My Victron MPPT reports higher battery voltage and goes into Absorption and Float prematurely.

Joined
Jun 24, 2022
Messages
7
Hi all!

I am new to the forum, so thanks for having me :)

I am experiencing an interesting issue on my off grid boat setup. I use a Victron MPPT charger for my main 24V solar array (2 x 170 watts). I also have a second charge controller with a portable 24v solar array ( 2 x 50watts) that I only use when needed.

I have a DIY 12 V 280AH Lifepo4 battery pack.

Attached a description of my setup:
Electric Diagram.jpg

Problem:
**********
Since a few weeks ago, I noticed that my solar input (Victron) is giving less than half of the power it normally does (normally I get next to 340w in optimal conditions). At the moment I am only getting around 100w to 120w in similar weather conditions. When looking closely, I noticed that even though my battery voltage is around 13.4V, the Victron MPPT sees 14.6V so it goes into float directly which seams to be reducing the output (due to the small to none voltage delta I am guessing).

1656073840414.png
In this picture you can see that my Victron shunt reports a voltage of 13.41V, while my MPPT shows the following:

1656073925766.png
As soon as I restart the device, it starts to work correctly for a few minutes, giving next to 200 watts, but then it comes back to this state again and stays there.

I've seen some posts that talk about a fuse on the output which isolates the MPPT from the battery so the voltage goes up to max. But In this case, how can I be getting any charge at all? the battery is being charged.. just slower.

I looked for errors, but it shows 0 errors:
1656074145607.png

I thought my shunt maybe was giving the wrong information, so I replaced it with a brand new one, but no change.

Anybody has any ideas? should I assume a faulty MPPT controller? I tried going to default settings already but did not help.

Thanks for any help!

Ignacio.
 
Last edited:

sunshine_eggo

Happy Breffast!
Joined
Oct 26, 2021
Messages
4,655
Location
USA
First, you need to validate these readings with a separate voltmeter.

Second, you need to check each and every connection in the system. Insure everything is torqued properly including cell interconnects.

With voltmeter:
Check battery terminal voltage. Should equal shunt voltage
Check voltage at MPPT battery terminals. Should equal MPPT reported voltage.

The MPPT will naturally have a higher voltage during charge due to the fact it is measuring voltage responding to the charge current. If this is excessive, it is because there is a loose connection or the MPPT voltage reading is incorrect.

If you can confirm the shunt voltage is accurate, you can enable DVCC on the GX and identify the smart shunt as SVS and SCS, i.e., the MPPT can use the shunt voltage and shunt current values instead of its own for better accuracy.
 

HRTKD

Boondocker
Joined
Apr 24, 2020
Messages
7,933
Location
Somewhere South of Denver
If you can confirm the shunt voltage is accurate, you can enable DVCC on the GX and identify the smart shunt as SVS and SCS, i.e., the MPPT can use the shunt voltage and shunt current values instead of its own for better accuracy.

I don't have any experience with the DVCC setting other than what I've read in the manual. In the case of the OP, I don't think DVCC is appropriate as a DIY battery is in use, not a CANBus battery. The documentation is a bit vague but DVCC looks like a cool feature.
 

740GLE

Solar Addict
Joined
Apr 30, 2021
Messages
1,138
Do you have any pictures of connections or the entire system?

Your one line shows no over current protection items, fuses breakers?

A cheap DC breaker can have substantial voltage drops.
 

rmaddy

Full-time Solar-powered Trailer Life
Joined
Nov 16, 2019
Messages
3,735
Location
USA
Your two charge controllers need to have their battery negative connections going to the shunt just like all of the other negative battery connections. Your diagram shows the battery negative connections of the charge controllers going to ground instead.
 

sunshine_eggo

Happy Breffast!
Joined
Oct 26, 2021
Messages
4,655
Location
USA
I don't have any experience with the DVCC setting other than what I've read in the manual. In the case of the OP, I don't think DVCC is appropriate as a DIY battery is in use, not a CANBus battery. The documentation is a bit vague but DVCC looks like a cool feature.

