Travato LiFePO4 Battery Install

[tbirdman]

Not sure why they would have a temp cutoff for charging. Temp sensor is a sub $1 part. If they do voltage sensing, temperature sensing should be a cake.

[booster]

It would be a low and high temp total cutoff from charging, not a compensation, and I would guess they would also have to have some other controls or restart ability added, also. It appears to be a good idea for lithium.

[tbirdman]

It would be simple to implement using a temp sensor like this and a small 4 bit micro which probably is on the BMS already.

[ptourin]

This is a test post - I just registered and wrote a long intro post - it may have headed into the ozone so I'm trying a small one. Davydd mentioned this thread to me on the LTV Unity MB forum on Sprinter-Source. I've browsed the thread and you're discussing the issues that concern me - and you're a good deal further along with them than I am, so I'm very interested to follow this. More later, if this posts successfully.

[ptourin]

I'm just in the process of doing a li-phosphate installation in my 2012 LTV Unity MB. I installed 2 SmartBattery SB100's 2 weeks ago - a shoehorn act, as they're a bit larger than my original flooded cells, but it worked well mechanically. I installed a Victron BMV-702 monitor last week, and I should have PD9160AL charger tomorrow. I'm right in the middle of trying to figure out several of the issues you're discussing, and so far I haven't been able to get any solid tech info from either SmartBattery or Progressive Dynamics, though I've talked to several people.

Victor has been my only real source of information at SmartBattery - I've found it basically impossible to get connected to any tech support personnel and I've never found their online chat to work. Here are the things he's told me that seem relevant to this thread:

* When I asked for details on the internal BMS, he replied that there is none - they rely entirely on very careful matching of the cells at manufacture. He said that internally there is only over- and under-voltage sensing that opens an internal relay to protect the battery.

* When I asked about chargers, he said the only ones he'd recommend were the PD9100L line, because they'd tested them and knew that they worked well with the batteries. I requested detailed information on battery/charger interaction but didn't get anything useful.

* I asked about alternator charging on the road, and he said that as the Sprinters put out 4.1-4.2v, they felt that I wouldn't go to 100% SOC but that the batteries would be entirely happy charging off the alternator. I'm not far enough along to test this, but when the engine is idling (the battery was at about 13.3v before starting), the monitor voltage reading moved to 13.7v and current moved from a slight drain to about +40A on the monitor. Please take all this with a slight grain of salt, as I'm new to the monitor as well as the batteries, and I don't want to state anything now as if it's solid fact, only to find out that I didn't understand something at the time <g>...

I asked for more detailed tech data on the batteries, but no luck.

I also haven't been able to get any detailed data on the charger. I did talk to a man when I called and requested tech support. He confirmed that the charger is a simple CCCV charger, supplying the rated current until the battery nears full charge, then supplying 14.6v at a current rate approaching zero. I asked what happened as the battery approached 100% SOC, and specifically, whether it had an internal shut-off as the current dropped to a specific setpoint. He said no, the current simply goes to zero.

So I've been very frustrated by the lack of solid info from the 2 sources that logically seem to be the ones to provide it. I'm very interested in the discussion of whether the batteries don't like being held at or near full charge, and whether continuous application of small charging current at 14.6v can damage the battery over longer periods of time. I do observe that the Victron has an internal relay that can close at a specified high voltage and open again at a specified lower voltage - or high and low % SOC, or a combination of the 2. The relay can also be set NO or NC. So it could be used to power a larger relay that breaks the 120v line to the charger when it's fully charged and connects again at some lower point. Not hard to do - I'm just not sure yet whether it's a good idea or just extra complexity.

If anybody on this forum is using a Victron monitor, I'd also be interested to discuss setting it up - I made my best guesses at what would work well, but I'm still at the beginning of this project, and am not sure about my config.

I'll have the charger installed over next week, and starting Aug 13 I'll have 10 days of vacation to watch the system and see how it behaves. It will be a good deal of driving and mostly dry camping. Right after that, my partner Jean will be taking a consultant job in the Boston area, and will be living in the RV until it starts to get cold. She'll have shore power, and I think the safest thing at this early point is for me to put a switch on the 120v supply to the charger and leave it off while she's there - simply top up the batteries and not use them.

I'm looking forward to reading everything in this thread more carefully and seeing where it goes. Sorry for the wordy post - but I thought that since my Sprinter-Source thread had been mentioned I should give you what little info I have so far.

[wincrasher]

Welcome aboard Ptourin! I think you and I are the only trailblazers on this combo. Right now I'm not set up to do any monitoring like you are, so can't help you with the Victron.

I'm not so sure leaving the charger plugged in with the battery at 100% is a great idea. Right now, mine is sitting with the shore power connected, but I've got a 1-2 amps of load going all the time. I'm semi-comfortable leaving it this way. If I store it with no loads, then I'm just going to disconnect the shore cord and hit the disconnect switch.

