Can I use a FN4A-EL Turbine Speed Sensor on a GF4A-EL?

Question

I have a GF4A-EL on my 98 Mazda 626 GF 2L and strongly suspect the TSS is defective. I can't find any sensors for my tranny, but I have read that the FN4A-EL is just an improved version of the GF4A-EL and was wondering if their turbine speed sensors are compatible, since there are allot of the former easily available.

The sensor visually look identical and are top mounted in the same way.

EDIT July 3, 2016

I just ran across this diagnostic sheet for the TSS on the Mazda 3 which also uses the FN4A-EL transmission and it looks the same as for the GF4A-EL.

Input/turbine speed sensor resistance

250—600 ohms (ATF temperature: –40—160 °C {–40—320 °F})

I've found the part numbers. The one for the GF4A-EL is FW01-21-550 and it looks like this:

The one for the FN4A-EL is FN01-21-550 and it looks like this:

Based on the testing procedures, it looks like the two sensors are electrically identical, and they also look like they are physically identical except for that metal tab which is used for screwing them down.

I'll bet that that metal tab can just be pulled off and rotated 90 degrees to turn a FN01-21-550 in a FW01-21-550.

Considering I can get the former for $14 shipped free from Hong Kong and the later I can only get from the dealership for about $100, I think I'll take the chance, and update this edit into an answer if it works.

Answer 1

So I installed the sensor and it works. No more hard, erratic shifting.

The thing with the old sensor is that it was failing intermittently, and maybe also giving incorrect readings. The hold light would flash sometimes and not others. When I looked at the data with a scan tool, I saw that it would typically give a reading lower than the RPM readings, sometimes allot lower, and then all of a sudden it would go to zero, stay there for a few minutes and then the signal would come back.

I think the thing to watch out for when testing these sensors is that you might get a perfectly good resistance reading and think the sensor is completely OK, and if you don't sit and watch the sensor for ten or fifteen minutes, or have a logging scan tool you could easily get a false negative and end up doing some really expensive repairs on the tranny when all it really is is a $15 sensor.

Here are the sensors next to each other:

And another view:

The only problem I ran into was that the wiring for the connector ended up very tight because the sensor ended up turned 90 degrees away from where it was supposed to be:

I basically had to put the connector on the sensor, then turn it and screw it down. The orientation of the sensor doesn't seem to matter, as I drove around for quite awhile, the hold light didn't come on and my shifts where very smooth and felt pretty good.