question

frankh avatar image
frankh asked

Using a SmartSolar MPPT 100|30 with a Common Port BMS and LiFePo4 batteries.

I need a definitive answer from someone that really knows what they are talking about.

It’s always being bandied about that the solar panels must be disconnected before the batteries, otherwise the Victron SmartSolar MPPT Controller will be damaged. Well, I find it pretty hard to believe that this unit is designed to self-destruct if its outputs are disconnected from the battery by any means; fuse, switch or otherwise. before the solar panels are disconnected.


I want to use a common port BMS with a LiFePo4 battery setup, and from what I’ve read, the BMS will break the battery connection to the MPPT at some point which, in turn, will damage the MPPT charge controller.

I do know the manual for the SmartSolar MPPT 100|30 states “First: connect the battery. Second: connect the solar array (when connected with reverse polarity, the controller will heat up but will not charge the battery). Torque: 1,6 Nm” but, scratching around the net I found the website pkys.com, and in their right-up on the Victron SmartSolar they state “You need to make the battery connections first so it knows what battery voltage you have. In the programming you can tell it the battery voltage and then in the future it won't matter what order you make the connections”. Now is this statement correct? If it is correct all this stuff about damaging the charge controller is just people talking parrot fashion, and there is no problem using a common port BMS on this charge controller; so long as one does the initial battery voltage programming of the charge controller first!


Any engineers responsible for designing these Victron charge controllers on this board??????????

MPPT ControllersLithium Battery
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

3 Answers
Guy Stewart (Victron Community Manager) avatar image
Guy Stewart (Victron Community Manager) answered ·

Hi @FrankH,

Disconnecting a battery from a charge controller that is charging is not recommended.

What occurs is the battery side voltage of the MPPT will spike, as it is raising the voltage and expecting the battery to draw down the voltage. When that is suddenly removed, for a moment there is nothing to constrain the MPPT's battery side voltage. It will shoot up, and there may be a false over voltage alarm generated and the unit may shut down. The MPPT may also handle it more gracefully, that would depend if there were other DC loads on the bus drawing down the voltage, or if the wasn't a lot of current going into the battery at the time.

Self destruction isn't part of the design - but neither is rapid changes like disconnecting a charging battery. It will certainly survive it occurring occasionally, but I would not make it part of the system design to have this as the control method of the MPPT's output either.

It is correct that the MPPT will do a system voltage detection on it's first start. I have been told by a Victron engineer that this algorithm does not use the solar voltage in any way, however for my peace of mind, I would still always connect the battery first (and double check the programmed/autodetected voltage with VictronConnect) before considering a system commissioned.

This is unrelated to regularly disconnecting the battery from the MPPT during charging though.

I think the FAR more likely is the risk to other DC components that are connected. Without the battery connected to moderate the voltage and smooth out the delivery of current, things like voltage will will swing wildly, very high, then very low then to it's target with a lot of DC ripple (presuming the other DC loads are less than the PV output).

A DC appliance that is rated for 24V suddenly getting 40V for a second could be enough to destroy it.

The MPPT itself can be isolated from rest of the DC bus, rather than have the battery itself disconnect from the DC bus, such as by using a Battery Protect just on the MPPT. Then the Battery Protect can be triggered via it's remote port.

Again this is not the recommended way of regulating the charge to the battery, it's just an emergency safety feature that is engaged when something else is going to blow up (like a runaway out of balance lithium battery cell at high temperature).

There is also a Remote Switch Port on many of the larger MPPTs, and this can be used to turn off the MPPT as required by the BMS. This can be used day to day without issues.

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

frankh avatar image
frankh answered ·

Hello Guy and thank you for your lengthly informative reply.


Your second paragraph talks about the MPPT "raising the voltage and expecting the battery to draw down the voltage". So, if we simply have a solar panel (plugged in last of all), connected to a Victron charge controller, connected to a battery and, on first connecting the charge controller to the battery, I then personally go on to program the definite battery voltage the controller should charge at; are you saying that an engineer has told you the output of the controller will exceed that which I have set it to if the battery is suddenly disconnected?


