webcraft
Well-known member
The bit I understood was the bit about building a plywood box.
- W
- W
270ah DIY LiFePO4 build
- Thread starter Poey50
- Start date
Kelpie
Well-known member
No you've got that wrong, plywood is out these days, it's aluminium plates and threaded rods for compression. Keep up ?
I'll show you my setup when we get back...
sailaboutvic
Well-known member
So basically you have to monitor your app to make sure the BMS is doing what it should do?
Kelpie
Well-known member
I'm not sure how else you would do it? It's not as if I have to actually do anything, I just check out occasionally, mostly for its function as a battery monitor.
sailaboutvic
Well-known member
Sorry I'm not criticising just working out the difference between having to rely on the BMS to disconnect if the batteries cells are <> there setting or having cells monitor like these guys are fitting which is more reliable.
Looking at my own situation with my 400ah the lowest I ever got is 35%dod no where near enough to bottom out and charging at 13,8v any one cell need to be way out to pop ( 13.8v cells 0.001diffrent) .
Being on board full time I monitor mine at less once a day like I would monitor other stuff like water ,
Thinking about it my BMS is more like my second defend me bing the first.
vas
Well-known member
MQTT / signalK are also means that will enable logging of cell performance on a minute basis (or shorter if you wish so!) on certain BMSes. Will be using SignalK for that personally.
Following email alerts can be configured assuming boat is online. Received a couple two weeks ago following 5days of miserable weather and a fridge running dragging the batteries under the limit I've set. So no sci-fi, and obvs v.easy to setup on a victron gx (or raspberry)
Thing is if a rule is indeed overrun, it will disconnect (or do whatever you've configured it to do)
So not quite sure what the discussion is about.
IMHO:
A. each of MPPT/inverter/B2B (whatever) are configured so that they wont allow over/under voltage at BANK level - they individually stop charges/loads - assume all of them are configurable (not always true!)
B. BMS instructs on over/under CELL level actions: disconnect of MPPT/inverter/B2B (whatever) - assuming all these can receive extra instructions from an external source to turn off
C. final line of defence is BMS disconnecting BANK using suitable means (contactor or internal solidstate approach) - means that A and B failed for some reason (most likely B wasn't feasible)
Note A & B battery remains online throughout the process. Vic, I wouldn't want me personally to be part of the monitoring/reacting process!
You may be liveaboard, but you don't even go out for a few hours? That would be more than enough to destroy the bank if something goes wrong!
tbh cell monitoring is a rather rough way of dealing with lifepo4 (again imho)
cheers
V.
Following email alerts can be configured assuming boat is online. Received a couple two weeks ago following 5days of miserable weather and a fridge running dragging the batteries under the limit I've set. So no sci-fi, and obvs v.easy to setup on a victron gx (or raspberry)
Thing is if a rule is indeed overrun, it will disconnect (or do whatever you've configured it to do)
So not quite sure what the discussion is about.
IMHO:
A. each of MPPT/inverter/B2B (whatever) are configured so that they wont allow over/under voltage at BANK level - they individually stop charges/loads - assume all of them are configurable (not always true!)
B. BMS instructs on over/under CELL level actions: disconnect of MPPT/inverter/B2B (whatever) - assuming all these can receive extra instructions from an external source to turn off
C. final line of defence is BMS disconnecting BANK using suitable means (contactor or internal solidstate approach) - means that A and B failed for some reason (most likely B wasn't feasible)
Note A & B battery remains online throughout the process. Vic, I wouldn't want me personally to be part of the monitoring/reacting process!
You may be liveaboard, but you don't even go out for a few hours? That would be more than enough to destroy the bank if something goes wrong!
tbh cell monitoring is a rather rough way of dealing with lifepo4 (again imho)
cheers
V.
sailaboutvic
Well-known member
Again not arguing just point a few things out .
Cell monitoring is basically what the bms is doing.
Going back to my own set up , there no way if I left my batteries unattended for any reasonable time for any damage to happen unless there a short between cells,
in which the bms isn't going to stop.
And only brakers or inclined fuse is going to stop damage further down the line.
Bms just keep set parameters in line and in some of the once people are using here deturment how much current ,
Mine doesn't it's my breaker that does that.
vas
Well-known member
not arguing either Vic, writing them down hoping for new ideas/experiences that would help me improve on that
fe, leaving the boat and having a leak or freshwater hose rupture meaning the freshwater pump will run indefinitely for say X hours (or overnight...) at say 10A, could cause a problem with already empty batteries..
mptt/b2b wont be able to help in this case!
and AFAIK the balancer board wont be able to drive a contactor to disconnect the lifepo4. So how is such a situation managed with their setup?
in my understanding their setup works for high voltage disconnect only
V.
