EDM-900 RPM detection failure

[DXB]

So  the RPM reading on my previously beloved EDM-900 went nuts last week.  It works ok at startup, assuming the plane has been sitting for a while.  Then at some point in cruise flight, the reading starts fluctuating wildly (it's definitely the monitor system - not the engine or governor).   Then it stops reading outright and indicates the big red X. I had two different mechanics look at it briefly on the rather long, 8 leg trip I took last week but they couldn't do much. The rpm sensor that goes in the inside bottom vent plug hole of my right mag (Bendix) is very hard to access and may require pulling the mag to check it fully.  It looked ok with a mirror and feels secure. One of my stops was at Oshkosh, where I got to ask Joe Polizzotto, the CEO of JPI, about it at their exhibit booth. He said it wasn't a common issue that he was aware of, but happened to have a new sensor on hand and gave it to me.   He also said that it could be caused by the magnets in an old mag getting weak - I'm not sure I totally get this - could the problem really be intrinsic to the mag?  My mags did have their 500hr service 6 months ago.  

Questions:

1. Anyone with similar experiences out there that got addressed?

2. Any other guidance to give my avionics shop when I take it back to them to swap the sensor?  

I will be very frustrated if the sensor gets swapped and I still have the problem- it can take up to an hour to show up in flight, so I may be back home before I know if the fix worked.  

[Browncbr1]

sounds like the lead connected to the mag may have come loose either at the mag or at the DB pin.  I had a pin that hadn't fully seated when the installer put in my 830.  It was for the oil pressure,,,  I pulled the DB connector and saw the pin wasn't sticking out as far as the others, a little push and click solved it, 

 

 

Edited August 1, 2016 by Browncbr1

[DXB]

8 minutes ago, Browncbr1 said:

sounds like the lead connected to the mag may have come loose either at the mag or at the DB pin.  I had a pin that hadn't fully seated when the installer put in my 830.  It was for the oil pressure,,,  I pulled the DB connector and saw the pin wasn't sticking out as far as the others, a little push and click solved it, 

 

 

Yeah we checked and rechecked that lead connector - pretty much because it's the only easily accessible part of the sensor - even put a zip tie around it t make sure it was staying together securely even though it looked ok.  No luck.

[PR474]

I had the same problem with the RPM on my JPI 830, installed in M20J. Replacing the sensor did not work, finally a new shop replaced one of the pins in the connector and have not had a problem since.

Paul

[Aerodon]

JPI do not use the best quality pins and connectors - I far prefer the barrel style crimp pins than the tape style.  But I guess when there are so many on each set, they do not want to spend so much.

I would definitely look at the connectors on the back of the 900, see if any pins have pushed out of the connector.

 

Don

 

[Marauder]

Dev - I would definitely check to make sure the connector and pins are secure. Connector J3 takes the sensor input to the unit.

Sent from my iPad using Tapatalk

[N601RX]

I would suspect the crimps on the end of the JPI harness where it plugs into the sensor. The wire may feel secure but there could be some insulation under the crimp. Other thing to look for are pushed pins in the connectors.

[DXB]

Thanks for the input y'all.  I hit the avionics shop next week and will report back on the outcome. Unfortunately it tends to work normally at startup and can take up to an hour in flight for the problem to appear.  So the outcome may simply be a lot of cursing around the time I make it home.  

[wishboneash]

My ammeter readings disappeared (bad ammeter shunt probe error) on my way back from Oshkosh. I was monitoring the battery voltage for the rest of the flight (stayed around 14V) and it was just that the sensor information was lost. Shop discovered a blown fuse in the sensor wire leading out of the shunt. Good thing those wires are protected, otherwise it could be a cockpit fire behind the panel! Seems like a one time event and can't find reason for it like bare wire chafing, but will keep an eye out out and check wiring again if it happens again.

[carusoam]

The sensor wire is measuring the tiny voltage drop accross the shunt.  If a sensor wire was to accidently ground it may be carrying a lot of current...

I Would be looking along the length of the wire from the fuse all the way to the JPI...  Something caused a short to blow the fuse.  That same something may return if you don't look for it and find it.

PP ideas only, not a mechanic...

Best regards,

-a-

[Aerodon]

On August 2, 2016 at 8:44 PM, DXB said:

Thanks for the input y'all.  I hit the avionics shop next week and will report back on the outcome. Unfortunately it tends to work normally at startup and can take up to an hour in flight for the problem to appear.  So the outcome may simply be a lot of cursing around the time I make it home.  

Send my a PM, you can borrow or buy my JPI RPM simulator, plug it in in lieu of the RPM transducer and it generates a signal of 2400rpm.  Lowest cost way of troubleshooting. Don 

[DXB]

So-   turns out it was the sensor itself (the type that goes in the vent plug hole of the mag).   That thing is a BEAR to get to on the C- no way to do it without pulling the vac pump.  Nothing was visibly wrong with the old one, but replacing it took care of the issue.

Forget upgrades, a working tach feels like a luxury after having it not working correctly for a while.

[carusoam]

Nice Pirep, Dev.

Losing reliable rpm or MP readings takes the fun out of power settings.

Best regards,

-a-

[DXB]

On 8/10/2016 at 8:25 PM, DXB said:

So-   turns out it was the sensor itself (the type that goes in the vent plug hole of the mag).   That thing is a BEAR to get to on the C- no way to do it without pulling the vac pump.  Nothing was visibly wrong with the old one, but replacing it took care of the issue.

Forget upgrades, a working tach feels like a luxury after having it not working correctly for a while.

Hopefully this is my final update on this miserable topic.  After after replacing the rpm sensor, it worked fine for several months.  Then it started reading erratically again, particularly at high rpm for takeoff.  I went back to the avionics shop - the pins in  the connector for the sensor were re-crimped.  It worked fine for another month and then the problem returned.  When I went to fly it home from annual, the issue was unbearably bad, and I returned to land.  The avionics tech at the shop simply applied some of this Stabilant 22 stuff to the connector:   http://ralaudio.com/stabilant-22-contact-enhancer-m-2.html

This appears to have completely resolved the problem!  I wish we had tried this in the first place.  I suspect the original sensor was never really defective, and this has been a pointless wild goose chase.  

[bdjohn4]

On initial installations note that the RPM probe (on slick mags) should be screwed into the air port closest to the motor (hidden under the vacuum pump. Do not screw it into the port that is easy to access (and further aft).

Also, to grossly test the probe before screwing into anything, have someone watch the EDM900 display while you wave a big magnet back and forth in front of the probe. Waving as fast as I could, I could register about 300rpm. 

'69 E-model with IO-390

Edited January 8, 2019 by bdjohn4