here is the problem i have, no fire. checked with a spark plug in all wires, spark plug holding on the block and no spark at all.

i started troubleshooting and it seems i am getting no signal coming from the PFC. There is one thing i am not sure though, would i be able to see the voltage on my voltmeter even if it's pulses that are coming out of the ecu? i included a pic of the electrical manual, i checked at pin 1J 1G and 1H and i have no signal coming from these when i crank the engine.

would i be able to see it with my numerical voltmeter or do i need some sort of frequency meter?

what could make the ecu not send any pulses to the igniter?

thanks for any help

 

The ECU fires the plugs quickly enough that a DC voltmeter would probably NOT detect it. You might see something if you set it to AC Volts, but the signal is a square wave so the reading won't be accurate.

If I remember correctly, one of the Mazda shop manuals shows a picture of the waveform. I don't have access to my shop manuals, but I'll check it and get back to you tonight with more advice.

-s-

 

hmm that's what i thought so it might be why i see no signal...

what else should i try then? would a cheap voltmeter with a needle work?

 

test light connected to the battery pos
probe the signal side of the coil terminal
it should flicker (signal coming)

check your crank sensors and air gap etc..

do the coils have power?
does the pfc/tach so a reading when the car is given a long crank??

 

From pg. F-152 of the Factory Workshop manual (pg 341 of the PDF),


At pin 1H (leading ignitor), the signal should be a 0-5V square wave, of varying frequency and duty cycle. At idle, each pulse will be ~20ms, with ~60ms between pulses. Frequency looks to be about 12 Hz at idle.


An AC voltmeter will display a pretty low voltage for this signal, probably less than 1V.


Rather than a test light setup, I would hook up an LED and a 270-ohm resistor in series, because it will respond much more quickly. The pulses are really short (0.02 seconds long), so you might not be able to see them with your eye. Since the duty cycle is so low (~20% at idle), you can use a 220 or even 180-ohms resistor so the LED lights up brighter. Remember that LED's are uni-directional, so be sure to hook it up correctly.

LED guide here:
http://www.kpsec.freeuk.com/components/led.htm



-s-

 

is there a way to see this from the pfc or the datalogit? how do you check the gap? what should it be? the sensors seems to be plugged at least...

what do you mean? i should see a rpm when i crank the car on the pfc? is that what you mean? i don't have an other pfc to compare, sorry.

thanks for the help guys, keep it coming.

 

a high impedence test light will do it no problem. i would not trust anything but a snap on or that quality but it has been battle tested a a reliable method in my book

scotty i am talking about going to the coil directly and seeing the ground state switch as the ignitor "fires" the coil. i like to start trouble shooting from the ends of circuits. sometimes you find it is just a bad connection etc...

my tach will start to move when the car is cranking. no tach signal prob a crank angle sensor prob. but i do have a full size batt so my cranking rpm is pretty good

 

ok, i have 12 v at the coils that is for sure, i checked with the rpm and i read around 190 rpm when i crank the car on the PFC so i suppose the crank angle sensor is correct.

i will try to find a multimeter with a frequency reading and see if the ecu really do sends the signal to the igniter.

do i have the right method of checking if i have spark? i plugged spark plug i had laying around that goes in an other car and holded it against a part of the engine to ground it and i saw no spark. I then tried it with one of the spark plug from the car (on from the rotary engine) and i still saw no spark.

do i have the right way of checking if i have ignition?
thanks again

 

The spark test is usually done by holding the spark plug near the chassis, but if your engine is grounded properly you should be seeing a spark.

I bought a Craftsman multimeter from Sears that measures frequency. It was about $20 USD.


-s-