Changing batteries

[merl]

I am saying it does not happen. What credible failure mode could there be?

To be fair Geek there is an issue with having a MPPT solar controller connected to the panels but NOT to a battery due to the high frequency switching spikes generated at the output of the buck coil, I covered it here on WC.

Solar controller question

That's one of the three I've tried and connected the panels first. No issues. Ok if the fuse is removed untill the battery is connected, all in the instructions, the makers know best.

wildcamping.co.uk

see post 23 and 26. In post 26 you'll see a circuit diagram of a typical MPPT solar controller with Zenner diode (D2) offering crowbar protection on the output clamping the output spikes to the zenner value of 24V, you can see it's rated at 5 watts so it's going to dissipate quite a lot of power and get hot. Personally I wouldn't want to rely on it for a long period of time while swapping out a battery, better to simply pull a wire coming from the solar panels out of the controller.
Cant see an issue with PWM types though

 

[Geek]

Without looking at those posts (I'll look later) I suspect that it's not describing something that will cause a problem when switching between batteries.

Generally, when I switch between battery banks, the disconnection is for a couple of minutes at most. I've done that literally hundreds of times with no issues.

If an mppt controller can't cope with power from the panels but not the battery, putting a fuse in the battery lead would be a risky endeavour.

 

[trevskoda]

Fuses protect cables. The fuse should be at the source end that might provide a high fault current. That means the battery end of the cable to the controller, not the controller end.

Votronic put one first so you can disconnect panel from unit before you pull a battery out, not for any other use.
But a switch would do the same thing.

 

[merl]

Without looking at those posts (I'll look later) I suspect that it's not describing something that will cause a problem when switching between batteries.

I guess it'd depend on how quickly you swapped the batteries but personally I think there are so many variables to 'switching between batteries' to give all encompassing advice that it doesn't matter if you leave the panels connected while you do the job. But the presence of a large 5 Watt 24V Zenner on the output shows that the output voltage of the controller is capable and indeed expected (subject to solar) to reach 24V while doing so, this would (depending on controller wiring) mean the system voltage would rise to 24V, again, probably OK but not worth the risk.
I see the clamping Zenner as short term protection (of the solar controller) from a situation with panels connected but no battery, yes it would protect the controller but only for as long as the Zenner remained working, at that 5W rating surely you see the risk of it getting hot and failing?

If an mppt controller can't cope with power from the panels but not the battery, putting a fuse in the battery lead would be a risky endeavour.

There is an issue with 'power' as suggested with the protection zenner getting hot and failing IF the battery is left disconnected for too long but there's also the issue of peak voltages, very short term transient high voltage spikes of just a few milliseconds will take out semiconductors.
As you've already said, a fuse is there to protect wiring so when pulling a fuse WRT this post it's simply just being used as a means of isolation, ie a switch and as Wildbus has already said and I'm sure you appreciate it'll depend on where the fuse is in the circuit as to it's effectiveness.
Ultimately Geek, you've swapped batteries multiple times without issue but you did ask the question "What credible failure mode could there be?" So I thought I'd point one out to you, not trying to be a smart arse or owt but as an electronics enthusiast I thought you might like the input, I know I would.

 

[Geek]

see post 23 and 26. In post 26 you'll see a circuit diagram of a typical MPPT solar controller with Zenner diode (D2) offering crowbar protection on the output clamping the output spikes to the zenner value of 24V, you can see it's rated at 5 watts so it's going to dissipate quite a lot of power and get hot.

I've now been able to look at that diagram.

Looks like this is a 12v only controller, and the diagram says it's for a 9v, 3w solar panel. I'm not sure how typical that is in itself: every one I've had has been 12v-24v and able to take 40 or 50v from the panel.

The way it's wired with the zener limiting it at 24v seems odd.

I suspect it's an additional safety feature in case someone connects a high voltage battery, because high voltage from the panel wouldn't make it past the mosfet labelled Q1, even if a 9v 3w panel was able to deliver 5 watts at 24v

Controller failure because of no battery regulating the output doesn't look credible to me, given the ratings of the input.

