, as their engines will not be the same as yours.

Larry
 
Last edited:
The thread topic is detonation potential in a stock 540 with 9:1 pistons and you offered data to support the OP's question about avoiding detonation. Your engine is far from stock in many regards. Therefore, IMHO, your detonation data is not helpfull to those of us with stock engines. I could offer that I have a small block ford that runs 11.5:1 compression and doesn't detonate on 91 fuel. However, that also wouldn't be helpfull.

Dynamic CR is directly related to detonation potential; Static CR is not. Dynamic CR includes the cams intake closing angle and rod length. how much the air gets compressed cannot be computed without knowing exactly when the intake valve closes. It seems clear you do not have a stock cam and therefore, your detonation experience will likely not apply to the rest of us that have one.

FYI, I have NO dissatisfaction that your engine is different. To the contrary, I am impressed with the performance. Only making the point that the differences make your data not useful to Carl and the others. It could lead them to think their experience will be the same and that can't be expected, as their engines will not be the same as yours.

Larry
Larry, the OP question was using 9:1 pistons in an IO-540, which is not stock. While much of my engine is also not stock, it does have 9:1 pistons, is essentially an IO-540-X (not a small block Ford, which I agree would be an apple to oranges comparison), and it is normally aspirated, so while I can appreciate your other comments, it is comparable.
 
Last edited:
Ron was good enough publish his data here from the Ly-Con dyno with the aim to demonstrate the detonation margin on his engine running on mogas. He went to some expense and trouble to do these runs to expand his knowledge in this regard. I certainly appreciate that.

He never made any specific claims about torque here, just offered up the data.

If the throttle is transitioning open from 2500 to 2700, MAP will be lower until it's WOT and so likely will be torque.

An observation: Tapered runners with bells can show substantial gains in power as they have reduced separation on the wall and improve velocity at the port. I've seen an 18% flow increase on the flow bench and an 8% power improvement on my dyno over a straight runner of the same length with no bell. (not a Lycoming)

Dave Anders has done additional tweaking of the basic SD manifold on his 360 through pressure measurement at the ports, refining runner shape and length. He's found fairly substantial gains over factory parts. Very interesting stuff and a huge investment of time to learn what works best. Kudos to Dave on that.

Pressure at the port is a better indicator of flow potential across the valve than plenum MAP. This is the true delta. I would expect some of the above ambient pressure here due to prop blast acting on the forward facing TB, it may not all be attributable to the SD intake system.
 
What I’m walking away with

Very good discussion, and data I never had before. Thanks to all for taking the time to respond.

I’m landing on staying with the stock 8.5 pistons. While I’m now more confident that timing and procedural changes can support running 9 pistons on Swift 94 or 93UL ethanol free gas, the data showing just 2% gain in engine efficiency translates to less than a gallon of fuel for a standard four hour hop. So the gain is not worth the trouble.

But - for all you IO-390-EXP119 guys I offer you have a very viable path to not be limited to whatever the 100UL avgas ends up costing.

Carl
 
He never made any specific claims about torque here, just offered up the data.

Actually he did, in post #46, then edited it out...but not before Larry quoted it in post #47, as did I in post #56. The original text was:

For my engine at full throttle (same manifold pressure set point), 2700 RPM will provide about 670 ft-lbs of torque to the prop. 2500 RPM will provide about 550 ft-lbs of torque. Those are measured data points and not theoretical. That's 120 fl-lbs of torque less at 2500 compared to 2700; torque is what is measured - horsepower is calculated.
 
Actually he did, in post #46, then edited it out...but not before Larry quoted it in post #47, as did I in post #56. The original text was:

For my engine at full throttle (same manifold pressure set point), 2700 RPM will provide about 670 ft-lbs of torque to the prop. 2500 RPM will provide about 550 ft-lbs of torque. Those are measured data points and not theoretical. That's 120 fl-lbs of torque less at 2500 compared to 2700; torque is what is measured - horsepower is calculated.

I'm not sure where you're going with this, but this was simply doing the math for Larry's statement from his previous post; I deleted it as I found it argumentative and pointless... pretty much where this thread is going. Most people would likely regard the provided data as torque addition above stock anyway.
 
Last edited:
Larry, the OP question was using 9:1 pistons in an IO-540, which is not stock. While much of my engine is also not stock, it does have 9:1 pistons, is essentially an IO-540-X (not a small block Ford, which I agree would be an apple to oranges comparison), and it is normally aspirated, so while I can appreciate your other comments, it is comparable.

I am stopping my posts here, but you must understand how the cam profile affects dynamic compression before assuming that your 9:1 engine is the same as a 9:1 engine with a stock cam before stating that they are comparable when it comes to detonation potential. Here is where the SBF comes in. With a stock cam and 11.5:1 CR, that engine would detonate constantly at WOT on 91 fuel. However, it has a very aggressive cam profile, with a late intake closing angle, and the dynamic CR is in the low 9's eventhough the static is much higher. I needed the higher static CR with that cam profile to get the dynamic CR into a good range.

