VSR rewiring

[pmagowan]

I would be interested in the results as I have the same problem in my head. Voltage is like a potential difference or a head of water, if flowing out (current) faster than flowing in the head of water drops. I can’t understand how the charger keeps the voltage high without actually charging the battery, no matter how slowly. Perhaps I have the theory wrong or something in the real world is having an unusual effect.

 

[lustyd]

Sorry if my previous post wasn't clear, but what I don't understand is how the MPPT could possibly maintain 14+V if I'm drawing more from the battery than its putting in, because then the battery is not "charging" in this scenario. My understanding is that the voltage (which is of course the same at the output of the MPPT controller as on the battery terminal) will drop when this happens. To put it another way: the current doesn't "charge" the battery with current subsequently "flowing out of it". If the MPPT controller is providing 7A and I draw 10A then the result is a battery that is discharging at 3A and it's voltage can't be more than 12.7V-12.8V. Or at least that is how I think it works.

EDIT: I'll test this on my boat tomorrow if I manage to get there before sunset.

Voltage of a battery drops with load, yes. This is because it's chemistry causing the voltage. Not the same process with chargers at all, especially ones designed to compensate and charge at a specific voltage

 

[PaulRainbow]

Sorry if my previous post wasn't clear, but what I don't understand is how the MPPT could possibly maintain 14+V if I'm drawing more from the battery than its putting in, because then the battery is not "charging" in this scenario. My understanding is that the voltage (which is of course the same at the output of the MPPT controller as on the battery terminal) will drop when this happens. To put it another way: the current doesn't "charge" the battery with current subsequently "flowing out of it". If the MPPT controller is providing 7A and I draw 10A then the result is a battery that is discharging at 3A and it's voltage can't be more than 12.7V-12.8V. Or at least that is how I think it works.

EDIT: I'll test this on my boat tomorrow if I manage to get there before sunset.

I have never seen this happen and i have fitted a lot of VSRs.

My previous boat, which i lived on full time, had a Victron VSR and solar panels. Most of the time during the Summer the solar power kept up with demand. Sometime, such as 2 or 3 cloudy days in a row, it didn't keep up. When that happened the domestic battery voltage dropped and the VSR opened. Not once did i ever have a discharged engine battery.

 

[lustyd]

Not once did i ever have a discharged engine battery.

That you noticed! I didn't see the VSR disconnect in two months last year during the day while cruising with solar. The bank did discharge though, and we definitely used more power than we put in some weeks. The MPPT kept the voltage above threshold while trying to charge, as it's designed to do.

 

[geem]

That you noticed! I didn't see the VSR disconnect in two months last year during the day while cruising with solar. The bank did discharge though, and we definitely used more power than we put in some weeks. The MPPT kept the voltage above threshold while trying to charge, as it's designed to do.

I just can't see how that can happen. Thr voltage measured at the battery is the battery voltage. If the mppt is keeping the battery voltage high enough, the battery can't be discharging. I don't like VSR as you can't adjust the threshold voltage settings. I always though the disconnect voltage was too low. I suspect that was the problem

 

[lustyd]

I just can't see how that can happen. Thr voltage measured at the battery is the battery voltage. If the mppt is keeping the battery voltage high enough, the battery can't be discharging. I don't like VSR as you can't adjust the threshold voltage settings. I always though the disconnect voltage was too low. I suspect that was the problem

The voltage is the voltage regardless where you measure it, you can't measure "battery voltage" without disconnecting the battery and waiting, that's chemistry

 

[geem]

The voltage is the voltage regardless where you measure it, you can't measure "battery voltage" without disconnecting the battery and waiting, that's chemistry

Exactly, so if the voltage is high enough, the battery can't be discharging

 

[lustyd]

Exactly, so if the voltage is high enough, the battery can't be discharging

Yes, it can. It’s the charger maintaining voltage and acting as a power supply, the VSR is simply measuring the voltage of the charge profile of the charging device, which is by design higher than the VSR cutoff.

 

[pmagowan]

If a charger is maintaining a higher voltage than the battery would then the charger is pouring energy into the battery and the battery must be charging, not discharging (irrespective of loads). If the battery is discharging ie pouring energy out at a faster rate than it is being poured in then the voltage (potential difference) will drop. In the water analogy the head of water will reduce. That is my understanding which may be completely wrong but I would like to see some controlled data. I just don’t know how a weak charger can maintain the voltage when the load should be pulling it down.

