Alternator vs Charger

[chris w]

The main reason was to save having a warning/charge light that needed disconecting when the engine stopped.

Errrrrrrrr......... what's wrong with using the ignition switch??????????????

[ditchcrawler]

Errrrrrrrr......... what's wrong with using the ignition switch??????????????

 

I would need two, one for the domestics and one for the engine battery/start unless I could find one with an isolated second contact on it. Any old road, what I have works with less wires, less connections, less bulbs and less switches.

[chris w]

I would need two, one for the domestics and one for the engine battery/start unless I could find one with an isolated second contact on it.

You wouldn't need two - you simply wire the plus side of both lamps to the ignition switch.

 

Any old road, what I have works with less wires, less connections, less bulbs and less switches.

............................and less useful information. That's a bit of a Luddite attitude I fear you have.

[ditchcrawler]

You wouldn't need two - you simply wire the plus side of both lamps to the ignition switch.

 

 

............................and less useful information. That's a bit of a Luddite attitude I fear you have.

I will have to think about the first statement, the second is half true

[Gary Peacock]

The newer alternators used by Barrus exhibited some strange behaviour when trying to do this and a number of workarounds are used to disguise this.

Edited December 23, 2008 by Gary Peacock

[Keeping Up]

You wouldn't need two - you simply wire the plus side of both lamps to the ignition switch.

Many circuit diagrams for doing this, show the use of a blocking diode in series with each lamp. I can't quite out why (apart from the fact that if the two batteries are at significantly different voltages, one lamp will glow dimly)

Edited December 23, 2008 by Keeping Up

[Gibbo]

If an alternator is energised via a charge warning light from a battery it isn't charging it can sometimes drive the internal regulator nuts with the output voltage jumping all over the place. If the battery providing the energising current is at a lower voltage than the one being charged then all is usually fine. But if the battery providing the energising current is a higher voltage it can fool the regulator into thinking the batteries are higher voltage than they are so it shuts down the field current. Then when it switches the field current back on it overshoots it. The output voltage can fluctuate by 3 or 4 volts with an apparently random time period.

 

The most fool-proof solution is to use the ignition switch to energise a small relay which then provides energising current from the correct battery. Sometimes a series diode in line with the D+ can help (as mentioned by Allan) but this only works if the battery voltages differ by less than about 1 volts.

 

This is in addition to the problem of one light glowing dimly as also mentioned by Allan. This too is cured using the relay method.

 

Gibbo

[ditchcrawler]

If an alternator is energised via a charge warning light from a battery it isn't charging it can sometimes drive the internal regulator nuts with the output voltage jumping all over the place. If the battery providing the energising current is at a lower voltage than the one being charged then all is usually fine. But if the battery providing the energising current is a higher voltage it can fool the regulator into thinking the batteries are higher voltage than they are so it shuts down the field current. Then when it switches the field current back on it overshoots it. The output voltage can fluctuate by 3 or 4 volts with an apparently random time period.

 

The most fool-proof solution is to use the ignition switch to energise a small relay which then provides energising current from the correct battery. Sometimes a series diode in line with the D+ can help (as mentioned by Allan) but this only works if the battery voltages differ by less than about 1 volts.

 

This is in addition to the problem of one light glowing dimly as also mentioned by Allan. This too is cured using the relay method.

 

Gibbo

I think I will remain a Luddite and just read the volts and amps to satisfy myself that the 24 volt batteries are being charged.

[chris w]

If an alternator is energised via a charge warning light from a battery it isn't charging.......................

 

Gibbo

I'm not suggesting one does that. I'm just suggesting connecting both lamp positives to the same ignition feed so they are both disconnected by the ignition switch when the engine is switched OFF. Once the alternator is energised, the D+ terminal is doing nothing anyway (unless an alternator/belt fails and then it does the job it's supposed to do, which is to indicate the fault via the lamp).

 

There is a small possibility that one lamp might glow until the domestic battery charges a little as it could have a difference of a volt or so across it; (that's why the diodes kill this as they drop 0.6v). I have the same setup (without diodes) and neither of my lamps glow.

 

Not having a warning lamp (and I also have audible alarms) could cause major problems if the belt driving the water pump fails and you have no audible/visual warning.

 

Chris

Edited December 23, 2008 by chris w

[Gibbo]

I'm not suggesting one does that.

 

But you are suggesting that.

 

I'm just suggesting connecting both lamp positives to the same ignition feed so they are both disconnected by the ignition switch when the engine is switched OFF.

 

And if they are both connected "to the same ignition feed" then they are, by definition, connected to the same battery.

 

Once the alternator is energised, the D+ terminal is doing nothing anyway (unless an alternator/belt fails and then it does the job it's supposed to do, which is to indicate the fault via the lamp).