I trust that Lycon has very successfully integrated all of thse modifications to maximize power production and from your numbers, it seems they are quite good at what they do. However, it's performance characteristics, including detonation potential are simply not comparable for the rest of us with stock parts.

Larry
 
Last edited:
Most people would likely regard the provided data as torque addition above stock anyway.

This is the point.

To have the torque value change that much over that rpm range seems very very unlikely. It might make sense for a 2-stroke "coming on the pipe".
 
I am stopping my posts here, but you must understand how the cam profile affects dynamic compression before assuming that your 9:1 engine is the same as a 9:1 engine with a stock cam before stating that they are comparable when it comes to detonation potential. Here is where the SBF comes in. With a stock cam and 11.5:1 CR, that engine would detonate constantly at WOT on 91 fuel. However, it has a very aggressive cam profile, with a late intake closing angle, and the dynamic CR is in the low 9's eventhough the static is much higher. I needed the higher static CR with that cam profile to get the dynamic CR into a good range.
Larry

Larry, you're focused on cam timing as the sole influence on dynamic compression ratio, which is fine, assuming you've controlled all of the other parameters. However, you are making assumptions, and then making statements of fact based on them that others may act on cited as credible advice. I agree with much of what you say on the forum, however we are at odds on this one.

There are many other factors that go into developing power in a combustion engine, and that can also affect dynamic compression ratio (DCR) if that's your concern. But let's run your theory through a logic process on my engine which has a static compression ratio (SCR) of 9:1. If cam timing was delayed such that DCR was less than a stock engine at 2700 RPM - which is what you've described above - all other factors being equal, the engine would also have a delayed onset of power, making less power at 2700 RPM, not more, due to not be able to fully utilize the fuel/air mixture - which would actually be backing up in the intake runners in this low RPM band. Additionally, at these low RPM's our direct drive engines run at, it is not possible to have a DCR higher than SCR, so we have to look at other factors which would support making more torque at 2700 RPM and the rate at which torque changes.

To simply look at a dyno curve and say the change in torque developed is related to cam timing doesn't make sense - especially not having a complete picture regarding the internals of the engine and any other differences that may exist. A few factors that I know are influencing power in my engine are intake runner velocity, port configuration, valve grind, ignition timing curve, fuel air mixture quality and density, just to name a few.

I trust that Lycon has very successfully integrated all of thse modifications to maximize power production and from your numbers, it seems they are quite good at what they do. However, it's performance characteristics, including detonation potential are simply not comparable for the rest of us with stock parts.

Larry

In the event my engine eventually blows up, I don't want the good people at Ly-Con to be blamed. They did my head porting and polishing and provided their beautiful diamond coated flat tappets, and then did the dyno testing. I get the blame for the rest of it.

This is the point.

To have the torque value change that much over that rpm range seems very very unlikely. It might make sense for a 2-stroke "coming on the pipe".

Steve...valid, calibrated data has been provided. Based on your tagline, you too are an engineer. Please take a look at all of the parameters in the data provided and not focus on the text of the debate, or just RPM and torque. There's a quote from Sherlock Holmes that applies here.

I don't find the direction of this thread constructive or helpful with regard to running 9:1 pistons on MOGAS at this point. Even more so, arguments that provided engine data doesn't compare to a stock engine, and therefore should be disregarded, ignores that this engine is experiencing stronger combustion events than a stock engine, which allows a simple conclusion that if an engine making more power can run without detonation on 91AKI fuel, then one making less power at a given RPM, having the same cylinder construction, piston compression ratio, and temperature constraints, can too.

The OP, Carl, has said he chose to go with a stock ratio based on a reportedly 6-8Hp increase, due to compression efficiency, not being worth the perceived risk. And although I chose differently, I respect his choice.

Have fun with the data and comments, but I too am done with this thread.

“You pays your money and you takes your choice.” Aldous Huxley
 
Last edited:
Larry, you're focused on cam timing as the sole influence on dynamic compression ratio, which is fine, assuming you've controlled all of the other parameters. However, you are making assumptions, and then making statements of fact based on them that others may act on cited as credible advice. I agree with much of what you say on the forum, however we are at odds on this one.

There are many other factors that go into developing power in a combustion engine, and that can also affect dynamic compression ratio (DCR) if that's your concern. But let's run your theory through a logic process on my engine which has a static compression ratio (SCR) of 9:1. If cam timing was delayed such that DCR was less than a stock engine at 2700 RPM - which is what you've described above - all other factors being equal, the engine would also have a delayed onset of power, making less power at 2700 RPM, not more, due to not be able to fully utilize the fuel/air mixture - which would actually be backing up in the intake runners in this low RPM band. Additionally, at these low RPM's our direct drive engines run at, it is not possible to have a DCR higher than SCR, so we have to look at other factors which would support making more torque at 2700 RPM and the rate at which torque changes.

To simply look at a dyno curve and say the change in torque developed is related to cam timing doesn't make sense - especially not having a complete picture regarding the internals of the engine and any other differences that may exist. A few factors that I know are influencing power in my engine are intake runner velocity, port configuration, valve grind, ignition timing curve, fuel air mixture quality and density, just to name a few.