Are your batteries inside the coach, or are your's mounted underneath and exposed to the cold? In Boston, those freezing nightime temps can creep up on you, so proceed with caution.

[booster]

Hi ptourin--welcome to the group!

Most of the things that you have questioned are right in line with what we have also been questioning and discussing.

Right at the top of my list would be the continuous 14.6 volts on the batteries, even when full. Almost everywhere recommends no float for lithium, and this is higher voltage than that, so IMO it is not the right way to go. PD is a very good company, so I don't understand their position on this. Maybe it was just easy for them to take a power supply and call it a lithium charger. Victron does the charge shutoff function with the relays, as you mention, which would certainly be benefit. I think that is what they recommend to use for a lithium system as I don't think that profile is built into their chargers. I know Magnum has a full cutoff profile that does the same thing in the charger, as well as a few other options. Wincrasher and I disagree on whether a load running would help the situation, with my opinion being that it would have zero effect on the battery, which would not even know the load was there.

If you do the control relay fix, be sure to look carefully at the cut in/out voltages, and if they are adjustable. Where those numbers are have to match the charger and will also determine how short of capacity you could be coming off of charge. One big advantage of the chargers that have that function built in is that you can set the charge voltage, how to end the charge cycle and do shutoff, and set the voltage for a recharge, all run off the built into the system battery monitor.

[ptourin]

Hi - nice to have found this forum and this thread! My batteries are internal, in a compartment under the floor where the top step of the side entrance is. I don't think it's insulated on the outside, but it wouldn't be hard to foam it if that turns out to be necessary. Operating temp is one of several areas that confuse me. SmartBattery's website has a chart that shows capacity vs. temperature, and it shows 80% @ 32F and 70% at -4F. That doesn't jibe with warnings that you can't charge or discharge below freezing. So I need to learn more here.

It appears to me that the internal SB100 over- and under-voltage protection is only disaster coverage, since the set voltages are 15.8 and 8.0. Supposedly I could run down to 8.0 with no cell damage, but I wouldn't want to make a practice of it. I don't think I'd want to run up to 15.8 much either.

So it comes down to using the Victron. It's a new device to me and I haven't yet triggered its internal alarm (flashing LED and buzzer) or the relay. They have separately configurable setpoints - lotsa settings! I don't know how much current the internal relay can handle, but I assume it'll handle the coil current for a larger relay whose contacts can handle 8-9A @120vac - I guess that relay could be either a 12vdc or a 120vac relay, but I need to settle in and read the Victron specs in more detail.

A much more basic concern is how to do the basic configuration for the Victron. There are a lot of settings - these are only a few of the more important. The manual has a section specifically about LiFePO4, and some of these numbers come from that section:

Charge efficiency - I set 99% for LiFePO4, according to their recommendation

Charged voltage - I set 13.3 for a start - Victron says 13.3-13.4 for LiFePO4

Low Voltage for SOC calculations - I set it to 12.8 though I may take it down to 12.5 later

Tail current at full charge - default is 4% of 200AH capacity - didn't change it

Peukert exponent - Victron says 1.10 for lithium batteries, but also 1.05 for LiFePO4 - I'm assuming this isn't a typo, and I set to 1.05

I have next to no real running experience so I don't know yet how this will work. As soon as I entered it all in, I showed around 90% SOC, and it's since dropped down to 85%. All the SOC, voltage, current, accumulated AH readings seemed to make sense. But when I started the engine, I started seeing SOC readings from about 2000 to infinity. I'm not sure whether I've picked bad setup numbers or whether you only care about the monitor readings when you're not driving or plugged into shore power.

So - lots to learn - hopefully some of you have been through this already and can save me from a few pitfalls. I'll be home this evening, should have the charger tomorrow, and will start thinking about installation

[Davydd]

You can discharge (draw power) from lithium ion batteries at below freezing. What you can't do without some damage is put a charge to the batteries when the batteries themselves are below freezing. Your batteries can be above freezing when the air temperature is below freezing. I found there is a considerable lag between air temperature and battery temperature and below freezing weather overnight with above freezing day temperatures probably will not drive your batteries below freezing while in use. The problem of course is you can discharge until fully discharged but then have no way to recharge your batteries. Chances are you will not be using your RV when it is too cold to charge in other than the most extreme circumstances. I doubt you would be doing that in a Unity. They aren't designed for cold weather use other than the occasional dip below freezing overnight.

[pattonsr]

I just read this thread and noticed at post 51 that Booster mentioned Magnum and their CCCV option. CCCV is just two stage charging and one could always make the Magnum do two stage charging but one has to study the manual to set it up.

All Magnum has done by offering CCCV is to make it easy to configure. There is no new charging logic. Lithium batteries are almost fully charged by stage one and one need only set a short time for stage two.

