flyer338

AOG four months now.

Recommended Posts

I am frustrated and out of ideas. I hope the brain trust here might be able to help.

After 8.5 years without flying after I sold my M20C, I bought a 1983 M20J. The plane has 5700 hours tt and about 600 since factory overhaul on a A3B6 engine in 2012. The previous owner replaced the right magneto with an EIS-41000 electronic ignition when the engine was overhauled. 
I rolled the pre-purchase into an annual at Top Gun Aviation in Stockton and hired Don Kaye to get me current and safe. 
After getting signed off, I flew around the Sacramento valley and to KFOT on the NorCal coast a couple of times. The plane has about 17 hours since the annual.

I was returning to Mather (KMHR) from my second trip to KFOT when as I reached 8900’ I detected a slight roughness in the engine. The roughness felt exactly like what one gets by leaning until there is a misfire. Adjusting the mixture had no effect. I also noticed higher EGT and lower CHT readings on the engine analyzer for the front two cylinders. I decided to divert to Eureka (Murray Field KEKA), as it is the only maintenance facility in Humboldt County. 
Long story short, they had the fuel servo overhauled by Victor Aviation in Palo Alto and the fuel divider replaced. This began 10/26/2019, and due to weather and other issues I did not get up to Eureka to bring the plane to its new home in Las Vegas until 2/17/2020.

It now has a new problem; within a minute after takeoff, the engine is intermittently rough and sounds like it is back firing out of the exhaust. The compression is fine, it passes the valve wobble test, the mechanicals are correct, the left (Slick) mag and harness have been replaced, to no effect. The problem cannot be duplicated during a ground run. The EGTs are even. It only occurs during full power (2700 rpm 18.8 gph fuel flow) on initial climb. If I shut down the left mag, it stops misfiring; if I shutdown the electronic ignition, it also stops misfiring.

Tomorrow, Kyle the IA, and someone I have known for 30 years, is going to ride along as we attempt to trouble shoot this problem in flight. 
Any ideas appreciated. 
 

  • Sad 1

Share this post


Link to post
Share on other sites

*Members that donate $10 or more do not see advertisements*

Brian,


Please tell us there is a handy JPI in the panel... (or other)

And you are about to post the downloaded files on Savvy.com

We have a few mechanics and engineers that are good at reading graphs and data...

As magical as your writing is... you probably won’t be able to keep up...

Then the next problem... nobody can remember all the wordy detail, like seeing a picture...

If you have a really strong engine monitor that collects all the data like where the plane is, MP, RPM, and a dozen other variables... things should work out swimmingly...

The engine had no response to mixture adjustment? Pull it out... push it in? Go directly to mechanic... do not fly...something is disconnected...

PP thoughts only, not a mechanic...

Best regards,

-a- 

  • Like 2

Share this post


Link to post
Share on other sites

I have a JPI, but I am not savvy enough to perform a download.. I plan to call savvy in the morning. Better late than never.

Share this post


Link to post
Share on other sites

Today is your lucky day...

You spent four months on the ground...

Time to make some headway...

What version of a JPI do you have?  830, 900, 930, 700...

What method do you want to use to get data out of it?  Usually the plane will have a socket installed or a memory card to plug in...  JPI has a  web site... with all their manuals...

They also sell all the hardware to get the data out of the device... software for doing this is free from JPI

Get data, then go to the savvy site... read upon how to upload your data...

Getting your data will be helpful for any mechanic that is going near your plane...

Are you familiar with laptop computers?

Best regards,

-a-

Share this post


Link to post
Share on other sites

I’m not a big fan of flying planes with known unknown issues...

When there are tools available to get the data...

And resources to figure out what the data says...

save your 30 year relationship with your mechanic!

At least check the mixture control continuity first...

You described no change when you operated the mixture control... this could be really bad... or just misleading...

How old are the engine controls?

Was the fuel injection system OH’d with the engine OH?

PP thoughts only, not one of those really good mechanics I described above...

Best regards,

-a-

  • Like 1

Share this post


Link to post
Share on other sites

A “backfire” from the exhaust (technically an afterfire as a true backfire is through the induction system) can be a symptom of a too rich mixture where excess fuel is unburned and ignites in the exhaust system. It might be running rough because it’s too rich. Operating on one mag slows combustion similar to leaning. 18.8 gph sounds a little high for an IO360 at sea level full power and this started after the fuel servo was overhauled. I would try leaning to somewhere between 17 and 18 gph during the climb and see if that has an effect.