 

[PaulRainbow]

That you noticed! I didn't see the VSR disconnect in two months last year during the day while cruising with solar. The bank did discharge though, and we definitely used more power than we put in some weeks. The MPPT kept the voltage above threshold while trying to charge, as it's designed to do.

I would have noticed, between the extensive monitoring, logging and the battery alarms.

 

[pmagowan]

Does ohms law not state voltage equals current times resistance? When a battery is discharging the resistance reduces and thus the current would need to be increased if you maintain voltage and thus a trickle charge would not manage it. I am getting above my pay grade trying to understand this! Obviously observed behaviour and theoretical behaviour can differ but usually there are “reasons”

 

[lustyd]

I would have noticed, between the extensive monitoring, logging and the battery alarms.

You seem to be missing the crucial battery % over time graph. Your voltage graph confirms a VSR would be closed all day though, so whether the battery drains or charges will be down to how much juice the solar is inputting.

 

[Baggywrinkle]

This is a graph of the charge voltage of a Victron MPPT - it is definitely lower when the batteries are discharged and drawing a larger current - in the bulk phase the VSR should not close until the voltage rises above the storage voltage - once closed, it's the disconnect behaviour that may be the problem.



My experience is that when a load is being drawn from the batteries that is much greater than the MPPT can supply, it can go back into bulk mode, because the voltage drops accordingly. This is assuming good batteries and constant MPPT output. As the battery approaches fully charged, the charger will end up between the float and absorption voltage supplying little current - which will keep the VSR connected.

One possible scenario for what @lustyd sees is that the MPPT isn't supplying a constant current due to cloud/shadows, and the resultant voltage variation is not enough to immediately disconnect the VSR - for example a cloudy day with varying and quick changing cloud cover, or swinging on a mooring with constant variable shade on the panels, and a battery bank in absorption or float.

There will be current flowing from the starter to the house banks while the VSR is counting down to a disconnect if the MPPT is not supplying enough current. The starter and house bank are not balanced, especially after a night of heavy house bank use.

A short across the batteries (effectively what the VSR does) will pull the voltage of the best battery down (starter), and the worst battery up (house) in the absence of a charge source - if the voltage they then reach is not enough to trigger an immediate cut-off of the VSR, then the disconnect timer starts ... if the MPPT comes online and starts charging then the voltage is pulled up and the timer gets reset. Then the MPPT goes offline and the timer starts again ... this cycles constantly and the starter is effectively a permanent, unbalanced, member of the house bank, until they equalise or the voltage eventually drops low enough to disconnect.

Basically what you have is a problem where the the VSR, due to its disconnect timers, is allowing a current to flow from the starter to the house bank permanently because they are in different states of charge - and the MPPT is regularly resetting the disconnect timer. Eventually they should balance but it's effectively a race to sunset.

... and at the end of the day, if the voltage remains OK then it is probably not a problem. In a modern car the smart alternator and recuperation system keeps the starter battery around 80-90% charged permanently.

 

[geem]

Yes, it can. It’s the charger maintaining voltage and acting as a power supply, the VSR is simply measuring the voltage of the charge profile of the charging device, which is by design higher than the VSR cutoff.

My smart shunt tells me if the battery is charging or discharging. If the loads are higher than the MPPT is supplying then the battery is discharging. If the loads are lower than the MPPT is supplying then the battery is charging. Once the VSR has engaged so the starter and house bank are functioning as a single battery then the same situation applies. You simply have a single large bank of batteries either being charged or discharged. This will happen all day long until you get sufficient discharge to bring the whole battery voltage lower enough to disengage the VSR. The problem with VSRs is they disconnect at too low a voltage. They should disconnect at at about 13.4v not 12.7v. This would leave the starter battery with a good charge

 

[andsarkit]

I have no direct experience but this Sterling VSR has a higher preset cutoff of 13.0V and connect at 13.3V. It is also programmable to other voltages if required.
Sterling VSR
A bit expensive for me and the programming looks complicated.
I would be happy for the engine battery to use a small amount of its capacity for regular domestic cycle use as long as it retained enough charge for engine starting.
This thread has been useful as I had not previously considered the engine battery partially discharging to the house bank through the VSR. For most people I don't think this is a real problem.

 

[lustyd]

For most people I don't think this is a real problem.

I completely agree, and I only noticed and thought about it after being long term on board and I suspect that's the case with Geem as well. For most weekenders it would be hard to discharge both banks because the problem is slow discharge over time. If you discharge fast enough for it to be an issue over a weekend then you might very well see voltage drop and a disconnect, but realistically you'd probably notice you were hammering the battery and start the engine to compensate anyway.

