What's new
Van's Air Force

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

Detonation w/Pmags

Walt

Well Known Member
Have a friend that is running an IO-540 angle valve with 10-1 pistons in a Glasair III. Brand new engine from Lycon with about 20 hrs. He installed dual Pmags with a switch to keep them in fixed timing for break in. He decided to try the advanced position on a cross country (against my warnings of operating that engine with advanced timing), at about 8K ft he started leaning for cruise, during the process he noticed #4 CHT started a rapid climb, he reduced power but by that time it was too late. Low compression on #4, borescoped and saw evidence of detonation on most of the pistons. All the cyl are now removed and visible signs damage to all pistons.

He initially called asking me to get him a new CHT probe :eek:

Just a warning to those running HC pistons and Pmags, advanced timing with lean mixtures and you're in dangerous territory,
 
Last edited:
Rule #4 of submarine work, you have to be smarter than the equipment you operate…..

The six cylinder pMag must be set during install for:
- Base timing (normally 20 degrees for angle head, 25 degrees for parallel valve).
- Max timing advance (I recommend no more than 9 degrees, but higher can be set).

The system does come with a switch to fix the timing at the base - a requirement for this pMag to be certified for the GA fleet.

But - a 10-1 piston IO-540 is not a standard parallel valve engine. He should have at least reduce the timing to the standard angle head setting, and carefully worked up to the max advance timing during flight testing.

What was the max timing set for this engine??

Carl
 
This is interesting and I feel bad for your buddy and his engine.

I've been arguing on this site and one other site for years now about the need for a cockpit display when running electronic ignitions. Something that shows the timing advance in real time. I had one 20 years ago when I bought my RV-3 with a Ted Rose ElectroAir system and the only unit on the market now that I am aware of is the one from Canada - SDS.

It goes back to way professionals measure risk in two ways: what is the likelihood of it happening and then what are the consequences if it does happen.
 
Rule #4 of submarine work, you have to be smarter than the equipment you operate…..

The six cylinder pMag must be set during install for:
- Base timing (normally 20 degrees for angle head, 25 degrees for parallel valve).
- Max timing advance (I recommend no more than 9 degrees, but higher can be set).

The system does come with a switch to fix the timing at the base - a requirement for this pMag to be certified for the GA fleet.

But - a 10-1 piston IO-540 is not a standard parallel valve engine. He should have at least reduce the timing to the standard angle head setting, and carefully worked up to the max advance timing during flight testing.

What was the max timing set for this engine??

Carl

Angle valve engine, base was set at 20 and max advance I believe set to 9 degrees.
Showing the timing likely won't help you, by the time you realize it's happening it's too late.
 
Last edited:
I had one 20 years ago when I bought my RV-3 with a Ted Rose ElectroAir system and the only unit on the market now that I am aware of is the one from Canada - SDS.
.
PMAGs with EIC will also display the advance in real time or you can change the advance in real time. And I am not sure if that would have helped the OP since the advance was expected and he had chosen it by turning the switch off.
 
Not sure the EI Commander works with the 6 cylinder pMags.

Highly recommend the EIC for the 4 cylinder engines though.
 
This is interesting and I feel bad for your buddy and his engine.

I've been arguing on this site and one other site for years now about the need for a cockpit display when running electronic ignitions. Something that shows the timing advance in real time. I had one 20 years ago when I bought my RV-3 with a Ted Rose ElectroAir system and the only unit on the market now that I am aware of is the one from Canada - SDS.

It goes back to way professionals measure risk in two ways: what is the likelihood of it happening and then what are the consequences if it does happen.

Lightspeed has the provision to display advance as well. But agree that it wouldn't have helped this fellow.
 
This is a curious topic. Has testing been published anywhere regarding advance levels on high compression engines? I’m sure an engine hot rodder has done it but I haven’t seen it.
 
This is a curious topic. Has testing been published anywhere regarding advance levels on high compression engines? I’m sure an engine hot rodder has done it but I haven’t seen it.

