2G Where does a 2g ECU get the Tach signal from?

[alice]

Does anyone know where the 2g ECU gets its tach signal for the cluster is? I know that the PIN in the ECU is 73 but where does that pin go to? Is it from the CAS or CPS or combination of the two?

 

[TsiJohn333]

The Tach gets its signal from pin 58 at the ECU, not pin 73. Pin 58 is the engine ignition signal and pin 73 is the Manifold differential pressure sensor.

 

[alice]

Thank you for the response! Where does pin 58 go to in the engine bay?

 

[TsiJohn333]

It goes to pin 4 at the power transistor in the engine bay.

This link shows where it is and what pin it is as well.

http://www.dsmtuners.com/forums/151528353-post8.html

 

[ed1380]

on the power transistor there is actually writing as to what each pin does. one of them says "tach"

 

[steve]

The Tach gets its signal from pin 58 at the ECU, not pin 73. Pin 58 is the engine ignition signal and pin 73 is the Manifold differential pressure sensor.

Nope, the signal on pin 58 is used to verify that the coils fired.
The ECU uses 88 and 89 for RPM, I'm guessing pin 89 from the crank is the primary RPM signal.

 

[TsiJohn333]

Steve, i'm gonna have to disagree with you on this one.

Pin 58 is the ignition signal and also is what the tach uses to get signal as well. My greddy turbo timer calls for me to hook up a speed wire. The speed wire hooks up into a tach signal. I tapped pin 58 and works amazing!

I've yet to try pin 4 though at the power transistor but i can verify that pin 58 at the ecu works.

 

[mike96]

Steve, i'm gonna have to disagree with you on this one.

Pin 58 is the ignition signal and also is what the tach uses to get signal as well. My greddy turbo timer calls for me to hook up a speed wire. The speed wire hooks up into a tach signal. I tapped pin 58 and works amazing!

I've yet to try pin 4 though at the power transistor but i can verify that pin 58 at the ecu works.

Correct. On the 2G eclipse that the white wire running from pin 4 on the power transistor is the Tack feedback for the ECU, the gage as well as the TCU if equipped. The wire splits somewhere in the harness and goes to pin 58 @ the ecu, the other one goes to one of the connectors behind the radio where it splits again one to the Tach the other the TCU connector.

 

[TsiJohn333]

Thank you Mike for verifying that for me.

 

[steve]

Sorry, I misread the original question. The signal on pin 58 is what the gauge cluster uses as a tach signal, but that's not what the ECU uses for RPM. It uses the signals from the crank and cam sensors.

 

[TsiJohn333]

Correct!

 

[Moonlight GSX]

Just to make sure I've read all this correctly a full RPM signal, not 1/2 a signal similar to that of the one at blue connector at the firewall, comes out of the ECU at pin 58. If this is correct then I should be able to wire that directly to my tach input at the gauge cluster, right? I ask this because I have deleted my PTU due to the coils I am using.

 

[brads]

Just to make sure I've read all this correctly a full RPM signal, not 1/2 a signal similar to that of the one at blue connector at the firewall, comes out of the ECU at pin 58. If this is correct then I should be able to wire that directly to my tach input at the gauge cluster, right? I ask this because I have deleted my PTU due to the coils I am using.

Pin 58 is the input to the ECU. Both the ECU and the tach gets the signal from the PTU. So now that you deleted the PTU, you need to generate a tach signal.

If you are using a CDI to fire the coils, it should have a tach output. Read instructions on how to wire it up.

If you are not using a CDI, and have instead picked up some coils with built in power transistors, then you need to generate a tach signal.

 

[Moonlight GSX]

Sorry, I misread the original question. The signal on pin 58 is what the gauge cluster uses as a tach signal, but that's not what the ECU uses for RPM. It uses the signals from the crank and cam sensors.

I was curious because of this statement, if the ECU is still getting it's RPM signals from the the crank and CAS signals then in theory I should be able to just wire it up from the pin 58 as according to the elecrical wiring FSM it shows a tap from the white wire (from pin 58/pin 4 at the PTU) going straight to the gauge cluster.

