Using the BMS to replace the 1-2-Both switch

The friendliest place on the web for anyone who enjoys boating.
If you have answers, please help by responding to the unanswered posts.
Joined
Jan 7, 2020
Messages
1,638
Location
Paris
Vessel Name
CHiTON
Vessel Make
Tung Hwa Clipper 30
I've asked several times for an explanation of "why keep lead" when one has a bank or two of LFP. Although I haven't gotten an answer, in some posts there is some assertion that LFP can't do it. The electrons from LFP are the wrong shape, color, scent, or something. My understanding is you just need enough electrons to the starter. A single LFP might not be enough. A bank can be.

If one keeps lead for starting, the standard setup is to have the ability to connect the house LFP batteries (using the Both setting) if the lead is dead (or too weak). I'm suspicious of that because the voltage of the LFP bank is likely going to be much higher than the lead starters. The LFP may start the engine despite attempting to rescusitate the 11.5V lead batteries at the same time. Is there really any reason to use the Both setting in this case?

Despite having a bank of start/house lead batteries for years (as did the prior owner), and now a bank of start/house LFP batteries for a year, I can see that having some kind of a "reserve" start battery might be handy. I have a lithium jumper on board, but I think there might be something much simpler. I can use the various BMSs as my 1-2-Both-Off switch. A state of the art replacement.

The "Off" position is simple. I can use Bluetooth to turn of any or all batteries. Same with setting 1-2-3-4-5 . . . using BlueTooth. But what I wanted to check out is whether the BMS allowed by to create the "safety" aspect of keeping start batteries in reserve. Turns out it is simple with my JBD BMSs and I suspect most BMSs.

My undervoltage cutoff was set by EcoWorthy at 10V and "awakens" at 10.4V. 10V to save damage to the battery, but I want a bigger margine, and I wouldn't mind having a "reserve" automatically set for my LFP. Even a reserve large enough to start my engine. It might allow my to use most of my amp hours (which are way more than a lead start battery) and still have enough to not need lead start batteries.

So I tried setting my undervoltage cutoff really high on a battery in my LFP bank. I turned off the other batteries and the charger. When the house loads dropped the voltage below the cutoff point, the battery shut down. When I BlueToothed the undervoltage cutoff to a new lower setting, the battery was back on. No need to "wake" the battery with a charge current. I simply set that battery (or all batteries) to cutoff at a voltage that is high enough to start my engine. I get the benefit of all that additional amperage in the LFP, plus keep a reserve to start the engine and recharge.

Also interesting was that I could also "wake" the shutdown battery by connecting it with one of the other batteries. Since they were at a higher voltage than the reawaken voltage set in my BMS, the undervoltage battery woke up (and was then charged by the other batteries.

The possible variations are complex. I could set battery #1 undervoltage cutoff at 13.1V. I could set batteries #2-6 to shutdown at 12.6V. If I ever used up all those amp hours without looking at the shunt, say, dancing in the moonlight on the flying bridge and the music suddenly stops, I could BlueTooth battery #1, turn it on by lowering the undervoltage cutoff parameter, and when it came back on line it would awaken batteries #2-6. I now would have the necessary MCA to start my engine and recharge the bank. I would get to use about 90% of a large LFP bank and always still have enough to start (without lead start batteries).

Need to play around with this a little more
 
I have separate hose and start banks. My system is set up to be fail safe if I do nothing. Should either bank experience a total failure I have a way of running everything on either bank and in an extreme emergency I can combine both banks.

I also have separate thruster(24v), generator(isolation safety) and crane banks(reduce cable runs).
 
My understanding is ABYC wants a mechanical shut off switch.
Also if you get a dedicated LFP start battery, as are coming available, great. IMO, I would not use batteries that were created for house loads to start the engine, run the windlass or davit.
 
The possible variations are complex. I could set battery #1 undervoltage cutoff at 13.1V. I could set batteries #2-6 to shutdown at 12.6V. If I ever used up all those amp hours without looking at the shunt, say, dancing in the moonlight on the flying bridge and the music suddenly stops, I could BlueTooth battery #1, turn it on by lowering the undervoltage cutoff parameter, and when it came back on line it would awaken batteries #2-6. I now would have the necessary MCA to start my engine and recharge the bank. I would get to use about 90% of a large LFP bank and always still have enough to start (without lead start batteries).