 

[merl]

Looks like this is a 12v only controller, and the diagram says it's for a 9v, 3w solar panel. I'm not sure how typical that is in itself: every one I've had has been 12v-24v and able to take 40 or 50v from the panel.

Doesn't matter what voltage the controller is intended, the circuitry is simply demonstrating the operation of a MPPT controller, there will be variations on a theme and the voltage from the panels is largely irrelevant. The output voltage with and without the battery attached is whats important here because that's the issue at hand.

The way it's wired with the zener limiting it at 24v seems odd.

WHY? It makes perfect sense! It's there for protection from transient spikes from the buck coil if someone leaves the panels connected with NO battery!!

I suspect it's an additional safety feature in case someone connects a high voltage battery, because high voltage from the panel wouldn't make it past the mosfet labelled Q1, even if a 9v 3w panel was able to deliver 5 watts at 24v

Completely wrong on several counts.
If a high voltage battery was connected to the output with a voltage exceeding the threshold zenner voltage it would simply blow the zenner, you cant expect to clamp a battery voltage down with a 5W Zenner!!
Why cant the panel voltage "make it past the Q1 Mosfet"? Surely you're not suggesting that Q1 remains 'hard on' to clamp the panel voltage to ground?

Well, I'll leave it at that, I'm not getting involved with this any more. Just trying to offer some advice but being 'right' seems more important to you than learning something.
Stick a fork in me, I'm done.

 

[trevskoda]

For forks sake ease up folks.

 

[merl]

For forks sake ease up folks.

Really don't know why I got Forking involved in the first place TBF Trev, could have spent my time doing something more productive like smashing my head against a brick wall.

 

[Geek]

As Merl has opted out, this reply is for anyone else reading this thread. He is simply misled.

That diagram is for a low power boost controller. As far as I can see, it isn't even mppt: two voltage comparators and two timers (with one used as an oscillator) aren't going to find the maximum power point

Every mppt solar controller I have seen is a buck controller, not a boost controller. They don't work the same way at all.

 

[trevskoda]

Bucking hell, more to ponder on.

 

[nabsim]

It is easier to follow manufacturers instructions though lol

 

[Geek]

Safer, perhaps, but not easier.

 

[trevskoda]

No instructions in Irish.

 

[wildebus]

Really don't know why I got Forking involved in the first place TBF Trev, could have spent my time doing something more productive like smashing my head against a brick wall.

When speaking to certain folk, you soon find the wall is a far more attractive proposition

As an aside, and this is not something I would recommend however, the Victron MPPT controller seems to work fine without a battery attached! I found this out in an interesting way as I run my Habitation electrics THROUGH the LOAD output of a Victron 100/20 MPPT. I do this as it allows me to both monitor and remotely control the power.

Back to the point, I was running stuff fine generally but the hab electrics would go off every night. Settings for Load output was set to "Always On" and could find no reason why this what happening and was totally discombobulated, and even posted a question on the Victron Community site (where just occasionally you do get some useful feedback other then suggestions to try what you have already tried multiple times already).
But one reply made me realise the power was lost when it was dark, AND connection to the MPPT was also lost at the same time. Controller losing power when it lost solar energy - which meant it had no battery power of course!
On removal and careful examination I found a dry joint on the Battery +ve, so the Solar Controller and the Battery had not been properly connected for around a month by the time I found this, and the MPPT Controller had been working like a power supply.
(I had checked the battery connection a few times whilst troubleshooting, but the nature of a dry joint means the connection is often remade when you push on it, so the very action of putting meter probes firmly on the terminals 'fixed' the connection whilst being checked, so never found the problem that way )

Victron sent me a replacement Controller and whilst waiting for it, I fixed the bad one anyway with a touch of solder and I had Hab power after dark . But that repaired MPPT controller has been working without a problem for maybe 3 years, including maybe 2 years since I fixed it, with no adverse issues.

 

[merl]

When speaking to certain folk, you soon find the wall is a far more attractive proposition

As an aside, and this is not something I would recommend however, the Victron MPPT controller seems to work fine without a battery attached! I found this out in an interesting way as I run my Habitation electrics THROUGH the LOAD output of a Victron 100/20 MPPT. I do this as it allows me to both monitor and remotely control the power.

