question

Troy avatar image

Why doesn't the documentation warn that multiple MPPT controllers don't play well together and is there a fix in the works?

I've got a 12v 320Ah Narada Lead Carbon battery bank, a Multiplus 1200/12/50 and 3 SmartSolar MPPT 75/15's connected to 3 separate solar arrays.

The problem I'm having is the MPPT controllers tripping over each other going into float WAY too early. So on the daily chart I'll have one controller yielding as much as 1300Wh while the other two will yield as little as 200Wh each. At the end of the day the battery is not fully charged either.

The dominant controller can be any of the controllers on any given day. The wiring and settings are the same for all three controllers. I guess it just a software race condition which one detects the 1A float first. As soon as is does that 1A load goes to the remain controllers.

Is it possible to have them share better. Does the BMV or Smart Sense Synchronise them at all. As it is they are operating at about 30% efficiency once two drop out.

Seems that only one MPPT controller can be the only charge source in the circuit. Otherwise once any one hits the 1A float threshold it go to sleep and doesn't wake up again. Leaving just one controller to do all the work and the battery no where near charged.

The situation is exacerbated as the Muliplus charger starts up when the generator is fired up. All the MPPT go to sleep because the generator is providing all the charge current. Then when it is turned off often only one charge controller comes back to charge and the other two stay on float even though the battery might only be 30% charged because the float voltage being reached does not wake them all up.

These MPPT charge controllers do not play at all well with other charge sources. Including their own kind. Is there any way to have them act as a simple power supply? That is just providing a voltage to battery and allowing the battery to simply absorb what it can? Or maybe get rid of the 1A threshold or have a timer that checks periodically if the battery can now accept greater than 1A and go back to absorption.

Basically modify the algorithm somehow so they're get back to work when they can. At the moment it's like having an employee that gets things done in the morning then goes to sleep and doesn't bother checking if more work has arrived because the other bloke is doing it all by himself and struggling at it as a result.

Seems theses things are too smart for their own goodand as a result are constantly tripping over each other. After which they don't get up and back to work again for the rest of the day! If it was an employee you'd fire them for not being team players! Lol

Is it possible for me to just move the charge leads from the battery connector to the output connector? If the MPPT can output 15A at say the float setpoint voltage to the output without a battery in the battery connection then that would be better charge controller than all the current "smarts" before the battery connection terminals!

Any solution would be appreciated.

Cheers,

Troy

MPPT - Solar Charge Controller
10 |3000 characters needed characters left characters exceeded

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

7 Answers
mvader (Victron Energy) avatar image
mvader (Victron Energy) answered ·

Hey @Troy;


Thanks for the detailed post. That was quite helpful for R&D.


And yes; there is a fix under way; actually two:

1) we’ll soon release an update that adds parallel charging: via bluetooth the chargers sync their operation: all together in bulk, absorption and float. Which will fix the issue you’re seeing. This update has been all implemented and is being tested right now.

Also, though you might not need it with the parallel charging feature in place; we’re working out to add a fixed absorption time; ie. having a user option to disable todays automatic reduction of the time based on start voltage.

I’d like to say more, @both, but can’t yet; please stay tuned.


11 comments Share
10 |3000 characters needed characters left characters exceeded

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

@mvader (Victron Energy Staff) Without the update, would the use of DVCC and SVS and the creation of a Bluetooth network between MPPTs including a Smart Battery Sense alleviate this situation without the firmware update?

Hi; that implies you have a Venus-device installed; right?


In that case; having ESS; which puts the Multi or Quattro in charge; would help for sure.


Having voltage sense wouldn’t help much I’d think. Still one or two of them would switch two float quickly once (after bulk) one of the other takes the grunt of the load. Test to make sure..


Anyway; real solution is paralleling and some other things we’re discussing (internally for now). I’d prefer to focus on that; hope you understand.

That is great news. Excellent. What I also think would be good is for there to be another condition test added to come back out of float. I understand the existing one is voltage dropping back below float voltage. I think a timer needs to be added to periodically check if the 1 amp tail condition is still valid and get back to charging.

It is inadequate to just do that check once and be done. As another charge source may come on for a while like a multiple Multiplus charger the MPPT current drops to tail trigger which is saying absolutely nothing about battery SoC.

As it is the MPPT algorithm makes these charge controllers incompatible with all the other charge controllers in your product line.