Need to play around with this a little more
Setting battery #1 at 13.1 is likely to lead to battery shutdown under loads since the battery generally operates in that range. So a little lower would probably be better.

Also not all batteries have JBD BMS's and access to the OverKill Solar app. So some wont be able to change the settings.

But in general, I think it's a viable option. You might experiment how low you can go for a cut off and still start.

Just to clarify, and Marco correct me if I am wrong, I dont think Marco is saying to eliminate the battery cut off switches since ABYC states a BMS is not sufficient for this. It sounds like he is just saying instead of isolating the start bank, he can create a reserve within the bank by setting low voltage cut off to maintain a small reserve.

You can also do this with a Victron BMV 712 shunt and relay or a smart battery protector other such devices with a contactor or an alarm etc. But changing the BMS LVC to leave a bit of room is viable IMO and can be done in seconds.

But in some cases, it's not needed. For instance, many have:
Generator and gen battery
Engine alt and engine start battery
Large lithium house bank with inverter to power various other chargers already plumbed in.

Thats a pretty basic set up and there is quite a few variables to access

Marco you are strictly battery, alternator and solar? No gen? And one single large bank?
 
Marco you are strictly battery, alternator and solar? No gen? And one single large bank?

I still have a 1-2-Both switch and have two 2 LFP battery banks. PO had installed 4 deep cycle lead acid in two banks and just kept in on Both. Deep cycle to start? OMG! Actually, MarineHowTo says that if you have enough deep cycle it is perfectly okay to use them to start. So I kept the same "outlaw boat" idea, only switching to enough LFP to start. I know some say that they won't do that until a manufacturer gives them the green flag (which will be in the form of a huge price tag).

I never ran the generator. PO said he hadn't run if for years and had thought about removal. He called it "Yanmar ballast." I paid a mechanic to come take a look and say he couldn't figure it out. No more money spent on something I didn't think I could find a use for. Cleared up a nice big space for LFP batteries. Much more useful. I actually have space for 2 more, but that would be redundant.

Nobody has yet explained what's wrong with using 2,200 MCA from my bank of LFP to start my engine. That is based on my reduction of the manufacturer's settings for max discharge. That number (as with undervoltage) also has a time delay setting (adjustable with the Overkill app). So with LFP, one can set the BMSs for 2x what's needed to start the engine. The ABYC let's you get away with not fusing the start cables (not recommended by many), but by using the BMSs, one can safely protect the cables, battery, starter, and boat. Turns off in seconds. Faster than running to the Off switch, assuming one figures out the need to turn it prior to fire breaking out. Also assumes, like the ABYS, that the short is always going to be after the 1-2-Both switch. Reverse fear mongering!!! Score one for LFP!

Choosing a low voltage cutoff to save a reserve for engine or generator starting (and still get the benefit of lots more usable LFP amperage) has another component that helps with the issue of voltage sag that you noted. One can also set the time delay. Say I've screwed up and run my batteries down. They have turned off at 12.9V. I can reset the batteries to 12.7V and they come back on with plenty of amps. The problem is that the voltage sag from the inrush to my engine starter may shut the battery(s) down again. I can either set the undervoltage even lower OR I can set the time delay to 5 seconds. My inrush is much shorter than that. Actually, my engine always starts in less time and it's only 400 amps.

The manufacturer's BMS settings are 1) to protect their product from warrantee claims, and 2) some kind of a compromise/guess as to whether they will be used in a garage, golf cart, RV, etc. Universal settings that are generally okay if one doesn't have Bluetooth and an access app. I don't see any reason to keep those parameters when more conservative and convenient settings are appropriate. Especially when the BMSs can, with a little thought and understanding, replace lead acid, DC to DC converters, isolators, etc. Simplify, simplify. Not for everybody. Some need reredundancy. They believe they need a special starter battery to start their engine with a starter battery that says it is for starting engines type of start battery for engines. See what I did there?
 
Provided you find the right set point...seems perfectly viable for your set up.

Marco...you're not an "in the box" kind of guy are you...lol.
 
