How are Volta units doing over time?

[booster]

Lithium systems are proliferating very quickly and products are, hopefully, maturing with the longer real world feedback that is out there by now. Recommendations are still all over the map for how to use and store them, but do seem to be drifting to more conservative recommendations of lower charge rates, SOC usable limits (or not), predicted life, etc. Prices have dropped so much that some of the life issues are nearly gone from being a major issue if shorter than the original claims as they are approaching AGM pricing.
Pretty much all the information we see here is in relation to the LiFePo4 chemistry batteries as that is what the industry has settled on, probably based on safety and long life.
There is still one early player in the high end systems around that is using the automotive style Lithium ion batteries, Volta. They are large, integrated, drop in systems so only Volta's responsibility so no extraneous multiple vendor stuff involved.
We have known for a long time the expected life of the lithium ion batteries in phones is rated at under 1000 cycles by some amount. 700 cycles appears to be commonly given. AFAIK the auto makers claim more often in years of life, I think. If Volta is aging like the phone batteries do and has the 1000 cycle life prediction, it would be lower than high end AGM batteries like Lifeline, at least at 50% discharge. Many of the LiFePo4 batteries used to claim 5000+ cycles but currently we see some in the 2000 if a large range of SOC is used so getting closer to the Volta style rating.

We have heard of some well designed early systems that are getting weak and at 80% original capacity at about year 8, which would require about 250 relatively large charge cycles per year to get to 2000, which is way more than many of us do in real use, I think. Most systems don't keep track of charge of cycles though (Volta may) so we don't know how many cycles they had on them.
The RV sites that I see, mainly here, don't really have any Volta users so hard to determine what is going on with them over time.

I did just look at the warranty on Volta currently as that can be a indicator of life sometimes.
The current Volta warranty is 36 months, 36K miles, or 1000 cycles, whichever comes first, which is really quite low, I think. The miles and months would come up very quickly for a reasonably large number of new van buyers who do a bunch of traveling in the very common touring style of near daily driving with short stops along the way. 120 days a year at 100 miles average per day would get you there. Even for us, who like to stay put in one place more than many folks, getting to near 1000 miles a week is not hard at all. 12 weeks a year of travel would get you to the 36K miles in three years.

The guess is that the Voltas are more likely to last to number of cycle claim of 1000 cycles (of unknown depth though so if shallow charging counts it could be not so good), but once it is 3 years old you have no warranty.

Lots of stuff has warranties that short and we don't worry about as we build that in to the decision making before buying, but replacing a Volta battery is probably quite expensive compared to the more common RV LiFePo4 batteries, might put it in a more risky category. Lithium batteries in RVs are touted for very long life, so many shoppers would probably be surpised.

Of interest to me in general on this topic is the last few years some automakers have switched at least some of the electric vehicles to LiFePo4 if what I have read is correct.
I am not saying Volta is a bad system, and it is very possible the integrated system is able to give as long or longer life in an RV type use, but with what I have been able to find I have no real idea of actual. When they came out they were far ahead of the pack in how they did it with the high voltage and fully integrated setup and that has not changed.
The fact that we haven't heard of major problems or recalls, is encouraging.

[Boxster1971]

I have the Volta 13,000 Whr Flex-pack battery in my 2024 Airstream Interstate. Bought it last September so haven’t had it long enough to comment on long term performance. I also have the Volta fully integrated Power Distribution Hub (PDH) that has been problematic. Mostly because only Volta can work on the components inside the PDH, like the inverter/converter, 48v to 12v converter, solar controller and alternator control unit. Many owners have complained about the PDH, especially the inverter on the Airstream forum. Volta claims they had many problems getting proper components for the PDH due to COVID induced supply chain issues.

I had to have the PDH on my van replaced in January by Volta because the charging from second alternator didn’t work. I doubt it ever worked from Airstream as the battery has to be at specific conditions to enable alternator charging, above 40F, below 85% SOC and engine above idle at cruising speed. It was a frustrating experience because I had to take it to dealer, who could only do a remote connection to Volta to troubleshoot the issue. Then wait two weeks for Volta to ship a replacement PDH to the Airstream dealer. They first wanted me to leave my van at dealer while they removed the PDH, shipped it to Volta in Michigan for repair and return. I said no way as everything else was working and I planned to travel to Florida for a winter break.

I’ve not seen any reports of problems with the Flex-Pack battery.

