Both battery banks receiving charge despite being isolated

[dluks]

I have a house bank and a starting bank, both of which can be isolated using a common four-position battery switch (OFF, 1 [house], BOTH, 2 [start]). I have solar connected to a Victron SmartSolar charge controller which has its negative lead connected to the main DC negative bus (to which both the house and starting banks ground to) and its positive lead connected to the "1" post of the battery isolator switch, i.e. the battery side of the house circuit at the switch.

In theory, unless I'm fundamentally misunderstanding something, the charge current from the charge controller, regardless of isolator switch position, should only be going to the house bank. My issue is that the starting battery's voltage nearly mirrors the house bank's when charging, e.g. it will read a high voltage, such as 14.6V, when the controller is in absorption mode. Again, the positive lead from the charge controller is connected to the switch on the battery side of the house circuit, not to the common post. I don't have a current-monitoring system in place so I'm relying on voltage readings for now.

My first guess is that the switch may be faulty and therefore when at the house position it is actually paralleling the two banks? Or could it be some stray current stuff (of which I'm woefully uneducated) due to the batteries being wired to the same negative bus? If it's a switch issue, how do I test this?

The two banks consist of different batteries (house uses two 12V flooded deep-cycle Trojans and start uses a single 12V flooded dual-purpose Bruteforce, and so my worries are that, if paralleled, one bank will be consistently (albeit slightly) under- or overcharged. I also don't have a great regulator for the alternator (which is wired to the common post of the switch), and so I generally pop the switch over to only the starter when starting and running the engine to avoid improperly charging the house bank.

Any thoughts?

 

[Davidasailor26]

A couple thoughts -

How are you getting your voltage readings, from a voltage meter on the panel or a multimeter? If from a panel then it is possible the meter itself is part of the problem. Verify the readings with a handheld multimeter.

My guess would be the switch too - either that it has somehow failed or that a previous owner improperly wired it. Disconnect the batteries and measure resistance across the various switch ports to see if it’s isolating things. Then start connecting one thing at a time and see if anything unexpected happens to the voltage readings.

 

[dlochner]

In theory, unless I'm fundamentally misunderstanding something, the charge current from the charge controller, regardless of isolator switch position, should only be going to the house bank.

If the switch is in the Both position, current will flow to both batteries.

Do you have an ACR or Echo Charger between the batteries?

How is the starter wired?

These two articles may be helpful.

1/2/BOTH Battery Switch Considerations

Making Sense of Automatic Charging Relays

 

[dluks]

A couple thoughts -

How are you getting your voltage readings, from a voltage meter on the panel or a multimeter? If from a panel then it is possible the meter itself is part of the problem. Verify the readings with a handheld multimeter.

I am using a voltmeter installed on the panel itself and I've confirmed the readings with a multimeter at the batteries directly. Didn't think about measuring resistance across the ports so I'll try that next.

 

[JoeWhite]

To test if the switch is at fault, lift the solar panel charger lead and the battery lead off of the switch lug then bolt them together and see if the voltages on the other switch lug has changed.

 

[dluks]

If the switch is in the Both position, current will flow to both batteries.

Do you have an ACR or Echo Charger between the batteries?

How is the starter wired?

Ah, d'oh, that makes sense for the BOTH setting since they would then be paralleled.

No ACR between them, but there is an echo charger from the inverter/charger (which is essentially never on unless I'm on shore power) running to the "start" post on the switch. Despite the inverter/charger being off, could that be affecting things? Now that I think about it, since the main I/C cable is connected to the house bank and the echo~charge cable is connected to the start circuit, would that mean they are now in parallel? I had assumed both the main and echo circuits would be open when the I/C isn't running.

The starter battery is wired simply, with it's positive going to the "start" post of the battery switch and its negative to the main DC grounding bus. The starter motor is wired to the common post on the switch so that I can crank the engine using one or both of the banks.

 

[dlochner]

I had assumed both the main and echo circuits would be open when the I/C isn't running.

I think not. Take your multimeter and check the connections there. There is likely current at the connections.

 

[dluks]

I think not. Take your multimeter and check the connections there. There is likely current at the connections.

Yep, the echo~charge line running from the inverter/charger to the start post on the battery switch appears to have been the culprit. Switching off the main inverter/charger switch opens that circuit and the starter battery is no longer receiving charge current from the solar charge controller. Thanks all for helping put the pieces into place!