Nobody has yet explained what's wrong with using 2,200 MCA from my bank of LFP to start my engine.
Since you referred to Marine how to
The reason most drop in batteries cannot be used for starting is the in-rush current. The in-rush of large DC motors looks like a dead short to the FET’s. Imagine sitting there and intentionally shorting your battery multiple times each day…..That is what starting your motor, running a windlass, electric winch or Bow thruster looks like to the FET’s. There are LFP batteries that can be used for starting but they are very expensive at this point in time.
So, until an LFP is sold as a start battery and warranted as such, even though my LFP bank has a continuous output rating above the starter needs I will not use them for starting the engine on a regular basis.
 
I would not use batteries that were created for house loads to start the engine, run the windlass or davit.
What batteries were created for house loads? My batteries came with capacity and output ratings. They were created to produce those ratings. Not difficult to stay within those ratings. That required a bank of LFP to start my diesel. I guess one could claim that a single LFP battery wasn't created to start the engine, but my LFP batteries (in a bank) were created to start my engine.

Interesting how LFP got this reputation. A gallon can of paint won't paint a house. But stating that paint in gallon cans shouldn't be used to paint a house is an odd conclusion. Turns out one only needs enough paint (too much is usually better) and the type of container is largely irrelevant. That was my reasoning using LFP batteries to do whatever. It only requires enough containers (i.e., batteries with X amps). And too much is also okay.
Marco...you're not an "in the box" kind of guy are you...lol.

I'm curious about golf carts with LFP and you have the experience. Carts can draw 200 amps, especially with 4 people going uphill. That is way more than my windlass. The golf cart LFP solution, like my solution, is simply to have enough amperage (or too much). Might require two batteries, might require four, but it is ultimately the battery(s) output capability, not its chemistry, that determines whether it can do the job. I see the benefit of LFP in a golf cart, but did you know it is used to run the engine!

Maybe one can't call golf cart engines engines because LFP can't be used for engines. Seems to me that a golf cart engine is similar to a davit or windlass. My windlass only draws 80 amps. I don't understand why LFP works so well and is so popular running golf carts all day up and down hill but boaters are concerned about using them on a windlass for 5 minutes (or a thruster for 5 seconds). LFP manufacturers even list golf carts as a viable usage. They must know that golf carts have engines (and house loads). Maybe golfers are soon to be disappointed en mass by a epidemic of LFP battery failures? Terror in Florida and Arizona. Probably not.

Still don't get why some believe LFP electrons are so different from lead electrons. The sparks look the same.
 
Read that MarineHowTo article why "most" drop in batteries shouldn't be used to start. Huge inrush current looks like a dead short to the FETs on a small LFP battery. One has to be aware of the size of the drop in battery and the dead short setting. Mine were set by the manufacturer at over 1,000A per battery. So for my bank (which I'm talking about), that is a approximately a 4,000 amp draw for a few milliseconds before a battery FET would shut down. The inrush current of my starter is about 400A for a few milliseconds, so a safety factor of 10. That is on my bank, not on "most" drop in batteries (unless they are also in a bank). I would not recommend using a single small drop in LFP to start. It might do it, but it also might cause problems. My starter motor's 400A inrush is also well below the manufacturer's set parameter for continuous current (1,000A for the bank). There is yet another middle overcurrent shutdown parameter on my batteries which I don't remember. Something like 600A for 5 seconds. That's 600A per battery. So the MarineHowTo articles concern for FET shutdown is if my starter drew 2,400A for several seconds. Not really a problem because my battery banks are also protected by 250A fuses. Clearly, the individual battery that Rod was using as an example isn't the same as my battery bank or any simple bank.

I agree that it is possible to use an undersized drop in battery and exceed its rating. Expensive LFP batteries with the word "start" on the sticker are beginning to appear. Even still, I doubt the manufacturers will be able to guesstimate whether their battery parameters are set for your application. I doubt that, for the same money, an official "start" battery can exceed a simple bank of LFP in either safety or utility. So far, the available "start" batteries appear to fit in a standard under-hood battery box. Most trawlers have the luxury of space such that little dedicated LFP start batteries don't make sense. They might never make sense, given the ease of creating a bank.
 
