Jump to content

Mixture too lean; misfires; 0.7 GAMI spread


Recommended Posts

1 minute ago, jaylw314 said:

There's no reason to put the JPI into lean-find mode for the GAMI sweeps.  You're going to be reviewing the data after the fact anyway, and having the display abruptly change is another distraction.  You can just put it in normalize mode when you're near peak, and you'll be able to see when all the bars are coming down since it's more sensitive

Yep, this is good advice. Just lean as slow and steady as you can until the engine is running really rough. Not until the first hint of roughness, but just before it quits at mixture cut-off. Then push it back up slowly too rich. 

Keep uploading and sharing the data. This is helpful for others just learning how to use an engine monitor as well. So thanks. 

  • Like 1
Link to comment
Share on other sites

3 minutes ago, jaylw314 said:

There's no reason to put the JPI into lean-find mode for the GAMI sweeps.  You're going to be reviewing the data after the fact anyway, and having the display abruptly change is another distraction.  You can just put it in normalize mode when you're near peak, and you'll be able to see when all the bars are coming down since it's more sensitive

For me LF mode was easier because I could instantly see at a glance when each of the cylinders had passed peak. I'm not as familiar with normalize mode and it doesn't seem nearly as obvious.

Link to comment
Share on other sites

Close up of the two inflight mag checks...

They are supposed to be identical in appearance... but they are not...

Two cylinders behave identically as the plugs get swapped on/off, top/bottom...

Two cylinders are completely acting differently as the plugs get swapped on/off, top/bottom...

See what I mean?

-a-

 

8EB0D477-9D1B-46BC-953B-101D9E07A16B.png

BEC52AD1-01FB-4707-A0FF-7A688094C6EC.jpeg

  • Like 2
Link to comment
Share on other sites

Just now, chrixxer said:

For me LF mode was easier because I could instantly see at a glance when each of the cylinders had passed peak. I'm not as familiar with normalize mode and it doesn't seem nearly as obvious.

Normalize mode is great for some individuals... same data presented in a different way...

For this conversation, let’s try to avoid doing things differently, unless it provides something useful...

Standardized presentations go along way to keeping as many eyes involved...

Best regards,

-a-

 

Link to comment
Share on other sites

5 minutes ago, chrixxer said:

For me LF mode was easier because I could instantly see at a glance when each of the cylinders had passed peak. I'm not as familiar with normalize mode and it doesn't seem nearly as obvious.

As you pointed out, the abrupt screen change means that you have lost context as to whether the EGT's have actually gone past peak or whether the JPI made a mistake. :huh:

Normalize mode just sets the current EGT's to their respective middle height, and increases the sensitivity of the bars to 10 degF.  That way you'll see the movement of the EGT's much more clearly when you change your mixture.

And just to clarify, @carusoam, it doesn't change how the data is recorded, so it'll still look the same on analysis

Edited by jaylw314
  • Thanks 1
Link to comment
Share on other sites

1 minute ago, gsxrpilot said:

Yep, this is good advice. Just lean as slow and steady as you can until the engine is running really rough. Not until the first hint of roughness, but just before it quits at mixture cut-off. Then push it back up slowly too rich. 

Keep uploading and sharing the data. This is helpful for others just learning how to use an engine monitor as well. So thanks. 

To be perfectly honest, leaning to that level of roughness ("just before it quits"), with the "less than precise" mixture control in the quadrant, after another F pilot may have just mixture-control-failure'd his way into a totaled airplane doing a GAMI test, and with my history (including preparing to ditch this very plane in the Pacific with a dead engine!) ... skeeves me more than it should. Maybe with some red wine and Xanax! (Kidding!!)

Link to comment
Share on other sites

1 minute ago, chrixxer said:

To be perfectly honest, leaning to that level of roughness ("just before it quits"), with the "less than precise" mixture control in the quadrant, after another F pilot may have just mixture-control-failure'd his way into a totaled airplane doing a GAMI test, and with my history (including preparing to ditch this very plane in the Pacific with a dead engine!) ... skeeves me more than it should. Maybe with some red wine and Xanax! (Kidding!!)