Skip

Share this post


Link to post
Share on other sites

Skip,

The takeoff and initial climb out fuel flow before this issue arose was 19.6 gph. I am wondering if the synchronization between the mag and the EIS is correct, but I cannot see how that might have changed.

Share this post


Link to post
Share on other sites

Skip, I think you may have made an important point and it is one I can test when I fly the plane in the morning.

Share this post


Link to post
Share on other sites

From the Comanche forum, a failure of the electronic ignition system.   I don’t know a lot of detail, other than less than 200 hours in service.
 

Clarence

2FCF3F6B-4558-484F-9761-2992D8B73B1E.jpeg

  • Like 2
  • Sad 2

Share this post


Link to post
Share on other sites

Just for kicks,  I figured out the rated power fuel flow for a Lycoming IO-360.

Data supplied with my factory rebuilt IO-360-A3B6 shows maximum fuel flow specifications at various airflows (AF= airflow in lb/hr, FF = fuel flow in lb/hr)

FF=52.6 @ AF=504

FF=86.3 @ AF=904

FF=95 @ AF=1000

FF is unspecified at rated power, but actual measured AF=1118.5 on my engine

This data projects a rated power fuel flow of 104 lb/hr. According to Aeroshell's data sheet for 100LL, https://www.shell.com.au/motorists/shell-fuels/sds-tds/_jcr_content/par/textimage_278c.stream/1519809888867/e1579e2bb13c53e309e13bad6be6881908514236/avgas-100ll-pds.pdf

the density is 0.718 lg/L or 5.995 lb/gal, so 104/6 = 17.3/gal/hr. My fuel flow normally reads about 17.2, FWIW.

This may have nothing to do with your issue, but I post it in case there is interest in IO-360 takeoff fuel flow.

Skip

  • Like 1
  • Thanks 1

Share this post


Link to post
Share on other sites

Hmmm, @PT20J calculated a lower fuel flow for 200 hp than my Owners Manual shows for my 180 hp Lycosaur at sea level . . . .

20200225_121140.thumb.jpg.92ea0b82a25f8d317bb3e1f6f7597948.jpg

Share this post


Link to post
Share on other sites
2 minutes ago, Hank said:

Hmmm, @PT20J calculated a lower fuel flow for 200 hp than my Owners Manual shows for my 180 hp Lycosaur at sea level . . . .

20200225_121140.thumb.jpg.92ea0b82a25f8d317bb3e1f6f7597948.jpg

I looked in my '78 J manual and it shows 18.4 gph at 100%.  My '94 J manual doesn't list anything above 75%. The specs I published are the current Lycoming acceptance test specs.

Skip

Share this post


Link to post
Share on other sites

+1 for savvy. There website has directions on download info for many types of analyzers.

You might also find Mike Busch’s books valuable.

On the “too rich front” I once had an air-intake obstruction that (speaking relatively) made for a too rich problem. One I removed the obstruction, everything was again good.

I think that I posted about it here on MS. I’ll see if I can find and post a link.

  • Like 1

Share this post


Link to post
Share on other sites
11 hours ago, M20Doc said:

From the Comanche forum, a failure of the electronic ignition system.   I don’t know a lot of detail, other than less than 200 hours in service.
 

Clarence

2FCF3F6B-4558-484F-9761-2992D8B73B1E.jpeg

That doesn’t look like it failed slowly either...

Looks like half a turn, and the teeth went from getting trimmed to getting cut off...

Thanks for sharing the pic!

Best regards,

-a-

Share this post


Link to post
Share on other sites

Update for N56338

It is still AOG. The basic problem is still the fuel system. This despite a red herring that worked into the analysis when one of the plug wire ends (#2) failed while checking everything else - valve wobble, compression, spark plugs, and so on. On Monday last week, after consultation with Electro-Air, we checked the signal generator for the electronic ignition and it meets specification; we also checked the plug wire resistance and found one wire out of spec. I ordered a set of wires from Electro-Air and installed them Wednesday. The missfire experienced previously was gone. However, the original problem was back and worse. The run-up was fine and the ignition/mag drops at 2,000 rpm were also fine. At take off the fuel flow was 18.8 gph (compared to 19.6 before this issue began) and the engine ran rough during the take-off roll. The runway is short and I continued the take-off and turned crosswind before getting over the bay. I continued to climb on downwind and shut off the mag and electronic ignition in succession and there was no change to the engine roughness. After reaching pattern altitude, I reduced power until the roughness stopped; this was at 20". I flew a normal pattern and landed.