In the event my engine eventually blows up, I don't want the good people at Ly-Con to be blamed. They did my head porting and polishing and provided their beautiful diamond coated flat tappets, and then did the dyno testing. I get the blame for the rest of it.



Steve...valid, calibrated data has been provided. Based on your tagline, you too are an engineer. Please take a look at all of the parameters in the data provided and not focus on the text of the debate, or just RPM and torque. There's a quote from Sherlock Holmes that applies here.

I don't find the direction of this thread constructive or helpful with regard to running 9:1 pistons on MOGAS at this point. Even more so, arguments that provided engine data doesn't compare to a stock engine, and therefore should be disregarded, ignores that this engine is experiencing stronger combustion events than a stock engine, which allows a simple conclusion that if an engine making more power can run without detonation on 91AKI fuel, then one making less power at a given RPM, having the same cylinder construction, piston compression ratio, and temperature constraints, can too.

The OP, Carl, has said he chose to go with a stock ratio based on a reportedly 6-8Hp increase, due to compression efficiency, not being worth the perceived risk. And although I chose differently, I respect his choice.

Have fun with the data and comments, but I too am done with this thread.

“You pays your money and you takes your choice.” Aldous Huxley

I have only made the assumption that your dynamic CR is different than that of an engine with a stock cam (likely lower) and that will represent differences in detonation potential between your engine and a stock engine. Yes, this IS an assumption, but so is your assumption that your engine will behave the same as a stock configuration. I have not tied anything about the the torque curve to this discussion of detonation potential. That was a different discussion where I questioned the validity of your dyno data. I remain convinced the lower torque seen at 2500 is due to the engine being at part throttle, not WOT, like it was for the 2700 measurements. You will never convince me that your engine actually has 25% less torque at 2500 than 2700. If it does, it was poorly designed and highly doubt Lycon would do that. The only comparative way to measure the output at 2500, would be to go WOT and turn down the governed RPM at the prop. I am assuming you cannot load down an aircraft dyno like one does with a car and therefore a lower prop RPM, with a coarser blade angle, is the only way to do it.

My desire is not to argue with you. You have taken steps to generously help others and I respect that. I only keep harping on this, as I feel it is wrong for others to accept your detonation experience as something they can expect for a stock engine and this thread will live on.

Larry
 
Last edited:
Steve...valid, calibrated data has been provided. Based on your tagline, you too are an engineer. Please take a look at all of the parameters in the data provided and not focus on the text of the debate, or just RPM and torque.

Yes, we did. Below I've re-posted your table from #59. The huge difference in torque between 2500 and 2700 was due to difference in manifold pressure, not RPM, and not cam timing. Note the time hack, 12 to 16 seconds between the initial capture and arrival at WOT. The dyno operator (or his software) was smoothly pushing up throttle and balancing it with load, per SOP.

If the operator had so chosen, he could do the max power run, then increase load (reduce RPM with the prop) to pull RPM down to 2500 without reducing manifold pressure. The result would be actual full throttle torque at 2500...and here it would not be 550 ft-lbs.

I don't find the direction of this thread constructive or helpful with regard to running 9:1 pistons on MOGAS at this point.

Fine, let's return to topic.

You've demonstrated probable freedom from detonation at a single benign operating point (2700, 30", 100 ROP, low temperatures). Although useful to know, it is important to recognize it's nothing like a standard detonation test, and I'm not speaking of the temperatures. I can easily accept the argument that a reasonable operator will never allow maximum CHT. I'm speaking of the standard test protocol, which is setting a stable manifold pressure and RPM, then sweeping the mixture from full rich to detonation onset, or lean roughness, which ever occurs first. If detonation, it is ranked in terms of severity. No one is much concerned with light detonation, but heavy is rapidly destructive.

Previously I asked you to consider what happens if you suffer a partial injector blockage. The answer is "The individual cylinder may tip into detonation due to going lean". That's why the protocol is based on mixture sweep.

Below, a 540-K, standard test. It shows light detonation beginning at about 135 pph, and ran away into destructive levels at about 128.
.
 

Attachments

  • Dyno Run.png
    Dyno Run.png
    268 KB · Views: 72
  • ScreenHunter_1109 Mar. 17 15.53.jpg
    ScreenHunter_1109 Mar. 17 15.53.jpg
    97 KB · Views: 75
The huge difference in torque between 2500 and 2700 was due to difference in manifold pressure.

You've demonstrated probable freedom from detonation at a single benign operating point (2700, 30", 100 ROP, low temperatures).

Yup, some of us already knew that...

2700 rpm (still hadn't reached peak torque) 30", 100 ROP and 420 CHT are not benign conditions for detonation...

The BMEP would be over 180 psi which is fairly impressive for a Lycoming at that rpm.

No sane person with $50K invested in a 540, equipped with modern engine monitors is going to run the CHTs up to 500F. Ron isn't Lycoming and obviously wouldn't be risking his new engine to satisfy your curiosity either... Ron wanted to find out if it was safe to run Mogas under his typical (and other pilots typical) operating conditions. He shared that here.