[booster]

Originally Posted by Boxster1971

I have the Volta 13,000 Whr Flex-pack battery in my 2024 Airstream Interstate. Bought it last September so haven’t had it long enough to comment on long term performance. I also have the Volta fully integrated Power Distribution Hub (PDH) that has been problematic. Mostly because only Volta can work on the components inside the PDH, like the inverter/converter, 48v to 12v converter, solar controller and alternator control unit. Many owners have complained about the PDH, especially the inverter on the Airstream forum. Volta claims they had many problems getting proper components for the PDH due to COVID induced supply chain issues.

I had to have the PDH on my van replaced in January by Volta because the charging from second alternator didn’t work. I doubt it ever worked from Airstream as the battery has to be at specific conditions to enable alternator charging, above 40F, below 85% SOC and engine above idle at cruising speed. It was a frustrating experience because I had to take it to dealer, who could only do a remote connection to Volta to troubleshoot the issue. Then wait two weeks for Volta to ship a replacement PDH to the Airstream dealer. They first wanted me to leave my van at dealer while they removed the PDH, shipped it to Volta in Michigan for repair and return. I said no way as everything else was working and I planned to travel to Florida for a winter break.

I’ve not seen any reports of problems with the Flex-Pack battery.

Thanks Boxster, great data point.



I hope they got it all straightened out this time.


I would imagine that the high end Victron integrated systems would be similar, although they probably allow and have more certified installers around the world so may not be factory contact only.



Does the Volta you have keep track of the history of charge cycles? It would be nice if it had both number of cycles and also the depth of discharge, especially if there were ever a warranty issue. It is the kind of information that none of us have on any system unless we keep meticulous manual records. Some solar controllers give total lifetime charging, if not reset along the way, and I think maybe our Magnum can do the same. But no place does it for the composite of all charging sources.

[avanti]

Originally Posted by booster

I would imagine that the high end Victron integrated systems would be similar, although they probably allow and have more certified installers around the world so may not be factory contact only.

I don't (yet) have very much direct experience, but by all appearances, Victron is much more open than the competition. You can even run their (open source) monitoring software on your own Raspberry Pi instead of their hardware if you like. Their (free) Victron-Connect web service also keeps a lot of history automatically and presents it with pretty graphs.

Direct access to their engineers is kept pretty tight, but there is a huge mutual-support community there when you need it.

Does the Volta you have keep track of the history of charge cycles? It would be nice if it had both number of cycles and also the depth of discharge, especially if there were ever a warranty issue. It is the kind of information that none of us have on any system unless we keep meticulous manual records.

Well, even a very basic Home Assistant setup makes monitoring that kind of thing really trivial.

[booster]

Originally Posted by avanti

I don't (yet) have very much direct experience, but by all appearances, Victron is much more open than the competition. You can even run their (open source) monitoring software on your own Raspberry Pi instead of their hardware if you like. Their (free) Victron-Connect web service also keeps a lot of history automatically and presents it with pretty graphs.

Direct access to their engineers is kept pretty tight, but there is a huge mutual-support community there when you need it.

Well, even a very basic Home Assistant setup makes monitoring that kind of thing really trivial.

Do those monitoring keep a charge cycle by cycle history of depth, or even better a continuous SOC graph which would be slick if it was easy to have it sort out things like DOD range total or such.


A very big issue for us is how to keep accurate track of the actual SOC as voltage is such an inaccurate method and parasitic makes the monitor not very accurate too. Any idea how Victron is handling that?



I have been more and more interested in that kind of history as I have played with our setup (much less sophisticated than yours or a Volta). What I am running into is just how much the parasitic losses in the batteries themselves (internal BMS with bluetooth times 3) messes up the battery monitor SOC over time (even relatively short times). Putting a number to the parasitic is kind of a moving target as it changes with whether or not the BT is on or asleep or in storage mode so is far from being the same all the time like you would figure in for lead acid batteries.


For those users that are OK with taking the charge cycles to totally full every time or fairly often, it matters less because you can get a consistent reset point for the monitor when you are actively using the batteries. If it is more sporadic use it gets harder. It is also a bummer that the Victron monitor does not appear to have a way to program in a parasitic loss like some monitors do.


I had thought I could achieve a good and consistent calibration point in the 85% SOC range by using voltage and tail current control to go to that point, but although it worked well in the constant temp of the shop, when I let it cool down to 45*F it did not give the same results. It appears that the tail amps, when in midrange, change a lot for any given SOC in relation to temperature.


I am currently testing to see if voltage a fixed time does a more consistent reference point for meter resets.


I was told by Current Connect, who sold me the batteries, that the BT used only .06amps, but that must be in sleep mode as I am sure I have a lot more than that. Worst is that the BT appears to go full time active whenever the batteries are discharging, so in use that is pretty much all the time. It would be nice to be able to turn it off and on manually as it is not needed probably 98% of the time.