Back to the point, I was running stuff fine generally but the hab electrics would go off every night. Settings for Load output was set to "Always On" and could find no reason why this what happening and was totally discombobulated, and even posted a question on the Victron Community site (where just occasionally you do get some useful feedback other then suggestions to try what you have already tried multiple times already).
But one reply made me realise the power was lost when it was dark, AND connection to the MPPT was also lost at the same time. Controller losing power when it lost solar energy - which meant it had no battery power of course!
On removal and careful examination I found a dry joint on the Battery +ve, so the Solar Controller and the Battery had not been properly connected for around a month by the time I found this, and the MPPT Controller had been working like a power supply.
(I had checked the battery connection a few times whilst troubleshooting, but the nature of a dry joint means the connection is often remade when you push on it, so the very action of putting meter probes firmly on the terminals 'fixed' the connection whilst being checked, so never found the problem that way )

Victron sent me a replacement Controller and whilst waiting for it, I fixed the bad one anyway with a touch of solder and I had Hab power after dark . But that repaired MPPT controller has been working without a problem for maybe 3 years, including maybe 2 years since I fixed it, with no adverse issues.

I did a similar thing. Can't remember why but I once had a bright idea of putting a diode in the output and then compensated for the voltage drop of the diode by increasing the output settings. All fine until the evening when the array stops generating, now of course the solar controller can't 'see' a battery voltage due to the diode is now reversed biased and and everything had to re programme the controller in the morning. Mumbled to myself " of course you tw@, that's never going to work, what a stupid idea"

 

[trevskoda]

I did a similar thing. Can't remember why but I once had a bright idea of putting a diode in the output and then compensated for the voltage drop of the diode by increasing the output settings. All fine until the evening when the array stops generating, now of course the solar controller can't 'see' a battery voltage due to the diode is now reversed biased and and everything had to re programme the controller in the morning. Mumbled to myself " of course you tw@, that's never going to work, what a stupid idea"

A zener diode will work or one for a rectifier.

 

[merl]

A zener diode will work or one for a rectifier.

Fraid not Trev

 

[wildebus]

I did a similar thing. Can't remember why but I once had a bright idea of putting a diode in the output and then compensated for the voltage drop of the diode by increasing the output settings. All fine until the evening when the array stops generating, now of course the solar controller can't 'see' a battery voltage due to the diode is now reversed biased and and everything had to re programme the controller in the morning. Mumbled to myself " of course you tw@, that's never going to work, what a stupid idea"

the setup works fine when the battery is connected and I am still running the hab electrics through the LOAD output

 

[merl]

When speaking to certain folk, you soon find the wall is a far more attractive proposition

As an aside, and this is not something I would recommend however, the Victron MPPT controller seems to work fine without a battery attached! I found this out in an interesting way as I run my Habitation electrics THROUGH the LOAD output of a Victron 100/20 MPPT. I do this as it allows me to both monitor and remotely control the power.

Back to the point, I was running stuff fine generally but the hab electrics would go off every night. Settings for Load output was set to "Always On" and could find no reason why this what happening and was totally discombobulated, and even posted a question on the Victron Community site (where just occasionally you do get some useful feedback other then suggestions to try what you have already tried multiple times already).
But one reply made me realise the power was lost when it was dark, AND connection to the MPPT was also lost at the same time. Controller losing power when it lost solar energy - which meant it had no battery power of course!
On removal and careful examination I found a dry joint on the Battery +ve, so the Solar Controller and the Battery had not been properly connected for around a month by the time I found this, and the MPPT Controller had been working like a power supply.
(I had checked the battery connection a few times whilst troubleshooting, but the nature of a dry joint means the connection is often remade when you push on it, so the very action of putting meter probes firmly on the terminals 'fixed' the connection whilst being checked, so never found the problem that way )

Victron sent me a replacement Controller and whilst waiting for it, I fixed the bad one anyway with a touch of solder and I had Hab power after dark . But that repaired MPPT controller has been working without a problem for maybe 3 years, including maybe 2 years since I fixed it, with no adverse issues.

What voltage output do you reckon you were getting from it while it was in its defective state?

 

[trevskoda]

Fraid not Trev

Ok me back to the books again.