Continental did STC testing on the engine I have which is the parallel valve 370 with 9.6 CR. Base timing is spec'd at 20 deg which is what I run (with Bendix mags) because I'm too chicken to go down the advanced timing road to save a little gas.

I used to run hot rods back in the muscle car days with 12.5 CR and I know if you advance the timing enough it will detonate. However back then with distributors it was easy to advance the timing until you heard it start to 'ping' then back it off just enough to eliminate it. Can't do that with the Lyc.
 
Angle valve engine, base was set at 20 and max advance I believe set to 9 degrees.

Walt, is it possible to confirm the 9 degree figure, and the MP/RPM/CHT at which the subject began leaning?

This is a curious topic. Has testing been published anywhere regarding advance levels on high compression engines?

...high compression angle valve engines..

The angle valve IO-540 K is the NA Lycoming used for detonation studies. Given enough heat and MP it can be pushed into detonation at will, with the stock 8.7 compression ratio and stock timing. Not specific to 10:1 CR, but there is a huge quantity of data running typical and minimum spec 100LL in the FAA's Swift fuel study. Search AR0853.
 
Walt, is it possible to confirm the 9 degree figure, and the MP/RPM/CHT at which the subject began leaning?

I checked with him, max was set to 29 deg, full throttle/2400 rpm 8000 ft.
CHT around 375 until the event which then went to around 460-470 from his recollection.
 
Last edited:
I checked with him, max was set to 29 deg, full throttle/2400 rpm 8000 ft.
CHT under 400 until the event which then went to around 460-470 from his recollection.

Too much timing for this engine and CR running ROP and even LOP. Results predictable.
 
I always set fixed timing. Especially with an angle valve. I am not looking to save 1 gallon per hour. I am looking for effortless hot starts while avoiding mechanical impulse couplers. I am way too slow to make ignition timing theory a full time hobby.
 
FWIW, the 6-cylinder's current Mode Switch function is poorly considered.

From the manual:

Start engine with Mode Switch set to VAR (plugs will not fire at cranking speeds in FIX Mode).

It means every operating cycle begins with the ignition in advancing mode. If a conservative fixed timing mode is desired, the pilot must remember to change the switch position to fixed mode after start. Humans being human, it's not going to happen 100% of the time.

I suspect the fixed mode requirement, if demanded by certification, was treated as an unnecessary ornament. That's obviously not true, give the Glasair example illustrated here. We'll probably see a fixed timing mode switch on the 4-cyl versions in due course. If so, let's hope they change the switching scheme to "one or the other", all the time.

FWIW, my own ignition is configured with fixed timing mode as the default, as the mode switch is located in a row with the aux pump, lights, etc. Advancing mode is enabled by moving the switch to a physical position perceived as an obvious ON, in accord with all the other switches in the row. And yes, it starts in either mode.
 
Last edited:
It means every operating cycle begins with the ignition in advancing mode. If a conservative fixed timing mode is desired, the pilot must remember to change the switch position to fixed mode after start. Humans being human, it's not going to happen 100% of the time.

Perhaps that could be mitigated with a DPDT start switch, one side of which engages the starter and the other side of which switches the PMag between VAR mode and FIX mode?
 
Great discussion, to which I'd like to add a couple questions for Walt, Ross, Michael, Dan and anyone that may have data/experience to add, plus some experience from my ROP/LOP testing. No intent to hijack, and hope this is value added.

Q's:

- Walt, do you have MP numbers for the test? I know you said WOT at about 8,000' in response to Dan's query, but knowing the MP would be a good data point. Seeing the trend and rate of increase in CHT vs Fuel Flow would be good too.

- Does a max increase of 9 on a base of 20 in the P-mag mean 29 is the max advance? (I know that may sound like a dumb question, but I also know the P-mag curves are proprietary, and we may not know where his timing actually was...that's not a shot at P-mags at all, just a data question).

- Does your friend by any chance have an O2 sensor installed, with a lambda or AFR readout? More good data if available.

- What type of spark plugs is your friend running, auto or aircraft? Is the heat range of the plugs correct?

