Top Overhaul, GAMI, CHT's etc...

[ReconMax]

Edit: 1998 M20J

I recently completed a "Top" overhaul. During the break-in, it became apparent that my injector setup was no longer valid. We swapped out an injector on the #4 cylinder that solved the biggest problem we were seeing with the #4 cylinder which was it was running too rich. Now with this swapped out injector, it is still running too rich but not quite as bad. It's still hard to manage though because if I try to run LOP, by the time #4 gets to 10 degree LOP, the other three cylinders have started flaming out.

I did a GAMI lean test and sent the data to GAMI. It turns out they want to change out three of the injectors so they sent me three new ones. 

Before we install them, I wanted to ask a question here about CHT. The #4 CHT is hard to manage. (never was before) I normally target a maximum of 380 for CHT. Currently, as I start to lean out the mixture, the EGT for #4 peaks first and the CHT easily will overshoot 380 if I don't hurry up and get to the lean side.

I'm not currently flying this way though because like I said, the other 3 cylinders start to get rough at this point. I am staying on the rich side and the #4 CHT is hard to manage there too. If I lean the engine such that the EGTs rise above 1300, #4 CHT goes over 380. So, I have to keep it pretty rich, burning about 13.5 GPH @ 7500 to to not exceed 380 on #4 CHT.

Everything lean of this, gets me into the problem of the #4 CHT or if I keep leaning such that #4 CHT comes back down, the other three cylinders are just running too rough

So, here is my question. Do you think the injector swap out will fix this? Is there something else we should be looking for? I noticed that barely cracking open the cowl flaps has a huge effect on the CHTs. I can do this and lean further. However, I don't really want to fly around like that. I can hear it. And, it might take a knot off. I'm not sure about that.

I have looked at the baffling and it looks good. Flashlight test and all. There is nothing about the baffling material that leads me to believe it's kinked or anything. 

I never had to work hard to manage temps before and I'd like to get back to not having to do it now. I'm definitely going to swap out the injectors as prescribed by GAMI. Just thought I'd check in here to see if there were any other thoughts on it. I've been told that sometimes a cylinder runs hotter due to differences in airflow. Maybe that's it.

 

Also, maybe once I swap the injectors, and get get all the cylinders peaking at the same time, it'll be easier.

 

Here is a pic of my engine monitor with me ROP. Notice I am already pushing the 380 and barely have EGTs 1300+. 

 

 

 

Here is another pic with me trying to get LOP. Notice that by the time #4 gets to 10 degrees LOP, the other EGTs especially #3 are already much cooler. (60-70 degrees LOP for #3)

 

I really want to have the option to fly LOP but it's not really even an option right now. Currently, I'm about ~50 into the Top overhaul and oil consumption is very stable at about 1 qt every 8-10 hours.

 

[ReconMax]

I don't recall getting them after we put the GAMI injectors in. The mechanic may still have them. Should I have tried to keep them? 

[ReconMax]

Just now, Evan said:

I'd talk with John-Paul at GAMI. Have him send you different injectors and go from there. If you get your GAMI spread below .5 GPH you should be able to run LOP easily. 

He already sent the injectors. Hopefully, we'll get them installed soon...

[ReconMax]

Current spread is >= 1.2. Before my top overhaul I had it down to 0.4.

[ReconMax]

18 minutes ago, Evan said:

Can you share your Savvy link to review? Not sure if you use them to analyze flights or not. 

Most recent GAMI lean test (three tests)  with mag test following final lean. 1 second sample rate...

https://apps.savvyaviation.com/beta/shared/flight/6781818/dcfbc8b0-e5c7-4107-b25c-1f08e4cbff6c

Correction on my 1.2 indication. It looks here more like 0.6 here but I could be reading it wrong. I had 1.2 in my head from a previous test run maybe. I'm relying on John-Paul's feedback after reviewing this test run. He indicated #3 was very lean (relatively) and eventually decided to lean out #4 and enrichen #1 & #3. Like I said, I now have the injectors. Just waiting to get them installed

#1 is going from C to A

#2 no change can't remember what is there now

#3 is going from H to J

#4 is going from C to A

 

 

[KSMooniac]

You could have multiple problems after an (invasive) top overhaul.  You should triple-check for induction leaks on all cylinders, do a GAMI Hi-Lo test, and check the timing of the mags.  All of these could contribute to what you're seeing IMO, especially if you didn't have big issues prior to the TOH.  Most Lycoming IO-360's can be made to run fine very LOP on stock injectors, too, so it should not take an heroic effort with GAMIjectors to get there.  I've got mine running stock with a GAMI Spread of 0.0.  All it took was swapping two injector positions after my overhaul, and verifying everything else.

[ReconMax]

4 minutes ago, KSMooniac said:

You could have multiple problems after an (invasive) top overhaul.  You should triple-check for induction leaks on all cylinders, do a GAMI Hi-Lo test, and check the timing of the mags.  All of these could contribute to what you're seeing IMO, especially if you didn't have big issues prior to the TOH.  Most Lycoming IO-360's can be made to run fine very LOP on stock injectors, too, so it should not take an heroic effort with GAMIjectors to get there.  I've got mine running stock with a GAMI Spread of 0.0.  All it took was swapping two injector positions after my overhaul, and verifying everything else.

I'll mention the idea of an induction leak. Is there anyway for me to tell or is that just something I have to leave up to the mechanic? I'm not familiar with the GAMI Hi-Lo test. Do yo have a link for this with information? 

