The attached diagram comes from this article: http://www.doityourselfrv.com/bring-van-camper-life-add-batteries-solenoid-stealth-rv/

Notice that the positive wires from the vehicle starting battery to the circuit breaker, and then to the continuous duty solenoid, and then to the aux (house) battery are specified as 10 AWG while the remaining main connections are specified as 2 or 4 AWG. I understand the importance of the large wire between the individual batteries in the aux battery bank, and also understand that the size of the wire to the inverter is dictated by the size of the inverter. I also plan to use 16 AWG wire from the fuse box to the solenoid.

What I am not completely convinced about is the 10 AWG wire. Please comment on that as well as any other parts of the diagram that pique your interest.

Thanks.

ETA: I will be traveling most of the day today, so don't be perturbed if I don't respond to your comments until tonight or tomorrow. Thanks.

 

The 10 ga. wire is properly protected by the auto-reset circuit breaker, but that is going to be the limit for charging the aux batteries. They will draw more than 10 amps and the breaker will cycle. Then there is the problem with modern alternators labeled, "Warranty void if used to charge a discharged battery". I built a 50 amp charger (200 amp surge) and I've never seen a healthy battery accept more than 35 amps, no matter how discharged it was, but the 90 amp alternators seem afraid you will use half their labeled capacity.

Then there is the boogeyman story about connecting unequal batteries. You know your aux batteries are going to be half dead after a 3 day weekend, and the solenoid connects them to the starter battery as soon as the engine lights up.

I think I'm exposing my ignorance.

 

#12,

You have put into words my muddled concerns. I think I should use 4 AWG for all the connections except the power for solenoid actuation.

Thanks, and it's good to hear from you.

ETA: I am planning to add two 100 watt solar panels to help hold up the house batteries.

 

What I did 30 years ago was use a couple of gigantasaurus diodes, one in series with each battery. Now, the alternator doesn't measure the battery voltage, it measures its own voltage. If you want to do the diode isolator trick, you have to start with adjusting the alternator.

 

How about an ideal diode circuit, such as this, with a low on-resistance N-MOSFET to isolate the batteries?
That should work with the standard alternator charge voltage setting.

 

That kind of drags me out of the stone age.
What? You can do that now?
Cool!

 

So, if I understand correctly, the diodes (ideal or otherwise) need to be connected directly from the alternator output, one to each of the batteries(?).

 

So, if I understand correctly, the diodes (ideal or otherwise) need to be connected directly from the alternator output, one to each of the batteries(?).

That's the way it worked in my 1973 Ford.
As long as the voltage drop is equal in the route to both batteries, they charge correctly.

 

As long as the voltage drop is equal in the route to both batteries, they charge correctly.

That would assume the voltage regulator is sampling the voltage after the diodes.
There's a problem with that for alternators with built-in regulators using regular diodes for isolation since I believe they sample the voltage at the alternator output (which I assume is why you mentioned changing the regulator voltage).
But the ideal diode circuit should work with either an internal or external regulator.

 

100% correct.