Corvette Forum : DigitalCorvettes.com Corvette Forums banner
1 - 20 of 21 Posts

·
Registered
Joined
·
22 Posts
Discussion Starter · #1 ·
The way I understand it, the tach reads a signal wire coming from the coil. If I'm wrong, please clarify. If this is true, or the signal wire comes from some place else, how much voltage is sent down that wire? I'm in the process of building microcontrollers that read all the old senders and such from our C3s and output them through a serial connection to an in-car PC hidden in the storage compartment. I'm so far able to read everything but the tach. If I knew how much voltage was being sent and what intervals at what engine RPM, I can make this part happen. But I'm having a hard time finding this stuff out easily. Any info will help me out a ton!

Thanks...
-T
 

·
Registered
Joined
·
3,100 Posts
79 should have HEI. One of the ways to troubleshoot HEI is to connect a 12v testlight to the tach terminal, crank the engine and see if the light flashes. No flash, possible bad module or pickup coil.
Not really an answer to your question, but may help you figure out how to measure voltage. May need an oscilloscope to read it- I don't think a VOM will respond fast enough.
 

·
Premium Member
Joined
·
4,668 Posts
If you have a meter that will register frequency (As in AC voltage) I think that would work better than a voltage. When the (-) side of the coil gets grounded through the ECM the voltage will go from 12 to zero but that will happen so quickly that the voltage may not show up as 12 volts. A digital meter will be plum screwed and an analog meter will just jump a little.

I think that the frequency of the signal will vary in direct proportion to the RPM. I'm not sure of this but it makes a great way to check the operation of the Optispark and I'm pretty sure any ignition would act the same. Good luck.;)
 

·
Registered
Joined
·
22 Posts
Discussion Starter · #4 ·
Thanks for your response. I'll look into how much voltage is being sent from the HEI module. I can sample it many times within one second with a microcontroller, but unless I know the max incoming voltage, my math will be off. Would be nice to know how much voltage GM officially says is being supplied by the signal, at what frequency, and how it relates to engine revolutions. With those figures, I can create the algorithm for the microcontroller rather easily.
 

·
Registered
Joined
·
219 Posts
The signal is a sloppy square-wave between 0V and 12V. Frequency is 4 pulses per crankshaft revolution.
 

·
Registered
Joined
·
22 Posts
Discussion Starter · #6 ·
That's perfect. Exactly what I needed. It does lead me to another question, though. Is the pulse the high end of the signal? What I mean is between pulses, is the voltage on the line 0V or 12V? I would think each pulse would bring 0V to 12V and back down at the end of the pulse, but wanted to make sure it was the inverse.

Thanks for that response. It was exactly what I was looking for.
 

·
Registered
Joined
·
3,100 Posts
Look for either Turbo84 or 69427- Send a PM-- he used to work for GM in that department.. Be THE GUY to ask!
 

·
Registered
Joined
·
636 Posts
Got your PM.

The Tach signal for an HEI is reasonably similar to that found in a normal points ignition, with the exception of the dwell period duty cycle. A points system has a constant 67% duty-cycle angle (30*/45*), while an HEI strives for a constant dwell time (of perhaps 4-5 mSec). These are primary current duty-cycles, and a voltage reading at the Tach/C- lead (via an oscilloscope) would show a reciprocal duty cycle. An oscilloscope signal trace on a points system would show a constant appearance over the RPM range (only a change of frequency), while the HEI signal would show a continually decreasing voltage duty-cycle because while the period is decreasing, the dwell time (C- at 1.5v) is attempting to stay the same 4-5 milliseconds. At high RPM the voltage duty cycle reaches a minimum, as the module preserves a short amount of time for the coil to dump its energy to the plugs. As you can see, the signal is reasonably simple, with the exception of when the dwell stops and the plugs fire. We can divide it up into two parts.

The Tach/C- lead sits at 12-14v during non-dwell periods (as the HEI Darlington power transistor is off and there's no current flowing through the coil inductance/resistance). During dwell the power transistor is on, and most of the supply voltage is dropped across the coil, with the exception of the transistor's saturation voltage (about 1.5 volts). So essentially the Tach is a PWM signal bouncing around between 1.5v (during dwell) and 12-14v between dwell. [ For additional technical minutia, there is a short time just prior to the end of dwell when, if the primary current has hit its calibrated amperage target, the Darlington power transistor comes out of the saturation mode and into active mode, and the collector voltage (same as C- voltage) will reach about 6-8 volts. This mode keeps the current limited to about 5.6 -5.8 amps, negating the need for a ballast resistor to limit the peak primary current. This short event is only visible on an oscilloscope.]
The main concern when connecting the Tach to any instrumentation is the inductive ringup/flyback voltage when the transistor shuts off (and the plugs fire). Depending on the plug gap breakdown voltage, the Tach/C- signal will spike up to 100-300 volts, and then decay with a bunch of ringing on down to 12v again. This is the point where some RC filtering (and perhaps some voltage clamping) is required to keep the Tach/instrumentation circuitry from being damaged from the high voltage spikes each time the coil/plug fires.

So, it's basically a a DC PWM signal with a high-voltage ringing/AC component added onto it.

Sorry for the longwinded description. It went a bit longer than I initially planned. :laughing:
 

·
Registered
Joined
·
22 Posts
Discussion Starter · #10 ·
Wow, thank you so much for that lesson. That was just the amount of detail I needed. Now I just have to put together some components to protect the signal input and then monitor it for the behavior you spoke of. Of course, there will be a lot of cursing and thrown tools in the process, but so it goes.