Or would I be able to tap a signal from the crank/CAS sensor and send that to the cluster? I was also curious because ECU manual for the AEM EMS Series II shows that pin 58 is an output for the tach (cluster I presume - according to the elec FSM)...

 

[steve]

I was curious because of this statement, if the ECU is still getting it's RPM signals from the the crank and CAS signals then in theory I should be able to just wire it up from the pin 58 as according to the elecrical wiring FSM it shows a tap from the white wire (from pin 58/pin 4 at the PTU) going straight to the gauge cluster.

And what do you think sources the signal that goes to ECU pin 58? Didn't you say you no longer have the PTU. It's what generates/sources the signal to the ECU and gauge cluster on a stock car.

See Brads comments.

Or would I be able to tap a signal from the crank/CAS sensor and send that to the cluster? I was also curious because ECU manual for the AEM EMS Series II shows that pin 58 is an output for the tach (cluster I presume - according to the elec FSM)...

I'm not sure offhand anymore what the 2G Crank and Cam signals voltage levels and polarity are. I seem to remember them being 5v and the Cluster is getting a 12v signal each time a coil is fired.

It might be an output on an AEM (not agreeing that it is) but it's an input on the factory ECUs.

 

[Moonlight GSX]

http://www.dsmtuners.com/forums/bolt-tech/439185-aem-high-output-inductive-coils.html#post153019949

Although mentioned here that I could use the crank sensor as a signal, I figured I'd ask about pin 58 here as it wasn't exactly mentioned in the other one other than, like you guys have said, I will lose the tachometer signal (as now fully understood in the FSM diagram for the PTU).

I took another look at the PTU diagram and was curious about the possibilities of running taps off of the four coils -12V (or maybe just one) into Pin 58 or tapping, as mentioned in the thread post, off the crank or CAS signal (or both?) and then sending both to Pin 58 and out back to the cluster...

It makes sense what needs to be done but there must be something I'm missing.

 

[brads]

http://www.dsmtuners.com/forums/bolt-tech/439185-aem-high-output-inductive-coils.html#post153019949

Although mentioned here that I could use the crank sensor as a signal, I figured I'd ask about pin 58 here as it wasn't exactly mentioned in the other one other than, like you guys have said, I will lose the tachometer signal (as now fully understood in the FSM diagram for the PTU).

I took another look at the PTU diagram and was curious about the possibilities of running taps off of the four coils -12V (or maybe just one) into Pin 58 or tapping, as mentioned in the thread post, off the crank or CAS signal (or both?) and then sending both to Pin 58 and out back to the cluster...

It makes sense what needs to be done but there must be something I'm missing.

The AEM coils uses a ground wire to the head. Is that set up internally for the spark current to flow through? If so, that could be a way to cleverly get a tach signal that functions like the stock PTU tach output.

Otherwise, you could wire OR the 4 trigger signals together into 1(2 signals into 1 if you are still using the stock ECM). This could be done with transistors fairly easily, or possibly just need diodes.

I see a problem with your system concept. It sounds like there is no knowledge of the dwell requirements for these, or the output at various dwells? Any idea on the total inductance? The OEM dwell climbs high at low RPM, what are these coils going to do? Seems like an iffy proposition. That is the whole reason for people using CDI, to eliminate the need for dwell tuning.

 

[Moonlight GSX]

The AEM coils uses a ground wire to the head. Is that set up internally for the spark current to flow through? If so, that could be a way to cleverly get a tach signal that functions like the stock PTU tach output.

Otherwise, you could wire OR the 4 trigger signals together into 1(2 signals into 1 if you are still using the stock ECM). This could be done with transistors fairly easily, or possibly just need diodes.

I see a problem with your system concept. It sounds like there is no knowledge of the dwell requirements for these, or the output at various dwells? Any idea on the total inductance? The OEM dwell climbs high at low RPM, what are these coils going to do? Seems like an iffy proposition. That is the whole reason for people using CDI, to eliminate the need for dwell tuning.