The EFI injector isn't like a Bendix injector. There are two 40 micron filters on the way to the fuel block and a fine screen with much larger area than the orifices, in the injector itself. Pretty well zero chance of dirt partially blocking an injector. Never seen or heard of that in the 850,000 flight hours our customers have accumulated. In any case, you'd get a high EGT alarm and at high MAP like this, the ECU would have the spark retarded as well.

EFI isn't like mags and a Bendix in many ways.
 
Last edited:
Since I have also run a 540 across the Ly-Con dyno recently, I just thought I'd add another data point to the "dyno methodology" aspect of this discussion.

Essentially stock D4A5 (8.5 CR) with ported heads and SDS EFI:
 

Attachments

  • image.jpg
    image.jpg
    313.7 KB · Views: 88
2700 rpm (still hadn't reached peak torque) 30", 100 ROP and 420 CHT are not benign conditions for detonation...The BMEP would be over 180 psi which is fairly impressive for a Lycoming at that rpm.

You mean like this stock TIO-540-J2BD? Detonation onset is around 364 HP, with BMEP there being 207 psi. MP is 42.5".

BMEP has very little to do with this. But hey, I'll restate and simplify. Ron's single operating condition assures others of nothing outside that condition. Without a mixture sweep, he has no idea how close to the limit it may be.

The EFI injector isn't like a Bendix injector.

It's just an example, so pick anything which makes it go lean.
 

Attachments

  • ScreenHunter_1110 Mar. 19 11.03.jpg
    ScreenHunter_1110 Mar. 19 11.03.jpg
    106.5 KB · Views: 63
Yes, we did. Below I've re-posted your table from #59. The huge difference in torque between 2500 and 2700 was due to difference in manifold pressure, not RPM, and not cam timing. Note the time hack, 12 to 16 seconds between the initial capture and arrival at WOT. The dyno operator (or his software) was smoothly pushing up throttle and balancing it with load, per SOP.

If the operator had so chosen, he could do the max power run, then increase load (reduce RPM with the prop) to pull RPM down to 2500 without reducing manifold pressure. The result would be actual full throttle torque at 2500...and here it would not be 550 ft-lbs.



.

The air-fuel ratio was not constant through the test either - leaning toward best power as RPM is increasing - so that would also explain the rapid rise in torque, along with the increasing MAP.
 
Since I have also run a 540 across the Ly-Con dyno recently, I just thought I'd add another data point to the "dyno methodology" aspect of this discussion.

Essentially stock D4A5 (8.5 CR) with ported heads and SDS EFI:

Fairly close to some Sky Dynamics torque figures I've seen for PV 540 builds.

Ron sent me a photo of the dyno setup. This was a load cell connected to a lever arm. Prop torque reacts the whole engine on this arm. Calibration is by test weights hung on the arm pivot point to the load cell. Dead simple and ought to be pretty foolproof.
 
The air-fuel ratio was not constant through the test either - leaning toward best power as RPM is increasing - so that would also explain the rapid rise in torque, along with the increasing MAP.

Yes, a standard dyno pull for max power. Juggle MP and load and when stable at the desired settings (here WOT and 2700), tweak mixture. As we've seen before, torque response to mixture is pretty flat in the best power range

The "TIMING TO 30" notation is also interesting, but I'll assume someone just forgot to change the note, i.e. there were also some 100LL runs with advanced timing.

To both points, and to the notion that this 540's torque curve was still rising at 2700, I'm attaching a dyno pull previously published by Sky Dynamics, for a equally tweaked 360. Note the typical torque curve, even with cam, porting, etc. Max torque ballparks around 2400, and remains near 2400 regardless of ignition timing.

Again, let's not fixate too much on questionable assertions. Ron has demonstrated a 540 PV can be run at 2700/30", with mixture 100 ROP and good CHT control. Thank you, good to know.
.
 

Attachments

  • SD Pull 25 Degrees.jpg
    SD Pull 25 Degrees.jpg
    52.3 KB · Views: 93
You mean like this stock TIO-540-J2BD? Detonation onset is around 364 HP, with BMEP there being 207 psi. MP is 42.5".

BMEP has very little to do with this. But hey, I'll restate and simplify. Ron's single operating condition assures others of nothing outside that condition. Without a mixture sweep, he has no idea how close to the limit it may be.

It's just an example, so pick anything which makes it go lean.

Detonation is related more to torque than hp. BMEP is highest at peak torque. If you look at data from Ricardo and many others, they speak of detonation onset in relation to BMEP as this is applicable to all engines and makes for easy comparison between types of fuels on the same engine type. AFR and timing have major effects on detonation tendency.

I submit that your data from a turbocharged AV 540 has even less to do with the atmo PV 540s that most folks fly here on VAF...

The chart simply confirms what we already know: High MAP and leaner than 12 to 1 AFR is bad with fixed timing and likely high IATs (turbo) and CHTs, so we just don't do that.

200ish BMEP would be very impressive for an atmo 540, not so much with a turbocharged one where VE can easily exceed 100%.

Do bonehead stuff like running at peak to 50 ROP, 500 CHT, 30 BTDC timing and 30" MAP on a hot day running MOGAS and you're GOING to be replacing broken parts.

