Worth converting alternator to battery sensed?

[GHA]

Is it really worth the bother, or any other vaguely relevant comments?


Engine is a Beta 35 fitted with a Iskra (letrika) 120A alternator. When the batteries are fully charged it will regulate at about 14.6v at the battery terminals, round about 60DegF from memory. Yesterday motoring it was putting out about 25a , battery terminal voltage ~14.2v, battery meter reading 14.5v (between +ve side of shunt and somewhere in the domestic LX panel. So I'm guessing across the alternator +ve & -VE was probably 14.6v. So....

1. Google came up with forum post hinting that it's easy to get at the sense wire going into the regulator on the back of an Iskra alternator, so easy to snip that and reroute to the battery positive terminal - can anyone confirm that's the case? The wording was a bit vague.
   http://www.canalworld.net/forums/index.php?showtopic=23223&p=386025
   " On the Iskra, you can just change the regulator which is screwed to the back of the alternator frame and then add the sense wire to the battery bank, easy peasy, you don't even have to remove the alternator from the engine."
   I haven't had the alternator off yet for a look. The alternator has a thermometer attached so keeping an eye on temperature/setting an alarm is easy.
2. How much to be gained? On a smartguage site there was a rough guess formula about how much power might be getting lost , something like... (Battvoltage -12.5)^2 / (ALTvoltage -12.5)^2 * 100 = percentage gain. Which fag packet calc suggested there might be chance of a 70% gain at that alterntor output. 14.2v should be before the battery acceptance starts tailing off, from watching the numbers with the mains charger on.
3. What would also happen though would be an increase at the switch panel - the mains charger is set to 14.8v (trojan t105's) without any probs so hopefully that won't be a problem.
4. Need to dig a bit into the wiring to see if it's possible to get less drop between the alternator & batteries, though 0.4v at 25A might not be too surprising with a 1/2/off switch and some connections on the way.

Any thoughts? Living onboard away from mains power every extra teaspoon of energy going back into the batteries is cause for celebration

TIA

 

[pvb]

The 14.6v reading at the fully charged batteries suggests that the alternator regulator is set at 14.6v.

I'm not clear why you have measured 14.2v at the battery terminals at the same time as your battery meter is showing 14.5v - surely your battery meter should be showing you the battery voltage?

A voltage drop of 0.4v at only 25A charge isn't acceptable, and you need to investigate the connections. However, if you're deriving that 0.4v from your battery meter (which sounds suspect), maybe you don't actually have a voltage drop problem.

I don't think you'll achieve much by converting the alternator to voltage sensing; this is usually done to overcome the inherent voltage drop caused when a splitter diode is used to distribute charging. If you want to maximise your charging efficiency, you might consider fitting an alternator-to-battery charger, which will boost the voltage at the batteries to maximise charging efficiency.

 

[GHA]

The 14.6v reading at the fully charged batteries suggests that the alternator regulator is set at 14.6v.

I'm not clear why you have measured 14.2v at the battery terminals at the same time as your battery meter is showing 14.5v - surely your battery meter should be showing you the battery voltage?

Bad wording, it's a battery monitor which reads at the shunt.

A voltage drop of 0.4v at only 25A charge isn't acceptable, and you need to investigate the connections.

Sure about that? It's still next to no resistance considering the junctions & switch to go through. Do you have reliable data showing the drop significantly less on other installations?
As mentioned in the post it's on the list to check anyway.

However, if you're deriving that 0.4v from your battery meter (which sounds suspect), maybe you don't actually have a voltage drop problem.

I don't think you'll achieve much by converting the alternator to voltage sensing; this is usually done to overcome the inherent voltage drop caused when a splitter diode is used to distribute charging. If you want to maximise your charging efficiency, you might consider fitting an alternator-to-battery charger, which will boost the voltage at the batteries to maximise charging efficiency.

Seems little to be gained spending money on a AtoB when the alternator already is regulated at 14.6v and the drop could be compensated for by the possible running in of 1 wire??

 

[lw395]

If the alleged 0.4V loss is real, it becomes nearly 2V at the rated current of the alternator?
Don't forget the drop is in both cables, so battery sensing the + side still leaves the drop on the - side. IS the shunt in the - side?
For starters, look at the cable dimensions and work out if the drop is credible at 25A.
If the drop is in a bad connection and not distributed in the cables, it's goingto get toasty if you push more current through....
I would consider measuring the actual drop directly using a tolerably good multimeter.
The two voltage readings may have errors, or averaging differently if there is any AC component present.

 

[pvb]

Bad wording, it's a battery monitor which reads at the shunt.

Call it what you want, it should be reading the voltage at the battery terminals, which you say is different. How did you measure the 14.2v at the terminals?

Sure about that? It's still next to no resistance considering the junctions & switch to go through.

Positive - it's too much.

 

[GHA]

Call it what you want, it should be reading the voltage at the battery terminals, which you say is different. How did you measure the 14.2v at the terminals

The meter is wired according to the manufacturers so shout at them.

14.2v is from a panel volt meter direct to the terminals, accurate as per a voltage reference and fluke multi meter.

With another probe looks like the battery monitor isn't calibrated very well, more like 1 or 200mV high at the panel with 15A charging. Both ve & -ve runs need checking as per 1st post.

