Solved - Tach signal from Cummins crank position sensor

[randy500]

I looked and couldn't find a good thread on getting a tach signal from the factory crank position sensor. I have searched the cummins forum, 4bt swaps and diesel resource to name a few.

After getting parts of information from numerous threads I was able to get an aftermarket tach to operate from the cummins sensor and I verified the accuracy of the output signal.

I didn't test the sensor but read that it produces I think 5 volts when not across one of the 2 gaps in the crank pulley/balancer, the gap drops the voltage to near zero, so essentially your should be able to get 2 pulses per revolution which is what a 4 cylinder engine would produce.

The crank sensor has 3 wires, black. orange and grey for my 95, these are the wire colors on the engine harness side, I'm not sure of the sensor harness colors but I read that they are colored, I do have a picture of the pin breakout which I will post later.
Black - Ground
Orange - 12v or possibly 5 volt from the computer, I'm not sure, I used 12v and the circuit works.
Grey - sensor output in 2 pulses per revolution.

The problem with the sensor output is lack of (amplitude, I think), in essence the signal strength is weak therefore the signal needs a boost.

I found a post in which someone built a "pull up" circuit to boost the signal output, I also found this same circuit on the Autometer tech website where autometer explained the circuit to correct a customers tach problem.

Essentially you put a resistor between power wire and the signal wire which boosts the signal change from the sensor pulses, the signal is now much higher when not dropping to zero over the crank pulley gap.

I tried operation a sunpro tach using the sensor output and it didn't work, I then wired a 194 bulb between the sensor 12v and the signal wire, this lit the lamp and produced a signal that operates the tach, I used the lightbulb as a resistor because in essence a bulb is a resistor which I think is small value. It was just something I had handy. I did ensure the sunpro tach was switched to 4 cylinder mode.
At idle, the tach reading was about 750 rpm which seems like a reasonable idle rpm, the tach responded to revving the engine and my stock engine only revs to about 2800 rpm or so which the gauge seemed to be reading.

At this point I really wanted to know if I was getting an accurate tach reading because I needed to calibrate my Autometer 2 5/8" diesel tach, I didn't know when I bought it that it would need to be calibrated....

More research found a fellow repairing 70's Camaro tachometers, the guy is some sort of electrical guru, he explained how he calibrates those tachometers using a signal generator. The signal generator outputs a variable sine wave frequency which he sends to the tach to calibrate it. A tachometer in its 4cyl 6cyl or 8cyl mode is expecting a reading in Hz = 60 / half the number of cylinders, the Cummins crank sensor output is 2 times per revolution so the frequency conversion Frequency * 30 = RPM.
I used a multimeter to measure the frequency of the crank sensor output, I was shocked I was able to measure this, I used a $50 multimeter from Home Depot that I have had for a long time, my meter did show a reading and the reading was linear with engine speed so I assumed was accurate and received from the crank sensor, it seemed weird though because I only needed one lead from the multimeter to get the reading.
At idle I was getting a reading of about 25Hz, using the conversion factor of 25*30 = 750rpm which agreed with the sunpro tach reading. I then compared the tach reading to my Hz reading at about 3000 rpm, the Hz reading was 98-100 so the tach and Hz reading seemed to agree and seem to be correct. I was now confident that I could use the sunpro and/or my Hz reading to calibrate the now uneeded, expensive, autometer diesel tach. (BTW, why would they sell a tach that needs calibration???, the instructions say lots of shops offer this service...where are these shops and why would I pay for that, gauges should be ready to operate) That was $160 wasted money for that gauge.

What sucks is I bought the diesel tach because I could not find reliable information on how to get a tach signal from the cummins engine to drive my factory tach or one of the aftermarket gas engine tachs which I have on hand, I guess the new tach looks cool because I matches my Pyro, fuel pressure and boost gauges but all I really needed was a tach signal.

So the real question is will the signal drive a factory gas tachometer....I believe it will, I don't see any reason why it wont. So now you ask, will it be accurate...

It would be accurate if I were able to get 4 pulses per revolution which is what a V8 gas tach expects, the solution is to cut 2 more notches in the factory balancer/pulley. Before I cut those notches I will hook up my factory tach with the "pull up" circuit to test it, it should read half RPM, I'm sure it will but will test it and post back.

[randy500]

So just before I tried all that stuff yesterday I found a post where a guy used this "pull up" circuit with a resistor on a Cummins install in a square body with a video of the factory tach operating, its on YouTube...I wish I would have found that last December when I was researching this and before I bought the Autometer diesel tach, I posted the above explanation and my test results as a culmination of what is probably 20-30 hours of searching and reading posts on Cummins crank position sensor output and getting a tach signal from the crank position sensor...whatever it was fun searching and fun seeing finding the bits of information and testing it out. I will find the squarebody tach video and post a link here.

[1972K20]

Good info....I'll be doing this soon.

[randy500]

Today I soldered a 10k ohm resistor across the wiring harness terminals that connect to the tach along with the previously described wiring to the crank sensor.
The tach worked and responded to rpm but as the engine rpm rose the rpm on the gauge decreased at the upper rpm range.
I decreased the resistance value by soldering in another 10k ohm resistor in parallel, giving me 5k ohms resistance. The tach responded better without falling off at the upper rpm range.
The tach read low compared to my aftermarket one so I added another 10k ohm resistor in parallel which caused little difference.

Im happy as the tach does work and I think it will work better when I add 2 more pulses to the circuit by milling 2 more slots in the balancer.
The tach I’m sure was never designed to be accurate with 2 pulses per revolution that it’s currently operating at.