[avanti]

Originally Posted by booster

Do those monitoring keep a charge cycle by cycle history of depth, or even better a continuous SOC graph which would be slick if it was easy to have it sort out things like DOD range total or such.

The Victron Cerbo exposes many dozens of parameters, certainly battery voltage and SOC to any software that would like to monitor or record it. Happens automatically in Victron Connect, and is very easy in HA.

I haven't really paid much attention to Victron Connect, but the following was readily available:

Untitled 5.jpg

HA has very powerful data recording and visualization features, and automatically records pretty much everything that it gets. By default, it saves everything for something like two weeks, but you can set a given parameter to anything you like. I haven't done anything with SOC yet, but mine is currently set to one week history. This isn't very interesting, since the van is plugged in and SOC hovers around 100%, but it makes the point:

Untitled 6.jpg

You can script any kind of summary statistics you like, driven by these and/or other data.

[booster]

Originally Posted by avanti

The Victron Cerbo exposes many dozens of parameters, certainly battery voltage and SOC to any software that would like to monitor or record it. Happens automatically in Victron Connect, and is very easy in HA.

I haven't really paid much attention to Victron Connect, but the following was readily available:

Attachment 14693

HA has very powerful data recording and visualization features, and automatically records pretty much everything that it gets. By default, it saves everything for something like two weeks, but you can set a given parameter to an anything you like. I haven't done anything with SOC yet, but mine is currently set to one week history. This isn't very interesting, since the van is plugged in and SOC hovers around 100%, but it makes the point:

Attachment 14694

You can script any kind of summary statistics you like, driven by these and or other data.

Very interesting indeed, and especially to me the holding at 100% full all the time as that has been more and more being defined as not a good idea. Maybe Victron knows it is not as big a deal as others are moving to. Kind of similar to wider temp range that Victron says is OK.



Of course both could also be a case of necessity in the real world for usability and keeping track of SOC accurately.


Do you know how your setup is determining the actual SOC?

[avanti]

Originally Posted by booster

Very interesting indeed, and especially to me the holding at 100% full all the time as that has been more and more being defined as not a good idea. Maybe Victron knows it is not as big a deal as others are moving to. Kind of similar to wider temp range that Victron says is OK.

As far as I know (which isn't a lot, yet) Victron doesn't have a built-in feature to keep SOC below 100%. I think a lot of people do it manually during storage, which is easy enough. My intention is to have a "storage" switch which HA will use to change the behavior. Just haven't gotten to it yet. In the mean time, I am not worrying about it between trips.

Of course both could also be a case of necessity in the real world for usability and keeping track of SOC accurately.


Do you know how your setup is determining the actual SOC?

Not in any detail. There is an all-knowing external BMS that communicates with both the Multiplus and the solar charger.

[booster]

Originally Posted by avanti

As far as I know (which isn't a lot, yet) Victron doesn't have a built-in feature to keep SOC below 100%. I think a lot of people do it manually during storage, which is easy enough. My intention is to have a "storage" switch which HA will use to change the behavior. Just haven't gotten to it yet. In the mean time, I am not worrying about it between trips.


Not in any detail. There is an all-knowing external BMS that communicates with both the Multiplus and the solar charger.

I assume the BMS does not have all the current going through it and controls external relays?


I also assume it probably has access to a shunt somewhere in the system.

[avanti]

Originally Posted by booster

I assume the BMS does not have all the current going through it and controls external relays?

I don't think that is true. My understanding is that the Lynx Smart BMS has a built-in shunt and does have all the current from all batteries passing through it.
It is used along with one or more "Lynx Distributors", which is a "smart" fused busbar capable of reporting things like fuse status to the BMS.

The BMS also has contacts that are capable of shutting down charge sources. However, I think this is intended as a backup--the main charge control is done via communication with the various chargers.

The batteries themselves also appear to have data connections to the BMS, but I am not sure what they do.

https://www.victronenergy.com/upload...art-BMS-EN.pdf

I also assume it probably has access to a shunt somewhere in the system.

You can add additional communicating shunts if you want to monitor things like the chassis battery, etc. But, a lot of load information comes directly from the Multiplus charger.

[booster]

Originally Posted by avanti

I don't think that is true. My understanding is that the Lynx Smart BMS has a built-in shunt and does have all the current from all batteries passing through it.
It is used along with one or more "Lynx Distributors", which is a "smart" fused busbar capable of reporting things like fuse status to the BMS.