 

But that isn't what happens. The reg switches the field off, the magnetic field starts to collapse and thus the output current falls. When the voltage (as measured on the D+ terminal) drops below the regulator voltage, the reg switches the field back on. What happens if it's connected to a higher voltage battery is that the reg switches the field off, the reg is then drawing no current, the voltage on the D+ terminal rises which can actually shut the reg down completely! In most instances it simply messes up the voltage regulation and results in severe pulsing.

 

It's been a common problem on certain engines (as Gary pointed out) and the only sure-fire cure is to use a relay to connect the charge warning light to the correct battery. A series diode can sometimes help as that reduces the voltage seen on the D+ terminal by one diode drop. But it isn't guaranteed to work under all circumstances.

 

It depends upon where the reg measures the output voltage. Most of them measure it at the D+ terminal not the B+ terminal as would be expected.

 

I have the same setup (without diodes) and neither of my lamps glow.

 

Aye, but that's the difference between having seen one or two installations in a certain manner that work and drawing the conclusion that the system is fine OR having seen hundreds where some of them don't work and then figuring out why. Taking a single sample often results in the wrong conclusion

 

Not having a warning lamp (and I also have audible alarms) could cause major problems if the belt driving the water pump fails and you have no audible/visual warning.

 

Couldn't agree more.

 

Gibbo

Edited December 24, 2008 by Gibbo

[Keeping Up]

Edit to delete strange multiple-posting effect

Edited December 24, 2008 by Keeping Up

[Keeping Up]

And even the single diode can cause problems, on just a single-alternator setup.

 

My ignition light glowed dimly when charging, because of the voltage across it that was equal to the forward voltage drop of the split-charge diode. I cured that by putting a diode in series with it - in fact the wiring diagram for my (Perkins) engine shows a diode in such a place. Then because someone on this forum pointed out that certain types of fault could be masked by the diode, I placed another diode across the first, but facing the other way. This setup worked perfectly for a couple of years.

 

Then I added another battery to the domestic bank (so I now had 4 instead of 3), and suddenly now had a problem of the voltage and charging current occasionally pulsing wildly, being led to do so by the Adverc controller. Changing the Adverc made no difference (thanks to Adverc for the loan of a spare), and after much fruitless diagnostic activity I removed the diodes. Hey Presto, no pulsing.

 

Of course I was back to having a dimly glowing light. But changing the bulb for one of a higher wattage cured that. I can nly assume that the additional battery subtly changed a time-constant for voltage decay somehwere, thus de-stabilising the system, but that's pure conjecture, and now it's workinbg OK again I'm not doing any more work on it.

[Gibbo]

And even the single diode can cause problems, on just a single-alternator setup.

 

My ignition light glowed dimly when charging, because of the voltage across it that was equal to the forward voltage drop of the split-charge diode. I cured that by putting a diode in series with it - in fact the wiring diagram for my (Perkins) engine shows a diode in such a place. Then because someone on this forum pointed out that certain types of fault could be masked by the diode, I placed another diode across the first, but facing the other way. This setup worked perfectly for a couple of years.

 

Then I added another battery to the domestic bank (so I now had 4 instead of 3), and suddenly now had a problem of the voltage and charging current occasionally pulsing wildly, being led to do so by the Adverc controller. Changing the Adverc made no difference (thanks to Adverc for the loan of a spare), and after much fruitless diagnostic activity I removed the diodes. Hey Presto, no pulsing.

 

Of course I was back to having a dimly glowing light. But changing the bulb for one of a higher wattage cured that. I can nly assume that the additional battery subtly changed a time-constant for voltage decay somehwere, thus de-stabilising the system, but that's pure conjecture, and now it's workinbg OK again I'm not doing any more work on it.

 

I remember a marina with a similar problem on an Isuzu with twin alternators. In the end they drove the charge warning lights from relays (normally closed contacts) energised from the D+ terminal and energised the alternators from resistors. That cured the problems. It's never as simple as it at first appears.

 

Gibbo

[chris w]

And if they are both connected "to the same ignition feed" then they are, by definition, connected to the same battery.

 

 

Gibbo

Gibbo, me old mucker............ that's nonsense.

 

Once the alternators are running, the voltage at the D+ terminal is dictated by the diode trio's feedback from the output voltage of the battery. So if the domestic batteries are discharged and their terminal voltage is only say 12.3v, then 12.3v is what will appear at the D+ terminal on that alternator (I have ignored the voltagecdrops across the field diodes and the diode trio for simplicity, but you know what I mean).

 

So what the OTHER end of the charge light is connected to (say 14.2v from the start battery alternator) is irrelevant save that the voltage difference (1.9v in this example) MAY cause the bulb to glow (it doesn't actually cause it to glow on mine as it happens). It won't affect that alternator's ability to charge its batteries.

