What's new
Van's Air Force

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

Engine stumble

walter

Well Known Member
I have flown 17 uneventful hours behind my Barrett IO-390X. It has run perfectly until today. On climb out I noticed the engine loose power or stumble, slightly and briefly, three times and I immediately turned back to land. This has not happened before.

Yesterday the wx was a bit snotty so I de-cowled and looked for any anomalies - none were found. This mornings temperatures were downright awesome at 80. This engine never runs hot, cylinders always below 330 and it's hard to get oil temps above 190. DA was about 2000 ft this morning and field alt is 600.

I pulled the chip from the G3X and pasted today's flight on savvy. The attachment is a snapshot from run-up (prop cycled 3 times), short taxi to runway and the very short flight to landing. Notice EGT's on 1,2 are divergent from 3,4.

I pulled all the fuel injectors and checked for blockage - nothing found. I have dual PMags and each PMag powers one spark plug per cylinder - eg cylinder 1 has a plug powered by the Left PMag and a plug powered by the Right PMag, same for all cylinders. I have not checked the timing yet.

This takeoff was done with fuel pump on throughout the short flight. As I was climbing out I started reducing rpm from 2700 to 2500 when I started to feel the slight power loss.

Later today I did a PMag ground check again and both were functioning properly on the ground but I did notice that it would not idle unless the mixture was pulled way way back. Boost pump was not on. I de-cowled again and found that the mixture control is moving correctly stop-to-stop.

After every flight I post my flight to savvy. Up til today the EGT's on all cylinders have been in sync and look very normal.

Looking for suggestions on what to look for next.

Screen Shot 2021-07-11 at 3.26.48 PM.png
 
I have gotten a stumble about 3-4 times, usually just once at about 1000ft during climbout. Seems to be related to warmer temps, but it feels like an air bubble in the fuel system or a little bit of water. I certainly don't think it's water. What fuel injection system do you have? I have an FM-200 and dual pmags. My stumble is just a very short cut that is gone before it can be picked up with my GRT even at 200 milliseconds resolution, but I sure hear and feel it, and I don't like it. I can't seem to "make" it happen when I try, either with another takeoff, or similar climb.

Any hints most appreciated!
 
Climb - Texas - was the boost pump on? Only thought, agree probably not water.

One check - look back at fuel pressure on savvy to see if it gave any corresponding change.

I was recently check pilot for my 10Friend and had the same thing happen but only once. We did 6 approaches.
 
Not uncommon full power right after take off to have a valve stick momentarily. Causes a shot of cold urine directly to the heart! Issue goes away in seconds, hard to diagnose. Some engines have valve guides pressed into the cylinders without a final ream. ECI used "ball-izing" to final ream the cylinders and give a bit of a knurled inside to allow better oil adhesion. Not Kosher.... but I try 1/2 qt. Marvel Mystery Oil, see if the problem disappears. If it does, most likely a sticky valve.

DAR Gary
 
the plot ...

is that a mag check between 10:00 and 10:30? Am I reading it right that there was about 200 rpm drop on the mag check?

Looks like you pulled power once airborne and then the EGT on #1 and #2 went crazy?

Could you show us the CHT's too?
 
Not a mag check. EMag checks generally result in nearly zero rpm drop.

Electric Fuel (boost) pump on the entire flight.

Here's more information with Fuel Flow, Fuel Pressure, CHTs and Oil Pressure:

Screen Shot 2021-07-11 at 6.30.36 PM.png
 
Last edited:
Emags - -

I am running dual emags on a o-360 and generally see 30-40 rpm drop on one side and 50 or so on the other. Zero rpm drop would make me question grounding....
 
I verified all the wire connections on both PMags after todays flight. Even pulled the retainers and made sure each set screw was tight.

As for the RPM drop I maybe get 10-20 rpm drop. Nothing substantial. Been the same on both planes I've built. (7A-8A)

I'm running a Precision Airmotive fuel injection system. On the ground I lean aggressively to get peak RPM and idle and taxi like that. Run-ups are full rich.

In the air for the first 10 engine hours I was mostly down low full power and full rich. Lately I've been flying LOP at reduced power settings.

The engine is broken in. Oil consumption is negligible. Since my last oil change at 5 hours I'm still at 7 quarts. (8 added during oil change).

This has been the smoothest running engine I've ever flown behind. Right up till the point where I commented this to my wife yesterday. I Jinxed myself.
 
Last edited:
Another data point.
Similar setup. Lycoming/Aerosport IO360. Precision Airmotive FI. Dual Pmags - retarded 1 tooth from TDC. My fuel pressure, engine driven pump constant 25-27psi. With boost pump on 30psi. Your FP looks low.
Mag drop for runup 100 to 150 rpm on each. 700+ trouble free hours.
 