- What type of fuel was he running? 100LL hopefully, but want to be sure it wasn't MoGas.

- Walt, Ross and Michael, you guys indicate this was almost an expected result. Is that because it's an angle valve engine with 10:1? Or because it's an AV/10:1 with Pmags? Interested in why you feel this way, and if it has applicability to parallel valve motors with 10:1.

Here's my semi-selfish reasons for asking all this (along with trying to add to the discussion here, and help find root causes of a damaged engine) ;)

Background: I have a Lycon IO-540 CB45 (narrow deck, parallel valve) with 10:1 CR. One mag (I've had both Bendix and Slick over the years), one Electroair EI. I run REM37BY massive electrode plugs on all cylinders, gapped for the mag on the bottom, and the EI on the top. Much of my testing was done with the mag timed at 25 and the EI at stock delivery setting (which is about 23). I have an EI timing readout and an AFR readout. My engine monitor is an older VM-1000, with no data logging, so LOP testing and injector balancing is quite the 20th century analog/slide-rule-ish process, with lot's of hand-written log sheets transferred into spreadsheets along the way.

I've done quite a bit of LOP testing and injector balancing over the past 10 years, most of which was done on X-C flights up at 10.5K to 13.5K. I've always been concerned about detonation "red zone", especially with my 10:1 engine, so higher altitude testing has been my norm (its where I like to cruise on X-C's too). Typical power settings have been WOT/2100 to WOT/2300 (2200-2300 is where most of my data is). WOT at those altitudes has been 19-21 or 22" MAP. With the mag at 25 and the EI MAP sensor on, and EI advance reading in the mid to high 30's, I'd never seen runaways or CHTs over 400 during any of the tests. I've felt that I had safe margins from detonation. Cylinders 2 and 5 used to peak first, and run a bit hotter than the rest (CHT), but over time, I made injector restrictor body changes that balanced things out nicely.

Flash forward to now, after an upgrade from a Bendix RSA-5 to an AFP FM300B fuel injection servo (a couple years ago). I hadn't done any recent LOP tests, and Steve Smith and I discussed that at Reno this year, as we chased cylinder cooling with a new nitrous system and water spray system installed. Our goal was to adjust external water spray to keep 1, 2, 5 and 6 cool, and bring up 3 and 4, while I slowly leaned towards best power for racing with nitrous (about 11.8 to 12.0 AFR). We just about had it nailed when my fuel pressure transducer failed on Sunday taxiing out...didn't get to test our last adjustment. 2 was the last cylinder to stay hot, and we didn't know if it was the governor blocking air, another airflow issue, a water issue, or a leaning issue. So Steve encouraged me to do some LOP testing on the way home to check that aspect, which I did.

While I was looking to see how 2 was acting on those recent post-Reno '21 tests, it seemed to fall back into line, and 5 became my hot spot. On one test, at 11,500/21.5"/2300RPM, it slowly crept to 425-430, but cooled right back to 380 on the lean side. 6 got to 400, and the others were all 395 and below. Concerned about that 430, we did a compression check and borescoped the engine, and saw no indication of damage. In fact, my airpark neighbors (builders and mechanics) were impressed with the condition of my 2000 hour motor (and compressions were in the 73-77 range, 5 being the 73). I also cleaned all the plugs and injectors.

One thing I'll fess up to is that I found all of my plugs (top and bottom) gapped for the EI. Dammit...that's too much gap for my mag plugs! I messed that up, and must have raced that way...probably left some power on the table. Hopefully I've not damaged my brand new mag! So far no indication of that.

So after assessing the engine, cleaning (and re-gapping) my plugs and cleaning my injectors, I set out to do some more LOP/injector balance testing. After a discussion with Don at AFP and Steve about race CHT and my recent LOP testing, I did a cruise power leaning test (10.5K/22.1" MP/2300 RPM) on one day, and a 75% (ish) test (8.5K/24.2" MP/2400 RPM) on another day. Mag and EI had been set at 21 for racing (MAP sensor off for racing with nitrous), but I reset both to 22.5 for these tests, and had MAP sensor on. EI readout was 30 on the first test, and 29 on the second).

