Another Crank / No Start RX-8

Chris Technical Support Specialist New York Posted   Latest   Edited  
Case Study
2005 Mazda RX-8 1.3L 6-spd
Crank / No Start

2005 Mazda RX-8 – Crank no start

As many of you have probably seen, RX-8’s have issues with crank no start conditions. Mazda has some tsb’s for Faster spinning starters, dechoke procedures and compression testing. What do you do when none of that seems relevant?

A mobile tech had recently called in with an RX-8 Crank / No Start issue. His customer was already in this pretty deep. His client followed Mazda Service Bulletin 01-048/06.

First, he followed Procedure “A”

Plugs were removed and the Eccentric Shaft Position (ESP) sensor, the rotary equivalent of a crankshaft position sensor, disconnected, then the engine was cranked until the chambers (the equivalent of cylinders) were purged of all possible fuel. The plugs were replaced, and the engine still would not start.

Then Procedure B was followed

About an ounce of 5W20 oil was introduced through each of the vacuum ports on the side of the intake per tsb instructions. Still a no crank no start scenario. So, on to engine compression

This is a rotary engine. Rotary engines have three combustion chambers per rotor. A regular compression tester makes for inaccurate results and there was not a transducer available. If the Schrader valve is removed from a standard compression gauge you can get a ball park idea of the compression in each of the pulses. We had somewhere in the area of 90-100 psi for each pulse on both chambers. This should be enough to start the engine. After all this initial testing and verifying of the previous steps, we looked to scan data. No RPM cranking.

The shop’s tech had made it about this far on his own but took several detours. This would include checking AC voltage at the ESP (CKP) circuits at the pcm. The tech had also cut the wires from the PCM to perform a resistance test through the circuit at the PCM. When his meter was used to check for resistance through the sensor with the wiring reconnected and the PCM plugged in, the car would start.


So we verified the scope pattern of the ESP (CKP) at the sensor, and also at the PCM. The pattern looked as in the service information.

So here we are with a good ESP(CKP) pattern at the pcm (back probing the connector after repairing the wiring) and no RPM signal in live data. When a multimeter was used to ohm out the wires the rpm signal returns and the vehicle will start and run. Obviously there is an issue with the module, but what?


Modules use Bias DC voltage on AC signal circuits. With a known good ESP (CKP) sensor disconnected on another vehicle we were able to determine approximately 2.5V across the terminals. This PCM had no voltage across the ESP circuits. When the meter was on the ohm’s setting it outputted approximately 1.5 V DC and this was sufficient to bring the ESP(CKP) sensor waveform into the module’s expected range.

Most importantly do not forget to check for the basics prior to performing service bulletins. Bulletins aid in diagnostics, but should not be followed blindly. If the live data was checked on initial inspection this would have greatly reduced the diagnostic time invested in this vehicle.

Chris Diagnostician

Now that's just cool, I don't care who you are. I do ohms tests so rarely I never once thought about the meter's output voltage, especially as a potentially useable bias voltage. I appreciate you sharing this.

Ð Awarded
Geoff Diagnostician

Chris, (above), took the words right out of my mouth. Crazy!

Ð Awarded
Matthew Instructor

Great info. Thank you for posting. I've worked on very few rotary engines but the basics never change. The bias voltage is very interesting too. Great stuff.

Ð Awarded