DVCC is using the GX to manage things. Rather than use a VE.Smart bluetooth network, which doesn't permit sharing between incompatible devices (GX, inverters, VRM, etc.), DVCC with the GX shares data and manages across all GX connected devices.

I've used it for over a year with lead-acid batteries (clearly no CANBus). If you want to limit charge current, but not limit MPPT output, DVCC is what you have to use. If you want to share BMV-700/702 data with other equipment, DVCC. Example, my 250/100 and array can output about 50A. That would damage my batteries as they are limited to 39A per Trojan; however, my batteries are further deteriorated, so I limit it to 15A. DVCC allows me to specify a max charge current (from all GX connected sources, including Quattro) of 15A. My only other option without DVCC is to manually specify maximum charge currents without any coordination. Setting the MPPT to 15A also limits charging + loads to 15A, and I'm going to want a lot more than that.
 

HRTKD

Boondocker
Joined
Apr 24, 2020
Messages
7,933
Location
Somewhere South of Denver
The Victron documentation was tripping me up. It implied that LiFePO4 without CANBus wasn't a good fit. But FLA has no CANBus (like you said), so how can it not be an issue there? <- rhetorical question there.

I'm not having charge issues, so my current mode is, "If isn't broke, don't fix it." If I had all my charge sources active at the same time, I could get about 200 amps, which is right about .5C for my 560 Ah LiFePO4 battery bank.
 

sunshine_eggo

Happy Breffast!
Joined
Oct 26, 2021
Messages
4,655
Location
USA
The Victron documentation was tripping me up. It implied that LiFePO4 without CANBus wasn't a good fit. But FLA has no CANBus (like you said), so how can it not be an issue there? <- rhetorical question there.

I'm not having charge issues, so my current mode is, "If isn't broke, don't fix it." If I had all my charge sources active at the same time, I could get about 200 amps, which is right about .5C for my 560 Ah LiFePO4 battery bank.

Once I get the NMC built (it's like fusion energy at this point, always X years away), DVCC won't be needed for charge management as my battery will be good for about 1600A, but I'll still use it for sharing BMV-702 current, voltage and temp sensing with the system.
 
Joined
Jun 24, 2022
Messages
7
Hi all,

Wow! that is a lot of information very quick!! thanks so much! I'll take these actions and come back with additional data.

Also I should have clarified (and maybe added it to my diagram) that I am running the GX software on a raspberry-pi so the shunt and mppt are connected via usb interface for VE.Bus. Not sure if this will still allow me to run the features described.

Your two charge controllers need to have their battery negative connections going to the shunt just like all of the other negative battery connections. Your diagram shows the battery negative connections of the charge controllers going to ground instead.
Hi there! that was just to simplify the diagram. All negatives in the system actually go via the shunt to the battery. Thanks for the feedback!
 
Joined
Jun 24, 2022
Messages
7
Since we've highjacked the thread, I'll just add that the Victron landscape is incredible. The integration and sophistication makes it great to work with.
I couldn't agree more! I discovered the victron world recently for my new boat and I am so happy with it! I cannot recommend it enough.

And I love that they allow you to run their official software on a ras-pi. Great consumer friendly policy
 

sunshine_eggo

Happy Breffast!
Joined
Oct 26, 2021
Messages
4,655
Location
USA
Hi all,

Wow! that is a lot of information very quick!! thanks so much! I'll take these actions and come back with additional data.

Also I should have clarified (and maybe added it to my diagram) that I am running the GX software on a raspberry-pi so the shunt and mppt are connected via usb interface for VE.Bus. Not sure if this will still allow me to run the features described.

Using VE.Direct to USB adapters or VE.Bus to USB adaptors? If so, then it's no different than VE.Direct/VE.Bus to GX. You're connected to the GX (Rpi) appropriately.
 
