G3X Kavlico Oil Pressure Weirdness

[JamesClarkIV]

Ok this is driving me nuts. My new RV7A has 20 hours on it, and everything is just lovely...except this darn oil pressure intermittent problem where the oil pressure gauge suddenly pegs off-scale high (data download shows 155 psi, in abrupt change from minutes of nominal readings)

Here is a description, and what I have tried:

First cropped up after ~12 hours on new engine:

Kavlico 150 PSI 3 wire sensor (GND, 5V, signal)
0.5V = 0 PSI
4.5V = 150 PSI
Mounted on transducer block on firewall. Plumbed to the engine with a restrictor valve on the engine side of the hose.

Gauge will read fine: 50-60 PSI during taxi, 70-80 PSI during runup.
It will react normally and then just suddenly jump to 150PSI+. When it jumps there, it usually stays there while I terminate the flight. After engine shutdown, it stays reading pegged 150 PSI+. After avionics power cycle withe engine off, it will sometimes show 150+ and sometimes clear the error and show 0 PSI. If it crops up, it likely crops up in the first 10 minutes after engine start. If it doesn't crop up, I can fly for 2 hours (2 times) and not have a problem.

The 2nd and 3rd time it happened, I landed, quickly removed the cowl and wiggled wires near the oil pressure sender...no change. I separated the connector near the oil pressure sensor (3 lines) and measured 0V GND, 5.01V, and 4.5 V on the sensor return line (this is with sensor disconnected)...Also, seem to have measured 0V GND, 5.01V and 0V on the sensor return line sometimes when I disconnect it (each time i disconnect, I do see a change from off-scale high to red X on display)...this made me suspect the sensor return line was somehow shorting to the 5V supply, or that it has a pull-up resistor in the G3X GSU73 that is pulling high when the gauge is disconnected.

So I spent 2 hours, digging under the panel and disassembled the connector shell at the GSU73 and checked all wiring...the only wiring i haven't check is if there is chafing where it goes thru the firewall pass thru since that is covered in fire seal. With the GSU73 connector removed, none of the 3 wires seems shorted to each other.

This leaves me thinking I should try replacing the $90 gauge. but this is weird failure mode right? Could this be a firmware glitch that is solved by G3X firmware update. Could this be an intermittent GND line? When the ground goes away the pull-up brings the line high and thus the off-scale reading?

I also plumbed a mechanical gauge in parallel with the G3X Kavlico sensor, and the readings during a full runup are within 1-2 PSI from 0 to 50 to 80. But, since it seems i need to fly to get the glitch, I haven't had the mechanical gauge plumbed during any of the actual glitch events. But I'm convinced it is not a real pressure event and it gauge or wiring...

Why would sometimes an avionics power cycle clear the error and return it to 0PSI if it was a wiring problem?

Any ideas would be much appreciated.

[Mich48041]

Yes, a bad ground connection could cause these symptoms. Or the sensor could be going bad. I suggest that you connect a voltmeter between the sensor ground and signal wires right at the sensor and monitor the voltage while flying.

[JamesClarkIV]

What should the voltage on the three wires be when the sensor is not connected but lines are still connected to gsu73 eis box

black=GND
Red=supply=5v
Green=sensor output=Floating? Or pull-up 5v by eis?

[Walt]

I have a spare OP sender I could loan you, if it fixes it we should be able to get it covered under warranty if its not too old, if it doesn't fx it you can just send it back, that would eliminate that variable.

Voltage with sensor disconnected, '0' on green, sensor connected with 0 pressure, .5v

Although I haven't tried it I see no reason why this wouldn't work; swap the sensor wires to the OP sender with FP sender wires then recalibrate the FP on the G3X to read like the OP, other than the name and values we put in the sensor inputs are the same. It will still read fuel pressure but it will actually be OP.

[BillL]

Originally Posted by Walt

I have a spare OP sender I could loan you, if it fixes it we should be able to get it covered under warranty if its not too old, if it doesn't fx it you can just send it back, that would eliminate that variable.

Voltage with sensor disconnected, '0' on green, sensor connected with 0 pressure, .5v

Although I haven't tried it I see no reason why this wouldn't work; swap the sensor wires to the OP sender with FP sender wires then recalibrate the FP on the G3X to read like the OP, other than the name and values we put in the sensor inputs are the same. It will still read fuel pressure but it will actually be OP.

And vice versa. Clever way to check both circuits and separate results from a bad sensor!! Nice thinking Walt. Since vibration probably does not affect the sensor, but being electronic, temperature might. You could use air pressure and heat gun or hair dryer to heat a bit. The sensor data sheet says it's ok to 125C.

[JamesClarkIV]

Ok. Far too many hours in the hangar tonight but finally may have found the culprit: enlisted help to watch the gauges while I hunted for wiring problems. Finally found that tugging/bending the large bundle of wires going into the 78 pin connector on the eis caused the oil pressure gauge to do is jumpy thing

Surprisingly it also disturbed the gauges for the manifold pressure and the fuel pressure. Checked my wiring diagram and found the following wires in common among those three gauges:
Three grounds spliced together and three 5v lines spliced together and also sharing the 5v pin with the two capacitive fuel level senders.

The 5v and ground splices used those heat shrink solder bands and both looked only partially melted solder. I heated both and got much better solder flow. Put it all back together and gauges now steady when flexing and tugging same connector wire bundle. Will fly tomorrow and test. But I'm hopeful it is found.

Thanks Walt for the offer but I will wait and test for a bit more. I had tried swapping gauges without recal just to see but didn't carry it to far or fly that way.

[JamesClarkIV]

Is there any risk that the two extra 5v capacitive fuel senders are over drawing the current on the 5v line?

[g3xpert]

Originally Posted by JamesClarkIV

Is there any risk that the two extra 5v capacitive fuel senders are over drawing the current on the 5v line?

Hello Jim,

As explained on page 23-63 of the Rev. W G3X Installation Manual, the GSU73 +5V transducer power is capable of sourcing 125 mA.

Each of your 3 Kavlico pressure sensors should be drawing less than 5 mA each, so you should have plenty of reserve for the capacitive sensor modules, but it is probably good to measure the current draw of these modules to make sure they are not trying to pull more current than available.

Thanks,
Steve