Given recent events, I'd have to agree with you on that :o  Leaning to the first sign of roughness would be plenty LOP for me...

Edited by jaylw314
Link to comment
Share on other sites

4 minutes ago, chrixxer said:

For me LF mode was easier because I could instantly see at a glance when each of the cylinders had passed peak. I'm not as familiar with normalize mode and it doesn't seem nearly as obvious.

Yes. For determining in flight, when you are LOP and how much LOP you are, the LF mode is perfect. I still use it after hundreds of hours of LOP operations. But for the purposes of the GAMI lean test or sweeps, you really don't care when LOP happens and how lean you are. You're just trying to generate data for download later. For that, just leaving the EDM in it's default mode is fine. And possibly even better for doing these very sloooooowwww, GAMI sweeps.

  • Like 1
Link to comment
Share on other sites

I updated the close ups...

One mag is doing something differently than the other...

Anyone agree / disagree?

It does this from beginning of lean to the end... where E4 is behaving differently than the others...

The onus is on Chrix... 

Focus on one challenge at a time.  Do the best with what you have.   Quadrants are a challenge.  Continue to Do your best...

mags are a challenge... do the best with what you have.

Mixture control is a challenge... do the best with what you have.

Best regards,

-a-

  • Like 1
Link to comment
Share on other sites

3 minutes ago, chrixxer said:

To be perfectly honest, leaning to that level of roughness ("just before it quits"), with the "less than precise" mixture control in the quadrant, after another F pilot may have just mixture-control-failure'd his way into a totaled airplane doing a GAMI test, and with my history (including preparing to ditch this very plane in the Pacific with a dead engine!) ... skeeves me more than it should. Maybe with some red wine and Xanax! (Kidding!!)

Fair enough ;) but I've seen plenty of pilots trying to do the LOP thing stop at the first hint of roughness... only to later find it's the same "auto rough" we feel anytime we're over water, mountains, or other unsuitable landing area.

I was doing the same thing until a more experienced pilot showed me... "that's not rough... keep pulling the mixture... now this is rough, so richen it back just a touch"

  • Like 1
Link to comment
Share on other sites

Three Mag tests on the ground... 

100 rpm drops according to graph.  

First single mag op gives holy cow messy data?  Just data or is that real?

Then as things settle in the second and third mag test...

1700 and 1600 rpm... real or fake?  Use a cell phone app if your tach isn’t reporting right...

Best regards,

 

82E7DD8F-9C28-493E-ACF8-E6DEAF8D0392.png

Link to comment
Share on other sites

31 minutes ago, carusoam said:

Close up of the two inflight mag checks...

They are supposed to be identical in appearance... but they are not...

Two cylinders behave identically as the plugs get swapped on/off, top/bottom...

Two cylinders are completely acting differently as the plugs get swapped on/off, top/bottom...

See what I mean?

-a-

 

8EB0D477-9D1B-46BC-953B-101D9E07A16B.png

BEC52AD1-01FB-4707-A0FF-7A688094C6EC.jpeg

You're right. Probably plugs, then, not mags? (If it was a mag timing issue, wouldn't all four be impacted?)

I'm embarrassed to realize I don't know this, but which mags are connected to which plugs on the different cylinders? I know they alternate.

Link to comment
Share on other sites

One thing at a time...

engineers that really like this stuff have a tendency to be this interested...

Most other people are challenged by this level of tech... don’t be surprised if you weren’t born with this skill...

 

Honesty in data delivery is your friend... don’t hide because you don’t know....  :) (so far you are doing quite well with what you have...)

Add this to the list of finding out what is going on...

Somebody mentioned rpm oscillations before... and this wackiness appears in the first mag test, circled in blue... 