 

Chris

Edited December 24, 2008 by chris w

[ditchcrawler]

Gibbo, me old mucker............ that's nonsense.

 

Once the alternators are running, the voltage at the D+ terminal is dictated by the diode trio's feedback from the output voltage of the battery. So if the domestic batteries are discharged and their terminal voltage is only say 12.3v, then 12.3v is what will appear at the D+ terminal on that alternator (I have ignored the voltagecdrops across the field diodes and the diode trio for simplicity, but you know what I mean).

 

So what the OTHER end of the charge light is connected to (say 14.2v from the start battery alternator) is irrelevant save that the voltage difference (1.9v in this example) MAY cause the bulb to glow (it doesn't actually cause it to glow on mine as it happens). It won't affect that alternator's ability to charge its batteries.

 

Chris

But one of mine would be at 14.something and the other somewhere near 29 volts. some 14 volts difference not 1.9V

[Gibbo]

Gibbo, me old mucker............ that's nonsense.

 

Once the alternators are running, the voltage at the D+ terminal is dictated by the diode trio's feedback from the output voltage of the battery. So if the domestic batteries are discharged and their terminal voltage is only say 12.3v, then 12.3v is what will appear at the D+ terminal on that alternator (I have ignored the voltagecdrops across the field diodes and the diode trio for simplicity, but you know what I mean).

 

So what the OTHER end of the charge light is connected to (say 14.2v from the start battery alternator) is irrelevant save that the voltage difference (1.9v in this example) MAY cause the bulb to glow (it doesn't actually cause it to glow on mine as it happens). It won't affect that alternator's ability to charge its batteries.

 

Chris

 

No.

 

If you put a voltage on the "other" side of the charge warning light that is higher than the output of the alternator the voltage at the D+ terminal will rise above the voltage that the alternator is putting on it at the time the reg switches the field current off. AT that time the voltage will rise and the reg will shuit down completely.

 

Try it (disconnect your PDAR). Wait for acceptance then put 15 volts (or in fact any voltage higher than the internal reg is set to) on the D+ terminal. The reg will shut down.

 

It's been a common problem on Isuzu engines that tried to run 2 charge warning lights form the same battery (see Gary's post). Beta use a relay energised from the ignition switch running a feed from the correct battery and they don't have that problem.

 

Gibbo

 

But one of mine would be at 14.something and the other somewhere near 29 volts. some 14 volts difference not 1.9V

 

Which would, indeed, shut the reg down. You'd have to use a relay to connect a charge warning light from the correct battery. Or just not bother with one as you seem to be happy with it!

 

Gibbo

[chris w]

It's been a common problem on Isuzu engines that tried to run 2 charge warning lights form the same battery (see Gary's post). Beta use a relay energised from the ignition switch running a feed from the correct battery and they don't have that problem.

 

Gibbo

Well I have an Isuzu 35 with exactly that arrangement and I have never had an issue. One alternator is running the start battery, the other the domestics and both charge lamps are connected to the start battery.

 

The regulator will switch off at around 14.2v which will be the same voltage as the other alternator more or less. Even if one alternator were as high as 14.8v and the other at 14.2v then the worst that is going to happen is that the charge light will have 0.6v across it. The VOLTAGE at the domestic regulator won't change just because there is a higher voltage at the OTHER side of the charge lamp.

 

Chris

Edited December 24, 2008 by chris w

[Gibbo]

The VOLTAGE at the domestic regulator won't change just because there is a higher voltage at the OTHER side of the charge lamp.

 

Yes it will and yes it does. That is what causes the problem that many others (including marinisers of Isuzu engines themselves) have experienced.

 

Gibbo

[chris w]

Yes it will and yes it does. That is what causes the problem that many others (including marinisers of Isuzu engines themselves) have experienced.

 

Gibbo

How?????????? The voltage at the regulator side of the charge lamp will be the output of the diode trio.

 

Chris

[Gibbo]

How?????????? The voltage at the regulator side of the charge lamp will be the output of the diode trio.

 

Yes, the output is from the diode trio. The diode trio can increase the voltage at D+ but it cannot reduce it precisely because it is fed from diodes.

 

The only thing that can reduce the voltage at D+ is loading between D+ and ground. When the reg switches the field off the only loading will be the regulator quiescent current which is usually in the region of a few hundred microamps. So the instant the reg switches the field off the voltage rises to match that on the other side of the charge warning lamp. Once that happens the voltage remains high because the reg never tries to switch the field on again because the voltage at D+ is higher than the reg wants it to be.

 

It caused all sorts of problems for engine suppliers with twin alternators.

 

Gibbo

[GUMPY]

deleted

Edited December 27, 2008 by idleness