Head Temp Sensor

[Rtrorkt]

could someone post a picture of where this sensor is located?

[EC900]

Head Temperature Sensor Replacement

[Rtrorkt]

Brilliant, thank you. EC

[Schulisco]

Also a pretty picture:



https://rennlist.com/forums/911-turbo-930-forum/276047-tech-help-installing-cyl-head-temp-sensor-special-tool-needed.html

Also here's picture about the sensors with their location and values to test:


taken from Heidi & Franny's YT video series about "Shamoo", their white 3.2 Carrera project:
https://www.youtube.com/watch?v=-2rvMKMqQ2k&list=PLESd2KXwHSVKKl3FxYSFB3rGKZTsfWpQp

Thomas

[spuggy]

Originally Posted by Schulisco

Also a pretty picture:


Thomas

Uhhm. That's, like, not correct.

This has been covered here several time before - including links to Bosch's datasheet (which really isn't very hard to find): Bosch NTC datasheet (PDF)

The nominal resistance of a Bosch NTC sensor is 2.5kOhm @ 20C +/- 5%.

At 0C, Bosch spec it as 5896 Ohms. Therefore, the acceptable range - to be within the stipulated tolerance - for that sensor at 0C would be 5615 - 6190 Ohms.

And both 4.4KOhm or 6.8KOhms @ 0C are out of spec.

It's not as important what the sensor reads at 90C; I've tapered my warmup cycle trim to nothing at 60C because the motor is warm enough it no longer needs any.

However, Bosch spec 89 Ohms, so the tolerance range +/- 5% is 84.75 - 93.45 Ohms.

[E Sully]

The head temperature sensor is a bit awkward to get to. Besides the location, a special socket is needed due to the attached wire. I cut a slot in a cheap deep socket to get it out. This turned into 2 sockets due to the matter that the new 2 wire sensor uses a different size than the old single wire sensor.

[john walker's workshop]

And don't cross-thread it going in and don't over tighten because it can snap off.

[wazzz]

Originally Posted by spuggy

Uhhm. That's, like, not correct.

This has been covered here several time before - including links to Bosch's datasheet (which really isn't very hard to find): Bosch NTC datasheet (PDF)

The nominal resistance of a Bosch NTC sensor is 2.5kOhm @ 20C +/- 5%.

At 0C, Bosch spec it as 5896 Ohms. Therefore, the acceptable range - to be within the stipulated tolerance - for that sensor at 0C would be 5615 - 6190 Ohms.

And both 4.4KOhm or 6.8KOhms @ 0C are out of spec.

It's not as important what the sensor reads at 90C; I've tapered my warmup cycle trim to nothing at 60C because the motor is warm enough it no longer needs any.

However, Bosch spec 89 Ohms, so the tolerance range +/- 5% is 84.75 - 93.45 Ohms.

Don't know for sure is this data is correct, but it is what appears in the Porsche Workshop Manual, which I think can be trusted.

Also, the Bosch NTC you refer to does not match the sensor used for CHT.
The one used for CHT has a non detachable lead with connector at opposite end, while the one you point to has connector with no lead. Thread size and wrench size are also different.

[spuggy]

Originally Posted by wazzz

Also, the Bosch NTC you refer to does not match the sensor used for CHT.

The one used for CHT has a non detachable lead with connector at opposite end, while the one you point to has connector with no lead. Thread size and wrench size are also different.

This is an industry-standard sensor. One of at least two (standards are great, there's so many to choose from, heh).

Bosch developed their own compound for their NTC sensors. They use the same compound across the entire range of NTC sensors they supply - and packaging variations such as grounding/# of leads, thread size, sensor reach etc etc make zero difference. Because the chemical composition of the thing inside that does the work - all the same.

Which is borne out by Bosch's own datasheet, the sensor calibration tables provided by multiple aftermarket ECU suppliers, and 10 minutes worth of high-school level observation with rudimentary tools like a DMM, IR thermometer and a stove/heat gun.

Other manufacturers such as Magneti Marrelli (as used by Ducati & Fiat at least, and I think some years of Land Rover), developed their own NTC compound. And their sensors have a different temperature curve and a different calibration table.

Originally Posted by wazzz

Don't know for sure is this data is correct, but it is what appears in the Porsche Workshop Manual, which I think can be trusted.

If the data presented in the FSM contradicts the datasheet from the manufacturer of the device that Porsche sourced (and I'm pretty sure it didn't last time I bothered to compare, it's just a little less detailed/precise than the one provided by the actual manufacturer) - then it is wrong.

MoTeC provide a calibration table as a preset for the industry-standard "Bosch 2.5K NTC sensor" that exactly matches the curve represented in the Bosch datasheet. I didn't even have to type it in.

[wazzz]

Well I'm not saying you are wrong. You seem to know a lot about those NTC sensors and their curves.
But the figures that you said are not correct in the pic posted by Thomas above (and that come from the H&F YT video) are the exact figures found in the Porsche Workshop Manual. I double checked them before posting.

[mysocal911]

Originally Posted by spuggy

This is an industry-standard sensor. One of at least two (standards are great, there's so many to choose from, heh).

Bosch developed their own compound for their NTC sensors. They use the same compound across the entire range of NTC sensors they supply - and packaging variations such as grounding/# of leads, thread size, sensor reach etc etc make zero difference. Because the chemical composition of the thing inside that does the work - all the same.

Which is borne out by Bosch's own datasheet, the sensor calibration tables provided by multiple aftermarket ECU suppliers, and 10 minutes worth of high-school level observation with rudimentary tools like a DMM, IR thermometer and a stove/heat gun.

Other manufacturers such as Magneti Marrelli (as used by Ducati & Fiat at least, and I think some years of Land Rover), developed their own NTC compound. And their sensors have a different temperature curve and a different calibration table.



If the data presented in the FSM contradicts the datasheet from the manufacturer of the device that Porsche sourced (and I'm pretty sure it didn't last time I bothered to compare, it's just a little less detailed/precise than the one provided by the actual manufacturer) - then it is wrong.

MoTeC provide a calibration table as a preset for the industry-standard "Bosch 2.5K NTC sensor" that exactly matches the curve represented in the Bosch datasheet. I didn't even have to type it in.

For the 911 3.2 DME ECM temp input, it really doesn't matter what the curve looks like other than:

1. 2.5K < cold < 3.5K ohms
2. engine at normal op temp - < 200 ohms

[fintstone]

I am confused. Porsche says that from 60 - 85° F (15-30C) the resistance should be 4.4 - 6.8 k.

But the linked Bosch datasheet says Nominal resistance: 2.5 k ± 5 % at 68F (at 20°C)...but does not seem to be the same sensor.

If I have surging idle and mine reads 2K with a cold engine (ambient temp at 25C)

The linked Bosch data says it should be 2,500 - 1,707 at 20 -30 C...so do I call it good...or do I use the Porsche info and replace it? Does anyone have a new one they can put on an ohm meter and see what it reads cold?

[wazzz]

Mine is good and reads 2.5 kohms at 22°C with engine stone cold, so I'm tempted to think yours is good too, as it reads 2K at 25C.
As Dave mentioned above, accuracy is not crucial for this application, so long as values are in range.

Here is a few values I once copied from this board:

0 degrees C – 4,400 to 6,800 ohms
15-20 degrees C – 1,400 to 3,600 ohms
40 degrees C – 1,000 to 1,300 ohms
80 degrees C – 250 to 390 ohms
100 degrees C – 160 to 210 ohms
130 degrees C – 90 ohms

[fintstone]

Thank you!

[Schulisco]