Test 1 showed #5 got to 396 CHT with all others in the 360's/370's. #6 has been trending slightly lean of the others, and is the first injector body swap that I will try next.

Test 2 showed #5 got to 405 CHT with all others in the 380's and 390's. Similar delta to #5 from the rest.

Don recommended this comparative test, to see if at higher power, the richest and leanest cylinders were changing. I see a clear trend that 6 is leaner than the rest, and a less clear trend that 5 richens at higher power.

I need more testing to establish more clear trends, but seeing this thread has my attention about doing LOP testing at 24 squared. That brings my post full circle, and thanks for bearing with me on all the above.

Is the experience of Walt's friend a "10:1 Angle Valve with P-Mag" issue?
Does his result translate to a 10:1 Parallel Valve engine with a mag and an EI?

Asking for a friend (my engine)! :D

Cheers,
Bob
 
Last edited:
It appears that the AV engine needs 3-5 less degrees of ignition timing compared to a PV engine to make max power.

You can run 3-5 degrees more timing when 50 LOP because of the much slower burn rate compared to ROP which delays the point of peak cylinder pressure (PCP) in relation to crank angle.

The lower the MAP, the more timing you can run before encountering detonation due to lower charge density.

With a 10 to 1 AV running at high MAP, ROP (fast burn) and 29+ deg of timing, you almost certainly will get detonation and the severity and propensity increases with high CHT and IATs.

Lower RPM has the tendency to increase detonation through both decreased chamber turbulence and the slower piston speed in relation to combustion speed, effectively advancing the point the point of PCP.

A slow lean pull isn't the best idea as you pass through a more dangerous zone, slightly to the rich side of peak.

If you look at the Lycoming dyno test data from their fuel testing, you'll see that a bit more timing does almost nothing for power but does increase CHTs and likelihood of detonation. It isn't worth it in my view to split hairs. Lys are not cheap to repair these days.
 
Last edited:
It appears that the AV engine needs 3-5 less degrees of ignition timing compared to a PV engine to make max power.

You can run 3-5 degrees more timing when 50 LOP because of the much slower burn rate compared to ROP which delays the point of peak cylinder pressure (PCP) in relation to crank angle.

.

Ross, why is there a difference between AV and PV for the same CR? I'm guessing there is a different combustion chamber shape that goes along with the different valve orientation? And the combustion chamber shape has a lot to do with the way the combustion event progresses, as well as whether it enhances mixing, so it has a significant effect on propensity for detonation?

It is a tricky issue for us LOP proponents. We need the spark advance to get good LOP performance, but don't want the advance when ROP. So when MAP is moderately high (24") it is good to pull fast to the LOP position and not dwell near peak. BUT...

What Bob Mills was trying to describe is a series of our observations that the GAMI spread of the engine seems to be rather RPM dependent, or at least fuel-flow dependent. One wonders if the flow divider characteristics change the distribution at higher fuel flow rates? So, for example, at typical high altitude LOP operations, there may be very little or no GAMI spread, because we have tuned all the injectors there, but at race power, where we would like all the cylinders to be running at the same air-fuel ratio, they may not be. It seems possible that at 2700+ RPM and WOT race power, Bob's cyl #2 may be leaning sooner than the others (and getting higher CHT), whereas at his typical cruise conditions, it is #6 that is leaning first (only slightly). But how do we test safely at those conditions? I suppose we could install 6 individual air:fuel ratio sensors.


One take-away is that we should probably confine our higher MAP testing to times when we have purple gas.
 
37BY plugs

“Background: I have a Lycon IO-540 CB45 (narrow deck, parallel valve) with 10:1 CR. One mag (I've had both Bendix and Slick over the years), one Electroair EI. I run REM37BY massive electrode plugs on all cylinders, gapped for the mag on the bottom, and the EI on the top.”

