Can a Square Wave Tachometer Drive be Powered by a Sine Waveform?

[KrakenMarine]

TL;DR Summary

Can an alternator's sine wave drive a tachometer that "requires" square wave?

I'm a marine engine mechanic, and as engine controls & sensor systems have gotten more complicated with current technology, my shop gets more & more requests for instrumentation & control system repairs.

I have a lot of trouble getting technical info from suppliers, so I have been starting to reverse engineer components & build gauge & control systems on the shop table for bench testing before installation on a client's equipment.

I am hoping someone can shed some light onto tachometer drive styles. At this moment, I am trying to resolve an issue with a redesigned wiring harness for an "old" Volvo Diesel engine in a boat.

The engine uses both a mag pickup sensor & an alternator AC tap (pre-rectified AC voltage). The mag pickup senses the fuel pump drive gear teeth, and sends this signal to a sealed component that somehow uses that signal to operate an on/off relay switch. The purpose of this is to govern the operation of a blower clutch that only should run in the mid-range of the engine RPM. Nothing to reverse engineer here, just a description of the sensor & why it is necessary.

For driving the tachometer gauge, the alternator's AC tap terminal is used. Nothing to reverse engineer here, just a description of the sensor & why it is necessary.

OK, now to the issue:

The instrument panel I've opted to use comes with a tachometer gauge that, according to manufacturer, MUST have a SQUARE WAVE input signal. AKA, the sine waveform from the alternator's AC tap terminal will not drive the device.

Well, I bought the panel, bought the mag pickup sensor they list as compatible, and bench tested it with an oscilloscope while running a variable speed motor with a gear on the motor with similar gear teeth sizing to the engine. Well, based on the o-scope, that did NOT produce a true "square wave", but rather, a very sine wave-looking waveform.

So it would seem to me that, although this may not be a TRUE sine waveform, it's closer to that than a TRUE square waveform... which would suggest that a tachometer gauge that can be driven by a square wave should also be capable of being driven by a sine waveform...... RIGHT??If my O-scope was also a waveform generator, or if I had a good alternator in the shop that had an AC tap terminal, well I would have tested this already & would be posting the answer! But I don't have that, so I'm hoping y'all can shed some light on this. Many thanks!

 

[jack action]

If you need a ±5 V signal, you can use a signal with a higher voltage (say, ±15 V), but limited to ±5 V, and get an approximate square wave. Is this something that could be happening here?

 

[KrakenMarine]

I guess what I am really asking is, without any crazy modifications, or additional electronic components, shouldn't this work? Plugging a "square wave drive" tachometer into an alternator's AC sine wave output?

I suspect this is the case, due to my oscilloscope readings that show the mag pickup producing a near-sine wave instead of a square wave as I was told it produces? I tested the parts the manufacturer said are compatible & produce square waves, and they do not produce a true square wave...

Am I right & the supplier is wrong? Will this work?

 

[anorlunda]

TL;DR Summary: Can an alternator's sine wave drive a tachometer that "requires" square wave?

The instrument panel I've opted to use comes with a tachometer gauge that, according to manufacturer, MUST have a SQUARE WAVE input signal. AKA, the sine waveform from the alternator's AC tap terminal will not drive the device.

That sounds unusual. Can you point us to the manufacturer's statement?

 

[KrakenMarine]

Bro.... If I could get technical information from the reseller, I wouldn't be posting to this forum.

If I could read or speak Korean, I would have gotten this info from the manufacturer.

Unfortunately I have to do all this reverse engineering to find out what the product can/will do. I was told this tachometer requires a square wave, but my O-scope shows the driving sensor producing a very non-square waveform that really looks like a sine wave.... so really I dunno what to believe except what I can test/verify myself.

 

[berkeman]

Bro....

Whazzup Bro! Welcome to PF.

If I could read or speak Korean, I would have gotten this info from the manufacturer.

Google translate is your friend!

 

[Baluncore]

If I could read or speak Korean, I would have gotten this info from the manufacturer.

If you can link to a document, Google translate can handle the Korean.

 

[KrakenMarine]

No documents to provide :-(

I am really hoping a forum member like Averagesupernova could weigh in....To further my point, if they tell me it requires square wave, but I measure the output of OEM compatible sensor & it does not produce a true or even remotely clean square wave... then I have to doubt everything I've been told until I verify for myself....

 

[Baluncore]

The instrument panel I've opted to use comes with a tachometer gauge that, according to manufacturer, MUST have a SQUARE WAVE input signal. AKA, the sine waveform from the alternator's AC tap terminal will not drive the device.

Can you attach or link to the document that makes that statement "requiring a square wave, and definitely not the AC tap terminal".

 

[KrakenMarine]

Can't link an email from the supplier, but I can quote from it:

"overall your “dual station” logic is sound with this implementation. However, for the SMX analog panel to function correctly for the TACH signal input, it must be driven by a magnetic pickup that counts the teeth of the engine flywheel as it rotates, it cannot accept any other input such as a “W” alternator pulse source, etc."

From the manufacturer's website, found after some Googling of component part #s, here is the instrument panel:
https://seafirst.en.ec21.com/Marine_Instrument_Marine_Instrument_Panel--3566038_3569790.html

And here is the tachometer gauge that is part of that panel:
https://seafirst.en.ec21.com/Marine_Instrument_Tachometer_Sensor--3566038_3569890.html

The only info that says this MUST have a mag pickup & NOT an AC tap signal from an alternator is 1) the email from supplier, and 2) the corresponding mag pickup on the manufacturer's web page.I can tell you with certainty, there are other tachometer gauges on the market that can accept all drive types: mag pickup drive, alternator, DC pulse generator. So it is a fact that it IS POSSIBLE to manufacture a tachometer gauge that CAN be driven by any style of input...... maybe this one can too?
Thanks for everyone who is contributing to this thread.