 

[PaulRainbow]

You seem to be missing the crucial battery % over time graph. Your voltage graph confirms a VSR would be closed all day though, so whether the battery drains or charges will be down to how much juice the solar is inputting.

It's just past midday and the solar panels are at 34.22v. The domestic batteries are taking 5.55a, after whatever the loads are consuming.

The solar controller is outputting 14.4v, yet the domestic battery terminal voltage is only 13.1v, why is that ?

The engine battery voltage is only 13.0v, despite being in parallel with the domestic bank, why is that ?

I'd rather you didn't keep telling me how i don't know how to tell if my engine battery is being discharged or that i wouldn't know if it was flat, thanks. After 50 years in the job i reckon i know how to tell if my batteries are flat. I have had enough occasions (being full time liveaboard for the last 7 years) where the domestic batteries have dipped enough to set the low voltage alarms off. Yet, when this has happened the engine battery has remained charged and isolated by the VSR.

If you are observing something different, that's very odd, something i have never seen. I'm sure there are plenty of other things that i have not seen too. As Geem is often saying, a VSR is not the be all/end all solution, which i totally agree with. So, if it doesn't work for you, i would consider:

1) Set the BEP VSR so that it only works when the engine is running (temporary fix IMO)
2) Fit a Victron Argofet.
3) If you are planning to fit Lithium fit a B2B charger.

 

[lustyd]

I'd rather you didn't keep telling me how i don't know how to tell if my engine battery

I didn't say that at all. I said you weren't showing us the proof you thought you were in the image, and you weren't and still aren't. 13.1 is not sufficiently low to disconnect most VSRs, so you are kind of reinforcing my point here.

 

[PaulRainbow]

I didn't say that at all. I said you weren't showing us the proof you thought you were in the image, and you weren't and still aren't. 13.1 is not sufficiently low to disconnect most VSRs, so you are kind of reinforcing my point here.

The image wasn't meant to be proof of anything, other than the fact that i monitor my electrics.

As usual, you want to just argue until the cows come home, so i'll leave you to that.

 

[LittleSister]

Voltage of a battery drops with load, yes. This is because it's chemistry causing the voltage. Not the same process with chargers at all, especially ones designed to compensate and charge at a specific voltage

The charger may well be designed to put out a specific voltage, but as far as I can see if the solar panels aren't putting out enough power (amps) to sustain that voltage, then the voltage must fall (unless perhaps, as I suggested earlier, the voltage is pulsed).

Imagine you have a battery disconnected from all loads but partially depleted and showing, say, 12v, and a controller or other charger set to 14v.. If you connect the controller output to the battery terminals, and disconnect everything else, the battery and controller output wires must (at the sort of power levels and distances we are discussing) be at the same voltage.

If the controller/charger can put enough power out to meet the draw of the depleted battery at 14v - let's assume that's 2A for simplicity - the voltage measured across the terminals of the battery will rise to 14v. If it is putting out no amps it will fall to 12v. If it is putting out 1A, it is not enough to meet the draw of the battery at 14v the voltage will fall to somewhere probably in the region of 13v.

If you connect a load drawing say, a fixed 1A (lets ignore variation with voltage), the controller/charger will now need to put out 3A to achieve 14v, if it's putting out no amps it will immediately fall to below 12v and reduce further as the battery discharges over time. If the controller puts out 1A, the voltage will fall to 12v but remain there (in the short term).

I fail to see how your controller can achieve the 14v (or whatever) it is designed to achieve when to is connected to a battery and/or other load if the solar panels are not putting out enough amps to meet the amount those loads draw at 14v.

If your solar panels are putting out only 0.5A at 14v (as I think you said earlier) and has a load (e.g depleted battery plus something else) on it of, say 5A, then any controller output connected to the battery and the load must fall below 14v, and very likely below the voltage at which the VSR opens (disconnects).

The only way for it to not fall below 14v, as far as I can see, would be for the controller's output to be pulsed rapidly, so that it was providing 5A, but only 10% of the time. The battery voltage would be fluctuating between 14v and whatever voltage the battery would stand at if not connected to the contoller/charger. If that pulsing were faster then whatever delay period the VSR is set to, then the VSR might think it was seeing a continuous 14v. Otherwise, it will open when it reaches it's opening (disconnecting) voltage (12.7v or whatever) and the engine start battery will not discharge further (other than actual engine stating and long-term self-discharge).

I will be happy to be shown I am wrong, but won't be convinced unless someone can explain why.

I suspect you may either be misreading what is going on, or have an engine battery getting towards the end of its life.