Like Steve said, the signal at the cluster requires a 12V signal which is also what the FSM portrayed in the schematic using an NPN style transistor. I figured since the PTU has been basically relocated inside each coil it'd be alright to use the -12V (ground) side of each coil or pair of coils to create the 2 PPR signal needed to "trick" the cluster into thinking it was getting a normal [stock] signal. The voltage at the cylinder head ground, if I'm not mistaken, would be equivalent to approximately 40kV @ 102mA (max).

The dwell time was considered, it took me a while but after Beau stated that my dwell time would be affected by these coils with the stock ECU I went and changed the ECU all together - I now own the AEM Series II module and will be using that to control my dwell time. As for the CDI polarmount had a different explanation as to why CDI needed to be chosen carefully, my whole intentions were to just try the coils and see how well they worked. I don't have a CDI box and the coils are directly controlled by the ECU and powered directly from the battery via a 40A relay to take strain from the ignition switch.

"DWELL:
When setting the dwell the following guidelines should be used:
•
Target a base dwell time of 3mS and
only
increase it when needed due to high cylinder pressures.
•
The maximum individual coil dwell “ON” time must no
t exceed 9mS at any time, regardless of engine RPM.
Exceeding this time will cause the coil to overheat and fail.
•
For continuous duty the maximum “ON” time must rema
in below 40% duty (on 40% of the time, off 60% of the
time). Exceeding this will cause the coil to overheat and fail.
•
For short bursts, the coil dwell can go as high as 80%
“ON” duty but these forays need to be short (under 5
seconds or so) and cannot be frequent. "

This is what the manual stated on the coils for dwell control.

"SPECIFICATIONS:
Output (no load): 40kV minimum
Output (50pF load): 40kV +/- 10%
Output Energy: 103 mJ +/- 7%
Peak Secondary Current: 102 mA +/- 10%
Arc Duration: 2.9mS +/- 10%
Turns Ratio 71:1
Maximum Current: 19 Amps
Maximum Battery Voltage: 17 Volts
Base Dwell: 3.0 mS
Max Continuous Dwell: 9 mS
but don’t exceed 40% duty cycle
Max Intermittent Dwell: 80% duty cycle, 5 seconds maximum
Mating Connector: Packard/Delphi 12162825 “Pull to Seat”
Mating Contacts: Packard/Delphi 12124075 “Pull to Seat”
High Tension Wire Terminal: HEI “spark plug top” Style "

These are the specs for the coils as well. Again as for the dwell control, that was brought into consideration and should handle fine with the control of my other ECU (AEM 30-6310). The only problem I'm running into is the determination of whether or not pin 58 on the Series II is an actual output for the tachometer or as you and Steve stated an input on the stock ECU.

Controlling them, I figure, shouldn't be too hard, as the again the new ECU should be able to control dwell. As for not avoiding dwell control - I'm still slightly convinced as to what polarmount stated in that same thread.

 

[brads]

These are the specs for the coils as well. Again as for the dwell control, that was brought into consideration and should handle fine with the control of my other ECU (AEM 30-6310). The only problem I'm running into is the determination of whether or not pin 58 on the Series II is an actual output for the tachometer or as you and Steve stated an input on the stock ECU.

Controlling them, I figure, shouldn't be too hard, as the again the new ECU should be able to control dwell. As for not avoiding dwell control - I'm still slightly convinced as to what polarmount stated in that same thread.

Sounds like there are at least specs on it, the bit I read in the other thread sounded like people were just guessing at what it should be. For reference though, the stock DSM coils can put out more than that. If I remember right, they are closer to a 78:1 turns, and capable of 100-120mJ. They take a little longer to charge than 3mS, but that is because they have higher inductance.