On the taxi-way, I did a couple of full power run-ups - fuel flow was 18.4-18.5 gph and I did not detect any roughness. The EGTs were all the same - level on the graphical readout.

I am satisfied there is no longer an ignition problem - it creeped in during the trouble-shooting, and it was fortuitous, because I might have been over the Sierras when the fuel system problem returned. Of note, the fuel flow had been set to 19.1 gph, I registered 18.8 on take-off, then 18.5 on full power runup. Previously the full power fuel flow through take off and climbout has been very consistent at 19.6 gph. The change in fuel flow, the change related to power setting, and the elimination of other possibilities, lead me to believe I need to have a second shop look at the fuel injection. I consulted Savvy Aviation and their consultant agrees.

The plan is the shop is going to verify the fuel pressure, independent of the panel gauge, check the finger screen in the fuel injection servo for contamination, and if those check out, the fuel system will be sent to a second shop recommended by the Savvy Aviation consultant.

  • Like 1

Share this post


Link to post
Share on other sites
41 minutes ago, flyer338 said:

Update for N56338

It is still AOG. The basic problem is still the fuel system. This despite a red herring that worked into the analysis when one of the plug wire ends (#2) failed while checking everything else - valve wobble, compression, spark plugs, and so on. On Monday last week, after consultation with Electro-Air, we checked the signal generator for the electronic ignition and it meets specification; we also checked the plug wire resistance and found one wire out of spec. I ordered a set of wires from Electro-Air and installed them Wednesday. The missfire experienced previously was gone. However, the original problem was back and worse. The run-up was fine and the ignition/mag drops at 2,000 rpm were also fine. At take off the fuel flow was 18.8 gph (compared to 19.6 before this issue began) and the engine ran rough during the take-off roll. The runway is short and I continued the take-off and turned crosswind before getting over the bay. I continued to climb on downwind and shut off the mag and electronic ignition in succession and there was no change to the engine roughness. After reaching pattern altitude, I reduced power until the roughness stopped; this was at 20". I flew a normal pattern and landed.

On the taxi-way, I did a couple of full power run-ups - fuel flow was 18.4-18.5 gph and I did not detect any roughness. The EGTs were all the same - level on the graphical readout.

I am satisfied there is no longer an ignition problem - it creeped in during the trouble-shooting, and it was fortuitous, because I might have been over the Sierras when the fuel system problem returned. Of note, the fuel flow had been set to 19.1 gph, I registered 18.8 on take-off, then 18.5 on full power runup. Previously the full power fuel flow through take off and climbout has been very consistent at 19.6 gph. The change in fuel flow, the change related to power setting, and the elimination of other possibilities, lead me to believe I need to have a second shop look at the fuel injection. I consulted Savvy Aviation and their consultant agrees.

The plan is the shop is going to verify the fuel pressure, independent of the panel gauge, check the finger screen in the fuel injection servo for contamination, and if those check out, the fuel system will be sent to a second shop recommended by the Savvy Aviation consultant.

Any fluctuation in fuel pressure needle?  Perhaps an air leak in the gascolator.  I have seen/heard of this on a couple of occasions, usually causing issues at altitude and at full power, but it could be at full power on the ground.  Basically look for a leak at the gasket, which may not result in a fuel leak.  Small air bubble get into the fuel system.  Kind of a long shot, but easy enough to look at without it costing too much.

Share this post


Link to post
Share on other sites

There is something weird here. How would the engine know to only run rough when you are flying versus static? When you did the full power run ups did you let it run for a few minutes to stabilize temps and power? Are you sure your fuel flow measurement is accurate?   19.6 gph seems high. But suppose that number represents the correct flow even if the measurement is not accurate. It would represent a fuel flow somewhere in the neighborhood of 200F ROP. The lower fuel flow that you are seeing now should result in a leaner mixture and higher CHTs. Did it? 

Skip

Share this post


Link to post
Share on other sites

Any luck with getting data out of the JPI?

  •  Looking for signs of sticking valves...
  • looking for signs that plugs are not always firing properly...
  • Looking for FF vs. rpm... and MP.

If the fuel system isn’t delivering FF... it is often readily obvious on some JPI devices...

Sometimes it is a money saver to use the data to select the next course of action...

