Possible to have pulses with constant voltage?

Question

tl;dr: I'm asserting that you cannot have constant voltage pulses in a dc circuit. The pulses will be variations in potential or current that some appropriate device picks up. I think I found misinformation on a car forum.

I'm wondering how a car tachometer works. On this LS1 engine forum there is a discussion. "Mr_Dude1" says that :

the LS1 tach signal looks like . . .just shy of 5v square wave, 4 pulses per every TWO revs. that is, 2 pulses per rev. like a 4 cyl car.

it is not a varying voltage... the frequency (how close the little bumps in the pic are to each other) is what changes.

He attaches a picture of a square wave pattern with no axis labels, so I'm not sure what is pulsing. I think there must be electrical pulses, which are either voltage or current, but I think these types of pulses are inextricably linked due to ohm's law. I have to do some surgery on the wirebundle from the ECU of my car to get at the tach wire, so I though I'd ask you guys first what's going on here.

I don't think I can use a multimeter to test this. If the tach wire is 12V on the car, I suspect that the multimeter will read either 0, 6, or 12V (or some other whacky value) depending on the algorithm used for computing voltage. I think an oscilloscope would be the proper tool here.

Answer 1

Tachometer on some very old cars, which are not equipped with ECU/ECM, works in a similar way. This is before early 80's. Some of the cars in the trasition era mid 80's to mid 90's with simple fuel system also connect the tacho in a similar manner.

Method 1 - get the signal from the alternator's winding

This signal is taken from one of alternator's windings on the AC side, before its rectifying diode. Without a load this signal would be sine wave, but having the very low resistance car battery the upper rounded part of the sine wave is flattened. So you get something near a square wave, or to be more accurate - a trapeze wave. Dependancy between this frequency and actual crankshaft rotation speed is not equal for all cars - it depends on number of windings of the alternator and ratio between diameters of crankshaft's pulley and alternator pulley.

Method 2 - get the signal from the ignition coil switch - petrol only, mostly with mechanical ignition

This method applies only to petrol engines and mostly when their ignition is controlled with a mechanism, not electronics. This is before ECU/ECM or in their first years. A signal is taken from the low voltage alternating side of the ignition coil. This signal's ratio is 2 pulses for each crankshaft turn for a 4-cylinder engine or 3 pulses for a 6-cyl engine. This is because the full 4-stroke cycle completes in 2 crankshaft turns, so glowing all plugs sequently till getting back to the first one takes 2 crankshaft turns. Signal shape will be like thin pulses (<5% duty cylce) with most of time positive and active pulse - pulled down to zero.

Method 3 - crankshaft sensor signal

This method is trasitional while moving from electromechanical engine control to fully electronic control. The crankshatf position sensor signal is passed both to ECU/ECM and to tachometer. It's shape is more likely to be as thin pulses with a very low duty cycle - 1% or less. Ratio is 1:1.

Nowadays

Modern petrol and diesel cars have crankshaft position sensor and camshaft position sensor for the ECU to manage the engine properly and it is the ECU/ECM's job to send the correct signal to the tachometer. This signal is a digital comunication from ECU to tacho's chip and then this chip drives a servo- or stepper- motor to turn the pointer (or displays animation on LCD).

That's what I've got from my experience. Please feel free to correct me where needed.