A simple one-off tail current check is not useful in any multiple charge source circuit as far as I can tell.

A periodic tail current check should fix all these problem right? As it will have another go at charging until the tail current is reached again. This could happen many times a day.

BTW. I think Victron have got great products and as a former R&D manager for a software development team I certainly appreciate the hidden complexity in product integration.

I'm glad my post was useful for your R&D staff. I remembered well how valuable feedback can be from switched on end users.

Keep up the good work. :-)

Hey @mvader (Victron Energy Staff),

Its been about a month since we were teased with an update that's in the works.

I don't fully understand Victron's release cycle to know about how long a bugfix/update turn around time usually takes.

Is there any new status on the parallel charging feature you can share?

Thanks,

Nate

Hi Nate, trust me; its being worked on real hard.

The time it takes to get something like this out depends on the situation; ie. how many other things are already at a stage that they can be released; do we push that all aside and delay it; or not; how involved is the fix; and so forth. In this case it was more than just fixing one obvious thing; so took long. Longer than I’d have prefered. I’m getting a bit impatient too.

the good news is that I expect a test version to be available by next week/the week after.

Lawrence@sunz.co.nz avatar image Lawrence@sunz.co.nz mvader (Victron Energy) ♦♦ ·

Is this firmware update availible yet? I have identical situation that would be good to fix

neoneddy avatar image neoneddy Lawrence@sunz.co.nz ·

I'm sure it's being worked on and tested. Better to have it work right than play fast loose and release patches.


I've not yet seen Victron make empty promises. I am involved in software development as well and it's a constant struggle between Urgency and importance.

Dark Horse avatar image Dark Horse mvader (Victron Energy) ♦♦ ·

@mvader

Any news on when this will be fixed. I see that Victron pulled the software patch 1.41 for these units. It would be nice to know that there is a fix coming soon. I am pretty sure I would not have just bought 2 100|50 Smart Solar if I knew they would not work together. I have not installed yet and I am thinking of taking them back. My thoughts are this should be the very highest priority second to none. Having invested quite a bit of $ in your controllers, cabling and solar panels it really sucks to think that they could only use 1/2 of my investment.

@mvader (Victron Energy Staff)
Do you know if 2.30~35: https://community.victronenergy.com/questions/11876/venus-os-v23035-available-for-testing.html has this fix to parallel charging via bluetooth where the chargers sync their operation, to test with?

Hi @Natebert. No That doesn’t have the fix. The fix is part of an mppt firmware update.


mvader (Victron Energy) avatar image
mvader (Victron Energy) answered ·

Hi all, an update:

we are about to release v1.42 for the mppts. If all goes well, next week.

Note that it does not have synchronised charging yet; that will come quickly there after. It does change the lithium default to a fixed absorption time.

What systems would benefit from synchronised charging?

- systems with an ess assistant do not: the mppts are controlled by the Inverter/charger

- systems with an intelligent can-bus connected lithium battery, such as BYD or Pylontech, don’t either. With those batteries there is no bulk, abs, float. Instead, the battery is controlling the show.

- systems with a more basic lithium battery, such as our own ones with the VE.Bus BMS, don’t need synchronised charging either; because of the coming fixed absorption time.


So what rests are systems with lead batteries. most cases it will work OK without synchronising the chargers. During bulk, the chargers will always share the load without any issue. With or without synchronised charging.

Once the battery is mostly charged, and absorption starts, then it will still work well without in most situations. But in some one of the chargers might end up doing a too short bulk time, letting the others do it. Such as @Troy depicted in the question above.

And for that situation we have been working very hard on synchronised charging.”, and as part of that having the set of chargers working with one tail current, instead of each on its own.

Its finished; but needs more testing and also some more changes in VictronConnect, which is why first v1.42 is coming without it (but with many many other improvements).

I do apologise that all this is taking much longer than I anticipated. We underestimated it.

8 comments Share
10 |3000 characters needed characters left characters exceeded

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

This is great news and looking forward to the synchronised update.

Looking forward to it. Here in the Caribbean our 660 watt panels are not keeping up with our 12V of 200ah daily usage unless I manually restart the controllers every time they enter float. We have Firefly AGMs running the manufacturers recommended charge parameters.

Cheers, RickG

@mvader


Thanks for looking at this and trying to fix the problem. I have many hats and one of them is software developer so I know it can be tough.

