Page 1 of 2

EGT dropout on M800 PnP w/OKM

PostPosted: Sat Oct 15, 2016 8:12 pm
by mrsaturn7085
Seeing random intermittent drop-outs of my EGT 1 sensor on a 2006 STI running M800 plug-in w/OKM. Signal is normally fine, but over the last week has been dropping to zero w/o good reason for 20-30 seconds at a time (triggering error warning light).

I know these sensors are fairly basic in operation... has anyone else had this happen?

Sensor is a K-type thermocouple with metal braided sleeve on wires, terminated with proper Alumel/Chromel pins.

Re: EGT dropout on M800 PnP w/OKM

PostPosted: Mon Oct 17, 2016 8:57 am
by Stephen Dean
Hi,

Are you using a MoTeC TCA to get the signal into the M800? or another device?

Re: EGT dropout on M800 PnP w/OKM

PostPosted: Mon Oct 17, 2016 9:38 am
by mrsaturn7085
I am using the amplifier built into the OKM (also called TCK in some white papers) which is attached to the M800 plug-in board. Board is a WRX9-10.

M800 Links 4 and 5 are open, link 21 is closed.

K-type connection enters ECU via pins A1 and A2.

Re: EGT dropout on M800 PnP w/OKM

PostPosted: Mon Oct 17, 2016 4:24 pm
by Stephen Dean
Hi,

I would start with checking the resistance of the EGT, you should have between 8~12 ohms across the two wires, if it is open, then the sensor or wiring is at fault.

Re: EGT dropout on M800 PnP w/OKM

PostPosted: Wed Oct 19, 2016 1:32 am
by mrsaturn7085
I will test this when I return to the USA next week - due to the intermittent nature of the problem, I have a feeling it may not reveal the issue. Attached is an image showing the i2 log of three drop-out events during one datalog.

Re: EGT dropout on M800 PnP w/OKM

PostPosted: Wed Oct 19, 2016 3:40 pm
by Scott@FP
How old is the probe, and what type, exposed, or sheathed grounded, sheathed ungrounded?

Re: EGT dropout on M800 PnP w/OKM

PostPosted: Wed Oct 19, 2016 4:32 pm
by mrsaturn7085
Scott@FP wrote:How old is the probe, and what type, exposed, or sheathed grounded, sheathed ungrounded?


10 months installed, with less than 20 hours run time on motor.

Closed sheath, 90 degree, K-type purchased from Innovate (p/n RTD-TE-104SH)

Unknown if grounded or ungrounded. Innovate was very little help when I called about it in the past- if I had to guess, I would expect it to be a repackaged Omega unit that Innovate bought in bulk.

Re: EGT dropout on M800 PnP w/OKM

PostPosted: Thu Oct 20, 2016 9:33 am
by Stephen Dean
Having a look at the data trace, where you load up the engine I would have also expected to see a rise in the EGT being read, even with a short run like that, that it doesn't appear to go above 456 degrees (hard to tell properly from a screen capture) and then drops down to a point below that displayed looks more like a faulty sensor than a issue with the M800 and OKM. Are you able to connect the sensor into a multimeter or similar and read the value that way?

Re: EGT dropout on M800 PnP w/OKM

PostPosted: Thu Oct 20, 2016 7:21 pm
by mrsaturn7085
I've now noticed that this drop-out appears during a different log I took a few days prior (though this did NOT cause the check engine warning... or perhaps I did not see it; error flags WERE triggered in the log file).

The peak temperature at the time of the drop-out, in both cases, was exactly 479 deg C. When the drop out occurs, temperatures measured are in the 476-478 deg C range. So I now know that there is an exact point when this is occurring. I will be able to check resistance this weekend.

This sensor is measuring TIT on a Subaru motor so the temperature climb/fall is a little slower than an EGT sensor located near the exhaust valves. The temperature DOES rise and fall with the behavior of the motor/driver, but it is not very fast. Because of the slow response, I would assume this is a non-grounded sensor, as well.

Re: EGT dropout on M800 PnP w/OKM

PostPosted: Thu Oct 27, 2016 7:00 am
by mrsaturn7085
Measured resistance = 3.2 to 4.1 ohm.

Per probe manufacturer (Exhaust Gas Technology), probe is a closed element, non-grounded K type.

Also, manufacturer confirms this is reasonable resistance value, though claims there is no standard resistance. I will be running the sensor up to 600 deg C with a torch tonight and comparing the M800's ability to read the temp. to that of a multimeter w/K-type input.