Eliminating a Battery Isolator

[R_C]

I have made a number of improvements to the DC systems on my trawler. I have three Trojan T-1275s for a 450Ah house bank with a 90A Iota charger and a 40A Xantrex charger, which was the first charger and is now a backup. I have a separate start battery for the Ford Lehman and an isolated start battery for the genset. Charging with the genset and shore power is sound.

I am now tackling our charging while under way. Ultimately, I plan to add a new alternator with an external 3-stage regulator. First, I want to eliminate the battery isolator, which causes undercharging of the house bank because of the voltage drop across its diodes. I’d like to send the alternator output directly to the house bank and use an echo charger to top off the start battery. The diagram shows the current alternator wiring. My questions:

1. There are currently five wires on the alternator output terminal. Should I have only one wire going to the house bank connected to the output terminal?

2. In the existing wiring diagram, the start battery is charged from the 10awg wire from the alternator output to the starter solenoid. Isn’t the 6awg wire from the Bat3 post on the isolator to the battery redundant and delivering a lower voltage?

3. Both the Iota and Xantrex chargers are currently wired to charge both the house bank and the start battery. When using an echo charger should I remove the start battery charging wires from the Iota and Xantrex and use only the echo charger to charge the start battery from all charging sources?

4. The existing alternator has an internal voltage regulator and terminals for ground, tachometer, excitation, and output. How is the battery’s state of charge sensed? Is it sensed through the connection at the alternator’s output terminal?

Thanks.

[osirissail]

1. Ideally there should be only one wire from the alternator output either to a battery or an isolator. The other wires constitute "hot wiring" systems directly off the alternator - but, if you follow your diagram they are actually "hot wired" to the "start battery." In any case all systems should either have fuses or c/b's protecting the wires and/or devices, which ever has the lower ampere carrying capacity.

2. Yes, the output is routed once via the 10 awg wire to the "start battery" and redundantly via the 6 awg from the "isolator."

3.The "echo charger" will have directions to specifically tell you the best way to order the charging priorities.

4. Internally regulated alternators operate by sensing the voltage at their output terminal. This is by a wire normally routed to a battery since an "isolator" is a one-way electrical "valve."

An external regulator has an additional specific wire that is used to sense battery voltage, therefore the alternator output can be directly connected to only the "isolator." Replacing the "isolator" with an "echo-charger" gets into a area of different opinions about how best to get the most from the alternator. Some recommend connecting the alternator output and external regulator battery voltage sense wire to the "house" batteries since they are normally the largest and most drawn down batteries. Once they are near or full the "echo-charger" or battery combiner" allows power to flow to the "other" batteries connected to the downstream side of the "echo-charger" or "battery combiner."

- - A "battery combiner" is a simple relay that closes at a preset voltage to allow power to flow to the "other" batteries, and opens to prevent loss of charge in the "other" batteries when the primary battery voltage drops below the preset value.

- - "Echo chargers" are a bit more complicated and are described in the manual/instructions for each brand. Here is the Xantrex manual: http://www.google.com/url?sa=t&rct=j...aYTt0g&cad=rja

[DeepFrz]

If you want to eliminate the undercharging, run that red 16 awg wire (going to oil pressure switch) from the house bank instead of from the alternator output. You would probably need to increase its size. Remove the red 10 awg that goes to your starter solenoid. That wire is bypassing your isolator diodes. Actually you may be able to use that wire to replace the 10 awg to the oil pressure switch. But I would run it from at least the output of the diode isolator.

Edit: I see osirissail beat me with a more detailed explanation.

[DeepFrz]

I see also that once you remove that 10 awg wire from the start cct. you would have to move the wires going to instrument panel and fuel shutoff solenoid to the battery side of the cct., best place is house bank +12 volt bus. Don't forget to fuse them.

[DeepFrz]

Oops, just realized I screwed up on this one. Please ignore the previous posts except that you should only have the one wire going to the alternator output. The 10 awg wire is there to make sure the start battery is completely charged. If you could get access to the sense wire going to the internal regulator then my previous comments would make sense and the sense voltage could be run from the house bank +12 volt bus.

