Monday, September 5, 2016

Lithium Battery Presentation

We've been happy enough with our Lithium Iron Phosphate (LiFePO4) battery installation that I volunteered to do a presentation at the Annapolis Sail & Power Squadron meeting, Sept 1, 2016. The powerpoint slides may (or may not) be self-explanatory. If you're interested, a PDF of the presentation can be found here:

I later did a revised version of the talk to the Marine Trawler Owner's Association. This version contains a few more details. It is 29.2MB.
Lithium Batteries as a Marine House Bank



Derek Alldritt said...

Terry, thanks for the this extensive write up. A couple of questions as we undertake a Lithium retrofit on our L46 cat.

One -- the RPi, what are you using to monitor the batteries with this device?
Two -- I didn't quite understand how you wired the disconnects for LVW/HVE's and it looked like a battery combiner in the photos, so can you add some insight to that? Are you using the BlueSea disconnect to disconnect battery from house on each event, or did you break out into load/charge bus bars?

Thanks in advance,

Terry said...

Sorry for the long delay in responding - I just saw your comment and don't know why I missed it.

First, the RPi is reading the values transmitted by the Victron BMV-702 battery monitor. The BMV transmits values every second and I interfaced its serial output to the RPi's serial input. I use a switch on the RPi to tell it whether the serial interface should be used to read voltages or if it should be a console interface for loading new software, etc.

Second, I am not using load/charge bus bars. The low voltage/high voltage (LV/HV) events on the HousePower BMS drop the high current relay, disconnecting the batteries from the system. There are separate Blue Sea Systems ACRs that connect the alternator output to the house+starting batteries. I have the alternator output going to the house bank first, and the ACR connects the engine batteries when the voltage gets high enough to trigger them. I use a diode in the ground circuit of the ACRs to raise their voltages by .6v, making them activate at 13.6v instead of 13.0v. Without the diode, the ACRs would stay energized almost all the time, because the house bank is at 13.2-13.3v most of the time.

On my list of things to do is to add a separate relay to disconnect the alternator drive should a LV/HV event occur.

Does that help?

Derek Alldritt said...

Definitely helps, thanks. We got our install done in December, and it's really working well for us. We went with LiFeBlue batteries and they contain their own shunt/disconnect and that really helped solve a lot of intermediary problems building our own disconnect would have raised. They also have a BLE interface a phone or tablet can connect to and view SOC, Amps, Voltage, Cycles, etc. I'd like to take both and find a way to aggregate via a RPi but that's on the list, nothing urgent.

Unique solution with the ACR's...I was wondering how and ACR/AVR without specific settings would be made to connect at a higher voltage, as we typically sit about 13.1-13.2v as you do, so that's a great solution.

Thanks for sharing. Such great technology, and I think in the coming years there are going to be a LOT of people upgrading to this solution. I have a whole Total Cost of Ownership presentation together. It's remarkable the savings alone with LiFePO4 just from reduced engine/generator hour running. The solution can basically pay for itself in under a year for most people is my overall conclusion!

Terry said...

Yes, it is great technology. The LiFePO4 technology is so much safer than the lithium batteries that have had fire problems. However, the higher current carrying capability means that we have to check that the connections throughout the discharge and charge paths don't build up too much heat (i.e. have high resistance). For example, if we run the genset at the same time as the engines are running, our batteries can easily accept 140A. I feel safer charging at less than 100A. And the alternators need temperature sensors so that they don't overheat and burn out.

Is your presentation on TCO online and could you leave a link to it?
Take care!

Derek Alldritt said...

We have a 150A charger and can pump 80A solar into the batteries - haven't found any heating problems. The batteries do start to get warm with ~230A going into them, and the charger needs supplemental cooling as it sits in a closed compartment, but the wiring specifically doesn't get overly warm. We ran everything as 4/0 from the bus bar to battery and only a few feet. Our charger is 120VAC so the existing 10/3 wiring feed was sufficient to support this for our short runs - less than an hour. I do have the option to back off the charger and have done it before just for lowered abuse on the batteries.

Haven't solved the alternator problem yet in that ours will keep putting out power even when batteries are full, however, they never push beyond 13.5v which means really all they're doing is floating so I haven't been overly concerned. I do need to solve the problem and disconnect somehow eventually. Our 80A Hitachi's backoff on their own with temperature, so I've also not worried about that as after about 30 minutes they're only putting out 30A at best.

Presentation isn't online as of yet. I need to get it there.