This doesn't make sense but I had this once before exactly as you describe. It happened maybe 3 times total over maybe a dozen flights. No other symptoms. It was around the time I usually replace spark plugs so I did that. Has not returned (~250 hours since.)

I know, what are the chances of two plugs hiccuping on the same cylinder at the same time? Honestly it was probably unrelated but it has not come back.
 
Savvy

You mentioned Savvy but I didnt see their reply. Did they get back to you? Once they run it through their software and someone reviews it they should be pretty good at suggestions no?
I have always had very good luck with their programs helping find a bug once they get the data file.
 
I'm using the free service so i don't get their valuable advise. Using their paid services is definitely an option.

Let's first see if the VAF Brain Trust has seen this before first...
 
My vote is the problem is most likely fuel delivery. The flow divider needs a minimum pressure to open, but should shut down all 4 cylinders in a short period of time, if that pressure is not met ( there is the possibility the event was short enough that all 4 did not see a cut off). Contamination in the divider itself might have blocked off 2 injector lines and not the other two (but generally contamination in the flow divider only affects one injector line).
 
#1 CHT

Interesting that #1 CHT went from hottest to coldest when the EGTs started going wonky. Having both CHT and EGT drop on #1 implies less fuel getting to it.

Maybe said before, but I'd clean the nozzle on #1, blow out the line on #1, and if nothing noted, open and clean the spider.
 
Savvy

SAVVY will ask that you fly their in flight test while logging. Mag check, gami check and induction leak test. I recommend you go ahead and knock this out before requesting a data check. I am a savvy pro client.


https://resources.savvyaviation.com/in-flight-diagnostics/


While flying their test flight outline and submitting is excellent, I have over the years submitted other flights data dumps to them for review when something wasnt 100%. I have found it really helpful as i had some cht sensors that were going bad, and another occasion where my one mag was starting to slide. Both cases were found and fixed on my schedule well ahead of any failures or fat OMG shop emergency service bills. Been a happy camper. With the engine info we now get from these new systems and their software, it is really helpful to those of us that arent engine types to ensure all is as it should be and we arent missing signals. Good luck
 
After every flight I post my flight to savvy. Up til today the EGT's on all cylinders have been in sync and look very normal.

Looking for suggestions on what to look for next.

View attachment 13518

What stood out for me was the fuel flow versus fuel pressure right around the 13 minute mark. As the power comes up looks like your FF comes up and your FP takes a pretty good drop down to about 19 psi or so. Almost looks like your fuel pump is not able to keep up with the fuel demand as you open the throttle. Hard to say, hopefully the SAVVY folks will spot something.
 
Not uncommon full power right after take off to have a valve stick momentarily. Causes a shot of cold urine directly to the heart! Issue goes away in seconds, hard to diagnose. Some engines have valve guides pressed into the cylinders without a final ream. ECI used "ball-izing" to final ream the cylinders and give a bit of a knurled inside to allow better oil adhesion. Not Kosher.... but I try 1/2 qt. Marvel Mystery Oil, see if the problem disappears. If it does, most likely a sticky valve.

DAR Gary
Thanks Gary - is it common to have a sticky valve on new engines? The two listed here both seem to have a lot less than 100 hours. I tried to make it happen today and it would not. I re-checked plugs and plug wires before this flight - all looked good. Put a new plug in anyway.
 
IMHO, it would be extremely rare and unlikely that a valve would stick on two cylinders at exactly the same time.

If that happened, time to get a Lotto ticket!
 
Update

I am in contact with Savvy and we are working through a few items.

I fuel flow tested the flow divider (spider) without the injectors in place to see if there was anything in the lines. There was sediment in each cup.

I did a second fuel flow test afterwards and there was no more sediment and the flows were nearly identical.

I know a ground run doesn't say much but I did an un-cowled, no boost pump (except to start) ground run afterwards. I did a very slow leaning until stumble then slow enriching. Followed by a shutdown of the left and right P-mags. Looks better but I'm not flying it until I figure out where the sediment is coming from or am assured there is no more.

I assidentally knocked over the #2 cup but it had the same amount of sediment.
#1, 2 and 3 top to bottom.
IMG_4536.jpg

Now here's something weird, #4 cup is green colored on first flow run, unlike the rest. All the cups were new and clean. It was clean blue on second run.
IMG_4537.jpg

Ground run.
Screen Shot 2021-07-12 at 5.12.50 PM.png
 

Attachments

  • IMG_4539.jpg
    IMG_4539.jpg
    279.6 KB · Views: 121
I am in contact with Savvy and we are working through a few items.

I fuel flow tested the flow divider (spider) without the injectors in place to see if there was anything in the lines. There was sediment in each cup.

I did a second fuel flow test afterwards and there was no more sediment and the flows were nearly identical.