Marco...why limit yourself on just LVC. If you found yourself in a real dire situation...you can also raise the current output limits for a "last ditch effort" start. There is plenty of current capability in the lithium cells. Its the BMS and MOSFETS that take the pounding. Even with a low SOC, a bank such as yours would likely be able to pull out one more start with some BMS parameter manipulation.

Even though I don't think you should incorporate this into a system design, it could save the day just knowing it.
 
Marco...why limit yourself on just LVC. If you found yourself in a real dire situation...you can also raise the current output limits for a "last ditch effort" start. There is plenty of current capability in the lithium cells. Its the BMS and MOSFETS that take the pounding. Even with a low SOC, a bank such as yours would likely be able to pull out one more start with some BMS parameter manipulation.

Even though I don't think you should incorporate this into a system design, it could save the day just knowing it.
That is an issue. Setting parameters that make sense when I set them but might not when I need to wake a sleeping battery. Turns out there is more than one way to skin a cat with these parameters. I was trying various scenarios and wondering which one would be easiest to remember in a storm when off a lee shore.

Not wanting to burn down the marina, I've been cautious on testing various shutdown parameters. The only thing I have read that sounds dangerous with my experimentation is setting parameters either very close or contradictory. For instance, if I set undervoltage shutdown at 12.8 and the "wake up" at 12.9. I might think that then it only takes a little applied voltage to wake up the battery. But what might happen is the battery shuts down at 12.8 with some tiny house load. Once shut down and the house load is gone, the battery at rest goes back up to 12.9V and turns on. House load drops it back to 12.8V and a cycle is formed. The FET can cycle at 60 gajillion Hz. And that generates heat (that's why the BMS has a large heat sink) and the FET can then fail. Either open or closed (not sure) but neither is good. There are a few scenarios where it is possible to set up "vibrating" parameters. The time in seconds or milliseconds to quickly shut down might not be a problem, but the time to recover may be too fast and create a loop.

I assume that is some of the reason why the set parameters, as viewed with the Overkill app, have both a voltage or current difference and a time setting. For instance, my manufacturer set the battery low voltage cutoff at 10V and the wake voltage at 10.4 volts plus a time delay of four seconds. Not likely to get a vibration setup with those voltage parameters, and if so, it could only be at four second intervals. Presumably, that is to reduce the possibility of heat buildup that the heat sink can't handle. My takeaway is don't get too cute with the parameters.

I guess if it's my thread I can't complain when I go off topic.
 
Read that MarineHowTo article why "most" drop in batteries shouldn't be used to start. Huge inrush current looks like a dead short to the FETs on a small LFP battery. One has to be aware of the size of the drop in battery and the dead short setting. Mine were set by the manufacturer at over 1,000A per battery. So for my bank (which I'm talking about), that is a approximately a 4,000 amp draw for a few milliseconds before a battery FET would shut down.
Then please say that in your posts, you made a generalized statement and then asked
Marco Flamingo said:
Nobody has yet explained what's wrong with using 2,200 MCA from my bank of LFP to start my engine.
I should have just replied, do what you want for your bank.
 
Marco, you have a fatal flaw in your plan here. You are counting on BT access to your BMS. But you do not have an assured nor supported access. The Overkill app is made by one vendor for their products, and just *happens* to work with the JBD BMS in your batteries. Further, EcoWorthy has specifically stated there is no supported access to the settings you are counting on to make your plan work. Overkill could drop support for their app, or make changes to it, or any number of other things such that you can no longer access your batteries, let alone make changes to BMS settings and write them back out to the BMS.

You’re a pretty detailed as well as thinking-outside-the-box kind of guy, and if you are fully aware of the limitations I just mentioned and are comfortable with that, please continue. But I am writing this to ensure others that are much less knowledgeable who are reading this and trying to learn don’t get caught.
 
I've asked several times for an explanation of "why keep lead" when one has a bank or two of LFP. Although I haven't gotten an answer, in some posts there is some assertion that LFP can't do it.

I stopped reading at the above point.

When I switched to LFP I kept my sealed lead acid start battery. It was always online and the house LFP fed into the start/windlass circuit, but the start battery could not provide house loads. The battery was a dump load for the alternator in case somehow ALL the LFP BMP shut them off. Plus at the time I just had not reason to get rid of the start battery (my LFP could start my engine,

This whole topic is among the LFP never go away ones.
 