 

[Baluncore]

Thanks for everyone who is contributing to this thread.

It takes a lot to stop us.

The use of an alternator AC signal to feed a tachometer may be deprecated for several reasons.

1. The alternator drive belt will always slip to some extent, so calibration will be unreliable.

2. The tachometer produces a drive current to the magnetic pickup, it is not expecting a huge low impedance voltage signal to return.

3. The input frequency will be about one tenth of that from a gear tooth sensor, so the RPM reading will be unstable.

4. The tachometer front-end will selectively pass the alternator third harmonic and reject the fundamental, since it is designed for a higher frequency tooth rate input.

5. The amplitude of the alternator signal may vary in harmonic and noise content, so depending on electrical load, may sometimes read a multiple of the actual RPM.

 

[Baluncore]

The tachometer is probably fitted with a dedicated VR receiver IC, such as the MAX9924-7 or the LM1815.
https://en.wikipedia.org/wiki/Variable_reluctance_sensor#Interface_circuits

 

[KrakenMarine]

It takes a lot to stop us.

The use of an alternator AC signal to feed a tachometer may be deprecated for several reasons.

1. The alternator drive belt will always slip to some extent, so calibration will be unreliable.

2. The tachometer produces a drive current to the magnetic pickup, it is not expecting a huge low impedance voltage signal to return.

3. The input frequency will be about one tenth of that from a gear tooth sensor, so the RPM reading will be unstable.

4. The tachometer front-end will selectively pass the alternator third harmonic and reject the fundamental, since it is designed for a higher frequency tooth rate input.

5. The amplitude of the alternator signal may vary in harmonic and noise content, so depending on electrical load, may sometimes read a multiple of the actual RPM.

Maybe I should put this project into perspective:

This engine needs a LOT more repair than a new instrument panel… I fired it up to test the supercharger clutch engagements at various RPMs, verifying by photo tach, and BOY did she belch a lot of white smoke. If I’d been the primary point of contact on this repair I’d have STRONGLY suggested we do some fuel system service…

A solution engineered to perfection wouldn’t fit well on this boat. Haha! Further, I see inaccurate tachs very frequently. And the boats still sell for the cost of a nice house. Save perfection for the next gen of Mars rockets. This just needs to be accurate +/- 50 RPMs.
To respond to Baluncore:

1 - this imprecision is totally acceptable (the original OEM arrangement is an alternator-driven tach)

2 - not being a trained electrical engineer, I think I smell relevant info here but would need you to elaborate

3 - engine tachometer gauges are equipped with dip switches or similar ways to adjust the # of pulses that = 1 revolution, so input frequency can vary greatly without issue (sidebar: alternator-driven tachs are incredibly widespread & a very proven arrangement)

And not sure if 4 & 5 are rendered moot by the widespread evidence of alternator-driven tachs of being utilized as effective tachometer drivers for engine applications. To sum it up, this just needs to make a gauge needle move with +/- 50 RPMs accuracy. This is in no way going to be used for machinery control, where precision could matter greatly, just for operator display data.

And the operator will be drinking a margarita & ignoring this gauge 90% of the time

 

[Tom.G]

It would be helpful if you can attach waveform photos (or even sketches) for both the mag pick-up sensor and the alternator "W" signal.
Please include both amplitude data and time or frequency data.

How many wires or connector pins on the sensor?

Also, it would be informative if you can include a photo of the sensor waveform at the minimum sensor output voltage needed to operate the tach display. You can vary the voltage by changing the distance between the sensor and the gear teeth.

Straight-forward alternative: replace the control instrument panel tach with one that works with the "W" signal.

Cheers,
Tom

[edit] p.s. White smoke = coolant in combustion chamber; hopefully just a head gasket!
[/edit]

 

[Baluncore]

My five points were not directed at your particular installation, but at the general case. It matters not whether you care, but the manufacturer's tachometer was not designed to meet all their specifications when driven by an alternator. If it works OK, test it and use it.

My point 2. A modern automotive VR receiver would transmit a small DC current through the transducer. If that DC current did not return from the VR coil, the On-Board-Diagnostics would report an open circuit in that VR sensor circuit. A short to ground would also be reported by the diagnostics. Maybe that feature is present in the receiver IC used in the new tachometer gauge. There are unknowns we do not even know about, but the manufacturer does. When they say "use a VR magnetic sensor", they do not want you calling them to say it does not work properly with your alternator signal.

p.s. White smoke = coolant in combustion chamber;

Unburned diesel fuel in a cold exhaust, or engine oil in a hot exhaust.

 

[Tom.G]

or engine oil in a hot exhaust.

Hmm... I've generally see engine oil come out blue. (Or maybe black in a severe case, but I don't recall for sure.)

I have seen white 'smoke' (steam or fog) briefly at start-up in very cold weather though.

Oops, just realized that's a Diesel engine, perhaps different diagnostics than a gasoline engine.

 

[Averagesupernova]

@KrakenMarine in post #10 you quote an email that says the tach must be driven by a sensor that counts teeth on a flywheel. I have to assume this is a three wire sensor that outputs a square wave. A two wire sensor is going to output a sine wave. I would also have to assume the supplier of the tach supplies the sensor as well. For whatever reason, they want a square wave or at least a signal that switches hard to ground like most three wire sensors will.
-
There are tach adapters out there that may be able to do what you want.
-
https://www.dakotadigital.com/index...id=1192/category_id=694/mode=prod/prd1192.htm