I know a ground run doesn't say much but I did an un-cowled, no boost pump (except to start) ground run afterwards. I did a very slow leaning until stumble then slow enriching. Followed by a shutdown of the left and right P-mags. Looks better but I'm not flying it until I figure out where the sediment is coming from or am assured there is no more.

I assidentally knocked over the #2 cup but it had the same amount of sediment.
#1, 2 and 3 top to bottom.
View attachment 13588

Now here's something weird, #4 cup is green colored on first flow run, unlike the rest. All the cups were new and clean. It was clean blue on second run.
View attachment 13589

Ground run.
View attachment 13586

That is A LOT of sediment to be found post servo. You should have a filter before the pump and another filter in the servo itself. I would check the servo filter first. If it is properly installed and clean, your problem is the servo, spider, or the -4 servo to spider hose. That sediment is WAY too big to get passed the servo filter/strainer (vaguely remember it being 30 or 40 micron). I would get that sediment into your fingers and determine if it is rubber. I am guessing that it is.

If it is rubber and you can't confirm definitively that it is coming from the hose, then in order to be safe, the servo and spider need to be overhauled, as rubber is breaking down somewhere and will only get worse. Numerous rubber parts and diaphragms in the servo and one rubber diaphragm in the spider. With that amount of debris, you are just asking for clogged injectors and erratic fuel flows.

Are you by chance running fuel with ethanol? How old are the servo and spider?

Larry
 
Last edited:
Pull the brass fuel filter that is in the fuel servo and clean it into a coffee filter. If there is no contamination in the filter, that would imply the contamination was in the hose between the servo and the flow divider or in the divider itself.

I would remove the top cover (careful here) from the flow divider and clean out any remaining sediment in the divider housing. I would also remove and flush the hose coming from the servo and flush it out. I would then, while the hose is off go controls full rich and full throttle and hit the boost pump for a few seconds as a final flush for anything after the servo filter.
 
Little time to spend on this due to workload but this afternoon I pulled the servo fuel inlet screen. It was clean as a whistle. Not a fleck or spec.

I disconnected the fuel line running from the fuel servo to the divider. Nothing came out of it either.

I'm starting to think that the the two plugs in the divider may be the culprit. The sealant they used is similar to the stuff I found in the cups when I fuel flowed the system. It crushes very easily with finger pressure and is solid black just like the sealant. The sealant is hard and crusty but crumbles when pressed. Normally on fittings like this I would use fuel lube and very sparingly on the outer threads. I didn't insert these, they were done by Superior or Precision, back when this was an XP400.
IMG_4550.jpg
IMG_4551.jpg

I fuel flowed again and got more trash - only finer and less of it. I will do it again in the morning and then another ground run.
IMG_4552.jpg
 
Little time to spend on this due to workload but this afternoon I pulled the servo fuel inlet screen. It was clean as a whistle. Not a fleck or spec.

I disconnected the fuel line running from the fuel servo to the divider. Nothing came out of it either.

Just as an FYI, the screen in the servo collects debris inside the filter screen not on the outside.
 
Little time to spend on this due to workload but this afternoon I pulled the servo fuel inlet screen. It was clean as a whistle. Not a fleck or spec.

I disconnected the fuel line running from the fuel servo to the divider. Nothing came out of it either.

I'm starting to think that the the two plugs in the divider may be the culprit. The sealant they used is similar to the stuff I found in the cups when I fuel flowed the system. It crushes very easily with finger pressure and is solid black just like the sealant. The sealant is hard and crusty but crumbles when pressed. Normally on fittings like this I would use fuel lube and very sparingly on the outer threads. I didn't insert these, they were done by Superior or Precision, back when this was an XP400.
View attachment 13627
View attachment 13626

I fuel flowed again and got more trash - only finer and less of it. I will do it again in the morning and then another ground run.
View attachment 13628

Best to pull open the divider and check. It would be a bit surprising for a manufacturer to improperly apply sealant, such that it would be exposed inside the divider. However, work is done by humans, so mistakes are possible, such as getting sealant of the face of the plug. I would not consider the sealant as the source until you can confirm that there is sealant exposed inside the divider.

Larry
 
Last edited:
No ground run this morning because of a fuel leak at the servo inlet. I have new O-rings on order and I bought enough to check the filter at annual for a few years.

There is no filter in the flow divider on the Precision EFII model I have (EX-5VA1) and I've been advised by Precision that if I suspect contamination I should have it serviced instead of pulling it apart myself, which makes sense. I know I can do it but common sense says it's worth the small amount of money to get it done right. I'll have them pull and reseal the plugs with the appropriate sealant as well.

I'll recheck the timing after reassembly. The lack of mag drop may be a wiring issue at the switches. Wouldn't be the first time I messed something up.

Thanks for all the feedback to date.
 
Back
Top