Mike and Greg have been flying PV engines on Mogas for some time with no issues as they simply avoid the conditions which are known to cause detonation. It's that simple.
 
Ron sent me a photo of the dyno setup. This was a load cell connected to a lever arm. Prop torque reacts the whole engine on this arm. Calibration is by test weights hung on the arm pivot point to the load cell. Dead simple and ought to be pretty foolproof.

Common setup these days...Sky Dynamics, Lycon, the M14 cell at Barrett, and the cells at GAMI all operate this way. The mechanical part is easy enough, but the math to convert the oscillating load cell signal to mean brake torque is not so dead simple. Can't merely log the peak of the output waveform, or even assume it to be a sine wave.
 
The air-fuel ratio was not constant through the test either - leaning toward best power as RPM is increasing - so that would also explain the rapid rise in torque, along with the increasing MAP.

Bingo. Yahtzee. ;) And don't forget ignition timing.

The data I provided was never intended to be a comprehensive test designed to replicate Lycoming's certification process, only to provide some requested test stand data showing an IO-540-D (or C) can run on lower octane fuel without detonation. I achieved that.

I have only made the assumption that your dynamic CR is different than that of an engine with a stock cam (likely lower) and that will represent differences in detonation potential between your engine and a stock engine. Yes, this IS an assumption, but so is your assumption that your engine will behave the same as a stock configuration.

The only significant differences with my engine are the intake, modern EFII (SDS) and porting of intake and exhaust runners on the heads - it breaths a lot better. I'd talk about the custom exhaust as well, but that's not included in Lycoming's certificate. So yes, that is a basic assumption I've made, that my IO-540-D series engine will behave similar to another one - especially when considering detonation...more on that in a minute.

This isn't my first engine. I've built, owned, and maintained numerous other personally owned internal combustion engines (although mostly Continental, and a lot of Fords as I think about it). Even more so, I've made a career of turning solid, liquid and gaseous fuels into power for over four decades - from splitting atoms with neutrons to burning bunker 'C', including just about every form of kerosene and naphtha running reciprocating engines that have piston cylinders large enough for a human to stand in, and yes, even AVGAS...but also solar and wind farms, large scale battery storage, hydro power dams, pump water storage and even a good ol' wood burner (only made 31MW, each hand crafted).

While I sincerely appreciate your time to provide me explanations to persuade me that my engine is not comparable to the OP's example engine, I have a great deal of experience that says otherwise, and the greatest argument is at the mercy of experience.

Fine, let's return to topic.

I thought we were actually done with this, but here we are. Glad to see we're back to discussing detonation.

You've demonstrated probable freedom from detonation at a single benign operating point (2700, 30", 100 ROP, low temperatures). Although useful to know, it is important to recognize it's nothing like a standard detonation test, and I'm not speaking of the temperatures. I can easily accept the argument that a reasonable operator will never allow maximum CHT. I'm speaking of the standard test protocol, which is setting a stable manifold pressure and RPM, then sweeping the mixture from full rich to detonation onset, or lean roughness, which ever occurs first. If detonation, it is ranked in terms of severity. No one is much concerned with light detonation, but heavy is rapidly destructive.

Previously I asked you to consider what happens if you suffer a partial injector blockage. The answer is "The individual cylinder may tip into detonation due to going lean". That's why the protocol is based on mixture sweep.

Below, a 540-K, standard test. It shows light detonation beginning at about 135 pph, and ran away into destructive levels at about 128.
.


You mean like this stock TIO-540-J2BD? Detonation onset is around 364 HP, with BMEP there being 207 psi. MP is 42.5".

BMEP has very little to do with this. But hey, I'll restate and simplify. Ron's single operating condition assures others of nothing outside that condition. Without a mixture sweep, he has no idea how close to the limit it may be.

It's just an example, so pick anything which makes it go lean.

Dan, thanks for providing your examples. In them, you've provided a K series engine, which is an angle valve model, and a J series engine - turbocharged no less, as comparative examples of how detonation can occur in an IO-540-D series engine. While others may disagree with your use of another model of engine as an example - I agree with your examples, mostly because detonation doesn't care about what internal combustion engine the fuel is contained within - or even an engine at all. Detonation is simply an unplanned combustion that resulted from meeting the needed pressure and heat conditions. And of course having fuel available.

In both examples you've shown that if you experience a failure in your injection system, it can result in an engine failure. And I wholeheartedly agree, this is why it is important for a builder to include full telemetry on their engine, and then use it. Especially if you're going to experiment with non-stock modifications. I believe that any pilot that is competent enough to manage their aircraft controls, should be competent enough to properly operate their engine. Not properly operating the engine is a problem, but then again so is turning your aircraft into a lawn dart if you forget to pull back on the control stick.

While you have referred to the Lycoming test standard and provided the importance of cylinder heat temperatures (CHT) and intake air temperature (IAT) to be tested at, you did not provide either CHT or IAT in your two examples where detonation occurred. Please update your charts to show these two trends so that we can see how temperature affected the events.

