EDM 830 temperature problems finally solved.

[squeaky.stow]

Although I doubt anyone else will ever experience this same problem, I am posting the fix anyway because after a year of tearing my hair out and driving the tech support people at JPI crazy,  I am just so happy to have finally resolved the issue.

i posted about a year ago that I was having unreliable OAT temperature indications on my EDM830. OAT was the indication that initially made me suspect something was wrong because it was easy to correlate to the analogue OAT gauge and the Aspen OAT. It would often overread by 20 to 30 C but it was not consistent.  Further troubleshooting revealed that every temperature indication on the monitor was doing the same thing. If the OAT was reading high, CHTs, EGTs and Oil Temp would also overread by the same amount. This was determined by monitoring a cold engine in the hangar for about an hour with the OAT probe immersed in ice water and watching all of the temperatures rise in lockstep. To make it really weird, I discovered that adjusting the cabin heat in flight had a direct effect on all of the EDM temperatures. I could literally make all of my indicated temperatures move up or down by 20-40C by adjusting the cabin heat. Moving the OAT probe out to the wing and running a new ground wire to the engine block, both suggested by JPI, didn’t help. 

So what was actually going on? It appears that the EDM 830 was originally installed on the right side of the panel but at some point during an avionics upgrade it was moved to the left side. This requred building an extension harness for all of the wiring. Whoever made the extension used regular copper avionics harness wiring for all of the thermocouple leads. Not a good idea! Apparently the resistance of copper varies quite significantly with temperature changes and the currents generated by thermocouples are so tiny that this can throw the indications way off. So every time I adjusted my cabin heat, it would change all of my temperature readings. 

I ordered a new extension harness made of proper thermocouple wiring directly from JPI and last week we removed the old extension and replaced it with the factory supplied one. Test flight today showed OAT exactly matching the analogue gauge and the Aspen, and no variation on any temperatures when I played with the cabin heat.

Whoohoo!!! Finally I can trust my EDM830! 

 

[alextstone]

That's a good PIREP!  Thanks!

[Aerodon]

that is one dumbass installer who did that.  Aerodon.

[carusoam]

Yeah, mixing copper in place of thermocouple specific wires is really bad...

Great pirep!

Now Find the guy that did that wiring job... he needs a friend.

What else did he install?

Best regards,

-a-

[rbridges]

Did you bring that up with the installer?

[squeaky.stow]

2 hours ago, rbridges said:

Did you bring that up with the installer?

It was done years before I bought the aircraft. I don’t think the avionics shop in question is still in business. Not worth the effort to track them down at this point. The good news is that when the temperatures went wonky they would always overread vs. underread. This means I have been running my engine even more conservatively than I had thought when it comes to CHT. With any luck the previous owners were doing the same. 

[rbridges]

21 minutes ago, squeaky.stow said:

It was done years before I bought the aircraft. I don’t think the avionics shop in question is still in business. Not worth the effort to track them down at this point. The good news is that when the temperatures went wonky they would always overread vs. underread. This means I have been running my engine even more conservatively than I had thought when it comes to CHT. With any luck the previous owners were doing the same. 

Oh, I thought it was something fairly recent.  That must have been a very frustrating road for you.  Congrats on fixing it.

[Austintatious]

man, I love stuff like this.  You say no one will ever have the same problem, but hearing your story gives more knowledge to draw upon when troubleshooting.  Thanks for sharing!