1) What is your rpm drop during the mag test on the ground... this is a known standard... and something isn’t behaving right while in the ground...

2) Review your Check list for run-up... See if you have a target mag rpm drop... it isn’t much, but should be consistent...

3) targets are expectations... so you have something to compare to...

4) The second and third mag test are showing something more consistent.... as expected, but still too large...

No need to fly until the mag rpm drops are figured out... they are showing up while you are on the ground...

No need to fly until the how to Best lean on the ground can be determined... E4 and E1 are doing something differently from what you want...

5) While doing your homework verify all the EGT sensors are the same, not a collection of various parts from different suppliers... where they are mounted are important... one exhaust tube has two holes... one can/may be in a weird spot... know where the ship’s gauge is compared to the JPI...

Take notes and share your next steps before heading out again...

BA20F7DF-FB13-4036-9A9C-CCBF7974625D.jpeg

Link to comment
Share on other sites

41 minutes ago, jaylw314 said:

As you pointed out, the abrupt screen change means that you have lost context as to whether the EGT's have actually gone past peak or whether the JPI made a mistake. :huh:

Normalize mode just sets the current EGT's to their respective middle height, and increases the sensitivity of the bars to 10 degF.  That way you'll see the movement of the EGT's much more clearly when you change your mixture.

And just to clarify, @carusoam, it doesn't change how the data is recorded, so it'll still look the same on analysis

Agree to disagree. When I'm in the middle of flight ops, and not so familiar with the EDM-830 for it to be second nature (this is like the 3rd flight when I've been able to actually use it), the LOP icicle screen makes it easy for me to see when the last cylinder has passed peak. Per Savvy's instructions:

Now lean very, very slowly (a vernier mixture control really helps) until the first cylinder reaches peak EGT ... Continue leaning very, very slowly until each cylinder reaches peak EGT

https://www.savvyanalysis.com/articles/in-flight-diagnostics 

Link to comment
Share on other sites

26 minutes ago, carusoam said:

Three Mag tests on the ground... 

100 rpm drops according to graph.  

First single mag op gives holy cow messy data?  Just data or is that real?

Then as things settle in the second and third mag test...

1700 and 1600 rpm... real or fake?  Use a cell phone app if your tach isn’t reporting right...

Two mag checks. What you're seeing there is me running my pre-flight checklist:

6:19 - 6:33, two prop cycles
6:33 - 6:49, both mags
6:50 - 7:17, left mag
7:18 - 7:37, both mags
7:38 - 8:05, right mag
8:06 - 8:20, both mags
8:21 - 8:43, throttle idle check
... etc.

RPM counts are pretty accurate. My A&P checked the mechanical tach optically recently, and I've used the iPad / iPhone apps to check RPMs "audibly" in flight periodically, to cross-check the JPI and the mechanical tach.

  • Like 1
Link to comment
Share on other sites

Note the Savvy example of EGTs...

Pretty hard to see each peak, but they are there.   The data verifies that...

Compare to Chrixxer’s where one EGT doesn’t reach a peak... and continues to drift higher in the flight...

Lots of room for improvement in the mechanics of how to lean Chrixxer’s plane with the JPI and quadrant...

Do it on the ground... download the data... look at it... note the peaks, note changes in rpm during mag tests...

If it doesn’t make sense let’s find out why...

Is it sensor related?

Is it quadrant related?

Is it real or fake data?

Is it PIC related?

Lets get a real quality run-up with two mag tests and a couple of prop gov tests with accompanying oilP drops...

do you hear the weird rpm drops? If you do, they are probably real... if the data is shifting and you don’t hear it... it is probably not real...

Then go on to the next steps...

The plane is talking.  Keep improving your data collecting skills. But, don’t ignore what the data is saying so far... there is room for improvement.