You've said that my data is problematic in that I did not include a full sweep of air to fuel mixture implying a failure can occur in the conditions I tested at, so please provide supporting documenation for your statement. Please include a test or study that shows an IO-540-D or C series engine, having 9:1 compression, CHT limited to 420ºF and standard ignition timing of 25º advance that demonstrates detonation. It must be shown to occur while running the OP's fuel requirement of 93UL or Swift 94UL, a Motor Octane Number rating (MON). You can use manifold pressure up to 31"Hg and any intake air temperature you think is plausible with AFR sweeps from, say 12.5 to 17. Or if you don't have that, an IO-540 running on lower octane MOGAS, such as 91 Anti-Knock Index octane that has limited power, i.e. MP and ignition advance reduced below 26" and 25º, respectively, while fuel mixture is not maintained between 100º ROP to 20ºLOP will work as well.

If you don't have the studies, this is all provable with math; I'll accept your mathematical analysis of this, as long as you show all of your work.

And Dan, unlike your instructions to me, note I said "please".
 
Last edited:
The "TIMING TO 30" notation is also interesting, but I'll assume someone just forgot to change the note, i.e. there were also some 100LL runs with advanced timing.

To both points, and to the notion that this 540's torque curve was still rising at 2700, I'm attaching a dyno pull previously published by Sky Dynamics, for a equally tweaked 360. Note the typical torque curve, even with cam, porting, etc. Max torque ballparks around 2400, and remains near 2400 regardless of ignition timing.

Again, let's not fixate too much on questionable assertions. Ron has demonstrated a 540 PV can be run at 2700/30", with mixture 100 ROP and good CHT control. Thank you, good to know.
.
You're welcome, however in spite of saying you're moving on, you continue to fixate on the assertions that you and others have made about torque and RPM, which are a distraction from the topic of detonation. The data was provided; you can accept it, analyze it or disregard it - your choice. But don't break out specific numbers without taking into account all of the affecting parameters.

The timing notation on the test run you're referring to is not an error. We did about 30 different test runs using timing from 24-30º. I previously posted that we did limited runs on 91AKI fuel at 24º advance, and the power achieved, at 30.2" MP.

Ken offered that we could go on up to 38º while running 100LL, but I declined based on not expecting to operate at those advances at such high manifold pressure (and the potential resulting cylinder damage and/or failure). I've witnessed a Lycoming at 42º, producing significant power improvements, but the engine was also turning at about 4,000 RPM. I do anticipate running around 38º ignition advance at altitude, however I would not do that on 91AKI fuel; I would on 94UL-MON though.
 
Last edited:
Do bonehead stuff like running at peak to 50 ROP, 500 CHT, 30 BTDC timing and 30" MAP on a hot day running MOGAS and you're GOING to be replacing broken parts.

Mike and Greg have been flying PV engines on Mogas for some time with no issues as they simply avoid the conditions which are known to cause detonation. It's that simple.

Well said. Excellent point Ross.
 
Dan, thanks for providing your examples. In them, you've provided a K series engine, which is an angle valve model, and a J series engine - turbocharged no less, as comparative examples of how detonation can occur in an IO-540-D series engine.

No. The 540K was presented as a graphic example of a mixture sweep. The J was offered to illustrate that BMEP alone doesn't tell us much.

In both examples you've shown that if you experience a failure in your injection system, it can result in an engine failure. And I wholeheartedly agree, this is why it is important for a builder to include full telemetry on their engine, and then use it.

Let's remember (as the old joke goes) roughly 90% of the pilot population ranks themselves above average. Point is, a little margin for error is a good thing.

While you have referred to the Lycoming test standard and provided the importance of cylinder heat temperatures (CHT) and intake air temperature (IAT) to be tested at, you did not provide either CHT or IAT in your two examples where detonation occurred.

Sorry, should have marked the source. FAA Hughes, i.e. data per standard protocol...one cylinder at max CHT, all other within 50F, 100+F IAT, oil temp at max.

You've said that my data is problematic in that I did not include a full sweep of air to fuel mixture implying a failure can occur in the conditions I tested at...

No, I'm implying there is a strong potential for detonation outside the tested conditions. The purpose of the mixture sweep to to determine how far outside, and severity.

The timing notation on the test run you're referring to is not an error. We did about 30 different test runs using timing from 24-30º. I previously posted that we did limited runs on 91AKI fuel at 24º advance, and the power achieved, at 30.2" MP.

So the data table you posted (#59) is not a 91 octane run?
 
Let's remember (as the old joke goes) roughly 90% of the pilot population ranks themselves above average. Point is, a little margin for error is a good thing.

While I appreciate the humor in that, and have long held that I'm doing my part to keep the average down... as you've provided, detonation can occur in an operating certified engine - very quickly. Not properly operating your engine, or your airplane, can cause safety margins to erode nearly instantly.

Attempting to dissuade people from pursuing changes from the norm based on a premise that you may make a mistake has no guarantee that you won't experience failures anyway, as your previous examples show.