The BMS also has contacts that are capable of shutting down charge sources. However, I think this is intended as a backup--the main charge control is done via communication with the various chargers.

The batteries themselves also appear to have data connections to the BMS, but I am not sure what they do.

https://www.victronenergy.com/upload...art-BMS-EN.pdf

You can add additional communicating shunts if you want to monitor things like the chassis battery, etc. But, a lot of load information comes directly from the Multiplus charger.

I was thinking the shunt might be in the Cerbo or standalone, but in the same place as the control makes sense. I tend to think, probably, based on the smaller system setups that all the current runs through the circuit board that is the BMS. I think the Lynx is in a box so they would be able to have a big enough shunt that way and also have the benefit of being close to BMS.


If the multiplus is measuring loads, it may also have a shunt of some time, or if they consider it not needing to be extremely accurate a inductive loop current measuring setup like some other chargers use. Our Magnum is very inaccurate for current when checking it from the internal charger output compared to the monitor shunt reading.


It is going to be very interesting as you get deeper into it all and learn more about how they do things.

[avanti]

Originally Posted by booster

I was thinking the shunt might be in the Cerbo

The Cerbo is just a nicely-packaged compute device (similar to a RPI) with a bunch of communications ports (Ethernet, WiFi, CANbus...) and a few general purpose relays. It also has ports for an optional HDMI touch screen. As I mentioned, its software can be run on a Pi.

[Boxster1971]

Originally Posted by booster

Thanks Boxster, great data point. . . .
. . . Does the Volta you have keep track of the history of charge cycles? It would be nice if it had both number of cycles and also the depth of discharge, especially if there were ever a warranty issue. It is the kind of information that none of us have on any system unless we keep meticulous manual records. Some solar controllers give total lifetime charging, if not reset along the way, and I think maybe our Magnum can do the same. But no place does it for the composite of all charging sources.

I think Volta keeps track of the history and charge cycles, but it is not available to users that I have determined yet. I’ve seen comments from Volta that they have the data if they think you are abusing the battery for warranty purposes.

[Boxster1971]

BTW - Volta uses top quality components from Victron, Dimension/Magnun, and APS. They just have then setup for their proprietary integration.

[booster]

Originally Posted by Boxster1971

I think Volta keeps track of the history and charge cycles, but it is not available to users that I have determined yet. I’ve seen comments from Volta that they have the data if they think you are abusing the battery for warranty purposes.

I think you are correct on this, and it was my first impression when I read how detailed their warranty was stated as cycles, miles, and time. I wonder if they are also tied into the vehicle history also to get the miles independently of the odometer. With the huge alternator load they can generate, I think it certainly could be because they would probably need to reduce load when the engine was cold, at low rpm, or getting hot.

Do they define to users what "abuse" would be in things like DOD or charge and discharge rates?

[avanti]

I just noticed that the Lynx smart BMS entry in the Cerbo menu has a HISTORY item, which displays the following parameters:

Total charge cycles
Number of full discharges
Cumulative Ah drawn
Minimum voltage
Maximum voltage
Time since last full charge
Synchronization count
Discharged energy
Charged energy

You can reset these statistics at any time.

I'm assume that all these statistics are also available to Home Assistant for long-term logging and further analysis.

[booster]

Originally Posted by avanti

I just noticed that the Lynx smart BMS entry in the Cerbo menu has a HISTORY item, which displays the following parameters:

Total charge cycles
Number of full discharges
Cumulative Ah drawn
Minimum voltage
Maximum voltage
Time since last full charge
Synchronization count
Discharged energy
Charged energy

You can reset these statistics at any time.

I'm assume that all these statistics are also available to Home Assistant for long-term logging and further analysis.

Thanks Avanti, great information.

It does give some insight as to some things I was wondering about.

IMO, cumulative AHs drawn is big one for overall life considerations, especially when combined with number of total discharges.

Number of synchronizations confirms they need and use them in their system control, it appears. It would be interesting to see how they trigger them, though.

Charged energy vs discharge energy is probably used to help with the accuracy on non synchronized recharges and is basically a charge efficiency. Depending on deep they dive into that, they could adjust the CE by depth of discharge even.

The charge efficiency thing is interesting to me in that would be the way for me to know what it really in our system. With relatively high and somewhat variable parasitic we have it is a moving target at best and the Victron monitor quits counting AH on recharge once it gets to 100% that is charge efficiency modified so I don't know if it returns more or how much if it does on synchronization charge cycle. Our Magnum can be set to automatically accumulate and adjust charge efficiency by some kind of past cycles averaging of out/in amps so I hope the Victron can do that also so it could be used in our midrange of SOC cycling method.