Or you could remove the isolator, run the alternator output directly to the +12 volt bus on the house bank and install an echo charger to the start battery. You could run that way until you install an externally regulated alternator and regulator.

[R_C]

osirissai,

Thanks for the explanations. I wasn't sure if the sensing occured at the alternator output or the excitation wire on the internally regulated alternator. Although they are the same now, I want to avoid a battery combiner for the possibility of different chemistries in the start and house batteries. I want to make sure that I have truly isolated the house bank from the start battery. With the alternator output only connected to the house bank that would appear to be the case.

The fuel shutoff solenoid relay, instrument panel, and excitation wire can be connected to the start battery. One possible side effect I see is the instrument panel's voltmeter wouldn't be showing the alternator's output voltage but that of the echo-charger. Is there any reason for concern there?

[R_C]

DeepFrz,

I did catch that problem. I don't want the house bank powering any circuits on the Lehman. Eliminating the isolator completely and charging the house bank with the alternator is my plan. The echo-charger looks like a good solution that will work now and when I replace the alternator and add an external regulator. Thanks.

[osirissail]

Originally Posted by R_C

osirissail,

[1] Thanks for the explanations. I wasn't sure if the sensing occurred at the alternator output or the excitation wire on the internally regulated alternator.

[2] Although they are the same now, I want to avoid a battery combiner for the possibility of different chemistries in the start and house batteries. . . .
[3] The fuel shutoff solenoid relay, instrument panel, and excitation wire can be connected to the start battery. One possible side effect I see is the instrument panel's voltmeter wouldn't be showing the alternator's output voltage but that of the echo-charger. Is there any reason for concern there?

[1] On an internally regulated alternator, you cannot get at the alternator "field" wire. This is the wire that senses the voltage and controls the output of the alternator. The internally regulated alternator only knows the voltage at its output terminal. ** This is why there is a warning that you should not operate such an alternator without a battery attached to it. In such a situation the alternator will sense "no voltage" and try to boost its output to maximum voltage, most probably burning out the diodes in the alternator.

[2] This is where you are going to get a lot of different opinions. Currently available batteries all have the same "chemistry" - lead and acid - but vary in how they deal with the liquid acid. Simple "liquid acid lead" batteries are the mainstay of the market and vary from thin plate "starting batteries" (a.k.a., car batteries) and thick plate - "deep cycle" batteries typically used in trucks and marine applications.

- - What I personally term "exotic" batteries include everything else such as AGM (Absorbed Glass Matt) batteries and "Gel" batteries. These types try to limit the movement and possibility of spilling acid by altering the free movement or the viscosity of the acid. But to gain their advantages, they also bring them the problem of being very sensitive to charging regimens/procedures.

- - So it is my opinion that if you want to use "exotics" you need to have good "external" regulation of the alternator and additionally "smart" shore power battery chargers and "regulated" solar & wind systems. This can add up to significant money above and beyond the significantly increased cost of exotic batteries over old fashioned "liquid acid lead" batteries.

- - But if you want to mix exotics and simple batteries I would suggest setting up your charging systems, alternator, shore power, and wind/solar to match the requirements of the exotic batteries that you chose. Normally the exotics would be used in your "house" system. So charging, sensing and regulation would be linked to the house batteries.

- - The starting batteries would then be "piggy-backed" off the house system for re-charging by the use of a battery combiner or "echo-charger." The purpose of these devices is to prevent "back-powering" from the starting batteries to the house batteries. As I explained earlier, a battery combiner is just a voltage sensitive relay that closes to allow charging, but opens to prevent one set of batteries from draining the "others." The "echo-charger" is a notch better in that it not only connects the starting battery the house batteries to allow re-charging but also "regulates" the amount of charging applied to the starting batteries. A very nice feature for which they charge extra money.