Barry

[booster]

Originally Posted by pattonsr

I just read this thread and noticed at post 51 that Booster mentioned Magnum and their CCCV option. CCCV is just two stage charging and one could always make the Magnum do two stage charging but one has to study the manual to set it up.

All Magnum has done by offering CCCV is to make it easy to configure. There is no new charging logic. Lithium batteries are almost fully charged by stage one and one need only set a short time for stage two.

Barry

IMO, the bigger picture with the Magnum and lithium is that not only does the magnum eliminate the float stage (CC/CV) charging, it also give you several options of how the charger operates after the two stages, like monitoring and back to float, or back to charge cycle. Put on top of that the option of using the BMK monitor to use return amps or SOC to transition to the full shutoff mode and monitor, and you get the opportunity to stop charging reliably before totally full, which could be very handy going into storage, or if you don't need all the capacity.

I think Magnum has always had most of these features available, but I don't know that for sure, as I just recently looked closely into it while working on our settings for AGM. To me, it was surprising to see that you could take all these features and combine them to get a really quite good lithium charger, especially when compared to nearly all the ones who call themselves lithium chargers.

[ptourin]

Interesting that they'll discharge safely below freezing but not recharge. You're right, we're very unlikely to be traveling when both day and night are below freezing. It would be tricky if we were caught in really cold weather - I may set up the charger so I can manually switch it on and off, but I can't switch off alternator charging while underway - at least, I don't think I can - I believe the alternator line hits the battery at a junction in the battery box before the power disconnect switch. Easy enough to find out - I've just never checked before.

[pattonsr]

booster,

I am not sure the Magnum BMK works properly with Lithium. The last time I spoke with them they told me not to use the BMK for SOC because they were not sure if it worked properly with Lithium. Things could be different today so I would talk with them and not assume it is ok.

I would just use a short time for stage two. You will end up at around 95% full. Lithium does not need to be fully charged like AGMs.

The use of a shunt to figure out SOC gets interesting when a second alternator is used. I believe ARV developed a custom solution to determine SOC since off the shelf solutions from Magnum, Outback, or Trimetric did not give them what they needed. But I may be wrong about this.

I've always worried that extensive driving starting out with a full-up large Lithium bank could be an overcharging nightmare. It would appear that ARV has a sophisticated solution that works since DavyDD has traveled extensively with no problems.

[Davydd]

I don't know what ARV does but watching the Silverleaf screen readout the charge amp input dies down to nothing when the batteries are fully charged. That situation works with solar, alternator or shore power. How that is controlled I am not sure. Silverleaf in my understanding is just a reporter.

[booster]

I actually have had several long conversations with Magnum about the BMK and how it works in combine with ARC50. I have a couple of other threads here on some of it.

I can understand why they would say that the SOC transition setting, and the SOC readout would not be accurate for lithium, mainly because it is also not accurate or good at getting other battery full either. The downfall of the SOC setting is that to confirm full charge they use a fixed 2% of bank size for return amps being met. My understanding (that I have heard) is that lithium would be way lower than that. Lifeline specs .5% for their AGM, by comparison, so it is also pretty useless for them, also. It is also more important for AGM. Because of that 2% the charge will stop early and reset the meter to 100%. I think the charge efficiency setting in the SOC settings only goes to 99%, and lithium is said to be higher than that, so that also could mess it up. That said, SOC after CC/CV would safely charge lithium, but just leave them a touch short of full, which probably isn't all bad.

We ran into trouble because I want to use ending amps only to transition out of absorption. They have that setting, but it has a glitch in that the charger does not go into a full charge cycle every time you plug in. It will go directly to float if the voltage is up at about 12.7v or higher, which is very common after driving (surface charge) or from solar. You could easily wind up in float when the batteries weren't really full. We will need to use the Bulk force feature to get around it, each time we plug in if the voltage is up.

The good thing for the CC/CV users would be that the charger does go into a full charge cycle at every plug in, which is the way it should be.

Personally, and I am no lithium expert by any stretch, I think the best way to charge lithium with the Magnum/BMK/ARC50 would be to use CC/CV with the voltage set at what the battery mfg said, and using return amps, set to whatever you wanted based on how close to full you wanted to get, to end absorption. The shutoff stage would be set to monitor, and rebulk, at whatever voltage you would want to be the lowest to have when plugged in (and how low you might be when you unplug). From what I have seen of the other chargers for lithium, this would come much closer to the charge profiles that I have seen from the battery guys, than the lithium specific ones.

