Crank Trigger Info

Before the trigger wheel was installed, the engine would run fine at the low end but I was having problems at high rpm. Much over 3500 rpm and it would not run right and those who know me, know I like keeping it above 3500 all the time, lol. Tomtom brought over a oscilloscope and we looked at the tach signal from the GM HEI and it was very noisy. The spike for the signal was barely above the noise, so I theorized that when the rpms went up the FI computer couldn't figure out the engine speed. This could have been because of the noise and the odd timing of the signal. Anyway, the trigger wheel sorted it all out.

I eventually put a 231 evenfire in and went to a 7 pin HEI module to control spark timing too. For whatever reason, that signal worked fine with the microsquirt I had at the time.

 

Using a tach signal for ignition timing actually just provides a signal by intrrupting a DC ground path. By using a 7 or 8 pin module the ECM is recieving the AC signal which usually works much better. This is the correct signl for GM ECUs. Some after market ECUs, such as Powerjection and Megasquit, show in the software that they are supposed to be able to read tach pulse instead of, or an ignition module signal. I would definitely try an ignition module installation before I would spend the time and effort to install a crank trigger. Powerjection shows it will accept a tach signl or an MSD 6a signal, I have no clue what 6a signl is. Adding a vacumn advance limiter not affect the timing once the throttle plate has been opened since the vacumn has disappeared. It will require changing the weights in the mechanicl advance in the distributor. If you choose to proceed with a crank trigger, make sure it puts out the proper number of pulses per revolution that your ECU requires. A crank trigger puts out the same AC signal that an ignition module puts out, only evenly spaced.

 

John have you tried a tach filter, HEI tach signals are noisy and a tach filter can help, you can make one or Summit has them. I'm using a stock GM HEI with a t-body set up off a 4.3L on a 225 oddfire. Without the tach filter the ecm is reading an erratic rpm and tends to over fuel the engine. Our EFI systems are different so I don't know how a tach filter will affect your system but they are cheap and easy to install. Just a stupid question have you checked your timing, both base timing and fully advanced and can you monitor fuel trim or o2 sensor data while driving?

 

Thanks for the advise guys. Look like I need to do some more research at Summit. I have never heard of a tach signal filter. That's why I started this thread though. As a group we know much more than any one indiviual.

 

I tried a filter with no improvement, although it might be worth a try.

I think I would see if you can put a 7 pin module in the distributor.

Willys43, I think the 7 pin still puts out a square wave DC signal for the ECU but it is a cleaner signal than what you get out of the 4 pin. The distributor does put out an AC signal that feeds the module. I think it is cleaner because the reference signal on pin R is separated from the circuit that the coil is on. On the 4 pin you are tied into the same circuit as the coil. However, I am not an electrical engineer so take it all with a grain of salt. Either way, the 7 pin does put out a clean signal (at least clean enough for megasquirt).

I went with the crank trigger just to eliminate the odd fire variable. The hall sensor I used also put out a 5v DC square wave signal.

I think that you could replace the 4 pin for a 7 pin. You can use pin R for the reference signal to the ECU. When controlling spark, pin B gets 5v to tell the module to control timing using a signal from the computer on pin E, but if there is no voltage on pin B (this is override mode that is used when cranking) it should behave just like the 4 pin. It would be interesting to see how it worked.

 

I'm out on the Barrett Lake Trail today and tomorrow, then out on the Rubicon on Thurs. so I won't be able to try this out until next week. But I'm looking at this for a 7 pin: https://www.summitracing.com/parts/pnx-d2001

I'm still concerned about the oddfire signal. Even if it's a clean signal it will still have that odd lope for the ECU to figure out.

 

I understand the Megasquirt computer can be told to time-average the tach pulses.

 

I am very jealous of your trip!
Good luck with the EFI

 

A tach signal filter is a custom built circuit.

i have one on my TBI set-up that been modified by affordable fuel injection. I don't know anything about electronic but here what it look like on my set-up.

 

"and those who know me, know I like keeping it above 3500 all the time, lol."

this is a Understatement !!!!

 

Jonah, drawings I have available to me show a "Signal converter" in the seven pin module. I just checked an 8 pin and it appears to be DC to the ECU. I have had problems in the past with Gm ECUs not being able to read a tach signal and 7 or 8 pin has corrected it. The oddfire signals may not be a problem for fuel only EFI. From what I have been able to find out the ECUs pulse the injectors once per engine revolution based on the number of timing signals it receives. I think that the ECU can count the signals correctly, it just can't predict proper timing for an odd fire because of the changing # of degrees between firing events. In Megasquirt software you have to tell the ECU how many degrees between the firing events, and that they alternate. How many signals per revolution on your crank trigger?

 

Yeah, if you look at the diagrams you can see how the signal for the ecu on the 7/8 pin modules is separate from the circuit that is running the coil. I am guessing that is why it is cleaner. I do know a factory gm electric guy, he would probably know in more detail.

I was never sure if the odd fire signal was really a problem. I think some of the earlier versions of megasquirt and the holley projection systems didn't have the ability to input the degrees between firing, there always seemed to be some confusion about that. I decided just to skip it and go trigger wheel right away.

I originally built my crank trigger for the holley projection which was set up for 8 cylinders (with no provision for changing it), so I made my trigger wheel with 4 pulses per revolution. When I put in the megasquirt I just told the ecu to expect 4 pulses per revolution so I wouldn't have to make a new trigger wheel. At that point I wasn't controlling timing with the ecu so it just had to figure out engine speed. I eventually put in a 252 (4.1L) evenfire in the jeep and ditched the trigger wheel for the 7 pin module so I could control timing with the megasquirt. That worked great.

Now I have a LR4 in the jeep and hopefully not changing it anymore for a while...unless I get a deal on a 6.0

 

Posimoto, Did you ever resolve you issue?

 

To tell the truth , I haven't had time to work on the Jeep. I just led a work party out on the Rubicon that took a lot of planning time. Before that I took a sub-contractor for Placer County down to the County line at the bottom of Cadillac and he did a GPS survey of the trail back out to staging. We now have a true centerline for the Placer side of the Rubicon. Then my compressor on the Jeep died and I had to buy a different model which required a completely new mounting bracket to be fabricated and on and on and on. Still staring at the 30 spline offset D44 housing I bought from tarry99 sitting on stands in the garage. The list goes on. I did some fruitless research on the Powerjection site trying to find out if their software supported crank trigger, but got nowhere.

 

Better there than here..............like $$$ in the bank!

 

Wow John, you have been busy. And thanks for all of your work on the trail!