external diode sizing to replace a Valeo Motorola EK 300

catmandoo

catmandoo

Well-Known Member
Joined
21 Aug 2003
Messages
1,803
Location
The Earth but normally in the place of the high st
Visit site
As you all probably know when you have multiple battery banks fed by alternator / alternators a battery isolator is interposed between the alternator and the batteries such that the charge is evenly distributed through out the banks with out one battery depleting another. This is done by arrangements of diodes in a box with fins for air cooling . Sure Power Inc is a major player in this field . Unfortunately the isolator imposes a voltage drop of about 0.7 volts which means that if the alternator regulator sense wire is local to the alternator outlet terminal then the regulator cuts out prematurely and does not charge the batteries fully etc .

To avoid this happening it is the practice to insert an external diode in the sensor wire such that the regulator sees 0.7 v less than the alternator out put voltage .
In my case I have a Valeo Motorola 70 marine alternator feeding a Sure Power isolator feeding 3 battery banks but unfortunately have lost the Motorola EK 300 diode which does this deception .
( a fast Sterling type 4 stage charging regulator is not considered in this query because I already have one and if it fails then i fallback on the alternators basic controls )

I am therefore looking for a replacement diode and wonder if any one knows where to get it or alternatively can advise what specification I should have when searching Maplin's catalogue to buy one
 
Silicon diode

if the diode is only used for sensing the battery voltage then current is very tiny. Any silicon diode will do the job. However a more powerful diode will have naturally heavier wires or stud connection so be more robust in the marine environment. something along the lines of 10 amp 200 volt is what you want.
As said though the tiniest 1n74 or similar will do the job especially if you can mount it in some robust manner. I am guessing that all diodes you consider will be standard silicon types with a forward volt drop of around .7volt. If you find a diode in any old electronics that you are willing to wreck check the volt drop on a Multimeter with diode test or feed a tiny globe from a battery via the diode and measure the volt drop across the diode. An old computer power supply or old plug pack type transformer are 2 sources of diodes.
good luck and seasons greetings olewill
 
I suspect a bit of confusion here.

If you have a voltage sensing wire from the battery, beyond the splitter, back to the regulator then you don't need a diode in it. This is the only satisfactory way of coping with the drop in a diode splitter.

Surely when you need compensation is if you are sensing at the alternator. However if you're doing that I think the diode is going to be carrying the field current in which case you need several amps current rating - a 1N5400 (3A) as an absolute minimum, better a P600 (6A) or a 10A rated one (which I can't find in Maplin's list).

I can't trace the diode type number you give to see its spec, but what does it look like? You can have a guess at its rating from its physical construction.

See http://www.smartgauge.co.uk/alt_mod.html for the circuit I am assuming and note the WARNING comment at the bottom of the page that this is not a satisfactory way of compensating for split charge diodes. I think this is correct.

The only other possibility i can see is if it is a zener diode inserted in the voltage reference inside the regulator at the alternator but that seems unlikely (and would also be subject to the objections raised in the SmartGauge page above).

The problem with either of these methods is that you are likely to end up with excessive float voltage on the batteries when they are fully charged, the current is low and the splitter is no longer causing a significant voltage drop but your compensator is still raising the alternator output voltage.

Apologies if I have misunderstood what you have got and am alarming you unnecessarily!


PS It's not one of these is it? http://www.cooperindustries.com/content/public/en/bussmann/transportation/recall_info_-_eu.html
 
Last edited:
As Troubadour says .

If the alternator has a sensing connection it can simply be connected to one or the battery banks, usually the house battery.

Its important that it is not disconnected from the charging circuit by a battery isolator switch so a sensible place to connect to is the output terminal of the diode splitter rather than to the battery itself.

Alistair Garrods book contains details of adding a diode to to a machine sensed alternator to fool it into generating at the extra 0.7 volts. I thought I had a copy of the relevant pages but cannot find it. However, the objections to using this method are explained in Troubadour's link.

A "fooling" diode should not be carrying the field current, that comes from the alternator via the field diodes but I do not know what current should be allowed for. No doubt Garrod suggests suitable diode.

If I find the reference to Garrod's method I will post it.
 
VicS said:
A "fooling" diode should not be carrying the field current, that comes from the alternator via the field diodes but I do not know what current should be allowed for. No doubt Garrod suggests suitable diode.
Click to expand...

See http://www.ybw.com/forums/showthread.php?t=115435&highlight=garrod post #8 :)

On the SmartGauge drawing that would mean putting it in the purple line just below the B- connection.

Are you sure this would work? I wonder whether the regulator will draw enough current on its voltage sensing connection to bias the diode into the normal operating range? You might have to fit a load resistor as well for the diode.

I think this would give a much more constant increase in voltage than the SmartGauge method and be even more dangerous! If the diode is in the field current connection as per SmartGauge, the voltage boost will be less when the field current drops.

Incidentally Sure Power's own instructions say that you must have an externally sensed alternator regulator, not a fudge like either of these methods.
 
Are you sure this would work?
Click to expand...
Well Garrod is I believe well respected.
Nigel Luther had used the method and quite often suggested it even if it took a bit of prodding, as you'll see from my posts on that thread, to get a proper explanation out of him.
 
Whilst adding a diode to "fool" the regulator used to be popular many years ago, surely it's much better just to get the alternator converted to battery sensing? This overcomes many of the disadvantages of the compensating diode method.
 
pvb said:
Whilst adding a diode to "fool" the regulator used to be popular many years ago, surely it's much better just to get the alternator converted to battery sensing? This overcomes many of the disadvantages of the compensating diode method.
Click to expand...

But that would cost money .... Not something Nigel Luther did if he could avoid it.
 
VicS said:
Well Garrod is I believe well respected.
Nigel Luther had used the method and quite often suggested it even if it took a bit of prodding, as you'll see from my posts on that thread, to get a proper explanation out of him.
Click to expand...

Sorry perhaps I wasn't clear. It will work if you add a bias resistor, and as I suggested in my first post I think a zener diode would actually be better (which would need a load resistor), I just doubt whether a modern regulator will draw enough current for it to work without a bias resistor, as shown in the book.

I also suspect that on many modern potted regulators you won't be able to access that connection to insert a diode at all, it will be straight onto the alternator terminal. I'm not arguing with the principle. You could say it's more elegant than the SmartGauge solution, and will give a more constant voltage.

Anyway neither is really right for the OP's purpose!
 
pvb said:
Whilst adding a diode to "fool" the regulator used to be popular many years ago, surely it's much better just to get the alternator converted to battery sensing? This overcomes many of the disadvantages of the compensating diode method.
Click to expand...

Exactly! It's the right way.
 
VicS said:
But that would cost money .... Not something Nigel Luther did if he could avoid it.
Click to expand...

Need it cost money? If you can get at the internal connection for the Garrod method, can't you feed the battery terminal voltage back to there instead of the internal feedback?

Halcyon are you looking...? You will know!
 
Troubadour said:
If you have a voltage sensing wire from the battery, beyond the splitter, back to the regulator then you don't need a diode in it. This is the only satisfactory way of coping with the drop in a diode splitter.
Click to expand...

Quite.

Indeed a diode to "fool" the regulator would not work properly, as the voltage drop depends on the current.
 
You must log in or register to reply here.

Share this page

Top