Someone was talking about the motorcycle coils being no good because many were designed for CDI. Not entirely accurate. Many of the ones people are using in their conversions, ARE in fact used in inductive ignitions on the motorcycle, not CDI. They just are very weak inductive ignitions. And that is the problem.

I forgot to add: It is definitely worth checking the specs on your AEM. I would suspect that tach pin 58 is an input, by default. So that it will coexist with the existing PTU. But you might be able to reconfigure it as a 12V tach output, I don't know how AEM controls their IO.

 

[Moonlight GSX]

Sounds like there are at least specs on it, the bit I read in the other thread sounded like people were just guessing at what it should be. For reference though, the stock DSM coils can put out more than that. If I remember right, they are closer to a 78:1 turns, and capable of 100-120mJ. They take a little longer to charge than 3mS, but that is because they have higher inductance.

Someone was talking about the motorcycle coils being no good because many were designed for CDI. Not entirely accurate. Many of the ones people are using in their conversions, ARE in fact used in inductive ignitions on the motorcycle, not CDI. They just are very weak inductive ignitions. And that is the problem.

I forgot to add: It is definitely worth checking the specs on your AEM. I would suspect that tach pin 58 is an input, by default. So that it will coexist with the existing PTU. But you might be able to reconfigure it as a 12V tach output, I don't know how AEM controls their IO.

Awesome, just the information I was looking for (mostly). I know someone did mention the stock coils being able to put out more but I wanted to try them without going over to a COP setup, kinda my curiosity I guess. But in addition figured running a sequential would be nice to learn too. As for the AEM I'm trying to get some insight on the forum they have to see if they know if that pin is truly an output or not, like you said I figured it'd be as the stock ECU as well but going of the spec sheet it stated it as an output so there was confusion - I'll have to do a bit more research on that (will let you all know once I do - if you don't beat me to it).

Thanks for the clarification on the matter, I think I might have been the only one on here to be dumb enough to try out AEM's "Smart Coils." In the end I at least removed the PTU all together, worse case scenario I'll have a misfire in one cylinder rather than no start at all.

 

[brads]

Thanks for the clarification on the matter, I think I might have been the only one on here to be dumb enough to try out AEM's "Smart Coils." In the end I at least removed the PTU all together, worse case scenario I'll have a misfire in one cylinder rather than no start at all.

Well once you try it out, now other people will know if its good or not. Even the specs they provide leave some questions about the performance of the coil. If that 103mJ is their peak power at 9A, then they are weak. If thats their guaranteed minimum power at 3A, they may be pretty strong. Hopefully the switch to AEM will be relatively smooth.

 

[gsx4g64]

Does anyone know what happened with this thread? Did anyone figure out how to get the stock tach/cluster to work once you have deleted the ignitor?

 

[TSiAWD666]

In a wasted spark setup if you are using coils with tach outputs you can try wiring to one from each pair (this worked for me), or if your coils don't have tach output you can use a tach adapter like Auto Meter 9117.

edit - now that I think about it, in non-wasted setup (I dunno what that's called off the top of my head ) you could probably still just wire to say cylinders 4 and 2, or 3 and 1, since you'll still get the two signal edges per revolution.

 

[Moonlight GSX]

In a wasted spark setup if you are using coils with tach outputs you can try wiring to one from each pair (this worked for me), or if your coils don't have tach output you can use a tach adapter like Auto Meter 9117.

edit - now that I think about it, in non-wasted setup (I dunno what that's called off the top of my head ) you could probably still just wire to say cylinders 4 and 2, or 3 and 1, since you'll still get the two signal edges per revolution.

The word you would be looking for is sequential ignition. As for what he said about wiring those, that's basically what I'll be doing, or trying rather, on my 4 coil setup. Basically tapping off the pairs and running them into one and using it for the signal. The only thing that is a bit confusing is the "output" that is stated on the AEM Series II instead of being an input like the stock ECU... so mine is currently in the works.

 

[gsx4g64]

Is it not possible to run a wire from pin 58 to pin 8 on the white 18 pin connector behind the radio?