R900 EGT calibration data.

[the_bluester]

I am chasing up for a mate. Is my assumption correct that the Temp inputs on an M84 use a 1Kohm pull up to 5V?

He has wired an R900 to AV6 of his M84 using an external 1K pull-up and I generated a table based on the R900 data sheet, the reading rises with a running engine as expected however at ambient temp it is reading about 200 degrees.

I will be able to get a look at his sensor next weekend but not the car, is it safe for me to assume a 1K pull-up in the calibration data to work backwards to sensor resistance? If the reading at ambient is sensible I need to send him chasing a wiring issue but I would like to eliminate a problem with the sensor first

[David Ferguson]

Any chance he connected the pull-up to 12V (or 8V) instead of 5V?

[the_bluester]

That one had not occurred to me. It is always possible. I will generate an ECU config for him that will effectively produce a reading of what voltage is on the input and get him to disconnect the sensor (I can recall when he was doing it that the pull up is in the patch loom he made not on the sensor side of the last connector)

If he does that and it does not produce a value equivalent to 5V then bingo. Otherwise I may have to get to the car and meter things out properly to make sure his 1K really is 1K and not a dud or a mis labelled 1100ohm or something silly like that. Or even just a mistake on his part, he does not have an electronics background so if he got the wrong resistor by mistake the colour code would not tip him off and I don’t know if he tested it before fitting.

[the_bluester]

Well, everything I can get him to test prior to me getting there with a decent multimeter indicates a dud sensor. I will check his wiring hopefully this coming weekend but the pull up is at least marked correctly for a 1KOhm and thinking on it it has to be to the 5V supply as a higher one would produce a lower temperature value not higher. The only thing I can think of is a dud resistor (High value) a bad crimped or soldered connection around the pull up or a bad sensor.

Is there a known resistance value for a very low temperature such as 25 degrees Celsius? The calibration values do not go that low and I am not sure if they have much meaning at that low a temperature, about all I might be able to do is get himt o remove the sensor and put the tip in boiling water so it can at least be tested at a known temp, working backwards from the calibration values to see what 100 degrees should be in ohms.

[Stephen Dean]

Is this one of the VW117 NTK sensors that you are using? if so, they do not read below 100 degrees C. This from both testing and NTK's own data.

[the_bluester]

This is the data sheet he sent me http://www.motec.com.au/filedownload.ph ... docid=5388 so I assume that is the sensor. He is seeing readings of 200 degrees and more at ambient, would I be right to assume that the NTK would settle around a 100 degree reading for anything lower?

It has readings that look too high at idle as well to me, somewhere into the 600 degree C range (4G63 engine) but it has not been tested further than that yet as it is a fresh and changed engine build and is yet to be tuned.

[the_bluester]

Looks like a bad sensor in the end so the Motec guys in Melbourne will presumably hear from him shortly (Bought direct from Motec)

At ambient temp the sensor resistance equated to about 220 degrees Celsius in a voltage divider with a 1K pull-up. That agrees with the reading he was getting. It was somewhere in the order of 17k ohms where it should have been about 33k

[the_bluester]

Can anyone comment on if that is too far out of range for one of these sensors to be considered serviceable? There are a couple of minor tolerance issues found.

Firstly when the sensor is removed, his pull up comes up at 987 ohms (Close enough I would have thought)
Second, the 5V supply measures 4.94, too much load on the supply? That alone would make the sensor read about 150 degrees with the sensor disconnected.
Finally, where my voltage divider calculation says that at ambient the sensor should be over 30K Ohm at ambient it is reading between about 17 and 22K depending on temperature.

The 5V supply might cause some reading issues at ambient but IMO should be irrelevant by about 400 degrees, before I tell him to talk to Motec, are the sensors inclined to be that far off cal at ambient? resistance wise we are talking about 30% here and it is not going to be such an issue at the other end of the resistance range. I just don't want to waste his time if the correct answer is "Yeah, they do that" and it can be expected to be accurate at useful EGT ranges.

[Stephen Dean]

Send it back please.

[the_bluester]

That was my advice to him, I will let him know that yours is the same. The error persists well into high temps. His car was dyno'd on the weekend and peak EGT according to the sensor was well past the "Destroy all internals" stage. The logs showed EGT peaking over 1100 degrees and that is post turbo.

Like everything else, there has to be a dud one come off the end of the production line now and again, just a PITA when it happens.