The 37BY plugs will increase your compression ratio slightly like a quarter of a point because of the extended nose shrinking the combustion chamber slightly. When running stock compression ratio 8.5:1 PV engines I sometimes use these plugs for increased power. (Documented on the Dyno )
Mentioning this because if your on the edge the aprox. extra 1/4 point could be significant.
BY plugs have other advantages as you probably know but your effective compression ratio is slightly higher than the 10:1 with their use.
As one of my mentors use to say “ Now your picking fly s..t out of the pepper.”
 
Last edited:
One wonders if the flow divider characteristics change the distribution at higher fuel flow rates?

The divider has a V shaped slot cut into a cylinder for each injector line. There is a disc that rides up and down the inside bore (movement actuated by fuel pressure which is somewhat linear with flow, though it is fully open well below max power). It's design is to allow a flow rate well above what is used in all but the idle level flow rates. The design calls for the injector orifice to meter fuel in those ranges. The injector can't reliably meter flow at the very low levels seen at idle, so the small bottom of the V is the metering point for low RPM operations.

As long as the divider is in good health and working as designed, it should have no impact on flow distribution across cylinders, as it should be allowing more fuel through the circuit than the injector does.

Larry
 
Last edited:
I would suspect the AFR differences at varying RPM are due to air flow differences in the manifold rather than fuel flow differences.
 
I would suspect the AFR differences at varying RPM are due to air flow differences in the manifold rather than fuel flow differences.

Me too. Whole lot of action going on in those empty tubes.
.
 

Attachments

  • Capture w notes.jpg
    Capture w notes.jpg
    166.4 KB · Views: 244
The divider has a V shaped slot cut into a cylinder for each injector line. There is a disc that rides up and down the inside bore (movement actuated by fuel pressure which is somewhat linear with flow, though it is fully open well below max power). It's design is to allow a flow rate well above what is used in all but the idle level flow rates. The design calls for the injector orifice to meter fuel in those ranges. The injector can't reliably meter flow at the very low levels seen at idle, so the small bottom of the V is the metering point for low RPM operations.

As long as the divider is in good health and working as designed, it should have no impact on flow distribution across cylinders, as it should be allowing more fuel through the circuit than the injector does.

Larry
Thanks Larry, that is super helpful. I never dug inside one.
 
I would suspect the AFR differences at varying RPM are due to air flow differences in the manifold rather than fuel flow differences.

Likely. Although, Bob has a cold-air induction manifold that is super open and all cylinders have very similar shape and length tubes. I can't attest that they are all the same length though, which would go to DanH's point that the acoustic tuning of each tube is a little different, so different cylinders will act differently at different RPM. If they are all the same length, then this effect would be minimized.
 
Likely. Although, Bob has a cold-air induction manifold that is super open and all cylinders have very similar shape and length tubes. I can't attest that they are all the same length though, which would go to DanH's point that the acoustic tuning of each tube is a little different, so different cylinders will act differently at different RPM. If they are all the same length, then this effect would be minimized.

So do I (Superior cold air) - and I'm running the SDS system which operates according to fuel pressure delta across the injector face and total injector dwell time for fuel delivery - and I can tell you for a certainty that my IO360 has different AFR's across the cylinders at 2200 rpm than it does at 2500. With the kind of minute control I have over the injectors I'm quite comfortable in my assertion that, at least in my engine, the airflow differences outweigh any fuel flow differences. There's a lot of resonance tuning that could be done with intake manifolds (as with exhaust) that would help, but that stuff is outside my knowledge base.

/thread creep
 
Last edited:
So do I (Superior cold air) - and I'm running the SDS system which operates according to fuel pressure delta across the injector face and total injector dwell time for fuel delivery - and I can tell you for a certainty that my IO360 has different AFR's across the cylinders at 2200 rpm than it does at 2500. With the kind of minute control I have over the injectors I'm quite comfortable in my assertion that, at least in my engine, the airflow differences outweigh any fuel flow differences. There's a lot of resonance tuning that could be done with intake manifolds (as with exhaust) that would help, but that stuff is outside my knowledge base.

/thread creep

Thanks. Good to know.
 