[3] On diesel engines (at least, old fashioned diesels) there is no need of electricity for the engine to operate. But it is nice to know how the systems are doing on the engine so we have electrical "instruments" - oil, temp, etc. I would suggest that these systems be operated from the starting battery via a sub-panel of circuit breakers or fuses. Since the alternator "output" is directly (hot-wired) connected to the starting battery, it is merely a matter of choice where the power supply wire to these instruments originates. Likewise, a voltmeter, regardless of whether it is connected to the alternator output stud or the starting battery, will show the battery voltage until the alternator output voltage increases above battery voltage (charging), then it will show alternator output voltage.

[jumpshowhigh]

You could also buy a better battery isolator that doesn't suffer from a large voltage drop. On my boat, I'm running a POWER-GATE battery isolator which is MOSFET based. Thing is amazing and totally works as advertised. No relation, just very impressed with its performance verses what I originally had installed which didn't let my batterys charge fully.

[R_C]

osirissail,

Thanks again for the detailed reply. I now realize that the existing isolator and the internally regulated alternator both contribute to the undercharging. The isolator's voltage drop is one issue. Even with a severely discharged house bank, the alternator is sensing the start battery's state of charge so it will never put out its rated amperage for any length of time in the current configuration. Now I wonder if this alternator can handle the heat when it begins running at higher output for extended periods.

To minimize new wire and wire lengths, I will run the existing 10awg wire from the starter solenoid to the oil pressure switch with an inline fuse. I can then connect the wires to the instrument panel and the fuel shutoff solenoid relay to the hot side of the pressure switch. If the pressure switch can't handle the two extra ring terminals (I'm away from the boat now) I can add a separate power post if necessary. I don't know how many amps the alternator excitation, instrument panel, and fuel shutoff solenoid relay draw combined to properly size the inline fuse. I'm guessing 10A should be sufficient. Perhaps I should use 20A which is closer to the max ampacity of a 20awg wire. Any thoughts?

A Google query of "alternator 12088N" returns information suggesting this is a Lucas alternator but I have been unable to find any schematics or datasheets. I wonder if the two unused connectors next to the excitation wire could be alternate field and sensing connections. Is it possible this alternator's regulator could be disabled so I can use it with an external regulator?

[OceanSeaSpray]

What you have here is a common problem with all old internally-sensed self-excited alternators. There is no magic wiring fix for it. Voltage sensing initially comes from your starting battery when the pressure switch closes, then a field builds up in the alternator, it kicks in and from this point on, it measures its own output. This is the problem, because the output is now higher than the battery voltage.

Inside the alternator, there is a small second positive rectifier that provides the power for the regulator and field winding.
If you want to make it sense externally, you need to disconnect the output of that auxiliary rectifier, where all three diodes come together, so it doesn't energise itself internally any more, so the sensing voltage remains battery voltage, after the blocking diodes.
Because there is a little variation between brands, models etc, it is a matter of looking inside the unit and working out the way it is setup. A lot of alternators have a D+ terminal for the charging light. The auxiliary rectifier feeds into this D+ and the charging light also connects there. It looks like it could be EXC on your unit, but I also see you have 2 more unused terminals. Tacho and something else?
If you disconnect the auxiliary rectifier inside, all the field current must be come from outside and you can't go through a light. What you have with just a pressure switch to the positive would already be fine.

I have modified many units this way to make them compensate for diode voltage loss. It is a zero-cost alteration.

Another option for you would be sourcing a modern, externally sensed alternator from a wrecker for example. A lot of modern units sense through a specific input on a 2-pin connector. One for the light, one for sensing. Unless both are fed with +12V, the alternator won't operate.

Third and most expensive way to go is using a MOSFET splitter like already suggested.

[smac999]

I didn't read the above stuff. but here are some facts

1. currently. your engine battery will drain as you drain the house battery. because you have a path from engine battery to starter to alt post to isolater to house battery. when the house battery drops to .7v below the engine battery, the engine battery will start draining into the house to charge it. this is a big screw up by the person who installed the isolator

2. I`m betting the 10awg wire was the original charging wire. and someone added the isolator and 6awg later. and wasn't smart enough to figure out the entire system to remove the old, and just added instead.

however if you removed the 10awg. you would no longer have power to the other wires on the alt. because it won`t flow backwards on the diode (on most anyways, some of the electronic ones will) so your gauges, stop, and alt ex. wouldn`t have power. the alt would never start. you`d never get power there, and then the engine wouldn`t stop either...