Poey50
Well-known member
Absolutely agree with this. I think it's perfectly fine to manually manage a lot of things. I'm always interested in what my LFP is doing and check several times daily. But I'm fallible - sometimes tired, sometimes distracted, sometimes forgetful, sometimes stupid and sometimes not there at all. I want a second line of defence that will automatically protect me and my boat from myself and that is what the BMS does.
sailaboutvic
Well-known member
In my set up ,not arguing either Vic, writing them down hoping for new ideas/experiences that would help me improve on that
fe, leaving the boat and having a leak or freshwater hose rupture meaning the freshwater pump will run indefinitely for say X hours (or overnight...) at say 10A, could cause a problem with already empty batteries..
mptt/b2b wont be able to help in this case!
and AFAIK the balancer board wont be able to drive a contactor to disconnect the lifepo4. So how is such a situation managed with their setup?
in my understanding their setup works for high voltage disconnect only
V.
My bilge pump bypass my BMS anyway which should do anyway ,
Otherwise turning off the breaker while away for any time would mean that the pump would not work .
Putting that to one side .
if I was leaving the boat for any time
( any time = more them two over night )
I would be turning off the breaker switching off ,
Turning panels off so nothing coming in or out .
Only thing connected is the pump .
The batteries would last out 30 hours constant use at 10A ( mine uses less ) and to be honest if the pump had to work that hard , I have more to worry about then the batteries .
But that be the same if I had LA batteries and there no BMS to look after them.
I'm not saying not to use a BMS but I'm not sure its that bad not to have one as long as on board and monitoring them.
As I posted before, for non full time live aboard it's different althought I think there probably not much in it as every one should be disconnecting all power while away.
Again in my case as I can only talk for my self if the batteries reach 12v or 14v my victron alarm goes off,
because my cells are balance non of my cells will vary at them voltage.
That's a good indication I'm within my limits,
Normally I check the cells when the kettle goes on in the morning and most days in the afternoon to see what's been happening.
If an alarm did go off , I be wanting to know why .
That me ,
I'm very interested in these guys what their plain is and still getting info how they plain to protect their new cells .
vas
Well-known member
ok, if you plan to turn off breakers leaving the boat, fine.
I've never done that and not planning to do so!
I really don't see why I should be disabling things when leaving and where do you draw the line! I mean I have a passerelle (without it I cannot get in, so needs to be on) Once you have one on, you have to re-think what's the point of stopping a few.
btw, I wasn't talking about the bilge pump going off (that would be indeed alarming!), I'm talking about fresh water pump running dry with an empty fresh water tank (easy to "achieve" with a ruptured hose within the pressurised circuit)
V.
I've never done that and not planning to do so!
I really don't see why I should be disabling things when leaving and where do you draw the line! I mean I have a passerelle (without it I cannot get in, so needs to be on) Once you have one on, you have to re-think what's the point of stopping a few.
btw, I wasn't talking about the bilge pump going off (that would be indeed alarming!), I'm talking about fresh water pump running dry with an empty fresh water tank (easy to "achieve" with a ruptured hose within the pressurised circuit)
V.
Kelpie
Well-known member
I disconnect the panels and isolate the lithium battery if leaving the boat for a few days. You never know how long you will be away for...
Poey50
Well-known member
I don't have anything bypassing the BMS as Vic does and I never disconnect the BMS. I have a storage setting on my SmartSolar MPPT of 13.3 volts for both absorption and float. It can stay on that for a weekend or a year if needs be.
sailaboutvic
Well-known member
The BMS themselves like yours are connected to the batteries , so no the BMS are not disconnected but the breaker which everything goes through is disconnected ,
The bilge pump as I explained is bypass the breaker and is connected onto the batteries with an inline fuse .
The LED on the module use very little , I sure at one time you give a figure.
Personally I'm happer shutting everything down if I'm away from the boat , alway have done even when I use LA batteries ,
Should you have a short some where,
I fire is easily had.
Would you not normally turn off the master switch ? I sure you do ,
So really what you saying is you leave your panels connect at a low Float charge ?
Why? Lithium don't need it as they don't lost .
Poey50
Well-known member
Ah OK - I'd misunderstood what you had bypassed. Yes, master switches are turned off. LFP does self-discharge around 3% per month depending on temperature (more in higher temperatures) but the main issue is that I keep the engine battery on a storage float charge of 13.2 volts from the LFP by way of the B2B. I've found that the storage setting on the MPPT keeps everything stable but there is more than one way of managing things successfully.
vas
Well-known member
annoying having to search down the pages for this thread to post something new every week or two maybe should ask for it to become a sticky...
anyway, testing bms with 3S2P 18650cells ? massive amounts of power running a 0.7A e/r light I have here on my desk @12V
things are ok. However I have an odd Q:
BMS can issue warnings and errors, fine.