It must take a lot of training to be fully up to speed on the Victron systems and fully knowledgeable in all the capabilities. I wonder if most of the installers are that far in to it or the Victron engineers are doing a lot of the setup as the defaults. The Wakespeed regulator we have can coordinate with a Victron serbo and various BMS units and I was astounded at how many settings were needed to do that both in the Wakespeed and Victron stuff. An entire high end integrated system must be daunting to understand and program.


On edit..


I just looked at the user manual to see the available information in the history in the Victron monitor. It is essentially the same as what Avanti listed so they much do the same kind of tracking in both. No automatic CE available, though, so I will have took and the energy in and energy out history at some synchronization cycles to get a decent average for when we are actually camping and accuracy is more important. At home I have the Magnum shore charger set to do a mid SOC charge and then shut off and go into the cyclical recharge mode as need based on a recharge voltage so no problem with killing the batteries at home.

[avanti]

Originally Posted by booster

Number of synchronizations confirms they need and use them in their system control, it appears. It would be interesting to see how they trigger them, though.

From Victron:

Automatic synchronisation

Synchronisation is an automatic process and will occur when the battery has been fully charged. The battery monitor will look at a few parameters to ascertain that the battery has been fully charged. It will consider the battery to be fully charged when the voltage has reached a certain value and the current has dropped below a certain value for a certain amount of time.

These parameters are called:

Charged voltage - the float voltage of the battery charger.
Tail current - a percentage of the battery capacity.
Charged detection time - the time in minutes.
As soon as these 3 parameters have been met, the battery monitor will set the state of charge value to 100%, thus synchronising the state of charge.

Example:

In the case of a 12V battery, the battery monitor will reset the battery’s state of charge to 100% when all these parameters have been met:

The voltage exceeds 13.2V,
the charge current is less than 4.0% of the total battery capacity (e.g. 8A for a 200Ah battery) and,
3 minutes have passed while both the voltage and current conditions are met.

[booster]

Originally Posted by avanti

From Victron:

Just like our monitor does.


A couple of issues we ran across when trying to set an 85% SOC synchronization point was the limitation of the tail current, which can't be set higher than 10% of rated battery capacity and that makes is so we have to reduce the charge voltage to keep from getting higher than 85% if we want to charge at 13.8v to get good charging speed.


The other thing is that the resolution of the "charged" voltage is only to .1v settable. What I have found is that the Magnum, solar, Wakespeed and Victron all treat that setting differently. They might mean 13.8v starts at 13.75v and goes to 13.85v or 13.0v to 13.9v depending on which one it is and on lithium .5v is a large difference in the charge you get at the same tail amps. With the Victron at least you have centralized control so all of them should do the same thing which would be really nice.


It is also possible you might run into the charging of the alternator stopping before you are totally full, depending on how low the tail current is set. We ran into a conflict that way with 10% tail current of 61 amps at 13.7v charge setting. When we got to idle speed the alternator output would fall below the 61 amps long before the running down the road speed would, leaving the SOC lower than we would get at speed. This is why we switched most of the charging except the synchronization to SOC from the Victron instead of using the Wakespeed using tail current. SOC control only looks at the charge efficiency modified energy (maybe just AH) to decide when to stop the charging.


We have found midrange charging to be much more difficult to setup than charging to 100% full all the time, which is pretty easy by comparison.

[avanti]

Originally Posted by booster

It is also possible you might run into the charging of the alternator stopping before you are totally full, depending on how low the tail current is set. We ran into a conflict that way with 10% tail current of 61 amps at 13.7v charge setting. When we got to idle speed the alternator output would fall below the 61 amps long before the running down the road speed would, leaving the SOC lower than we would get at speed. This is why we switched most of the charging except the synchronization to SOC from the Victron instead of using the Wakespeed using tail current. SOC control only looks at the charge efficiency modified energy (maybe just AH) to decide when to stop the charging.

Well, remember that we have a DC-AC-DC setup, so the Multiplus thinks it is on shore power (I have the AC input limited to 13A, which is how you control the draw from the engine alternator. This is conservative, but good enough for now). Also, with the Transit's dual alternator, I get full power from the alternator even at idle. This turns out to be the main virtue of the Transit dual alternator scheme.

We have found midrange charging to be much more difficult to setup than charging to 100% full all the time, which is pretty easy by comparison.

I am thinking that with Home Assistant controlling things, I will just write a little script that will monitor SOC and simply turn off charging at the setpoint, and turn it back on below some hysteresis point. Maybe it isn't that simple, but seems like a good place to start.