Non of which addresses the question, there will always be a drop so what are the pros/cons of battery sensing an alternator?

 

[pvb]

The meter is wired according to the manufacturers so shout at them.

I thought you were asking for advice, but your replies suggest otherwise.

Non of which addresses the question, there will always be a drop so what are the pros/cons of battery sensing an alternator?

In systems with high voltage drops caused by splitter diodes, voltage-sensing restores the normal charge voltage at the battery terminals. However, your system has high voltage drop caused by other reasons, which need resolving before you continue. If you really have 0.4v drop at only 25A charge, this is totally unacceptable.

If you want to maximise your charging, you need to increase the alternator voltage, which you could do with a "smart" alternator regulator or an alternator-to-battery charger. The latter has some benefits as it will create a true multi-stage charge regime, with a proper float voltage. It also has an unboosted output for the starter battery. Both options include voltage-sensing control.

 

[Minchsailor]

As I understand it you are using 2 different meters - the battery monitor and a multimeter. Unless they are high quality instruments, recently calibrated, (the BM especially is unlikely to be) almost certainly they will not agree. At 14V an error of each of +/-0.2V does not seem to be unreasonable.

Use the multimeter to do both measurements, then get worried if there is still an 0.4V difference.

 

[BabaYaga]

Need to dig a bit into the wiring to see if it's possible to get less drop between the alternator & batteries, though 0.4v at 25A might not be too surprising with a 1/2/off switch and some connections on the way.

I think sensing voltage at the batteries (house or engine) and a 1/2/both/off switch can be a dangerous combination.

 

[GHA]

If the alleged 0.4V loss is real, it becomes nearly 2V at the rated current of the alternator?
Don't forget the drop is in both cables, so battery sensing the + side still leaves the drop on the - side. IS the shunt in the - side?
For starters, look at the cable dimensions and work out if the drop is credible at 25A.
If the drop is in a bad connection and not distributed in the cables, it's goingto get toasty if you push more current through....
I would consider measuring the actual drop directly using a tolerably good multimeter.
The two voltage readings may have errors, or averaging differently if there is any AC component present.

OK,close as i can get without loosing limbs the drop looks like 100mV in the positive & bit less in the neg at 12.5A so likely the runs can be cleaned up to get it lower. Might up the cable dia at some point , think there is some in a locker somewhere.

Anyway, all that side was going to get looked at as per first post - "Need to dig a bit into the wiring to see if it's possible to get less drop between the alternator & batteries,"

(With apologies to PCB, the drop does seem high.)

Anyone know much about Iskra alternators? Can you actually get to the sense wire?

 

[GHA]

I think sensing voltage at the batteries (house or engine) and a 1/2/both/off switch can be a dangerous combination.

Ta,will look into that.

 

[BabaYaga]

Anyone know much about Iskra alternators? Can you actually get to the sense wire? [/COLOR]

Cannot claim to know much, but this photo shows the regulator/brush unit of an Iskra alternator (70A for a Beta 20hp, 10 yo). I think the sense wire is probably the yellow one. The green soldered on one of the brushes is the field wire from an Adverc regulator.

 

[GHA]

Cannot claim to know much, but this photo shows the regulator/brush unit of an Iskra alternator (70A for a Beta 20hp, 10 yo). I think the sense wire is probably the yellow one. The green soldered on one of the brushes is the field wire from an Adverc regulator.

Brilliant,thanks. So might be worth investigating further as well as going through the rest of the system.

 

[William_H]

The realistic way to check volt drop in any wire switch etc is to attach one DMM (digital multimeter) lead to the alternator out terminal use the other probe firstly to the battery pos terminal. On low voltage range this will show actual volt drop in wiring and switch of the positive lead. (when charging 25 amps or at least a lot) Do something similar with one lead on the alternator case and other to battery negative terminal. If you get a lot of drop on pos line try similar probes across the terminals of the switch. Hopefully zero as switch contacts are mde but any problem will show up there.

However volt drop in the charge line is not so significant as one might imagine. The drop in volts at 25 amps will reduce voltage at battery so reducing charge current which reduces volt drop. So it levels out a t a somewhat reduced charge current. However the high charge current will not last long due to battery inherent voltage rising so volt drop lessens.
So you are always going to get the battery up to alternator reg set voltage eventually. It will just take a little more time. Probably not much especially when batteries are reasonably charged anyway.
Understand that diodes drop a finite .7v regardless of current so will stop batteries from fully charging. So a different (worse) problem to resistance in the switch or cable.
So my guess is just accept system as is and not much gain in going to battery volt sensing. Assuming you don't have spitter diodes. olewill

 

[Poecheng]

Anyone know much about Iskra alternators? Can you actually get to the sense wire?

Have a look at this
http://bennersadrift.blogspot.co.uk/2011/04/modding-iskra-100-amp-alternator.html

 

[GHA]

However volt drop in the charge line is not so significant as one might imagine.

You're right there, I just tried switching the acceptance voltage between 14.8v (recommendation for the t105's) & 14.4v on the mains charger. An amp or 2 less at 14.4v but not as much as I had expected.
So probably not worth the bother, though worth looking into as it highlighted the cables could do with upgrading at some point.

Thanks all