One warning is cell imbalance, also fine.
Problem is how do you define imbalance?
in diyBMS the easy and fast solution is to check if any cell is passive balancing, and if so, issue the warning. However that means that once charging and reaching the top (especially on shorepower or generator) inevitably there will be cells balancing, but not being imbalanced to one another.
If I got that right, wouldn't it make better sense to issue the warning when indeed cells are imbalanced (as they may most likely be when emptying!)?
Assuming the above make sense, the big Q is: on prismatic LifePO4 cells, what value would be the one to set for an imbalance warning, and what for imbalance error ???
warning is just that, error means some action: throttle charging if it happens during charging, or turn off loads if happening on an empty bank.
what do the ones that have studied all that have to say? any ideas?
btw that's the datalog of the last 24h with a 3h discharging session last evening, followed by an overnight slow charging at 12.5V which from 8:00am started balancing (as can be seen by the cell temps and the PWM balancing algorithm data at the bottom). Stopped charging just after 10am and started a second session of discharging, half way through it realised that one of the two cells in bank three wasn't making contact to this stupid holder, so at 12:00 you see a jump in voltage and system is going strong will probably be 6h before a cell drops to 3V.
cheers
V.
maybe should ask for it to become a sticky...anyway, testing bms with 3S2P 18650cells ? massive amounts of power running a 0.7A e/r light I have here on my desk @12V
things are ok. However I have an odd Q:
BMS can issue warnings and errors, fine.
One warning is cell imbalance, also fine.
Problem is how do you define imbalance?
in diyBMS the easy and fast solution is to check if any cell is passive balancing, and if so, issue the warning. However that means that once charging and reaching the top (especially on shorepower or generator) inevitably there will be cells balancing, but not being imbalanced to one another.
If I got that right, wouldn't it make better sense to issue the warning when indeed cells are imbalanced (as they may most likely be when emptying!)?
Assuming the above make sense, the big Q is: on prismatic LifePO4 cells, what value would be the one to set for an imbalance warning, and what for imbalance error ???
warning is just that, error means some action: throttle charging if it happens during charging, or turn off loads if happening on an empty bank.
what do the ones that have studied all that have to say? any ideas?
btw that's the datalog of the last 24h with a 3h discharging session last evening, followed by an overnight slow charging at 12.5V which from 8:00am started balancing (as can be seen by the cell temps and the PWM balancing algorithm data at the bottom). Stopped charging just after 10am and started a second session of discharging, half way through it realised that one of the two cells in bank three wasn't making contact to this stupid holder, so at 12:00 you see a jump in voltage and system is going strong will probably be 6h before a cell drops to 3V.
cheers
V.
Poey50
Well-known member
annoying having to search down the pages for this thread to post something new every week or two
anyway, testing bms with 3S2P 18650cells ? massive amounts of power running a 0.7A e/r light I have here on my desk @12V
things are ok. However I have an odd Q:
BMS can issue warnings and errors, fine.
One warning is cell imbalance, also fine.
Problem is how do you define imbalance?
in diyBMS the easy and fast solution is to check if any cell is passive balancing, and if so, issue the warning. However that means that once charging and reaching the top (especially on shorepower or generator) inevitably there will be cells balancing, but not being imbalanced to one another.
If I got that right, wouldn't it make better sense to issue the warning when indeed cells are imbalanced (as they may most likely be when emptying!)?
Assuming the above make sense, the big Q is: on prismatic LifePO4 cells, what value would be the one to set for an imbalance warning, and what for imbalance error ???
warning is just that, error means some action: throttle charging if it happens during charging, or turn off loads if happening on an empty bank.
what do the ones that have studied all that have to say? any ideas?
btw that's the datalog of the last 24h with a 3h discharging session last evening, followed by an overnight slow charging at 12.5V which from 8:00am started balancing (as can be seen by the cell temps and the PWM balancing algorithm data at the bottom). Stopped charging just after 10am and started a second session of discharging, half way through it realised that one of the two cells in bank three wasn't making contact to this stupid holder, so at 12:00 you see a jump in voltage and system is going strong will probably be 6h before a cell drops to 3V.
View attachment 127657
cheers
V.
A common figure for maximum allowable voltage difference between cells - 'delta voltage' is 0.3 volts. Some BMS's do a disconnect at that setting.
vas
Well-known member
any ref to that to study Poey? and is that whilst balancing or whist discharging? guess the latter!
Poey50
Well-known member
It's just a figure that seems to keep cropping up. I assume it applies on both charging and discharging although more likely to show up on charging as the separate cell charging curves diverge at the knee.
vas
Well-known member
cheers, I'll try to implement it in diyBMs and see how it goes (in testing... cells to arrive late Jan )
