Massive engine problem

Jeff Blakeslee wrote:

Chris,

Can you reproduce this problem with the propeller control in the manual position? 

I never tried. But I tested with preslected 5800 U/min and with this the prop will stay at min pitch (its hardware limited at 5500 U/min). There should be no influence by the controller with preselcted 5800.

But yes, I could test this as well...

Chris, I don't know what equipment you have access to but if you had access to an oscilloscope you could set it up so that channel 1 is connected to a loop of wire wrapped around the insulation of one of the ignition A HT leads, channel 2 connected to a loop around one of the ignition B HT leads and then trigger the scope from the tacho output. You could then see the relative ignition timing of the 2 circuits whilst the engine is running correctly and compare it with the timing when the engine is running rough.

This might eliminate any form of ignition timing problem. I accept that you have swapped the ignition modules and this is likely to be a pointless test but at least you would know that you don't have an ignition problem.

I think that you probably have some fuel related problem but I am not sure how to easily investigate that. You almost need a lambda sensor in one of the exhaust down pipes from each side. You can then see if the mixture was wrong on one or both sides. I would have thought that the EGT might indicate this but I don't have any experience of using them.

Here's a thought: I noticed from your videos that when the fault occurs the rpm drops down to a similar speed each time. Could you disconnect spark plugs from one cylinder and run the engine to see if the max rpm is about the same as the max fault rpm? If it is then perhaps that indicates a fault with a single cylinder. If that is the case then maybe you have a mechanical problem with one of the cylinders. Perhaps a sticking valve.

Kevin Stewart wrote:

Chris, I don't know what equipment you have access to but if you had access to an oscilloscope you could set it up so that channel 1 is connected to a loop of wire wrapped around the insulation of one of the ignition A HT leads, channel 2 connected to a loop around one of the ignition B HT leads and then trigger the scope from the tacho output. You could then see the relative ignition timing of the 2 circuits whilst the engine is running correctly and compare it with the timing when the engine is running rough.

This might eliminate any form of ignition timing problem. I accept that you have swapped the ignition modules and this is likely to be a pointless test but at least you would know that you don't have an ignition problem.

I think that you probably have some fuel related problem but I am not sure how to easily investigate that. You almost need a lambda sensor in one of the exhaust down pipes from each side. You can then see if the mixture was wrong on one or both sides. I would have thought that the EGT might indicate this but I don't have any experience of using them.

Here's a thought: I noticed from your videos that when the fault occurs the rpm drops down to a similar speed each time. Could you disconnect spark plugs from one cylinder and run the engine to see if the max rpm is about the same as the max fault rpm? If it is then perhaps that indicates a fault with a single cylinder. If that is the case then maybe you have a mechanical problem with one of the cylinders. Perhaps a sticking valve.

Yes I have access to an oscilloscope but i have no idea how to use this magic box. ;) 

Anythink related to a spark plug or to one side of the ignition circuit I had eliminated by switching off each circuit during the vibration occours with no effect. Sorry for not pointing this out before.

Chris

Jeffry Stetson wrote:

A valve problem here could be from air in the lifters from an incomplete initial purging of the oil system. Since the engine and oil system started dry, I'd do another, very patient and thorough purge of the oil system. Before purging, I'd also pull off the valve covers and check for mechanical damage, like for example broken springs, missing spacers and such.

I think this bleeding air out of lifters only exists in rotax's collective heads. They used this as a way to avid a recall on the series of 912uls engines that I have/had, where the valve spring retainers crack in half and destroy the engine. Their procedure is to really carefully run the engine after an oil change, pull the valve covers off and physically check them. If they haven't fully bled the air out what is the procedure? Run it again.

And if that is all it takes to resolve the air that somehow got past the oil and into your lifter then how could that still be the case after the engine has presumably run for 20 minutes by the time you are in the air?

Some discussion and link to FAA report.

https://teamkitfox.com/Forums/threads/12430-Rotax-912-valve-spring-retainers-failing

I am salty because it cost me half of my Bristell's $40k aud engine's life and Rotax would not even reply to my emails asking about it. They should have suggested we change retainers for $300 which would have saved me $20k,.

To the OP, I haven't read every suggestion but you should consider fitting a clamp on wideband sensor to the aircraft temporarily to see what it is actually doing with mixtures.You can buy a budget motorsport one for a couple hundred bucks.