Best regards,

-a-

Link to comment
Share on other sites

I am wondering whether we’re taking about both a fuel and spark issue   

 

And just to confirm did the spider and lines ever get either replaced or thoroughly inspected?  I’m trying to work out conceptually whether high fuel pressure after the servo was replaced could was associated with high or low flow. If the servo was replaced or metering was down-adjusted because of a downstream obstruction/ foreign body you’d end up potentially normalizing the pressure and decreasing the flow too much. Tells me I need to study Bendix failure modes a bit better :-).   My overall question is if a piece of something landed in the wet part of the fuel servo where else did it go?

Baby bottle test will be useful.  Do them with nozzles on and off.  Make sure no kinks or cracks in lines while you’re there. 

make sure that the spider was opened and inspected.  

If you want to test flow through lines you can remove individually and test their flow rate roughly by attaching to a flask / funnel / vessel with a known volume of liquid and time how long it takes to flow past  for each line (like you’d do for a paint viscosity test).  

There has been mention of a potential for induction leak in the thread- just do the savvy profile to help rule out  

for a 4-banger lycoming you can swap injectors to get a better gami spread.  

Given your somewhat bad luck don’t touch without a trusted mechanic :-) 

 

Also, once we know the diagnosis this would be a good one for the savvy newsletter 

 

 

Link to comment
Share on other sites

21 minutes ago, chrixxer said:

Agree to disagree. When I'm in the middle of flight ops, and not so familiar with the EDM-830 for it to be second nature (this is like the 3rd flight when I've been able to actually use it), the LOP icicle screen makes it easy for me to see when the last cylinder has passed peak. Per Savvy's instructions:

Now lean very, very slowly (a vernier mixture control really helps) until the first cylinder reaches peak EGT ... Continue leaning very, very slowly until each cylinder reaches peak EGT

https://www.savvyanalysis.com/articles/in-flight-diagnostics 

True 'nuff, and the lean find screens do adjust the EGT sensitivity to 10 degF/bar as well.  I will add, though, that as you richen the mixture on the LOP lean-find screen, eventually the icicles all go off the top and to see any visual data you have to go back to the main screen, which shows EGT as 40 degF/bar IIRC

Link to comment
Share on other sites

A thread on mag rpm drops...

Uniform, 75 to 100 rpm drop for Lycoming four cylinder engines seems to be the norm...

The messy first mag rpm drop still needs an explanation...

Next up... finding which mag and wire goes where...

Best regards,

-a-

Link to comment
Share on other sites

22 minutes ago, carusoam said:

Honesty in data delivery is your friend... don’t hide because you don’t know....  :) (so far you are doing quite well with what you have...)

Add this to the list of finding out what is going on...

Somebody mentioned rpm oscillations before... and this wackiness appears in the first mag test, circled in blue... 

What is your rpm drop during the mag test on the ground... this is a known standard... and something isn’t behaving right while in the ground...

Review your Check list for run-up... See if you have a mag rpm drop... it isn’t much, but should be consistent...

Mag check RPM drop is a bit over 100 RPMs, but within spec ("Neither magneto should drop off more than 125 RPM when operated individually nor should the difference between the two exceed 50 RPM").

... Or so I thought. I just went through the JPI data and did my own tabulation. The left mag's average RPM is 126 RPM below "both," so, 1 RPM beyond what the book says it "should" drop off, and fluctuates ± 6.5 RPM. The right mag's average RPM is 111 below the average for "both," and fluctuates ± 12.5 RPM. Something to keep an eye on, but I'm not super concerned at this point (and I suspect, as I address everything else that's going on, this will clean up a bit, too).

Screen Shot 2019-05-23 at 12.35.53 PM.png

  • Like 1
Link to comment
Share on other sites

6 minutes ago, carusoam said:

A thread on mag rpm drops...

Uniform, 75 to 100 rpm drop for Lycoming four cylinder engines seems to be the norm...

The messy first mag rpm drop still needs an explanation...

 

Um, I explained the first "messy ... RPM drop," it had nothing to do with mags.

Link to comment
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.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.