Latching relays for protecting LiFePO4 batteries

[exMaggieDrum]

Originally Posted by senormechanico

I must be a caveman using LiFePo4.
I leave my solar on all the time (440 watts).
I leave the fridge on all the time (insulated with Aerogel).
I leave my ultrasonic barnacle system running all the time (average 800 ma).
I leave the LED backlight for my MPPT controller on all the time (mounted in the companionway steps which makes a great stair nightlight).
I leave indivdual LED cell voltmeters running all the time (under companionway steps adding to stair nightlight and readable from the V berth).
Total draw average? I really don't know exactly. Maybe 1.5 amps or more, but the 200 aH battery bank hits float almost every day. 13.8 bulk, 13.8 absorb time zero, 13.2 float.
CleanPowerAuto BMS just sits there doing nada but insurance only.
This is at the dock, unplugged for months at a time.
When we're cruising, we have lots of electrical stuff to use power including radar. watermaker, wifi, computer, wife's 1650 watt hair dryer, VitaMix (1400 watts for daily smoothies) and an ElectroScan every time we flush.
We never take a shorepower cord, it's just extra weight.

Never a problem for several years.
Are you guys really worried about 5 milliamps?

SenorM,
I actually agree with you on this. I want to use the Blue Sea switch, rather than just an EV200 or equivalent contactor, not because of any millamps saved, but to have the capability to manually switch it and to manually override it at the switch if any of the electronics fail. I expect (or hope rather) that the HP BMS will work flawlessly forever (not) and I will never have an issue related to it.

I do plan (and hope) that my charger settings will stop any overcharging (like you do). One of my criteria is whether the Admiral can understand how to use the dadgum system to have battery power even under emergency situations. Some of the designs don't do that IMHO.

My complexity has been, admittedly, in reaction to all the discussions of how the system can fail and kill the cells. By far the biggest investment I will have is in the cells. The rest of this complexity is fun in a way to put together for now but complexity is terrible on a sailboat especially while cruising offshore which is our intention. I plan to install everything with the possibility to wire around the contactors or use them independent of the BMS and just hope that everything else works as advertised.

You and some others who don't bother with all this and have actual experience with it, and it is working fine, give me a lot of comfort that my investment will work as I need it. So my backup is to just remove (in place) the BMS involvement if need be, such as if it fails and I still need to use the boat (a very nice concept).

A question though: You say your BMS is just insurance. Does that mean you have installed it in its simplest form (relays driving contactors)?

Thanks for your comments. Right now I am trying to be an information sponge before I actually put anything together. I still don't have cells in hand and my recent surgery will delay any installation for at least a couple or more months. And since I like the intellectual exercise it doesn't bother me if I scrap half of my current design before I put something in. At least that is what I have consciously decided to do. I have the time and would rather do this than cut grass. Of course I'd rather be sailing but that has to wait too. I am doing several big boat projects in the background and having fun with those too. I did this before - spend time putting together systems that work without constant twiddling after going offshore. It worked. And the Admiral loved to keep track of the batteries via DC monitors, which will always, always be powered up even when I am not on the boat.

[exMaggieDrum]

Originally Posted by isaacchacon

OK.

I got the circuit.

Sorry that i couldn't make it simpler, but i put together my master design and then i may work on a simpler version that shows only the 7713 part.

this configuration drains 0 amps from the battery when opened. ( ALready tested).

IN my circuit pay attention to the 7713 on the left, and the ELK912 that is closest to the 7713. Also pay attentions to the conexions of those 2 and the involved switches / diodes / fuses.

I am using another ELK912 relay for cascading Over Voltage but that gets out of this topic.


To provide power :

press Normally OPen momentary switch 1 don't release it. you'll hear how the elk912 closes the ground for the 7713.

press Normally OPen momentary switch two. you'll hear a loud blast for the 7713 closing itself.

release the Normally OPen momentary switch 1.

In my case, i have to wait about 5 seconds until the BMS initializes and changes its under voltage from zero volts ( alarm state) to +12 v (Normal state)

ONce the LCD of my BMS powers , then i can release the Normally Open momentary switch 2.



Now, to completely turn off the system, so it doesn't draw any power, i'll just press the Normally Closed Momentary switch 3 and the 7713 opens itself which in turns opens the elk912 .


Hope this explanations helps to understand my already tooo complex circuit.


P.S. when i decided to go lithium I have not a single clue of the COMPLEXITY that it is. It's beeen a rouughhhhh road but i finally got my system 100% working !!!!!!!!!
Attachment 126966

Thanks for sharing your working installation design. Very informative. I'll take a detailed look at how you did it all.

[mbartosch]

I've been following this discussion while I was away working on the boat. Limited time and infrequent Internet access kept me from posting, but better late than never.

Just some random information in case it helps anybody here:

The P Channel MOSFET you've been talking about is essentially an electronic relay which can be set up so you get a switched + 12 V.
Most devices (including the HP BMS) feature N Channel MOSFETs as output relay drivers. These components delver a switched source of ground instead.