Thank you again so much!

-Tony
 

·
Registered
Joined
·
22 Posts
Discussion Starter · #11 ·
@Turbo84,

I'm also having a hard time tracking down how much voltage an oil sending unit puts out. The way I understand it is that it puts out a variable amount of voltage based on oil pressure. Is there a chart somewhere that shows the voltage for the psi ranges?

Thanks once again!
-T
 

·
Registered
Joined
·
636 Posts
@Turbo84,

I'm also having a hard time tracking down how much voltage an oil sending unit puts out. The way I understand it is that it puts out a variable amount of voltage based on oil pressure. Is there a chart somewhere that shows the voltage for the psi ranges?

Thanks once again!
-T
I'm not aware of any publicly available chart. Most of the simple sensors are just variable resistors that change due to temperature or pressure. The ECMs or gauge panels have a pullup resistor (to the 5Vcc rail or 12v line) and the sensor voltage is read by an A/D circuit or the gauge itself. Outside of reading the sensor voltage and comparing that to a gauge reading to get the slope/equation, I'm not much help here.
Sorry, that's the best I can do. Post some pictures of your instrumentation setup sometime.
 

·
Registered
Joined
·
22 Posts
Discussion Starter · #13 ·
Thanks for your input. Let me give you a bit of background on what I'm doing here. I'm in the middle of a C3 resto mod, where I created a new dash and console from fiberglass and deleted all stock guages. The car will have a single 8" touch screen in the middle of the dash that displays virtual guages, manages a music library, downloads news, traffic, provides GPS navigation, etc. I know there are other packages on the market that do this, but I own a software company and naturally want to do it my way. Besides, the others on the market don't look as good and they interface with OBDII sensors. Our old C3s obviously predate OBD. So what I have so far is a microcontroller that reads various sensors and reports back to an in-car PC that displays the guage cluster you see in the screenshot below:



It's a bit barren right now, as I havent added in some of the controls like volume, navigation, etc. This is just a shot of the guages themselves so far.

I havent added oil pressure to the screen yet, but that's my next goal. Here's each guage and what I plan to do with each:

  • Fuel: reads resistance. I have a table that shows resistance range for empty to full tank. This is working live in the cluster you see.
  • Battery: This just reads voltage from the battery. Simple enough and works in my bench tests.
  • Oil Pressure: This is where I am now. The way I understand it to work is that it puts out a variable voltage depending on psi. I can't find much information on this, however.
  • Coolant temp: I found a chart that shows resistance-to-temperature and this is working in my bench tests.
  • Outside temp: Very easy to do with simple electronic temp sensors.
  • Cockpit temp: Also very easy and same process as above. This is also working in my bench tests.
  • Speed: Instead of installing a pulse generator on the transmission, I've decided to use GPS for this. I've performed similar tests and I'm confident this will be accurate enough.
  • Tach: This is another big hurdle for me. The tach signal obviously goes through some huge variations as pointed out by Turbo84 in this thread. I'm still trying to figure out how to properly protect the circuit from large spikes and how to effectively count the pulses from the microcontroller. I've also seen that for diesel engines, there are sensors that read alternator rotations and convert that to engine RPM. That might be an option if I have too many problems with the tach signal. But stock tach signal is my goal.

I'm trying to stay as close to original sensors as possible, so anyone else wanting to make this swap only has to purchase the microcontroller and not a bunch of sensors, too.

I think in the end, I'll have a nice package with the electronics needed and the PC software to render cool guages and just a nice automated interface for our cars. Well, for those that don't mind putting some technology in their C3. I understand that the majority of people will want to remain stock and believe that such an upgrade isnt an upgrade at all, but takes away from the nostalgia of these cars. To each his own, I guess. But I love the idea of bringing modern day technology to our old cars. I was watching an Audi commercial a few days ago where they talked about the things the onboard computer can do, and I just laughed. My C3 will be doing all of that and possibly more. :laughing:

Thanks to all who have listened to my crazy ideas and contributed with technical information.

EDIT: I just noticed that I had the fuel and battery sensors disconnected in my bench test when I took that screenshot. Ignore the fact that I'm on empty and have no voltage!

-T
 

·
Registered
Joined
·
3,100 Posts
That will be slick when you get it all figured out. :D

I don't know much that will be able to help, but the temp and oil pressure gauges (stock) work on a variable ground scheme. 12v to the gauge, then the ground is supplied by the sender.
 

·
Registered
Joined
·
22 Posts
Discussion Starter · #15 ·
That does help, thank you. I appreciate you guys taking the time to talk about this stuff. The how and why of the stock sensors is an elusive subject. If I run into more questions, which I undoubtedly will, I'll reach out to you guys again.

Thanks!
-T
 

·
Registered
Joined
·
552 Posts
Michael, I have an idea. Why don't you throw my tach signal into a torn-down tach's circuit board and sample the gauge face signal instead of actual coil pulse?

Yours truly
K.I.T.T.
 

·
Registered
Joined
·
22 Posts
Discussion Starter · #17 ·
That's an idea I hadnt thought about. I have the original tach from my 79 thats being used as the prototype for this endeavor. I'll have to disect it and try that out. Excellent idea! I'll post my results once I'm done.
 

·
Registered
Joined
·
22 Posts
Discussion Starter · #19 ·
@Dojo: I wanted to see your thread detailing your frame-off restoration, but all the photo links are broken. Do you have them up anywhere else? Would love to see that 74 getting all spiffed up. 74s are my favorite C3.
 
1 - 20 of 21 Posts
Top