I would move those 3 wires to the starter post. (or a bus powered from there, or engine battery after engine battery switch, which you have right? not in pic) all fused of course.

then there are bettery battery isolators. acr`s are easy and work. the promariner isocharges are pretty neat. these have a alt reg sense post to best control an external reg. otherwise you have to pick a bank. and voltages on both will normally be different. (unless acr)

what is charging the gen battery? I would move one of the charger leads to the gen batt. or wire the 3rd lead if 3 banks.

[roetter]

Some internally regulated alternators have an external voltage sensing wire connector.

I have an internally regulated Mitsubishi 115A alternator that came with the Volvo B2-40, and it has an external sensing connection. The pink (almost white) wire going to the lower right of the alternator is the voltage sense wire.



I have installed in a double throw switch to connect the sensing port either:
- to the battery terminal (=high amperage)
- or to the output from the split diode (=medium amperage) as the voltage drop in the rather long charge cable to the battery is quite significant at high amps.

I can also manually route the voltage sense to "at the alternator" for even lower output.

The alternator is set to 14.2V at the sense port.

Sensing at the battery terminal will drive the alternator at full output (minus heat regulation) until 14.2V is reached at the battery terminal. I have measured over 15V at the alternator with this configuration.

Sensing at the split diode output regulates the 14.2V to that location. The voltage drop in the long charge cable can run up to almost 0.5V at high amps (very empty battery). So I see only 13.7V at the battery until the amps drop and the voltage slowly climbs to 14.2V at the battery terminal as the amps go to near zero.

Sensing at the alternator adds another 0.7V voltage drop, so at the terminal I may only see 13.0V at high amps (0.5V in the cable and 0.7V at the diode).

I have Lithium batteries with very low internal resistance and they tend to overload the alternator, so I use these methods to throttle back the output of the alternator and not kill it.

[R_C]

I did eliminate the battery isolator shortly after I started this thread and all is working well. I now have a single fused wire from the stock alternator output post to the house bank. Yes, an external smart regulator would be better but with solar panels, we rarely leave an anchorage (even after a week or more) with the house bank more than 25% discharged. And we aren't under way daily so the alternator isn't doing a lot of charging.

I added a fused wire from the starter solenoid to a power post next to the oil pressure switch for all the other wires originally connected to the alternator output post. And I fused all branch circuits at the power busbar.

I added the echo charger to keep the start battery topped off when any charging source is active (solar, genset, alternator). The genset battery is completely isolated and is only charged by the genset's alternator. The sketch below shows my current charging configuration.

[OceanSeaSpray]

A lot of us talked about way of keeping the diodes, but fair enough if you are going to have one bank only.
There are a few aspects you might want to think about here:

• Fusing B+ alternator leads is risky. Some will say you should, some the exact opposite. If your fuse blows for any reason, you will destroy the rectifiers and regulator in your alternator - so probably hardly worth fixing. If you don't fuse it, well, there are too many real issues. Alternators cut back on temperature and fail if disconnected under load, so you are not protecting the alternator itself from burning out. If it shorted internally, there are enough small wires, diodes etc inside to blow and open the circuit. If you really want to fuse it, you should do it well above its rating, not at 80A for a 70A rating. It could blow that fuse starting on a cold day without anything being wrong.
  I personally wire them in direct and I use heavy battery cable from the B+ post. Otherwise you just can't get the current out of them, especially when internally sensed like yours. A very short cable could arguably get you there, but it will get quite warm.
• Same, unless very short, the 6AWG wire to the busbar is unlikely to carry 100A. It will lose to much voltage. You need to use heavy cable if you are going to start shifting 80, 90, 100A.
  On a mains system, you can plan with voltage losses of 5-10% at times, and you still have plenty left, but on a 12V system, one volt goes a long way between fully charged and near flat.
• Why don't you disconnect the positive out of the bank? A single ground disconnect can disconnect multiple banks with a single switch, but you are down to one bank and it is not good practice anyway.