Marco, you have a fatal flaw in your plan here. You are counting on BT access to your BMS. But you do not have an assured nor supported access. The Overkill app is made by one vendor for their products, and just *happens* to work with the JBD BMS in your batteries.
Turns out that the "one vendor" that makes the program that happens to works with the JBD BMS in my batteries is JBD. The Overkill app is the app I've used, and I'm happy with it, but the JBD Tools app works. I haven't tried them but other functioning apps listed on D.I.Y. Solar are Xiaoxiang, SmartBMS Utility, AOlithium, Carplounge BMS, Daly BMS, and probably others. There are even some "pay for play" apps. One requires payment for the "pro version" that allows changing parameters (which other apps do for free) and one can supposedly do phone notifications of fault conditions. I'm happy with the free app. I don't see the BMS Bluetooth concept going away. It might get complicated to choose which one to use.
 
Where did you get the JBD Tools app? The only thing I could find through their website was a link to what was *supposed* to be the Android and iOS apps, but just yielded a zip file from Xiaoxiang with a .apk file in it. There are a couple of tools on the Apple App Store by Xiaoxiang related to BMSs, but you need the $10 paid version to be able to write to the BMS.

But notwithstanding any of that, these are tools for JBD’s BMSs, as in external BMSs. Once it has been incorporated into a battery by a manufacturer, it is now anybody’s guess. You have no idea, other than anecdotal observations based on experimentation, what may be custom tailored, or restricted in some way, or any other possible modification that they (eg EcoWorthy) may have done to customize it. Which in my mind make all bets off when it comes to being able to use it as a tool for my discretionary use. Not for me, no how no way. BTW, evidence of what I just said is the settings that were in the battery that Will tested in the YT video were significantly different than what were in mine when they arrived, which was only a few months after his test. These were the ones related to max discharge current etc. (I don’t have them all handy right now and not going to take the time to try to find them). I absolutely require a higher level of trust and dependability and reliability and repeatability in what is a critical system in my boat. YMMV (and obviously does).

K, I’m stepping back now. This thread, and actually several others like it, seem to be turning the whole lithium battery topic into a bit of a hobby project, or science experiment. It reminds me of the very early days of satellite TV here in Canada, where we didn’t have any. The solution was to get a Dish or DirecTV setup, run it through a pseudo US address to get it to work. And then the endless hacking of the cards. It was a joke if your only objective was to watch some TV. There are wayyy too many things on my boat demanding my attention and money, so I can’t be in a non-stop fooling-fiddling-f***ing around with my battery setup/configuration/setting. I set up my EcoWorthy 280 bank with advice from EW tech support, Balmar support and Victron (lotttsss of reading and asking), and have a system that performed beautifully all summer. All using pretty reliable and dependable architectures - fusing, switching, configuring and protecting. There was a bit of tweaking and detailed observing for the first couple weeks after install, and since then, set and forget….just what it needs to be IMO.

Cheers, good luck, and here’s hoping the only smoke I see from south of the border is a forest fire 😱😏
 
K, I’m stepping back now. This thread, and actually several others like it, seem to be turning the whole lithium battery topic into a bit of a hobby project, or science experiment.
One of the issues with these budget batteries is the lack of good programming. The JBD BMS has to be programmed when it's powered up upon build. It very common for these budget batteries to have incorrect parameters applied for the build details. If one was savvy, it would not be unreasonable to check and then tweak various parameters. But of course, you really need to know what you are doing.

In some testing, especially for discharge, ratings are routinely exceeded during "youtube" testing and then deemed to be "powerful" batteries. Now we know that such behavior is actually an abject failure and safety hazard. I was always of the mind that a battery should do exactly what it is spec'ed to do and that any exceedance was a quality failure. We should take those published parameters and spec our bank size, loads and Ah etc

I have a Vatrer battery that they sent me to make a video and it seems to generally be ok. But I noticed the cell overvolt was set to 3.8 volts and during testing battery voltage rose to 15.2 volts before shut down. I chose to change those parameters via the OKS app to 3.7 and 14.7 since there is not good reason to approach those elevated voltages. and I consider the Vatrer to be a better budget battery

I guess I am trying to say...that if you buy a budget battery it might actually be worth going over some of the more common settings to at least verify for correctness.
 