 

read up on engine out procedures.... be prepared... push hard on the yoke...

PP thoughts only, not a mechanic... or CFI...

Best regards,

-a-

Share this post


Link to post
Share on other sites

When the fuel servo was done was it flow tested with the injectors and fuel divider?

A baby jar test may be in order.

A bad plug.

A bad induction tube

When I was reading the initial post again I was thinking fuel system

Air leak in fuel lines.

One both pump are putting air into the fuel system

Share this post


Link to post
Share on other sites

I had intermittent symptoms that seemed to point to fuel flow, and after checking ignition and injectors I sent my injection system off to a specialty shop for some R&R.  When the system returned from the shop things seemed to improve.  Then 1-2 years later, after an off field landing, a tear down of the servo revealed that a small piece of torn o-ring had been floating around for some time and eventually moved to block fuel flow.  The shop that worked on my servo responded that they didn't open that section of the system.

Per the NTSB report "a fragment of the O-ring was located on the surface of the mixture control valve plate".

https://www.ntsb.gov/_layouts/ntsb.aviation/brief.aspx?ev_id=20120721X23053

 

 

  • Sad 1

Share this post


Link to post
Share on other sites

Below I have organized my reply based on the individual comments and suggestions. Thank you all very much for offering the benefit of your collective knowledge and experience.

Takair

Any fluctuation in fuel pressure needle?  Perhaps an air leak in the gascolator.  I have seen/heard of this on a couple of occasions, usually causing issues at altitude and at full power, but it could be at full power on the ground.  Basically look for a leak at the gasket, which may not result in a fuel leak.  Small air bubble get into the fuel system.  Kind of a long shot, but easy enough to look at without it costing too much. – So far there is no sign of the fuel system having air leaks. I will see that this is checked.

PT20J

Are you sure your fuel flow measurement is accurate?   19.6 gph seems high. But suppose that number represents the correct flow even if the measurement is not accurate. – I think this number represents the correct full power fuel flow (29”, 2700 rpm, full rich) even if it is not accurate. That the fuel flow is not stable across time is what concerns me. For the 16 or so hours prior to this problem appearing, the fuel flow was very consistent at 19.6 gph at full power.

Carusoam

Any luck with getting data out of the JPI? – Not at present. I am still in the upward learning curve on this plane. I plan to arrive in Eureka on my next trip with equipment and cables necessary to download the JPI. Down the road, I may replace it with a more modern unit that has a better interface.

     Looking for signs of sticking valves... – Valve wobble test done and it is within specification. The is no sign of corrosion and internally the engine is very clean – the previous owner changed oil and filter every 25 hours since the factory overhaul in 2012.

    looking for signs that plugs are not always firing properly... – All of the plugs have been removed, cleaned, electrically tested, and visually inspected, and they are all relatively new. There is no sign of a malfunctioning spark plug. I have a complete set of replacement plugs and if necessary, I will install them.

    Looking for FF vs. rpm... and MP. – See my comment in response to PT20J.

Yetti

When the fuel servo was done was it flow tested with the injectors and fuel divider? – Yes.

A baby jar test may be in order. – It was done and the flow from each injector is uniform.

A bad plug. – See response to Carusoam above.

A bad induction tube – There is no sign of an induction tube leak, but a pressurize the intake system and soap bubble test has not been done. I do not think this is a likely culprit because an induction leak is more likely to manifest a symptom at part throttle when the pressure differential between ambient and the intake is greater.

Air leak in fuel lines.  There is no sign of an air leak, but it is something that will be checked further.

Share this post


Link to post
Share on other sites
2 hours ago, flyer338 said:

Air leak in fuel lines.  There is no sign of an air leak, but it is something that will be checked further.

IIRC, there's an SI or SB from Lycoming about using a clear piece of tubing between the servo and flow divider during a ground run to look for bubbles.

  • Like 1

Share this post


Link to post
Share on other sites
2 hours ago, EricJ said:

IIRC, there's an SI or SB from Lycoming about using a clear piece of tubing between the servo and flow divider during a ground run to look for bubbles.

Precision Airmotive (current manufacturers of RSA fuel injection systems) has a DVD on troubleshooting that includes this test. Call them and they will send it to you free of charge. 
 
@flyer338 didn’t mention whether CHTs are higher now than with earlier higher fuel flows. Concern would be that if fuel flows are now too low, engine may be too lean at full power reducing detonation margin. 

Skip

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.