Camshaft RPM sender - how does it work?

[cheesegas]

Like most pre-blue-era Isuzus in the late 80s/early 90s I've seen, my Isuzu 3KC1 has a tacho fed from a pickup unit on the camshaft cover rather than from the alternator.

 

It's never worked since I've had the boat as the pickup unit had been trodden on at some point and the wires were broken off level with the top of the potting compound. Well...I like a challenge and I wanted to get it working. I carefully heated and dug out enough of the potting to solder on two new wires, and then covered it in epoxy. 

 

This unit screws into the top of the camshaft cover (OHC engine) and its tip sits less than 1mm above the number 1 cylinder exhaust valve lobe when it's facing up. There's no reluctor ring like I expected. That cam lobe doesn't seem to be magnetic, at least not enough to attract a thin steel thing.. 

 

So, I got it all hooked up and nothing happens at all...bear in mind that no components of this are proven working!

 

The pickup has a DC resistance of 800-1m ohms, and when the engine is running it doesn't show anything on a scope or frequency meter. The tacho dial looks like a conventional one, -+12v in and a 3rd wire connected to one wire of the sender. The other wire goes to negative. Between the sender wire of the gauge and negative, there's 1v. 

 

I've attached a photo of mine. It looks very very similar to pickups fitted to truck gearboxes to send RPMs to a tachograph (not an RPM one!) to record mileage: https://www.google.com/search?q=tacho+sender

 

Anyone know how to test it and what sort of signal I should be expecting from it? And should whatever it's near be magnetic? I don't think it's a Hall effect sensor as two-wire variants of these are a pretty new thing. Definitely inductance or magnetic based.

 

[Tony Brooks]

Typically, two wire inductive sensor has a coil wound around a small bar magnet inside the case (that is your 800 ohms) Each time a ferrous object passes close to the tip of the sensor the magnetic field around the coil becomes more diffuse or tighter packed so the coil produces a voltage. You typically end up with a sine wave output with the frequency proportional to the engine speed. Try testing it with the meter set to frequency (Hz). Try touching the tip of the sensor to some steel to see if it is magnetic.

 

Three wire sensors use a hall effect sensor, but use three wires.

[cheesegas]

3 minutes ago, Tony Brooks said:

Typically, two wire inductive sensor has a coil wound around a small bar magnet inside the case (that is your 800 ohms) Each time a ferrous object passes close to the tip of the sensor the magnetic field around the coil becomes more diffuse or tighter packed so the coil produces a voltage. You typically end up with a sine wave output with the frequency proportional to the engine speed. Try testing it with the meter set to frequency (Hz). Try touching the tip of the sensor to some steel to see if it is magnetic.

 

Three wire sensors use a hall effect sensor, but use three wires.

Thanks Tony, it is indeed magnetic! That explains how it works. Unfortunately I'm not getting anything out of it, not on a meter set to frequency nor on an oscilloscope.

 

However...I found this site here which sells similar sensors, and the resistance is 1.2k which roughly matches up with mine. Unfortunately it looks like you need a special tacho to go with it - most don't mention inductive pickups so I don't think their input impedance is high enough. Most generic ones also seem to have a speed ratio of 1 minimum, as this is camshaft driven it needs to be 0.5.

 

https://wema.co.uk/products/tachometer-magnetic-pickup?variant=29803765447

 

Wema also sell a rather expensive tacho dial which can be set to a speed ratio of 0.5 and works with inductive sensors though!

[Tony Brooks]

My view would be to use a standard alternator sensed tacho as long as the alternator has a W terminal. Even if not, a half competent auto electrician could solder a fly-lead to a diode to do the job.

[cheesegas]

Just now, Tony Brooks said:

My view would be to use a standard alternator sensed tacho as long as the alternator has a W terminal. Even if not, a half competent auto electrician could solder a fly-lead to a diode to do the job.

...I'm about to update my Zeus thread as to why this isn't possible, hence the long route to get a working tacho! Have a look in half an hour or so.

[Quattrodave]

A peugeot engine I worked on recently had to have the tach sensor screwed in a specific amount to geta stable reading. With engine running I wound it in by hand till second person shouted the rev counter was working IIRC I went a further 1.5 turns before it was stable, then tightened the lock nut...

I believe to test with a multi meter you'd need it set to AC.  Personally I'd use the tach sensor over alternator, once working would be accurate and reliable...

[cheesegas]

Just now, Quattrodave said:

A peugeot engine I worked on recently had to have the tach sensor screwed in a specific amount to geta stable reading. With engine running I wound it in by hand till second person shouted the rev counter was working IIRC I went a further 1.5 turns before it was stable, then tightened the lock nut...

I believe to test with a multi meter you'd need it set to AC.  Personally I'd use the tach sensor over alternator, once working would be accurate and reliable...

Thanks, I'll try unscrewing it a little. Currently it's around 1mm away from the cam lobe.

 

As mentioned, I've measured it with an oscilloscope - think of it like a fancy AC multimeter where you can see changing voltages mapped out on a live graph!

[Bod]

"So, I got it all hooked up and nothing happens at all...bear in mind that no components of this are proven working!"

The wires to the tacho are intact? and connected to the tacho?

Very basic questions I know, but that level has got me going at times.

 

Bod

[cheesegas]

Just now, Bod said:

"So, I got it all hooked up and nothing happens at all...bear in mind that no components of this are proven working!"

The wires to the tacho are intact? and connected to the tacho?

Very basic questions I know, but that level has got me going at times.

Yep, tacho dial's getting +-12v and the wire from dial to sender and from sender to negative are all intact and tested with <1ohm resistance. Looks like it's possibly a bad dial or misadjusted sender unit.