There are three commonly accepted methods to address safety, the first is through engineering - which is through Lycoming's design, and well thought out modifications builders (experimenters) perform. The second is administrative controls, which is what you are doing, knowing or unknowingly, by advocating restricting the use of lower octane fuel while identifying the manufacturer's fuel recommendation, testing and concern for ensuring a larger margin from detonation. It is worth noting that 100LL wasn't the initial design fuel for the IO-540, but rather 100/130, although you could also use 110/145, which was intended for the supercharged engines of yesteryear. I am going further and recommend another administrative control to allow lower octane unleaded fuel. The third method - well, carry a fire extinguisher, and maybe a parachute.

I'm saying unequivocally that you can operate a IO-540-D or C series based engine on MOGAS with constraints, and certainly 94UL MON fuel. Does operating on MOGAS have a smaller margin for detonation when compared to 100LL? Yes. Can it be done with good controls? Also yes.

No, I'm implying there is a strong potential for detonation outside the tested conditions. The purpose of the mixture sweep to to determine how far outside, and severity.

Dan, the problem with your implication is that it creates doubt in the reader's mind without actually showing proof that detonation will occur within the conditions I've provided - when just the opposite is possible, and in my experience, true. I've stated that the engine ran fine and showed no evidence of detonation on post testing inspection, which included about 30 test runs using mostly 100LL, but also 91AKI fuel.

We did vary injector loading significantly, aka "mixture sweep", which changes AFR over the course of the tests, as was noted by Steve in an earlier post. I just don't have all of the charts and documented data that you want to post here, but then again, you haven't presented data demonstrating detonation occurs under those same operating constraints.

In my previous post, I provided recommended constraints for running MOGAS on a 9:1 compression engine, but also caution that if one will operate LOP (which I do), the move from ROP to through the peak AFR zone to LOP must occur quickly. This is referred to as the Big Mixture Pull by the LOP crowd. If possible, I recommend doing so at or below 65% power.

So the data table you posted (#59) is not a 91 octane run?

I provided two tables and described both of them in that same post. Note the 30º and 24º ignition timings on the tables...the same timings you previously said you assumed were in error. Again, it's not a comprehensive FAA commissioned study on engine operation, simply a few data points of the requested test stand data to show operation can occur without detonation.

Summarizing, operating on MOGAS will have a smaller margin from detonation when compared to 100LL. It be can be done well with the proper setup and operation of the engine defined in this thread.
 
Last edited:
I'm saying unequivocally that you can operate a IO-540-D or C series based engine on MOGAS with constraints, and certainly 94UL MON fuel. Does operating on MOGAS have a smaller margin for detonation when compared to 100LL? Yes. Can it be done with good controls? Also yes.

Agree.

Dan, the problem with your implication is that it creates doubt in the reader's mind without actually showing proof that detonation will occur within the conditions I've provided

No. Again, and as above, I happily accept your finding of no detonation within the conditions provided.

I provided two tables and described both of them in that same post.

Ahh, I missed the 100LL notation for the 30 advance table. Thank you for pointing it out.

Summarizing, operating on MOGAS will have a smaller margin from detonation when compared to 100LL. It be can be done well with the proper setup and operation of the engine defined in this thread.

Again, agree. The only real difference in our respective opinions is that I think the amount of margin should be better defined before calling 9:1 and 91 a safe combination for the community at large.
 
Again, agree. The only real difference in our respective opinions is that I think the amount of margin should be better defined before calling 9:1 and 91 a safe combination for the community at large.

That's a fair statement, and I appreciate your advocating for the community at large.

A great way to advocate is providing additional knowledge - and knowledge is power. In this case, the power to make an informed decision. With the understanding that the OP was trying to make a decision on whether to use 9:1 or the stock 8.5:1 compression pistons on either 93 or 94UL, which is likely to become a much more prevalent fuel (Swift, G100UL, or just basic 100LL without the lead, which is essentially 94UL MON), a demonstration showing that constrained operation can be performed on 91 AKI, which is has a MON rating of about 88+/-2 (lots of affecting variables), should give some comfort to a builder considering this modification - that continued operation on 94 MON fuel is very achievable. I would be disappointed to learn that a builder was dissuaded based on a perception that their interests are not achievable, and to later find out it was entirely possible and had a missed opportunity for greater engine efficiency and performance.

If folks haven't seen this, there is a Service Instruction authorized by Lycoming for a long list of engines, which does include operation on 93AKI - specifically for IO-540-D engines, which I presume would have been tested using Lycoming's worst case scenario testing protocol. Notably, the 91 AKI column is blank, however, there are also STC's available to run 91AKI for some of those same engines (also noted are that STC's are not required for experimental aircraft, but is another reference point).

If you're not operating your engine "like a bonehead" (love that one), then you should have have no issue operating the above engine on 94UL, and can actually use MOGAS such as 93 AKI - per SI 1070AB, or even 91 AKI with constraints, albeit at the caution of the builder.

I'll leave you with one last "pot stirring" thought; you can run 12:1 compression, or an engine with significant manifold pressures, some at or over 42" with turbo or super charged engines on the same 100LL fuel. Been there, done that. Folks in the racing community still do it all the time - and still have considerable margin to detonation remaining. There's a LOT of margin built in - and that's great, but in reality, we're only talking about 1/2 a point of compression increase for the example engine - and that's not a lot, and can easily be compensated for by another point of octane (not a linear relationship, but close enough). I posit that if you can pass Lycoming's certification test on 93AKI fuel with 8.5:1 compression, you will have no issue running 9:1 on 94 MON.
 