I had a flood on my boat last year and had to choose a new system as the entire battery management system was destroyed. I am glad I choose victron.


Cheers,


Looking forward to the updates.

It seems v1.42 has now been released as promised. Just waiting for some sun..

OffgridSA avatar image OffgridSA mvader (Victron Energy) ♦♦ ·

Will there be a firmware update for the MPPTs - BlueSolar VE.Can 150-70 and 150-85.

As they also need the same improvments so we can set fixed absorb times along with with other improvments .

Hi OffgridSA, maybe, maybe not. We don't know yet.

Any update on this fix of the MPPT controllers being able to synchronize with each other? I was on the verge of replacing my whole blue sky system with Victron, my current system of 3 MPPT controllers is daisy chained and they plan just fine with each other.

StraightUp avatar image
StraightUp answered ·

Premature skips to float time seems to be a hot issue presently, even for people with only one controller:

https://community.victronenergy.com/questions/7809/why-was-absorption-only-45-minutes.html

https://community.victronenergy.com/questions/1856/how-to-remove-smartsolar-automatic-absorption-time.html

But that aside for now - logically, here's how I would expect things to proceed in this order, with multiple controllers:

- All controllers start in bulk, providing max amps for their given panels.
- Voltage rises accordingly, but so long as it's still below the Absorption voltage for all controllers, all should continue to stay in bulk until around 80% SoC.
- Voltage hits Absorb volts, controllers detect it more or less at the same time, and switch to absorb mode.
-Voltage holds @ absorb V, current starts to taper across all units and batteries demand for amps lessens.
-As the battery fills, depending on individual wiring/connection characteristics, first two, then one unit ends up able to supply all the current the battery can absorb at the Absorb voltage. The controllers with the most resistance between them and the battery sequentially drop below the 1A threshold and drops to float (OR exceed their timeout - see below).
-Eventually the battery fills and the final controller drops to float.

Sounds like this isn't what you're seeing, though.

So, to figure out just exactly why your controllers are tripping we may need some more information. Specifically when each controller is going into what stage? Eg are they all at least hopping from bulk to absorb at the same time (roughly)? Is one already in float befoe the others have left bulk? And most importantly - do they all turn on at the same time?

On that last one, I bet they don't.


The critical factor here, I suspect, is the same one being discussed in those other threads: the detected battery voltage at "sun-up" determines length of the absorb phase.

Each MPPT unit uses the below table to calculate the standard absorption cutoff time. It does this calculation, supposedly, based on battery voltage measurements, at the time when it first detects the solar panel volts are 5V higher than the battery ie first decent sun on the panels.





Here's the thing: if you have 3 different controllers on 3 different PV arrays, I'd bet you'll find that one array is gaining volts just ahead of the other, turning its controller on and starting bulk charging before the other two. This will in turn bump up the battery voltage, meaning when the other two turn on, they see a higher batt V, assume the battery is more 'charged' and thus auto-select a shorter absorption time.

I suggest you watch your controllers and voltage at next sunrise, and see if one is turning on before the others. Even if it's just a few seconds, that could be all it needs to bump the volts and fool the other controllers.

If this is the case, as far as a fix goes....I'm honestly not sure. Others seem to be struggling with this too, namely the MPPT units don't seem to have any function to simply set a fixed, single absorption time.


websiteabs.png (152.6 KiB)
5 comments Share
10 |3000 characters needed characters left characters exceeded

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

Thanks @StraightUp. That was a very helpful post. I've done a bit of digging into the lead carbon batteries I've got. At 12.6v they are only about 70% charged.

So I'm going to max out the absorbtion time. That should help a bit. Well done. :-)

@mvader I think you will be seeing more advanced lead carbon batteries coming into service because of their sulfacation resistance and improved cycle life. Victron even have them in the product line now. They can have different SoC voltage levels to other lead acid cells.

Is there any way to set in the controller the SoC low and high voltages? The high needs to be 13v for these batteries for the absorption multiplier to be calculated correctly at start up.


Hi Troy, its not settable right now; I'll take the comment home though; thanks!

vic_user_2837489 avatar image vic_user_2837489 mvader (Victron Energy) ♦♦ ·

As good as your products are, I am really dissapointed about the lack of multiple MPPT synchronization. Particularily due to the lack of documentation on this topic. Just browsing the forum helped me understand that a FW update is needed.