Note that MOSFETs are great devices, but if they fail they typically fail with a "closed circuit", this means that they will permanently conduct electricity instead of cutting the circuit. This is different from typical relay failure. MOSFETs come in "electronic component" layout and require soldering in PCBs and dedicated cooling when switching higher loads.

In addition there are solid state relays which essentially use MOSFETs to mimic a normal relay and which come in a form ready for installation in a real world circuit. Crydom produces some nice NON-latching solid state relays which will use virtually no power when activated. SSR also need cooling when using them for higher loads.

Classical latching relays are great devices. I consider using some of them as well in the future (in particular for my portable battery bank). During my research I found two which look quite interesting to me:

http://www.te.com/prodimages/pdf/130-C.pdf

Or this collection: NCR INDUSTRIAL LTD.
In particular the NRL709F looks nice for small load applications such as solar chargers.

[senormechanico]

MOSFET's are indeed no driving current devices, but still drop voltage under load.
In my home I have an Enphase grid tie with a homebrew DC recovery system for grid down events. This includes a couple of 40 amp Crydom SS switches in parallel which divert DC to my LiFePo4's. With full sun and hungry battery, there is noticeable voltage drop.
I will replace them with a EV200 powered by the solar array and "grid loss" switching system.

[exMaggieDrum]

Originally Posted by mbartosch

I've been following this discussion while I was away working on the boat. Limited time and infrequent Internet access kept me from posting, but better late than never.

Just some random information in case it helps anybody here:

The P Channel MOSFET you've been talking about is essentially an electronic relay which can be set up so you get a switched + 12 V.
Most devices (including the HP BMS) feature N Channel MOSFETs as output relay drivers. These components delver a switched source of ground instead.

Note that MOSFETs are great devices, but if they fail they typically fail with a "closed circuit", this means that they will permanently conduct electricity instead of cutting the circuit. This is different from typical relay failure. MOSFETs come in "electronic component" layout and require soldering in PCBs and dedicated cooling when switching higher loads.

In addition there are solid state relays which essentially use MOSFETs to mimic a normal relay and which come in a form ready for installation in a real world circuit. Crydom produces some nice NON-latching solid state relays which will use virtually no power when activated. SSR also need cooling when using them for higher loads.

Classical latching relays are great devices. I consider using some of them as well in the future (in particular for my portable battery bank). During my research I found two which look quite interesting to me:

http://www.te.com/prodimages/pdf/130-C.pdf

Or this collection: NCR INDUSTRIAL LTD.
In particular the NRL709F looks nice for small load applications such as solar chargers.

Thanks once again for sharing technical expertise. I looked at several of these types of relays but had trouble reading the specs. For instance, the Tyco specs said it required "Must operate current" of 1.3A. I assume this is the amps required to switch the relay to a closed, latched state. It did not give the amps required after that but perhaps it is zero so no spec required. I couldn't find that it said that specifically.

One issue I have is that I want the relays I install to default to open, not closed, even though I would like an override if necessary for some. I have purchased an SSR for the AC input to my inverter/charger but I won't be activating it unless I am on shore power so don't care about how many mA's it needs to stay closed.

The main issue with these solid state relays is the low amp rating on them, at least for the load and main bus disconnects. Have you used these for activating your contactors on your system?

Side note: When I was reviewing your excellent schematics, especially the relay circuits, I thought I had found an inconsistency because you had your main bus contactor switched by the relays rather than direct by the BMS. I understand now why you did this.

I have learned more than I ever thought I would about relays with this project.

[mbartosch]

Originally Posted by exMaggieDrum

I looked at several of these types of relays but had trouble reading the specs. For instance, the Tyco specs said it required "Must operate current" of 1.3A. I assume this is the amps required to switch the relay to a closed, latched state. It did not give the amps required after that but perhaps it is zero so no spec required. I couldn't find that it said that specifically.

Yes, I also understand it that way. The relay will activate (i. e. change state) if at least 1.3 A runs through the coil.
It's a coil. Current will run as long as you apply a voltage to the coil terminals. As it's a latching relay, you can stop doing so once the relay has changed state. According to the specs this happens after 50 ms. The specs also says that max. coil pulse time is 100 ms, so the coil of this relay is not safe to operate permanently!

One issue I have is that I want the relays I install to default to open, not closed, even though I would like an override if necessary for some. I have purchased an SSR for the AC input to my inverter/charger but I won't be activating it unless I am on shore power so don't care about how many mA's it needs to stay closed.

You mean default closed (NC), right? NO is the default for most relays.
I'd completely ignore the power a SSR draws. It's not worth the effort IMO.

The main issue with these solid state relays is the low amp rating on them, at least for the load and main bus disconnects. Have you used these for activating your contactors on your system?

No, too expensive and they really need cooling when running higher currents.

Side note: When I was reviewing your excellent schematics, especially the relay circuits, I thought I had found an inconsistency because you had your main bus contactor switched by the relays rather than direct by the BMS. I understand now why you did this.

I have learned more than I ever thought I would about relays with this project.

