Howell Odd Fire V6 Fuel Injection Kit Install

I agree with Mjrafan says about needing only 3 pulses per each crankshaft revolution but I will say this also, most aftermarket crankshaft tone rings for this era ecm have 6 nodes. The crankshaft in my Chevy v6 engine has a tone ring built into it with 6 nodes.

3 more pins added to your crank may not be a bad experiment.

 

My brain fart was leaving the power for the map and tps un plugged..... I misread the instructions and pulled that pin out of the ecm connector...... and left it out......
I was supposed to tie into it for power to the crank sensor. I tied them together and all was good again.

Testing with my multimeter, A11 and D2 are both grounds.

funny thing is....
Howell’s instructions say
C14= 5v reference..... power into sensor
B3 = low reference....ground into sensor
B5 = rpm signal into ecm...... output from sensor

If I hook the sensor ground up to B3, the sensor will not work. I tried it in two different ECM’s.
If I ground the sensor to the firewall, ECM housing... ect... the sensor works.

 

Prior to the crank sensor install the block learn was 125-130, integrator was 128.
As soon as I fired it up on the crank sensor it threw a lean code. Block learn suddenly was in the 150’s. I had to increase the fuel pressure from 12.5 psi to 15 psi to get the numbers bouncing around 128.
It’s pretty much impossible to hold a steady idle now too. I have done the IAC relearn a few times and that has not helped.

The chip is a “225” chip from Howell.

 

I’m thinking that 6 pulses per revolution will allow me to back the fuel pressure back down to the 12 range.

 

3 more crank nodes......
Just that fast!

 

I think this a fair assumption especially with the added information you provided when replying to Mjrafan. And of course, pay attention to your BLM and INT numbers.

 

6 nodes.... that’s allot of fuel....
Engine was way rich, black tail pipe smoke rich. I backed the fuel pressure down to 12 psi but it was still too rich. At 12 psi integrator was 90, blm was 110.
I’ll go back to 3 nodes tomorrow. I have a chip provided to me by Howell that has a “richer” tune on it. I’m going to throw that chip in tomorrow with 3 nodes on the balancer.

 

I hate to ask how many hours you have into this(ducks quickly). Thanks for all of this your patience of Job is impressive.

 

...... I try to stay positive because I am learning too.....
But I feel like way too many.... It has ran good, but never as good as I expected it should. Probably some guys coming from poorly operating carb setups would be happy..... but my MC2100 still ran better than this Howell setup ever has. There have been high points to the journey, but it just leaves lots to be desired.
I have joined the Gearhead EFI forum. There seems to be a bunch of guys over there who have scratch built their systems. I am hoping they will bounce some ideas off me.

The Holley Sniper 2300 2bbl is starting to look attractive to me. It is 580 CFM, supports up to 350HP, uses wide band 02 and with fuel system is cheaper than this Howell kit.

 

This is a VR sensor, no? There is some good discussion of VR signals in the Megasquirt documentation. If you need to signal-condition to a square pulse, all the circuitry is laid out.

 

Everything I am finding says that the sensor I am using is a Hall sensor.... but maybe it's not the same as others.
I have a Hall sensor on order from DIY autotune, with the accompanied resistor it will accept 5v input and send out a 5v signal.
Hall effect threaded body crankshaft position sensor

 

Hang in there Kyle, D2 and A11 are the low ref. (grounds) GM use for the 5v ref. circuits, grounding the crank sensor through one of these pins may give you a cleaner rpm signal. It sounds like you not getting a clean rpm signal into the Ecm, do you have a lab scope?

 

I do not have a lab scope.
I will tap the sensor ground into D2 or A11 and see what happens.
I am still wondering what is causing the B3 terminal not to do what it is supposed to. I tried it on both ECM's I have and neither worked when grounding through B3.

 

And to think I got a jeep because they are easy to work on. My head hurts.

 

A thought just occurred to me....

I was thinking about efi systems, and all of a sudden i recalled Posimoto's build thread years ago, when he was setting up his dauntless with efi. After banging his head against a wall for a loooong time, he finally realized that his aftermarket intake had a bad casting, and was causing a vacuum leak through the porous aluminum!

What I'm trying to say is that maybe something else is causing the problems? Did you change anything else when you added efi?

We all appreciate you documenting things here, and share your frustration that it isn't going as smoothly as you'd like. You'll get it figured out!

 

When I added the EFI, the only thing I did to the jeep was add the fuel return.

 

Well darn....

 

What exactly do you expect B3 to do? B3 is not a ground. The ignition control module based on rpm either sends a 0 volt signal or a 5 volt signal to the ecm at B3 to tell it to control ignition timing or not.

 

Howell is claiming that B3 is where the sensor should ground.
I’m work on connecting the sensor ground in with the TPS and IAC ground now.

 

I tapped the crank sensor ground into D2, one of the grounds used by the MAP and TPS sensors, yesterday afternoon. Grounding the crank sensor externally was obviously causing some interference issues before. The engine runs really nice on the road, but my futzing around with the fuel pressure yesterday now has the idle screwed up. Integrator numbers are bouncing from the 110's up to high 130's, BLM is about 124. I think this is causing my idle fluctuation because the computer is trying to compensate for too much fuel.
I'm no expert..... just my thought.