Joined
Jun 24, 2022
Messages
7
Do you have any pictures of connections or the entire system?

Your one line shows no over current protection items, fuses breakers?

A cheap DC breaker can have substantial voltage drops.
I do have some mid priced fuses in line.. will check! thanks for the feedback!
 
Joined
Jun 24, 2022
Messages
7
HI there! quick update.. I enabled DVCC and indeed now the MPPT gets the voltage from the Shunt and applies the correct charging profile.

I still need to verify all my physical connections and fuses for any high resistance or loose connections but in terms of settings, it is looking good now.

Thanks for the help!
 
Joined
Jun 24, 2022
Messages
7
I think the mystery has been resolved.. the positive cable coming out of the MPPT was getting very hot so it was increasing the resistance.

Here is the offendfer.
WhatsApp Image 2022-06-25 at 2.39.57 PM.jpeg
Now I have replaced this part with a properly sized one and disabled the DVCC feature (for testing). Now the voltages are almost identical again.

Thanks all for the good feedback! Very interesting to see DVCC feature in action. I might leave it permanently on. I was using the BT networking feature before.
 

sunshine_eggo

Happy Breffast!
Joined
Oct 26, 2021
Messages
4,655
Location
USA
Glad you got it sorted. Connection issues are the #1 cause of voltage drop.

The GX is the only way to manage both your MPPT and inverter charging simultaneously. I don't believe there's any way to manage the Orion as it doesn't have a VE.Direct port.

And the EPEVER is always a wildcard. :)
 

HRTKD

Boondocker
Joined
Apr 24, 2020
Messages
7,933
Location
Somewhere South of Denver
I suggest that there is a lesson here: Enabling DVCC may make things work, but it may not be the panacea that you think it is. If the system is having problems, DVCC could very well mask those problems and lead to some interesting failures later.
 
Joined
Jun 24, 2022
Messages
7
I suggest that there is a lesson here: Enabling DVCC may make things work, but it may not be the panacea that you think it is. If the system is having problems, DVCC could very well mask those problems and lead to some interesting failures later.
In this case that is very true.. I would have never found out in time and have a thermal runaway event.. that would be a very bad day..

Being able to see the voltage delta is actually an interesting metric for resistance on the circuit.
 

sunshine_eggo

Happy Breffast!
Joined
Oct 26, 2021
Messages
4,655
Location
USA
IMHO, the lesson is, "do this stuff right."

It's my understanding that it's not just DVCC but any VE.Smart network that's feeding the MPPT voltage/current/temperature sensing will create the same situation. I have a SBS feeding my MPPT voltage and temperature data, and I KNOW that the MPPT temperature and voltage data reported by VC are identical to what the SBS is reporting even with many Amps inbound.

The question becomes, "do you want to use open circuit voltage or do you want current influenced voltage for operation?"

If the latter, then there's no point in having a VE.Smart network at all.

IMHO, better to engage in best practices and ensure proper wire gauge and connections are present rather than disable useful functions.
 

HRTKD

Boondocker
Joined
Apr 24, 2020
Messages
7,933
Location
Somewhere South of Denver
IMHO, the lesson is, "do this stuff right."

It's my understanding that it's not just DVCC but any VE.Smart network that's feeding the MPPT voltage/current/temperature sensing will create the same situation. I have a SBS feeding my MPPT voltage and temperature data, and I KNOW that the MPPT temperature and voltage data reported by VC are identical to what the SBS is reporting even with many Amps inbound.

The question becomes, "do you want to use open circuit voltage or do you want current influenced voltage for operation?"

If the latter, then there's no point in having a VE.Smart network at all.

IMHO, better to engage in best practices and ensure proper wire gauge and connections are present rather than disable useful functions.

I absolutely agree. I just wouldn't want someone working around an actual problem by enabling DVCC.
 
Top