I'm glad I could help

[exMaggieDrum]

Originally Posted by mbartosch

You mean default closed (NC), right? NO is the default for most relays.
I'd completely ignore the power a SSR draws. It's not worth the effort IMO.

As I said, I am still learning much about relays. But I have found that the automotive little square relays are most commonly NO (called 4-pin or A configuration) and the 5-pin ones, where you can select NO or NC using the extra pin are commonly available too. I don't recall seeing a NC 4-pin relay but that doesn't mean they aren't available as I am sure they are somewhere. I like the little elk912 solid state control relay recommended above because it has both NO and NC capability.

The EV200 relays I believe are NO, as are the Blue Sea ones, including the 7713 which I am going to use (right now).

What I don't like about that is the BMS must be working for the downstream contactors to be on and this means constant coil current unless they are latching types, which may be pulse opened/closed or require constant current. The 7713 is the later but it magnetically "latches" per the vendor but it seems it must really be a two coil device where it only uses "less than" 13mA to keep closed. They market it as magnetic latching though. It will "automatically" open when the coil current is stopped.

I certainly can't guarantee I really understand everything I just said so if anyone sees anything they can restate for me please do. Still learning.

[gpeacock]

So lots of info on connecting a BS 7713 relay to the BMS and the 7700 was mentioned here a few times but nobody has posted a circuit that uses it.
I have a Bluesea 7700 solinoid and a Housepower BMS which has a +,- to power a relay/solinoid.
Using what I have on hand (automotive relays, the 7700, and various resistors and a few 470uf capacitors) I need to set this up so when the +,- circuit on the housepower BMS is pulled to ground, the 7700 gets a momentary + signal and disconnects the house bank.
Anyone able to help? I've tried a few of the options here:
https://www.the12volt.com/relays/relaydiagrams.asp
I know the answer is somewhere on that page but dont really know enough about circuits to pick the correct combination.
Thanks
Gary

[isaacchacon]

Originally Posted by gpeacock

So lots of info on connecting a BS 7713 relay to the BMS and the 7700 was mentioned here a few times but nobody has posted a circuit that uses it.
I have a Bluesea 7700 solinoid and a Housepower BMS which has a +,- to power a relay/solinoid.
Using what I have on hand (automotive relays, the 7700, and various resistors and a few 470uf capacitors) I need to set this up so when the +,- circuit on the housepower BMS is pulled to ground, the 7700 gets a momentary + signal and disconnects the house bank.
Anyone able to help? I've tried a few of the options here:
https://www.the12volt.com/relays/relaydiagrams.asp
I know the answer is somewhere on that page but dont really know enough about circuits to pick the correct combination.
Thanks
Gary

I put it together with a diagram in an RV that i was assembling

My slow but hopefully successful NV conversion

[gpeacock]

Thanks isaacchacon. I just checked out your post on the other forum and although I don't have access to make the pics larger, I understand what you did. I hadn't thought of a flashing relay. I should be able to pick one up here in mexico if need be.
My main problem is that because I have the start bank connected on the load side of the 7700, I can't just do what everyone else seems to do and assume no power gets to the BMS etc after a High or Low voltage event. This is good because it keeps my charging sources happy with fewer relays but bad because even with a flashing relay I'd still be hitting the 7700 orange with + voltage at some regular interval (better than constant voltage with a normal relay I guess).
Anyone know how I could send 1 ping and only one ping to the 7700 with simple parts I can get at an Autozone in Mexico?
Gary

[alctel]

Thanks guys, this is a great thread.

I'm also trying to work out what relays to use with my Orion JR BMS and 8x100 cells in a 2P4S config.

I think i have a lot more options than the guys using the HouseBMS do - almost TOO many options

[alctel]

So it appears the signal wire from my BMS is the same type as the HouseBMS - a pull down type.

I'm curious why people have gone with the 7700 or 7713, rather than the 9012. Is it just because of the manual switch + less power or are there other reasons?

[gpeacock]

Originally Posted by alctel

So it appears the signal wire from my BMS is the same type as the HouseBMS - a pull down type.

I'm curious why people have gone with the 7700 or 7713, rather than the 9012. Is it just because of the manual switch + less power or are there other reasons?

For me, the 7700 is what I had so Im trying to make it work so I dont need to spend any more money. For others they want it to use less power so they use the 7700. If I didn't have one already I'd go with the 7713. Way simpler set up and minimal power use. I do like the idea of being able to manually override if the bms fails which is why I'd choose it over the 9012.

[alctel]

Ok so I went with the 7713 for the LVC and then a pair of RA-700112-DNs for the solar and alt cut off.

To pull up the output from the BMS I am going to use a STP10P6F6 and a resister, same as Stan Honey.

[gpeacock]

So I've worked out the circuit in theory. All I need to find is a time delay off relay now. Anyone know of a simple one that isn't 60 bucks? Less than a 1 second delay is all I'll need.
Once I get one and test it, I'll post the circuit so others dont have to go through the pain I just went through.
Thanks
Gary