The third paragraph you say "It will certainly survive it occurring occasionally" A battery disconnect that is.

The charge controller is designed to take a battery disconnect reliably or it isn't, there should be nothing occasional or accidental about it. Considering that fuses and breakers do sometimes get fed-up with life and just break from old age/fatigue, the controller could blow on the first sudden battery disconnect if it is only designed to occasionally survive.


Regarding the forth paragraph. What exactly is the engineer saying? Are they saying the charge controller won't dump the solar panel voltage out of its output?

I'm not trying to upset anyone but I'm only interested in facts here and not what anyone feels.


Paragraph six & seven. If, on a battery disconnect, the charge controller cannot be relied upon to keep its-self under control and stay within the charge voltage it was manually set to by the user, I would say an electronic crowbar circuit is needed to suppress any excessive voltages so as to defend all the other attached electronics from the charge controller.

Once again you are stating what you think but, what is the fact? Can your engineer tell you and us all what actually happens on the output of a Victron SmartSolar 100|30 charge controller when, it has had its charge voltage manually set at the outset, and then goes on to suffer a battery disconnect; for what ever reason, at some point in the future?


Paragraph eight. My understanding of a Common Port BMS is that minus output of the charge controller goes to P- of the BMS and B- of the BMS goes to the negative of the battery thus, disconnecting the charge controller which, is what started me off on trying to find out what really happens when a charge controller is disconnected from the battery with the solar panels still connected.

All the other electronics would be powered strait from the battery via a battery protect, to protect the battery from being over discharged, I wouldn't use it as you suggest.


When it boils right down to it, what I want to know is; if a Victron SmartSolar 100|30 charge controller is connected to a battery then, has its charge voltage manually set then, has the solar panels plugged in then, at some point in the future and for whatever reason the battery is disconnected while the panels are still plugged in; what actually happens at the output of the charge controller? Does it go all stupid or does it stay sensible?

Facts are what I need from an engineer that works on these things; what people think or feel are of no use when it comes to answering my question.


All the best and thanks for your efforts,

Frank H.

6 comments
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Hi @FrankH,

I have got a few more comments back from the Victron MPPT hardware engineering dept:

  • The MPPT should NEVER break due to a battery disconnection. Even if that happens often.

  • If no other load is applied to the DC bus when the battery disconnects, a short voltage overshoot may occur.

  • The overshoot should only last a second or two, if it occurs, then the voltage will stabilise at the voltage set point of the current charge state (eg float, absorption), including temperature compensation.

  • Be careful if also using a remote battery sense option, if the MPPT sees a variation between the battery sense and the voltage at the terminals, it will adjust output to compensate (up to + - 2V)

  • The MPPT is not designed to be a power supply, it is a battery charge controller. So fast load variations may lead to voltage dips and spikes when the battery is disconnected and loads turn on and off. The MPPT output voltage will stabilise when loads are stable though.

  • Using the MPPT with a BMS like this that disconnects the battery is allowed by Victron, from our equipments perspective. In fact our own SuperPack lithium batteries operate on this same principle, where the battery disconnects itself internally to protect itself.

1 Like 1 ·
frankh avatar image frankh Guy Stewart (Victron Community Manager) ♦♦ commented ·

Thank you very much for your reply and the info on the remote battery sense option.

I have to say that the feedback you have from the Victron engineering dept. is what I hoped for and expected. I just couldn't believe all this tosh about 'disconnect solar panels first not battery or you'll destroy the MPPT'.


Thanks again and all the best,

Frank.