[Rincewind]

16 hours ago, cheesegas said:

Anyone know how to test it and what sort of signal I should be expecting from it? And should whatever it's near be magnetic? I don't think it's a Hall effect sensor as two-wire variants of these are a pretty new thing. Definitely inductance or magnetic based.

 

Depending on what you are attempting to do here I may have a solution.

Question... are you wishing to repair / restore the revcounter back to its former glory? i.e. as per original spec?

Or... do you simply want a revcounter that works electronically and is constantly accurate, i.e. not connected to the alternator in any shape or form as

Quattrodave has suggested?

If its the former then without physically examining the equipment you have, I may not be of much help, but if its the latter and you want a project to pursue, then I can point you in the right direction as some years ago I made, using a variable reluctance sensor (VRS), my own digital revcounter that is accurate to approx. 1 or 2 RPM at high engine speeds. (This accuracy was sufficient for my purposes, although you could make it even more accurate if you have the will to do so).

 

As you already know, the variable reluctance sensor that you seem to have will output a sine wave signal (if it is working).  Assuming that the signal is to be read by a microprocessor, that will most likely be housed inside the tacho instrument, the sine wave will have to be converted to a square wave prior to being fed into the microprocessor.  When counting pulses, microprocessor's do not like sine wave signals and work best with crisp low to high square wave voltage signals.  The point to note here is that, from the reluctor sensor to the tacho instrument, the cable you use should be of the shielded type with the braid grounded in order to prevent electromagnetic interference from other electrical equipment, e.g. alternators, corrupting the signal. - This can be important!  (Less so important with Hall effect sensors as they are less susceptible to electromagnetic interference ).

Hall effect speed sensors are also fitted with permanent magnets so "testing" the sensor by seeing if it is attracted to ferrous metals is not conclusive evidence of which type of sensor you are dealing with, however the number of electrical connections is a pretty good indicator.

 

If it helps, I just measured the resistance of a spare reluctance sensor I have at home and it measures 777 ohms but do not read much into this as the resistance can vary widely from manufacturer to manufacturer.  Also, be aware that the output voltage of a VRS can vary from as little as 8 volts to over a hundred volts depending on speed / make of sensor.   e.g. The reluctor on my engine (over the flywheel teeth) at 642 RPM and producing1038 Hz, gave an output voltage of 152 volts - (both figures recorded on an oscilloscope). And that was at tick-over speed!

Hall effect sensors normally output a voltage close to the supply voltage.  i.e. If the input voltage = 5 volts then output voltage = 4.5 volts or thereabouts. 

 

[Tracy D'arth]

Do you really need a super accurate tacho on a diesel engine in a boat?  Most of us are quite happy with a guestimate one running off the alternator.

 

In fact, do we need tachos at all?

Edited March 26 by Tracy D'arth

[Tony Brooks]

5 minutes ago, Tracy D&#x27;arth said:

Do you really need a super accurate tacho on a diesel engine in a boat?  Most of us are quite happy with a guestimate one running off the alternator.

 

In fact, do we need tachos at all?

 

I think you obviously do if you are into trying to get into the latest tech a "cheap" way. Not that £800 is cheap, but it seems to have significant drawbacks when trying to marry lithium batteries to automotive alternators. I am sure an alternator sensed revcounter could be driven by a start type reluctor bolted to the front pulley nut and a sensor on a bracket. Just pleased I am old and happy with basic technology.

[cheesegas]

44 minutes ago, Tracy D'arth said:

Do you really need a super accurate tacho on a diesel engine in a boat?  Most of us are quite happy with a guestimate one running off the alternator.

 

In fact, do we need tachos at all?

No need for a tacho at all - it's never worked for the 4 years I've had the boat and I've never felt the need for one. This is mainly a vanity exercise as I have a week off work when I should be derusting and painting bits on the boat but it's weeing it down every day... With most external regs which can switch to float mode, alternator driven tachos are effectively disabled when it's in float, especially if there's solar. It's just fortunate my boat happened to come with a non alternator sensed tacho.

 

 

33 minutes ago, Tony Brooks said:

I think you obviously do if you are into trying to get into the latest tech a "cheap" way. Not that £800 is cheap, but it seems to have significant drawbacks when trying to marry lithium batteries to automotive alternators. I am sure an alternator sensed revcounter could be driven by a start type reluctor bolted to the front pulley nut and a sensor on a bracket. Just pleased I am old and happy with basic technology.

This isn't a cheap way of doing it - that would be hybrid, which ironically doesn't suffer from this issue! As I'm sure you've read in my other thread, an alternator sensed tacho is unreliable with any regulator like this - Wakespeed, Balmar etc. The Zeus doesn't need a tacho input to operate but it does rely on tacho input for one of the optional features. 

 

It's worth noting too that a 'marine' alternator would suffer from exactly the same issue as this. Balmar regulators have a similar tacho keep-alive function which needs to be used with their alternators; it's less configurable than the competitors and works poorly.

 

Many high output installations do not suffer from this issue though as there's a separate alternator for the starter battery, which the tacho and regulator takes its signal from. If the field of the domestic alternator is dropped, it doesn't matter. However, I don't have the room in my installation to fit a second alternator, hence doing it all from one.

 

Basic technology and lead batteries are fine if you have shore power and/or don't live aboard, but lithium batteries in the winter for a liveaboard without shore power are literally life changing. I've tried marina life and I hated it, I'm happy to pay for the technology and do the research needed to fit lithium batteries etc - it's definitely at the forefront of things so there's still shortcomings people are unaware of! My job includes a lot of getting various things speaking various protocols to talk together and electrics too, so this keeps me on the edge.

Edited March 26 by cheesegas