I just hope the 3 classic blue mppt 100/50, VEdirect connected to my Venus are supported by that announced FW update too - and not just some smart solar ones over bluetooth?

Are you able to confirm?

Looking forward to the release

Agree that it would be excellent to get similar solution to VE.direct devices connected to for example CCGX. I personally have Easysolar and classic 100/30 MPPT, neither of which has bluetooth. Hopefully this is technically feasible.

Yes I would like to know the same. Fixed absorb time update and multiple mppt working as a team. Firmware update is it for all mppt units or just latest Bluetooth models?

Djingis avatar image
Djingis answered ·

This question is drawing some attention on different Facebook groups discussing issues with multiple MPPT:s connected in parallel.


Can @mvader (Victron Energy Staff) confirm that this is solved for the SmartSolar MPPT:s using the synchronized charging via Bluetooth in the v1.47 firmware?

8 comments Share
10 |3000 characters needed characters left characters exceeded

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

I’m still waiting for this promised update which really should have been integrated before the product was launched and sold.

It’s been many months since v1.47 was released without the Sync feature, and at that time we were given this statement below:

“And to let you know, there is one more significant improvement coming, Synchronised Charging for the SmartSolar range, over Bluetooth. Its implementation has been finished, and is currently being extensively tested”

@solgato, you must be mistaken - SmartSolar FW version 1.47 hasn't even been released to VictronConnect yet -it's only currently available through the VRM or the direct FW downloads on Victron Professional- so it certainly wasn't released months ago without the sync feature.

Solgato avatar image Solgato Justin Cook - Bay Marine Supply USA ♦ ·

Sorry, I meant Firmware v1.42/1.43 which I believe is up to 1.46 now. Was released back In Sept and was supposed include Bluetooth Synchronization for MPPT chargers, but when released the statement I posted above was made saying that it was not included and that it was still being worked on.

The statement was made on the Firmware & Software page under release notes about Victron Connect 5.9 features and improvements.

The Change Log from what I saw doesn’t mention or address the feature.

Maybe I missed something?

@solgato, no, you probably didn't miss anything, it's just the "is currently being extensively tested" bit took longer than expected; my equipment was in on the beta testing of it and it took many many iterations to get it just right. Now, VictronConnect is up to 5.16 and sync'd charging still isn't there, of course, but since SS FW 1.47 is available on VRM and Professional, I'd expect the new firmware to be part of VC 5.17.

@Djingis 1.47 contains the SmartSolar bluetooth-synchronized charging feature (all devices must be on a shared VE.Smart Network); though 1.47 hasn't been released to VictronConnect yet, you can update your connected devices through VRM or by loading in the firmware update manually through a wired connection.

wolfandy avatar image wolfandy Justin Cook - Bay Marine Supply USA ♦ ·

Is it possible to directly link to SmartSolar MPPTs on a VE.Smart Network for synchronized charging? Or do I need a BMV or Smart Battery Sense in the mix?

@wolfandy, to the best of my knowledge you will still require either a Smart BatterySense or a BMV-712 to successfully set up a VE.Smart Network between SCC's, but I will test that on my bench later today and -if I discover differently- I will report back here. If I don't report back, the result is that the VE.Smart networking requirement of a sending unit in the mix (the BMV-712 or SBS) is still in place.

wolfandy avatar image wolfandy Justin Cook - Bay Marine Supply USA ♦ ·

Thanks a lot, @Justin Cook. I appreciate it!

Waterskier1 avatar image
Waterskier1 answered ·

I have updated both of my MPPT 100/50 to Ver 1.46. I set the Absorption time, under Advanced, to 6 hours. I'm still having problems with one or the other dropping back to float much sooner. Here is a screen shot from VRM showing one controller still in Bulk, current rising, and the other dropping to zero in Float. Both controllers have the same length of cable to the batteries, which are 4) Lifeline AGM. Is there something I'm missing. I just added the second set of panels and controller to increase charging time, and it's not really happening.


mppt-delta.jpg (75.8 KiB)
6 comments Share
10 |3000 characters needed characters left characters exceeded

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

Has the lazy controller actually switched over to float?

It does not look like it has, as at 13:15 a cloud floated over your system an the lazy controller woke up and started to contribute again.