0 Likes 0 ·
anthony-sennett avatar image anthony-sennett Guy Stewart (Victron Community Manager) ♦♦ commented ·

Hi Guy,

Thanks for the info provided here - very informative. I would like a point clarified:

If no other load is applied to the DC bus when the battery disconnects, a short voltage overshoot may occur. The overshoot should only last a second or two,

What happens if there is a load applied to the DC bus when the BMS disconnects? Does the MPPT absorb the impulse caused by the battery opening the circuit (e.g. using a snubber circuit) or should I take extra precautions?

Cheers,

Anthony

0 Likes 0 ·
rickp avatar image rickp anthony-sennett commented ·

Good morning @Anthony Sennett,

I was having similar concerns when Guy pointed me to this thread. I’m not sure it will suit your situation, but I decided to locate the breaker in the circuit directly out of the SmartSolar, so that anything downstream would be disconnected when the breaker trips. My understanding from the information above is that anything remaining connected to the SmartSolar output might either: a) help absorb the effects of the disconnect, or; b) be damaged if it cannot handle a momentary spike in voltage. It occurred to me that the safe bet was to isolate the spike between the SmartSolar and breaker.

Rick

0 Likes 0 ·
Trina avatar image Trina rickp commented ·

Not a direct response to the post above but would look odd if I commented on the OP question when it was so well answered.

I do want to point out that even in small installs it really is not that much more complex to add an in-line fuse from your PV's or unplug that ONE side of the PV's while moving or changing a battery.

I appreciate the technical side of how this relates to a lithium BMS for custom setups though! Thanks!


0 Likes 0 ·
airtime avatar image airtime Guy Stewart (Victron Community Manager) ♦♦ commented ·
Guy,

I'm replying to an old (and very good) thread that almost completely answers my questions on this same topic. One further detail I'd like to ask. Can Victron quantify the magnitude of the voltage overshoot that occurs if the MPPT output is disconnected from the battery, as discussed above? And more specifically, in a 24V system, could any overshoot exceed 35V?

The reason I ask is that I'd like to use a BatteryProtect on the output of a SmartSolar 100/20, under control of my BMS. I want to use the MPPT's VE.Direct port to connect to a Cerbo GX, not use it for remote on/off. My battery is 24V LiFePO4. The BatteryProtect specifications say max input voltage of 35V. So as long as any voltage spike on disconnect does not exceed 35V, then the BatteryProtect should not be damaged on disconnect.

Appreciate your help in answering this question.


0 Likes 0 ·
markdt avatar image
markdt answered ·

Guy:

I have a very similar question to the person who posted back in February. I just ran across this situation and could use your help in evaluating what to do.

I had a failure of the main battery switch that connects my house battery bank to the DC loads on my boat. My SmartSolar MPPT 85/100 is connected on the "load" side of this battery switch, so if the battery switch opens completely, the MPPT controller would be supplying it's output into some DC loads on the vessel, but without batteries connected. This is what happened recently.

With the batteries disconnected, the MPPT controller was short-cycling between a fairly high output voltage (something on the order of 30 volts....it was a little hard to get a precise measurement) and shutting down. I'm guessing that's either because the voltage went too high, or because the load current was too high for the controller without the batteries to supply current. Either way, the output of the controller would jump up for a few seconds and then shut off. In the hurry to understand what was going on and prevent damage, I didn't record the error messages on my color control GX, and I can't find any longer term logs that might have the errors stored in them.

I've read the earlier replies and I think I understand the problem (I'm a retired electrical engineer). However, I'm wondering now if I need to add some external over-voltage protection on the output of my MPPT controller to avoid the momentary over-voltage situation you described, which could be harmful to some of the DC loads.

I'm wondering about installing a high-current zener diode on the output that could "clamp" the short over-voltage spikes to avoid damaging equipment if this happens again (and/or someone turns the battery switch off without first disabling the MPPT controller).

Can you advise me on how to best protect the DC loads from this potential over-voltage condition?

Please note that moving the MPPT controller to the battery side of the switch might be a good solution but physical constraints make it almost impossible to implement.

Suggestions?

Mark

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.