New engine roughness appearance at end of great runs

Engine is a O-320-H2AD, t mod, 40 minutes tt since OH, Ellison throttle body, stock Bendix D3000 dual mag. At the tail end of the runs prescribed by Mahlon, I went from full power to 1000 rpm. Everything was excellent until about 30 seconds into the 1000rpm run, where it started running rough. It was the first hint of any roughness in any run. Now having looked at the data from the Dynon at 1hz, In the 30'seconds of the event, the roughness is seen in rpm, the average came up by about 50 Rpm and it moved plus or minus 40rpm or so. At the same time, all the EGTs started slowly climbing together, and climbed 55F during the roughness, from 1111F ave to 1166F ave, all within 10F. Three CHT stayed stable, #3 climbed by 10F, they were in the 340-360F range. All other measured elements showed no changes. During this period I made no throttle or mixture changes. Full pressure was solid at 4.5 psi. I shut it down with mixture at idle without it smoothing out. OAT was 74F, fairly humid.

I'm looking for ideas where to look. Thanks, Rick

Your EGTs went up with a big power reduction. My guess is that it is way over rich and the burn is slow, carrying over into the exhaust. You don't say what type of ignition you have. If it is adjustable then retarded timing would produce the same result.

Clarification and ignition

My ignition is a D3000 dual mag, fresh overhaul, stock tempest plugs, timed at 25 deg btdc, per spec.

The timeline of the rough run: start, establish good numbers for 15 sec at 800rpm, then 5 minutes at 1400rpm, stable, mag check, completely normal with 50 rpm drops, then to WOT, stable at 2180-2190 rpm for 20 seconds(Sensenich fixed, this seems on target) then down to 1000rpm. The first 30 seconds of 1000rpm is stable and smooth. The second 30 seconds is where it gets rough, the average rpm comes up by 50rpm, varies by +/- 40rpm in roughly 6 second intervals, and the EGTs come up, all together. The only other flyer is CHT of #3 also climbs, none of the others do.

I'm going to go looking for an intake leak, I think. It's appears something changed. I actually considered ice, but I think it was too warm out at 74f oat. It was fully cowled on this run. I have done zero tuning yet, everything is out of box Ellison settings.

I have the same engine combo on a pusher but with EIs. You probably just need to lean the mixture. I would guess you have some carbon-fouling on the plugs. It's good practice to lean on the ground as much as possible. It saves gas and keeps your plugs cleaner.

A full power run with full rich is also too rich but needed to avoid detonation.

The Ellison is a great product. Lean on the ground as much as you can. Lean to peak power when passing 4000' climbing. Lean as much as desired above 8500'. Lean below about 2350-2400 RPM at any altitude. Stay lean in the pattern except at WOT. Lean, lean, lean.

Mixture

The OP states 40 minute since overhaul, ground runs, not flown yet. You DO NOT lean the mixture until rings are seated, especially given the near sea level conditions.

jrs14855 said:

The OP states 40 minute since overhaul, ground runs, not flown yet. You DO NOT lean the mixture until rings are seated, especially given the near sea level conditions.

Click to expand...

This is knowledge without understanding. Yes, run rich at high power settings to keep temps down. Otherwise, there is no problem with leaning during low power ground ops or when CHTs are reasonable, even while breaking-in an engine.

Richening is simply another aid to control heat during break-in. When high CHT or detonation are not a problem, then lean. No need to sit at idle in full rich fouling plugs.

I would check Ellison filter.
Also, could be fuel vaporization. The engine compartment gets very warm with full power run on ground.

My first guess would also be idle mixture. At 1000 RPM you're getting into the transition circuit (not adjustable), but the idle mixture should still dominate. My MA4 had an initial setting of 2.5 turns out from seated (carefully seating it, very little force). This was too lean and I ultimately ended up somewhere in the 3.5-4 turns range. If you're not sure whether you're lean or rich, you can start there. You can also research idle mixture testing, but I struggled to get good readings in my set up.

I also agree with the above poster. Absolutely no reason to run your idle mixture at anything other than the optimum setting during break in. Running rich will give you only headaches with no benefit.

Good luck,

Larry

working theories

Thank you all for your thoughts. Based on all the input I've gotten, I'm looking at:

1. Carb ice. go ahead, laugh. 73F ambient and 43F dew point it's right there, and right where the Ellison is (cough) known for a sensitivity. Where this falls apart is the rising rpm.

2. Fuel restriction. RPM up (leaning), but out of nowhere? Rise in EGT says no. Interesting with the data, ~5psi of pressure at 1400, 2.5psi at ~2200, the 5.2psi at 1000 rpm? My pressure sensor is DOWNSTREAM of the throttle body.....hmmm.

3. Intake leak. Since rpm went up, maybe indicating leaning, maybe? Again, EGT went up, so it has conflicting data.

4. Ignition dropout. While EGT went up as an ignition dropout often signals, the rise in RPM is contrary. I had done a mag check two minutes prior, and all was normal/expected.

5. Vapor "lock". resulting leaning gives way to an rpm rise, but conflicting with a rise in EGT.

6. Mixture. Larry, the only reason I'm not going there first is I have a 40 seconds of rock solid idle, then something changes with no control changes from me. Regardless, I have evidence on idle cutoff to suggest it's running really rich, so I leaned it back a bit.

My plan:
A: pull the plugs, see if there is any fouling. With 40 minutes TT on new plugs though...CPL, pulled bottom plugs, no fouling, somewhat black, pretty uniform, no wet oil.
B: Nut the engine. Maybe something is loose (vac). CPL, nothing found on the intake or primer loose or not sealed.
C: Pull the collator and Ellison screen. Maybe it's junk from assembly. If it is, Tom, we're gonna talk (kidding)
D: Run a single ignition run. See is if looks similar.

Other ideas before I go deep? Nutting is lunchtime Tuesday. From the race car days, nutting is a kind of a challenge.

Rick

rjcthree said:

6. Mixture. . . .I have a 40 seconds of rock solid idle, then something changes with no control changes from me. Regardless, I have evidence on idle cutoff to suggest it's running really rich, so I leaned it back a bit.

Click to expand...

Plugs don't foul instantaneously, of course. I would guess the combination of too-rich at idle and a bit of oil getting past the unseated rings is affecting your idle RPM. Yes, it could be something else but the first thing I'd try is to lean aggressively as soon as you retard to idle. See http://www.avweb.com/news/maint/182900-1.html

Root cause...maybe

I have checked plugs, clean. I check for leaks s/tightness, no suspects. Today I checked fuel filters in the and air and in the Ellison. Both clean. I put it all back together, turned on the fuel pump and got fuel draining out of the throttle body, a stream of it. That's not right. Called Ellison, they were very helpful and we got it cleared. I'll do runs tomorrow to confirm.

I probably saw the extra fuel draining at the end, but with the oil leak, I missed as significant. I'll report back when I know.

Runs well

Looks like this issue is solved. Runs well, now to setting idle, etc. the runs showed CHT4 as higher than the rest buy 30-35 degrees, which is different from the past runs. Always something. I think I also over leaned the idle mixture, no rise on red knob, runs much worse at 600rpm. Back to square one. But that's ok, I'm learning.