Really, I am assuming basic stuff like mag timing and induction leaks would be things that the mechanic is very aware of and paid particularly close attention to. That said, anything is possible. 

We did have the entire fuel system sent off and rebuilt (pump,  servo, etc.) as part of the work too. This may account for the need to switch out injectors more than just the top overhaul. 

 

[jaylw314]

Induction leaks should cause fuel balance that changes between partial throttle and WOT.  If your #4 is peaking late at partial throttle, but the problem goes away or diminishes at full throttle, it could be an induction leak

If you have the stock injectors, putting them back in while you're waiting might be informative.  Realistically, IO-360's really should be pretty darn well balanced to begin with, with their distributed air intake (not sure what the name for that is) as opposed to Continental setup.  Even with stock injectors, my GAMI spread at cruise is about 0.6 gph

[KSMooniac]

I couldn't find an easy link quickly, but here is a copy-n-paste from a good Beechtalk.com thread:

"(Assuming normally aspirated engine):
High Test: Climb to an altitude 9-12K or so and run a normal GAMI test at wide-open-throttle. Note the manifold pressure and RPM while you do this. Record all the data (either manually, or using the engine monitor recording function [better, if you have it]).

Low Test: Then, ideally on the same flight, descend 6-8K feet and repeat the test, with the same RPM and with the throttle set so as to yield the same manifold pressure as you recorded during the high test. Record the data using the same process as above.

If the engine shows a discrepancy in the relative order of cylinders peaking, that is often indicative of an induction leak. (During the high test and at wide open throttle, there is little difference between ambient air pressure and air pressure in the induction system, so a leak has little effect. During the low test, with partial throttle, there is a difference between ambient air pressure and pressure in the induction system [ambient pressure being higher, of course] and that will help reveal an induction leak.)

One other way that you can observe a large GAMI spread is by running the test too fast, so make sure you're doing it nice and methodically."

[ReconMax]

31 minutes ago, KSMooniac said:

I couldn't find an easy link quickly, but here is a copy-n-paste from a good Beechtalk.com thread:

"(Assuming normally aspirated engine):
High Test: Climb to an altitude 9-12K or so and run a normal GAMI test at wide-open-throttle. Note the manifold pressure and RPM while you do this. Record all the data (either manually, or using the engine monitor recording function [better, if you have it]).

Low Test: Then, ideally on the same flight, descend 6-8K feet and repeat the test, with the same RPM and with the throttle set so as to yield the same manifold pressure as you recorded during the high test. Record the data using the same process as above.

If the engine shows a discrepancy in the relative order of cylinders peaking, that is often indicative of an induction leak. (During the high test and at wide open throttle, there is little difference between ambient air pressure and air pressure in the induction system, so a leak has little effect. During the low test, with partial throttle, there is a difference between ambient air pressure and pressure in the induction system [ambient pressure being higher, of course] and that will help reveal an induction leak.)

One other way that you can observe a large GAMI spread is by running the test too fast, so make sure you're doing it nice and methodically."

Interesting... Thanks for the info. The concept sounds reasonable...

[Ragsf15e]

If opening your cowl flaps is helping that much, I’d be looking for an airflow issue.  Is the back metal baffle really flush against the back of 4 or is it allowing airflow?

Another thought, at your power setting and OAT, 370 isn’t that hot.  Was it above that in the climb because you get a lot less airflow in the climb, or were you just staying really rich to keep it cool?

Id balance the injectors within 0.5 and look at airflow.  

[ReconMax]

Yes. I am following the GAMI procedures for conducting the Lean test followed by the mag test. So, I lean one more time and do the mag test for the time indicated in their documentation. 

[ReconMax]

1 minute ago, Evan said:

Something is going on with your #3. It looks like #3 intermittently is firing. What kind of plugs are you using? This coupled with it being the leanest cylinder, you might have an issue with an induction leak. 

I don't recall the plug types. I'll mention the possibility of an induction leak to the mechanic and see what they say and check back in here. It might take a couple weeks to get the mechanic's attention on all this.

Thanks for looking and sharing your observations. Mooneyspace is a great resource. 

 

 

[ReconMax]

20 minutes ago, Ragsf15e said:

If opening your cowl flaps is helping that much, I’d be looking for an airflow issue.  Is the back metal baffle really flush against the back of 4 or is it allowing airflow?

Another thought, at your power setting and OAT, 370 isn’t that hot.  Was it above that in the climb because you get a lot less airflow in the climb, or were you just staying really rich to keep it cool?

Id balance the injectors within 0.5 and look at airflow.  

All the baffling looks good. I don't see any issues. It is odd that cracking the cowl flaps a tiny bit improves the temperatures so greatly. Those images were taken from cruise. I'll continue to discuss and investigate airflow issues with the mechanic. Thanks for the input.

[ReconMax]

23 minutes ago, Evan said:

Your GAMI spread is consistently .6-.7 with #3 being the leanest and #4 being the richest. After looking at the flight data again, it seems #3 might be misfiring because it is almost 120 degrees LOP. 

Here it is from Savy

Your problem is really a lean outlier, not a rich outlier. 

Also, you are leaning a tad to fast for the test. Try to slow that down some as you can get false peaks if you go to quickly. 

Report back when you get the new injectors in.

Thats also what John-Paul concluded though ultimately he decided to lean out #4 and also enrichen #1 with #3.

 

I’ll report back when we get it done. 
 

Thanks again…