Yes, it had switched to Float. Here are two more pic's, each with the voltage/current and MPPT State for each controller.

mppt-258.jpg (129.1 KiB)
mppt-256.jpg (114.0 KiB)

HI Waterskier,

That is interesting. It does look like the lazy controller has a lower absorb voltage setting to the other controller. Could you please check that both controllers have the same settings?

Are they the same model controller? Just asking as I wonder if the lazy controller wakes up a little later than the other.

Do you have a bmv?

Both are Victron identical (except for year) MPPT 100/50 SmartSolar charge controllers. I did set the "lazy" controller 50mV higher on Absorption, in an attempt to keep in in Absorb longer. I can't perceive any difference though.

I've got two concerns: 1) Why did the "lazy" controller go into Absorption so soon, and 2) Why did the "lazy" controller go into Float when the other controller is still in Bulk - it hadn't even reached Absorption and the "lazy" is in float!

I have set both controller to Fixed Absorption time of 6-hours in the Advanced settings. Neither should drop out of Absorption for 6-hours unless the tail current drops below 1.5 Amps (even tried dropping the "lazy" controller tail current down to 0.5 Amps with no change).

I do have a BMV 700 and a CCGX. I don't think, but am not sure, that any of the BMV settings effect the MPPT controllers, only my MultiPlus - I may be wrong.

The above charts are with the system running at minimum load on Solar only (no Shore Power or Generator).

Any questions, comments or suggestions are quite welcome.

You should set the absorption voltage the same for both controllers.

What I am seeing is that once the lazy controller goes onto absorb at 14.75v, it will reduce its output current. Then the other controller catches up, it pushes the battery voltage up to 14.8v, and the lazy controller has nothing to do, other than go to float.

I can reset the lazy controller back 50mV, but that is where it was and there is no change in operation. The first controller (set at 14.4 V absorption) and the lazy controller (set at 14.45 V absorption). I set it slightly higher in an attempt to keep it from going into absorption prior to the first controller.

It is not clear at all why the lazy goes into absorption while the other is still in Bulk, and stays in bulk for over an hour more. Shouldn't both go into Absorption at approximately the same time/voltage? Please note that at this time the battery SOC is still below 85% Both controllers should be getting the same voltage/current and other data via the VE.Bus, so the difference isn't due to monitoring circuitry differences. I understand that there are component tolerances, but I wouldn't think they would be this much.

Also, why is the lazy controller current dropping off while at the same time the other controller is still in bulk and the current is rising? I would expect the current graphs to be similar to each other. Note that the first controller has about 900 watt rated panels while the lazy controller has only 360 Watts rated, so I wouldn't expect the currents to be the same, but would expect the curves to be somewhat similar.

Thinking out loud, why would the lazy controller start limiting current when the other controller is seeing the same voltage, and is running at max current? There clearly is something different between the controller, but their input should be exactly the same.

I'm also confused as to why the voltage rises so early in the charge cycle. I know it's a function of battery voltage, but I'm surprised that battery voltage is 14.5 volts when the SOC is below 80% and I'm not charging the 600 AH bank that fast - no where near 0.5 C.

Jadedgate avatar image
Jadedgate answered ·

I have just installed (4) 100/30 mppt and (4) 150/35 mppt each attached to their own individual solar panel. I have updated the firmware on each mppt and it says 1.39. I do not see where to download v1.42 to solve the issues with synchronization. Is this now fixed?

Share
10 |3000 characters needed characters left characters exceeded

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

Dark Horse avatar image
Dark Horse answered ·

What state are we at with this right now. I have just bought 2 Smart Solar MPPT100|50 charge controllers to carry 2 280 watt panels each.


Will these not work together to charge my batteries properly? Am I going to see them fighting with each other but not doing the job they are supposed to do?


When I look at this web page it seems it was possible back in 2013 using the 150/70 charge controller.

https://www.victronenergy.com/blog/2013/11/15/synchronizing-multiple-mppt-15070-charge-controllers-2/

I have Color Controller GX.

3 comments Share
10 |3000 characters needed characters left characters exceeded

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

Really hoping to have an answer for this soon as well. Holding back on adding to the rest of my system until this is sorted.
I also have a CCGX.

I run a pair of 150/100 MPPTs with 12 x 180w panels (6 in series per each MPPT), and they have been working great together on the latest firmware, etc.

Below is a screen shot from VRM of a couple days just last month and their performance together.


capture.png (243.5 KiB)

Thanks Natebert.