Again, personally, I would not use the SOC transition or meter reading for anything. We won't be with the AGM system when we get it in. It is much easier, at least for me, to see how many AH we are down and go from there. We know how much the solar will be able to in various conditions, we will know how much and how quick the big alternator will put back in, so knowing what capability we have left is pretty easy.

pattonsr's concern about overcharge off the alternator is very valid, and one of the huge things that I think gets overlooked in nearly all systems, not just lithium. Accurate charge control that gets the batteries full (non lithium) without overcharging (all) is essential for long battery life and having maximum capacity. We will have the typical 3 different charging systems, shore charger, solar, alternator. Here is how they will each work to give accurate charge control.

Shore charger--Magnum MS2000 that will transition out of absorption based on pretested return amps of the BMK shunt.

Solar controller--Blue Sky 2512 that will also transition out of absorption based on return amps off the shunt.

Van alternator--DC Power Eng 250 amp will run at about 14.4v minus whatever the line losses to the batteries will be (will be tested after system installed). We will have a cockpit mounted inductive ammeter at the batteries to be able to see the amps to the batteries. Our separator is a pure manual Blue Sea, so I will be able to just shut off the alternator charging to the batteries when the amps read the return amps that would be the same as for both charger settings.

This setup SHOULD be able to assure we can get 100% from any combination of charging systems without overcharging, as all the sources will exclusively be controlled by the actual return amps to the batteries. If all works out as planned, it will be very accurate, trouble free due to very few components, and easy to use in real life--time will tell. I have all the preliminary testing of the Magnum and batteries done, and most of the battery mounting hardware fabricated, so we should be set to install when we get home from our fall trip.

[ptourin]

Booster, would you explain a bit about your SOC readout paragraph? I just took a peek - the Victron monitor defaults to 4% tail current, but it's configurable - and I don't know what it should be set at. I have charge efficiency set to 99%, which was Victron's recommendation for LiFePO4's - and the highest that I can set it. I'd be happy to understand how the tail current setting works. Victron didn't have any recommendation for LiFePO4's.

[booster]

Information on things like the right settings for tail current (return amps) are very hard to find or understand, like a lot of stuff is, for lithium. The good news is that it appears to be really non-critical for lithium because they don't have get full to be kept in good condition. This charge profile I posted earlier showed 3% of "rated current", whatever that is.

I have seen other numbers that vary all over the place, from very low to almost 10%. Lithium is still in it's infancy, for sure.

Davydd has stated that ARV setup his system to stop early at the top and bottom, leaving some cushion from getting all the way full, and most battery mfgs do say to not go to full all the time, or store there, so it makes sense. Setting the tail current higher would just leave more room for cushion on the top end. It would also cost you capacity, though.

The setting of the tail current will really only mess up the SOC reading in relation to the true "totally full" condition. For lead acid we all want/need to get to totally full, so it is a problem for those batteries. With lithium, all it is going to do is give you a false 100% full reading that is really something less than 100% depending on the tail current that is sent. Again, it will cost you some capacity, so if that is an issue, you may need to determine exactly what your tail current is, and set very close to it.

Almost all the lithium manufacturers say not to totally discharge their batteries, and most say not to always go full, with Smartbattery seemingly being the exception. I don't know why that is.

[ptourin]

The project is creeping along. PD sent me the wrong charger; finally got the correct one today. I found that there's Victron tech support available from a company in NH - so I'm going to write up my plan and send it to them, then call and see what I can learn about config settings for the monitor. I found out that the monitor relay contacts can only handle 1A @ 60V max - was hoping to use them to switch a 120vac relay coil, but now it'll be a 12vdc coil - I'll use that to switch the PD charger. More later, as this progresses - I'm gone this weekend, so it all has to go together next week, as we leave with the RV on Thursday night of that week.

[wincrasher]

Just an update. I've been away 2 weeks. During that time, I've left the shore power connected with the refrigerator on 110v and the Fantastik fan ventilator running. So the charger has been running at 14.6 volts constantly this whole time, I assume putting out around 1-2 amp to cover the load of the fan and the CO detector.

When I shut everything down and disconnect the shore power, I see 14.3 volts on the guage. Before, when I'd charge over night, it would be 13.4 volts when I disconnected.

So what is this telling me? It is above the curve from SmartBattery for 100% full.

[ptourin]

When you disconnected shore power, how long did you wait before measuring the voltage? Since the charger was pulling the battery up towards 14.6 while it was running, I'd expect the battery voltage to measure something in the 14's if you measured immediately after disconnecting. But if you wait a few hours and measure again, I'd expect that it would be down near 13.4 or so.

So - what do you actually see? I'm just guessing, without any on-the-road experience yet. But I'm wondering if you're seeing the LiFePO4 equivalent of surface charge and whether it'll dissipate over an hour or so. I don't know if this exists for li-phosphate batteries, and if so, what it's called - I'm a newbie here. Interested to hear.

My PD charger is going in over the next few days, and I have a 1-1/2 week trip starting next Thursday night, so I'll soon have some observations from that trip.