Thanks Don.

Now I am remembering the AFP innards. Slots cut with a wire EDM as opposed to the triangular holes in the Bendix.
 
Really excellent discussion! Walt, humble apologies for the thread drift, but thanks for allowing it to be a springboard for the follow-on points.

Circling back to the original concept, having seen Ross's post that a 10:1 AV engine at high MAP and 29 degrees will very likely reach detonation, I'd like to ask, how high is too high? 24"? Higher or lower? And how much timing would one need to back out, in order to safely lean through peak at that same power setting? The 3-5 degrees that you mentioned (24-26 degrees?)...would that be safe? I'm sure its not cut and dry, but perhaps a rule of thumb would help other's avoid Walt's friend's troubles.

I've owned one plane with an AV Lyc. It was a TIO-580, but it was 6.5:1 for the turbocharger. I raced it at 67" MAP, but ran dual mags and leaned very carefully. I actually couldn't get the fuel flow high enough for the power I wanted to run, despite boost pump and mechanical pump changes to chase higher flow rates. Probably needed bigger pipes (#8 instead of #6). Ross will likely chuckle, as he's helped some of our Gold racers get far more MP and FF than I ran, along with the SDS and its cylinder tuning capabilities. The results showed it too...I was a distant third to Andy running SDS. But timing was back at 20, and it was not a great LOP flyer outside of racing, so I never really pushed into this danger zone with leaning.

The discussion of fuel flow/flow dividers, and airflow issues is great info! As Steve said, my AFP FM-300B and SkyDynamics cold air induction typically provide excellent flow. Good info from Greg on your AFR differences at different RPM. Greg, by any chance, do you have data that shows what happens to AFR in each cylinder (especially 2, the one I'm working on now), as you increase from 2200 to 2500?

The other curve ball in my engine performance, and this temp chase throughout the week in Reno, is the addition of N2O. I installed a system from AFP that injects both fuel and N2O into a spacer just behind the FM-300. There is one nozzle on each side, and each adds both fuel and N2O right into the cold air inlet plenum. Varying the jets in each line allows choosing a 50HP, 100 HP or a 150 HP shot. I used the 50 inserts. While my recent LOP test without nitrous showed the slight tendency of cylinder #2 to lean earlier at higher power (24 squared vs 22 or 23 squared), it was not as marked at is was at full race power (about 27 squared in Reno). Could it be that higher speeds, leading to more ram air pressure, could be forcing the air/fuel mixture towards the back of the cold air plenum, and #2 just isn't seeing quite as rich a mixture. I'm just musing out loud at this point...I know the time increments between cylinder firing are minute, but I'm wondering how the flow might be affected by all of this ram air/nitrous and other monkey motion going on. And wondering what, if anything, might be done (inlet pipe tuning comes to mind, as was mentioned earlier).

Dan, interesting graphic. What is the term "bleed" referring to in it (MP higher or lower than bleed)?

Thanks again for the discussion!

Cheer,
Bob
 
Last edited:
Circling back to the original concept, having seen Ross's post that a 10:1 AV engine at high MAP and 29 degrees will very likely reach detonation, I'd like to ask, how high is too high? 24"? Higher or lower? And how much timing would one need to back out, in order to safely lean through peak at that same power setting?

A perfectly stock angle valve can be moved in and out of detonation onset at will, with stock timing. Just supply the wrong conditions. I'm sure the same is true of the parallel valve, but I don't have comparable data. The manufacturers treat detonation info like a state secret.

Accurately knowing the limits for your race setup will require an off-season dyno session with a vendor equipped for detection. George Braly has a great setup, although Sky Dynamics might be a player, and Lycon should be.

For the average person, the takeaway from Walt's report should be two-fold. One, you can't just bolt stuff to an engine and assume everything will be OK, no matter what the vendors say. Two, with apologies to the poor schmoo who got the hard lesson, moving the knobs without first engaging the brain has consequences. In fairness to the nice folks at EMag Air, we should all remember it was possible to conduct the test without breaking the engine. The FAA, Lycoming, and Mr. Braly, for examples, do it routinely.

