Dancing Tachometer - Batteries Fully Charged?

[demonboy]

Good morning,

We have a Perkins Prima M60, 80amp alternator and a cheap-ass Turkish external alternator regulator.

An on-going 'problem' I have is that the tachometer needle bounces between zero and whatever revs I am doing, with the battery light flickering in time. I read once that on some systems this is an indication that the batteries are fully charged, and I have to say this symptom does seem to coincide with a fully charged battery bank. However my suspicion is that the crappy regulator, which is nigh on impossible to set up the float stage on it, is not managing the charge coming from the alternator and switching it on and off as it flickers between 99% and 100% charged.

Does this sound familiar to anyone? Can anyone assist?

Thanks in advance.

Jamie
s/y 'Esper'

 

[gregmlucas]

This may not help but...

I had a similar problem on a Volvo 2002 circa 1991. After several attempts to fix the problem by different people, it was pinned down to a dry joint in a fuse box - apparently a design fault. Now rectified by a replacement, newer design fuse unit.

Obviously, you don't have a Volvo but could it just be something as simple as a dodgy connection?

 

[jdc]

Conditional stability of regulator

... on some systems this is an indication that the batteries are fully charged, and I have to say this symptom does seem to coincide with a fully charged battery bank. However my suspicion is that the crappy regulator, which is nigh on impossible to set up the float stage on it, is not managing the charge coming from the alternator and switching it on and off as it flickers between 99% and 100% charged...

I think this is quite possibly the case, and indeed is due to poor regulator design. Forgive me if I'm teaching you to suck eggs, but here is what I think is going on:

The main 3-phase output of the alternator is rectified to DC and comes out of the B+ terminal, whereas the Tacho is connected to the 'W' output of the Alternator. This gives a low power sine-wave (ie AC) output. The Tachometer measures and displays the _frequency_ not amplitude of this signal, which has a fixed ratio to engine rpm, but this AC signal still has to be a few Volts for the tacho to work.

Both outputs are regulated in amplitude (ie Voltage) by the current through the field windings, so if this falls to zero the Voltage on the W output may fall very low, and the tacho stop working and so read zero.

The Voltage output by the alternator on both the B+ and W outputs should not reach zero even if the batteries are fully charged because the regulator should still regulate for some Voltage, even if no curent is flowing. But it sounds like your regulator is only marginally stable, and the feedback circuit - who'se function is to maintain a constant output Voltage - goes into oscillation aka 'hunting' (no need to go into Nyquist stabilty criteria here), supplying some current to the field winding, then when the output Voltage rises to be too high it shuts down the field current to zero, then since the Voltage it's trying to control is now too low it puts current through the field winding again, and the Voltage now rises so... repeat ad infinitum at maybe one cycle per second.

Does this make sense?

 

[demonboy]

By reading your post v-e-r-y slowly, yes, it makes perfect sense. Thank you for the breakdown, that's very useful. It also helps me in my decision to purchase a better reg.

One other thing: would this problem have anything to do with the fact I could not turn OFF my engine yesterday, or is that a separate problem with the relay switch? I had to depress the stop button a number of times before it would turn off, but bare in mind the engine hasn't been used for three months...

 

[FlyingDutchman]

I have the MD 22 A, which is the same engine and this spring I also had a problem with stopping the engine after the winter lay-up. I had to stop it a couple of times with the hand operared lever on the injection pump, but after taht it works just fine. Maybe the valve sticks a bit?

 

[demonboy]

Yes, I think it was just 'one of those things' after non-use. I'll continue to monitor it as I turn the engine more frequently over the next couple of months. Thanks for the reassurance that someone else has the same problem