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??

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.

Are you suggesting that reducing detonation potential by going from WOT to part throttle is a fallacy? Please support that assertion with some data please.

Larry
 
I think the "fallacy" highlighted is:

"...Don't change compression because it reduces detonation margin and you will want that margin when you hit a seagull and your CHT runs away..."

Agree, following that train of logic quickly concludes that we should stop flying because its just too unsafe.
 
Last edited:
That's how I understood the comment it too. My point remains that we should be careful how much we eat into safety margin. This thread is specifically discussing detonation margin.

What failure rate is acceptable? Can we tolerate a certain event happening once in 10,000hrs? What if we have 100 different things we're worried about happening once in 10,000hrs? I picked birds as an example because I've hit more than my fair share. Could have picked a clogged fuel filter and high power lean condition. That's happened to me too. I also had a fuel servo go crazy rich and drown the engine in fuel at high power. I can honestly say I've never taken off with cowl plugs installed, but that does happen too. Things break and stuff does go wrong. It's nice to know there's room to spare in safety margins before a bad day turns worse.

I get it, this is a balancing act, we can't fly A340s, modern quad-jets, and be extremely risk adverse. There are also some aircraft running 12:1 pistons on 100LL, and they can't have much detonation margin left either, but they're not crossing oceans with that. In effect, I'm defending the detonation test conditions mentioned above. While it sounds extreme on the surface and nobody would ever intentionally operate their engine under those conditions, it's not difficult to foresee a situation where similar conditions could be inadvertently encountered. Is that a risk you are willing to take? Only you can answer that question, just make sure it's an informed decision.

I'm paying very close attention to this thread because I love the idea of increased efficiency of higher compression and reduced fuel cost of 91 octane. But I'm also cognizant of risk vs reward. What are the failure modes, the chance of the failure occurring and the consequences of the failures? What is the reward? Is it worth it? Not everyone will have the same answer.
 
With any single engined aircraft, there are multiple single points of failure awaiting if your luck has run out...

If it makes you feel any safer, you can specify 7.5 CR pistons to give wider detonation margins on mogas.
 
But that's the magic question. How much detonation margin is left with 91, 93 or 94 octane fuels on 7.5 pistons? 8.5 pistons? 9.0 pistons? There is a standard detonation test and experience shows it offers a reasonable level of safety. From what I've read and understood, nobody has conducted this detonation test with 91, 93 or 94 octane and 9.0 CR. Maybe it will pass the test, and I'd like to know too.
 
You won't know without extensive dyno testing across the whole range of IATs, CHTs, AFRs, RPMs, and ignition timing with various fuel octanes and CRs. Many runs to document all of that and rather a lot of time and dollars.

Beyond the realm for most experimenters and I doubt Lycoming has an interest in doing it either.

Perhaps Mike or Greg will share what they're doing with their engines.
 
Last edited:
You won't know with extensive dyno testing across the whole range of IATs, CHTs, AFRs, RPMs, and ignition timing with various fuel octanes and CRs. Many runs to document all of that and rather a lot of time and dollars.

Oh come on...

The issue is 9.1 CR and 91 octane. No one seems to be proposing unlikely stuff like 10:1 and 87 octane. Lower compression and/or more octane is less critical. So, only one combination to check.

So far, the only objection to testing per the standard is the mandated CHT and oil temperature...i.e the assertion that no reasonable owner with current engine instrumentation will let temps reach maximum. So set some arbitrary CHT and oil temperature believed to be practical, perhaps 425F and 210F. Keep the IAT number at 100F, as it really is common.

Then three runs:

2700/sea level or better MP, basically the same Ron was doing at Lycon, but with a mixture sweep and a piezo or combustion pressure sensor to rank detonation intensity.

Same for 2500/28, simulating the climb out RPM pull which many folks habitually do.

Last, 2400, maybe 2300 for the sixes, and 22", an 8000 ft cruise setup.

If the cruise setting is clean, maybe a fourth pull with lots of advance dialed in for the P-mag faithful.

Look at the time hack on the FAA Hughes plots. Each run is 60 to 70 seconds. Do every run twice, and it's still less than a hour.

I don't buy the idea that a pro dyno operator doesn't have access to detonation sensing. George Braly and others have been doing it a long time. I'll even venture it's now within the realm of smart homebuilders.

Does it still seem like too much trouble? The simple answer is go with 8.5 CR and work on aero improvements for the 2% (or less) efficiency gain.