FedEx just dropped off my new EcoWorthy 280Ah battery. Immediately looking at its parameters with the Overkill/JBD/Xiouxiang/whatever app, they are not identical to my prior batteries (which themselves weren't consistent). Likely close enough such that EW didn't fine tune the parameters, but it does matter to me, since I am responsible for my vessel and crew. It would be irresponsible to not adjust BMS parameters to address several safety issues.*

One of the most noticeable ones is the overvoltage cutoff at 15 volts. My other batteries were consistent at 14.6V. Now it is possible that this particular battery was built by EW is a different manner and therefore has a different voltage than other LFP batteries. Or 15 is close enough for the manufacturer and/or JBD. Either way, I don't need to ever charge to 15V and will reduce that. Even the present "resume" setting of 14.4V is higher than I will go with charging.

The BMS "event counts" viewable through the apps were also interesting. It had already logged 1 BMS undervoltage time, 1 Single cell overvoltage time, and 1 single cell undervoltage time. Something to fret about or normal initial counts for a BMS? Some might rather not know this info. I took a screen shot and went on to enter safer BMS parameters.

* LFP fear mongering works both ways.
 
Marco, have you contacted EW tech support and asked them WHY? their batteries are not set the same.
Sure glad I am oblivious to how my eight batteries are programmed. Since they talk to each other they know more than I do.
 
Installed my new 280Ah LFP that is "dedicated" to the induction stove. Preinstallation charging on shore with the new Eco-Worthy charger went fine. Installed the battery with the Anderson clip leads accessible in case solar isn't enough. Now that it is installed, I went back and looked at the factory-set parameters to see if they meet my "stove" requirements. Shutdown parameters aren't that important, as it won't be connected directly to the alternator. Running down through the trigger conditions, the first odd one is batt over voltage shutdown @15V. Haven't seen anywhere that recommends going over 14.6V, and that is just to get the last little dribbles out of the LFP charging. The release condition is 14.4V, but if there is an overcharge issue, that might simply let the charging start up again until the 15V shutdown occurs over and over again. Will change that parameter.

The next odd one is the battery's undervoltage trigger condition @8.8V. Again, nowhere have I seen an LFP low voltage cutoff lower than 10V. The release condition of 10.4V is the same as my other EW batteries. The individual cell overvoltage is 3.75V. This is the same as my other batteries (before I lowered them), but again a different setting than I have generally seen for LFP. I will lower that to 3.65V, as is usually recommended. The cell undervoltage is 2.2V, which when X 4 cells, equals 8.8 volts (the battery's oddball under voltage trigger). I will bump up the individual cell undervoltage to match my new battery undervoltage setup, i.e., 2.5V per cell to match my other batteries.

One last difference is at the bottom under the "Level 2 Protections." Level 2 overcurrent (charge? discharge?) is 160A but with no time parameter set on this battery. My other batteries all have this same time parameter set at 320mS. I'm not even sure what this Level 2 parameter does, as the primary charge protection is set at 220A for 32 seconds and the primary discharge is set at 220A for 32 seconds. I don't think this is for a quick 160A draw, as these have a higher continuous rating and I have started my engine (@400A inrush) with one.

Screenshot_20251002-154012.png


The rest of the parameters seem to match my prior EW 280Ah batteries. This battery only has one thermometer. One of my other batteries has 3 and all 3 temps appear on the app. Not that it matters. In fact, it's arguable that even the parameters I'm changing might not matter. A 15V overcurrent event may never happen. And it's possible that a battery could survive a 15V overcurrent event. Without looking at these parameters, either because of no Bluetooth or with Bluetooth and a lack of interest, these batteries reset themselves in a matter of seconds and the event is over. Still, it is odd that the parameters vary from battery to battery.
 
Last edited:
I still have a 1-2-Both switch...
I do too, but it's disconnected at the moment. My plan is to use it to start my generator from the main engine start battery if needed, but running the big cables up to it seem to be hardly worth it. If the generator won't start, I'll jump it from my jump pack like I do my truck. Or at least that makes sense to me right now.
 
Back
Top Bottom