I would be disappointed to learn that a builder was dissuaded based on a perception that their interests are not achievable, and to later find out it was entirely possible and had a missed opportunity for greater engine efficiency and performance.

Do you agree with a previous post estimating the difference at about 2%?
 
My first engine was stock due to concerns about 100LL going away, that was 20 years ago.
My new engine 2 years ago, 9.6CR, I'm pretty sure at this point 100LL or equivalent will be around long past my demise, the extra power is sweet.
 
The factory Lycoming tests are for worst scenario conditions and were developed when a single analog CHT and EGT probes were common. In this past era, pilots may not have had a good idea where the engine was operating as far as mixture and CHT limits.

With the typical Garmin and Dynon glass today, pilots have equal engine information available as compared to most dyno cells. There is no excuse to be probing the dangerous edges of engine operation these days and pilots are more educated about safe and efficient engine operation than they were 25 years ago.

Monitoring has changed and engine controls have changed if you are running programmable EFI/EI as well, giving more flexibility with regard to efficiently and safely running lower octane fuels.
 
Instrumentation won't solve problems that occurs enroute. For example, bird smacks into the cooling air intake and those cylinders shoot up to 500°. It would be nice to be comfortable knowing the engine won't be detonating to destruction. Granted we'll have the power pulled back and running very rich while looking for a runway.

I do agree we can cut into the margins a bit with better instrumentation, but where do we stop? Unexpected things happen where we might wish we had lots of safety margin.
 
Instrumentation won't solve problems that occurs enroute. For example, bird smacks into the cooling air intake and those cylinders shoot up to 500°. It would be nice to be comfortable knowing the engine won't be detonating to destruction. Granted we'll have the power pulled back and running very rich while looking for a runway.

It won't be detonating with the power pulled back running full rich and the instrumentation will inform you it's running hot. We're talking about the 99.999% operational aspect here, not what you'll be doing the other .001% of the time after a bird strike to the cooling air inlet.

Anything over 420F and you should be reducing power and/or lowering the nose or richening if you can, if you want max engine life and minimal chance of detonation.
 
Instrumentation won't solve problems that occurs enroute. For example, bird smacks into the cooling air intake and those cylinders shoot up to 500°. It would be nice to be comfortable knowing the engine won't be detonating to destruction. Granted we'll have the power pulled back and running very rich while looking for a runway.

I do agree we can cut into the margins a bit with better instrumentation, but where do we stop? Unexpected things happen where we might wish we had lots of safety margin.

You solve the "detonation to destruction" problem when you pull the throttle back, even if you were running 87.

This is kind of a slippery-slope fallacy though.
 
I took a bird in a cooling air intake. It was a sparrow, so had no effect on the engine. Also a Piper Navajo, so twice the chance of it happening. Luckily it wasn't the seagull I took in the prop a 18 months ago. Personal record is 3 bird strikes in one flight, went through a flock and had 3 small blood stains to clean off the plane including 1 still wedged in the nose gear.

Yes we should keep the cylinders below 420, but sometimes things break, stuff happens and we don't have a choice.
 
I took a bird in a cooling air intake. It was a sparrow, so had no effect on the engine. Also a Piper Navajo, so twice the chance of it happening. Luckily it wasn't the seagull I took in the prop a 18 months ago. Personal record is 3 bird strikes in one flight, went through a flock and had 3 small blood stains to clean off the plane including 1 still wedged in the nose gear.

Yes we should keep the cylinders below 420, but sometimes things break, stuff happens and we don't have a choice.

Holy cow! That sucks!

I had a circling group of buzzards move over top of me in the pattern the other day. Grateful for my skylight where I saw them and then boogied on to another airport.
 
Looking forward to this answer as well for going from 8.5 to 9 pistons.

Carl

From Design and Tuning of Competition Engines, Phillip H Smith. Pg 23 -24

Air Standard Efficiency
Based on the foregoing we can calculate a useful efficiency figure for what might be called a " perfect " engine. This figure is known as Air Standard Efficiency, or a.s.e. ....

In calculating a.s.e. we assume that both induction and exhaust strokes take place at atmospheric pressure...

Under those conditions the a.s.e is given by the formula:

a.s.e. = 1 - (1/r) ^ 0.4
where r is the compression ratio
and ^ is to-the-power-of

which leads to an a.s.e
of 58.4% at 9:1 compression ratio
and 57.5% at 8.5:1 compression ratio

So the increase in efficiency would be (58.4 – 57.5)/57.5 = 0.0156 – or 1.6%

While one can argue that the “perfect” engine does not represent reality, that difference should not appreciably invalidate the conclusion of how much efficiency is gained from this 6% increase in compression ratio.

Originally I found a calculator online that gave the 2% estimate. This time I went back to the foundational math.

Entertainingly I found a typo in the reference book on page 24 - the calculation for 9:1 is wrong in the table.

-Bryan
 
Back
Top