Break.

For your amusement, this quote from Kevin Cameron, one of my favorite technical writers:

I have sometimes thought of what goes on in that hot end gas as being similar to action in a bar full of keen sports fans. The preflame reactions I described above are the little squabbles that develop between highly argumentative persons. The analog of temperature is the average number of drinks they have had. And the big question is, when, if ever, does simmering argument break down into a brawl?
 
Last edited:
...Perhaps Mike or Greg will share what they're doing with their engines.

Cant speak for Greg but I pull a bunch of timing out (17 degrees at TO power), effective cooling, spot on mixture distribution, and full engine monitoring. I have run my real world "worst" case scenario (well, "bad" anyway) enough times to no longer be concerned. Now I just go fly. Does not matter the air temp, heat soak times, or grade of fuel. Everything just works.
 
I don't buy the idea that a pro dyno operator doesn't have access to detonation sensing. George Braly and others have been doing it a long time. I'll even venture it's now within the realm of smart homebuilders.

I suspect that a lot of trial and error with known detonation is required to find a good place to mount a knock sensor on these engines. Very few rigid flat spots on the head and barrel; Even the cyl base is not flat. I suspect the boys in ADA spent a good amount of time finding a reliable frequency and spot to pick it up. Seems most of the OEM auto makers are putting them on the block somewhere.

Hard to call this in the realm of experimental work. Maybe the previous poster observed where Lycon put their sensor for his dyno runs.

Larry
 
Last edited:
Oh come on...

The issue is 9.1 CR and 91 octane. No one seems to be proposing unlikely stuff like 10:1 and 87 octane. Lower compression and/or more octane is less critical. So, only one combination to check.

Dan, I was answering the question posed by David Z: "How much detonation margin is left with 91, 93 or 94 octane fuels on 7.5 pistons? 8.5 pistons? 9.0 pistons? There is a standard detonation test and experience shows it offers a reasonable level of safety. From what I've read and understood, nobody has conducted this detonation test with 91, 93 or 94 octane and 9.0 CR. Maybe it will pass the test, and I'd like to know too."

My background is engine development, engine building and dyno testing, even built my own engine dyno from scratch. If you want to have a COMPLETE picture of what David asked above, you need to to do the same sort of testing that Lycoming has done and which you were previously advocating for.

If you want to know the effect of CR and detonation margin with all variables including octane, you'd be doing DAYS of testing with 3 different engines.

If you just want to know at 9 to 1 on 91 octane, the task is much simpler. We already have a pretty good idea of what that involves from folks already doing something similar. Michael gave some idea on what works for him.

For only a slight power loss at high MAP, you can run mogas by retarding spark under those conditions. This has the added advantage of reducing CHTs in the climb on a hot day. Some even run LOP in the climb above 5000 feet by pulling the prop back and LOP while flying formation.

Once MAP is low enough, timing can be re-advanced for best efficiency.

I've always advocated to be conservative. Mogas quality is more variable than 100LL. Better to chop a couple more degrees of timing off and suffer a few hp loss than get close to the edge. I think Michael is doing just that and doesn't worry. Mojave isn't a cool place in summer so that's a worst case scenario as far as IATs and CHTs go.

For folks using mags or EIs with non-programmable timing curves, I'd think twice about 9 to 1 pistons, in fact, I'd come out and say don't do it if you plan to run mogas. The cruise compromises resulting from the required ignition retard at high MAP would be noticeable.

Higher CR is useful if you fly high a lot where ambient pressure is low. You want to squeeze that thin air as much as you can up there. Dave Anders was running 12 to 1 at one point and high altitude performance was awesome (100LL). You can limit MAP and timing down low to take advantage up high. Hopefully with the next gen EFI, we'll be able to raise CRs even more and retain detonation margins as they do in the automotive world where we see turbo engines running over 15 psi boost with 10.5 CRs on 87 octane fuel. We won't see those kinds of numbers on a standard Lycoming combustion chamber of course but I believe we'll be able to push CRs a bit higher than what we generally see now.

Ron, Greg and Michael have shared some of their mogas experiences here. It can and is being done already and actually for a very long time (39 years) on many hundreds of aircraft using the Petersen STCs with fixed timing. Nothing new at all like you make it out to be. https://www.autofuelstc.com/approved_engines_airfames.phtml

A point more compression isn't a big deal with variable timing EIs.
 
Last edited:
Back
Top