Dan, interesting graphic. What is the term "bleed" referring to in it (MP higher or lower than bleed)?

You're looking at the output from a fast deltaP sensor, one leg tapped to the intake port via the primer hole, and one leg tapped to the butt end of a bleed air rail feeding shrouded injector nozzles. You'll recall the nozzles bleed upper plenum air into an annulus around the fuel stream just after the restrictor, to promote atomization of the stream as it enters the port. The goal of the experiment was to determine how often, and to what degree, does the nozzle suffer a reversal of bleed air pressure, something Don Rivera wished to see. Here there is some pressure reversal, for a very short period. The fun part is the ancillary data derived at the same time.

In the context of this discussion, bleed air pressure is assumed to be a constant, relatively steady value. That may not be completely true due to pressure oscillation in the propeller wake, and would probably vary quite a lot based on cooling air intake design. However, it's a close enough assumption here. It means the pulsing pressure seen in the plot is the wave action in the intake tract for that cylinder.

Below I've added two notes. Each little graph square is about 20 degrees. The positive waves are arriving at the port at intervals of about 117 degrees, mostly a function of intake tract length and diameter, and only optimum for one RPM. I've used spark as a marker, which allows me to identify TDC, and BDC, which lets me approximate intake valve open and close. Note that the wave is arriving twice during the intake period, once as the valve gets well off the seat, just after TDC, and again late in the stoke, as the piston nears BDC. Those positive pressure periods help cylinder filling...the whole purpose of the tuned runners used with the horizontal intake sump. Way back when, some engineer at Lycoming got it pretty close.
.
 

Attachments

  • Capture w notes.jpg
    Capture w notes.jpg
    171.1 KB · Views: 149
The other curve ball in my engine performance, and this temp chase throughout the week in Reno, is the addition of N2O. I installed a system from AFP that injects both fuel and N2O into a spacer just behind the FM-300. There is one nozzle on each side, and each adds both fuel and N2O right into the cold air inlet plenum. Varying the jets in each line allows choosing a 50HP, 100 HP or a 150 HP shot. I used the 50 inserts. While my recent LOP test without nitrous showed the slight tendency of cylinder #2 to lean earlier at higher power (24 squared vs 22 or 23 squared), it was not as marked at is was at full race power (about 27 squared in Reno). Could it be that higher speeds, leading to more ram air pressure, could be forcing the air/fuel mixture towards the back of the cold air plenum, and #2 just isn't seeing quite as rich a mixture. I'm just musing out loud at this point...I know the time increments between cylinder firing are minute, but I'm wondering how the flow might be affected by all of this ram air/nitrous and other monkey motion going on. And wondering what, if anything, might be done (inlet pipe tuning comes to mind, as was mentioned earlier).

Cheer,
Bob

Once you start shooting fuel in the intake plenum (i.e. the common area before the runners), you now experience the issues the carb guys do - way too small of a plenum and poor air flow distribution, etc. Your case is even worse than a carb, as the fuel is injected into stable air, where a carb injects it into the venturi, which helps a bit with atomization. While I don't understand the engineering of why, I do know that many with carbs experience it. Just looking at a modern automotive intake, shows that engineering principles matter here. An interesting experiment would be to see what happens at the same 27 squared combo (ROP) without the NO/fuel addition at the plenum. Does the lean condition on #2 lessen or remain the same. If it goes away, you are left with the same issues that plague all Lyc carb users. I suspect someone may have played with this and found some potential tricks to lessen the effect.

If you have the space, adding another spacer between the NO extentsion and the plenum will help to increase the plenum area and improve mixing, though it is critical that you match the walls closely. Remember the old hot rods with the carb(s) sticking out of the hood? Same concept here, though those also increased the runner lengths and made them more vertical as well as enlarging the plenum. Adding spacers between the carb and the intake manifold is a time honored trick to improve mixing and intake efficiency in the hot rod community.

Larry
 
Last edited:
As Dan correctly points out, the dyno with knock sensing capability is your friend here.

There are many inter related factors here- CHT, IAT, timing, MAP, CR, squish, piston shape, fuel droplet size, fuel octane, ADI amounts etc, cylinder to cylinder AFR variation etc., so there is no easy answer to as how much timing is safe- it depends as they say.

As far as the top Reno Sport Class turbo race engines go, most have multiple non-stock parts in them and have been dyno developed over several years. Pinnacle and Lycon obviously know what they are doing, having taken all the Gold Class wins over more than a decade where Lyconentals have been running.

The top engines are making a lot more power than the ones running even 20 mph less given the cube hp function and lowering prop efficiency.

As related to stock engines where ultimate hp is less important, it simply doesn't make sense to push timing that last few degrees for very minimal gains and higher risks. Conservative will serve you well here.

I will again mention that 30-50F LOP and low MAP (less than 25 inches) with CHTs below 370 and IATs below 100F isn't a place where most Lycomings have a high likelihood of detonation but as Dan said, you can still do it under the right or should I say wrong, conditions.
 
Last edited:
Dan, Larry and Ross, thanks for the replies and info...very good stuff!

Dan, makes sense now on the bleed vs MP. Don and I recently discussed all this as well, and we may end up trying turbo rails like I had in the Super Glasair...if fuel distro appears to be the (or an) issue. This discussion of intake air issues may slide that back in priority order, but it's still on the table. Another system of systems approach...as always. ;) I spoke with a Sport racing couple today, and they recommended George as well. I talk to Ken at Lycon often, and look forward to chatting with George. He's close by in OK too, so that's a good opportunity. Thanks!

Larry, interesting points. I will compare 27/2700 data with and without N2O...good call. On the spacer install...I actually had to use a 3/4" spacer between the FM-300 and the N2O/fuel plate, because the butterfly was impacting the nozzles. Don made that for me just before leaving on vacation (I was running out of pre-Reno flight test weeks...what a great guy, eh). But that doesn't buy me extra space between the nozzles and the plenum...you have me thinkin. I believe I can fit another 3/4" or 1" spacer in behind the N2O/fuel plate. I have the room to the cowl inlet...but I need to check the clearance on the lower cowl bottom and side clearances to the nitrous and fuel hoses and nozzles. It's pretty tight up there. Yay...another new project, er I mean opportunity! :D

Ross, concur on Lycon and Pinnacle, and the great engineering. You've been a big part of that success as well...just want to make sure that gets mentioned too! :) I was one of those 20mph back in Gold (or 30+ in some races) for a few years, and knew the quantum HP leaps (and $) needed to catch them. Its fun territory!

When Don and I discussed LOP testing at 24/2400, I expressed my concerns about detonation. We planned and executed the test to be done at the highest altitude I could get 24". That turned out to be 8500' that day. As I mentioned earlier, I had advanced my mag timing back up from 20 (in Reno) to 22.5, and did the same with the base timing on the Electroair. But I did turn the MAP sensor on for the EI before that test flight, so it did go to 29 during the test. Perhaps a better (and more safe) approach would have been to leave the MAP sensor off, and essentially run fixed EI timing, like I did in Reno. Might be a more valid comparative test too. I watched CHT closely for rapid trends during the test, and saw none...and actually the CHTs stayed relatively cool for the power setting, given the leaning exercise.

The original post from Walt really got my attention, after having done what, on the surface, appears to be a similar exercise. Really good to know that what works with one engine/fuel/ignition system may not be apropos for another. I'd still be very interested to know what MP and timing actually was when the gent ran into detonation. It certainly reinforces my caution about leaning at moderately high to high power. Good info guys...thanks again!

Cheers,
Bob
 
Last edited:
It would be interesting to hear what manifold pressure, RPM, CHT's and EGT's were. Was he below 75% power?

He tried to lean? A new engine I'd stay 100-150F ROP. Problem is you peak EGT first. So if below 75% power I'd lean to just see EGT start to move and leave it. Too late now.
 
Back
Top