Cheap LiFePO4 BMS?

[nicknorman]

16 minutes ago, Dr Bob said:

Agree with Mike. Price. £160 whereas the Tyco relays are £30-£70. My motorised switch takes very little power. I dont worry about power drain at all - we were away in Asia most of August and it hardly dropped in power.

 

Your strategy for charging/discharging - you have to work out for what you want to achieve. I went for a very simple system, ie batteries under the bed so wont go below 0°C and if they do they are out of the circuit. Tom, MP and I all use an LA as a dump so there is always a battery in circuit to power the appliances with the engine running. Tom uses 2 switches, one on the charge side and one on the discharge side. The more complex you make it (ie not having an LA) and the more you try to cover all eventualities the more foolproof (or idiot proof) it has to be. Its far cheaper to use the Tyco relays if you are going to have more than one switch. I got the motorised switch as I didnt have the skill to build a circuit to activate the latching relay....the BEP switch was very simple to wire up (once Tom showed me how to do it) but is £100 more.

If you are not going to have an LA in circuit then it does become more complex and issues like you are talking about have to be solved.

 

The whole point of this more sophisticated system is to not require LA batteries, which I consider a low tech fudge (not that that is a bad thing, but if everyone goes for that there is no need for a more sophisticated system!).

You talk about tyco relays - which ones precisely? And which BEP switch is it that you have? What about the sterling one?

 

What is a little frustrating is that one can get MOSFETs which can pass /switch several hundred amps and have an on resistance of less than 1 milliohm for a few quid, but my knowledge of high current electronics is minimal.

[peterboat]

1 hour ago, nicknorman said:

Can someone remind me what the issues with the motorised isolator switches is please?

 

Also thoughts on this point please: there needs to be an isolator for when there is an extreme high or low cell voltage to give ultimate protection to the battery. The battery would be completely disconnected.

 

But what about when the battery temperature is below 0 and the engine is started? One wants to keep the battery connected for domestic loads. One also wants to supply boat domestic loads from the alternator. So how should the system handle this? Should it set the alternator regulation voltage to a low value so that, whilst the batteries don’t receive any charge, domestic loads can be supplied by the alternator? (if so what voltage?)   Or should it just isolate the battery and connect the alternator directly to the boat’s services without a battery in circuit (likely to get lots of ripple I suspect)? Or should it isolate the alternator from the batteries and boat services, whilst leaving the battery to supply boat services (requires a second motorised switch)? Or what?

Most of us have the batteries in the cabin to keep them snug and cosy ?

[nicknorman]

3 minutes ago, peterboat said:

Most of us have the batteries in the cabin to keep them snug and cosy ?

A well designed system needs to cater for all reasonably foreseeable circumstances. Even as a live aboard, it is not unforeseeable that you might go away eg on skiing holiday, come back after a cold week to a chilled boat with low batteries that need charging. Obviously for a non-live aboard the probability is higher.
 

Those very expensive batteries need to be protected against even an improbable event.

[MoominPapa]

2 hours ago, nicknorman said:

Can someone remind me what the issues with the motorised isolator switches is please?

 

Also thoughts on this point please: there needs to be an isolator for when there is an extreme high or low cell voltage to give ultimate protection to the battery. The battery would be completely disconnected.

 

But what about when the battery temperature is below 0 and the engine is started? One wants to keep the battery connected for domestic loads. One also wants to supply boat domestic loads from the alternator. So how should the system handle this? Should it set the alternator regulation voltage to a low value so that, whilst the batteries don’t receive any charge, domestic loads can be supplied by the alternator? (if so what voltage?)   Or should it just isolate the battery and connect the alternator directly to the boat’s services without a battery in circuit (likely to get lots of ripple I suspect)? Or should it isolate the alternator from the batteries and boat services, whilst leaving the battery to supply boat services (requires a second motorised switch)? Or what?

I think to be able to use the alternator to carry domestic loads when the batteries are cold, you need to be able to measure battery current and control the alternator excitation sufficiently accurately that you limit the charge current to a low value. Charging cold batteries is not verboten, but the allowable rate is low because the rate at which the Lithium ions diffuse is markedly reduced, and you can end up with undesirable side-reactions if you try and charge faster than the rate-of-diffusion allows.

 

Disconnecting the Li and running the loads across an LA battery and the alternator  is much easier to achieve.

 

MP.

 

ETA. The same control requirements apply to other charge sources (mains charger, solar) of course.

Edited November 16, 2019 by MoominPapa

[Dr Bob]

59 minutes ago, nicknorman said:

A well designed system needs to cater for all reasonably foreseeable circumstances. Even as a live aboard, it is not unforeseeable that you might go away eg on skiing holiday, come back after a cold week to a chilled boat with low batteries that need charging. Obviously for a non-live aboard the probability is higher.
 

Those very expensive batteries need to be protected against even an improbable event.

If you have an LA in the circuit, then just isolate the Li's ....or they will be isolated by the auto disconnect....and when you get back run the engine and eberspacky thingy...and stove...till you get some heat back in, the reconnect the Li's. Simple. Maybe not elegant but simple.

I will post details of the tyco relay tomorrow unless someone has the details to hand. My switch is the BEP 701md If I remember correctly.

[Dr Bob]

1 hour ago, nicknorman said:

The whole point of this more sophisticated system is to not require LA batteries, which I consider a low tech fudge (not that that is a bad thing, but if everyone goes for that there is no need for a more sophisticated system!).

 

A low tech system using LAs is simple and that is why it is attractive. What is still needed though is a way to control the alternator and if possible a module to activate latching relays as I said a few pages back. 

The more sophisticated a system becomes, the more 'plug and play' it has to be and then that involves a lot more development to cope with the vast variation in systems out there. The low tech approach really is simple and it works....but would be simpler if we had a decent alternator controller.

[nicknorman]

1 hour ago, Dr Bob said:

A low tech system using LAs is simple and that is why it is attractive. What is still needed though is a way to control the alternator and if possible a module to activate latching relays as I said a few pages back. 

The more sophisticated a system becomes, the more 'plug and play' it has to be and then that involves a lot more development to cope with the vast variation in systems out there. The low tech approach really is simple and it works....but would be simpler if we had a decent alternator controller.

It’s understanding what the requirement is for the latching relays that I’m currently seeking.

[MoominPapa]

10 hours ago, nicknorman said:

It’s understanding what the requirement is for the latching relays that I’m currently seeking.

Latching rather than conventional? SImply power draw.

 

MP.

 

[Dr Bob]

This is the tyco relay that I think MP and Tom are using. I dont know much about them as I havent got one. MP can advise.

https://www.te.com/commerce/DocumentDelivery/DDEController?Action=showdoc&DocId=Data+SheetV23130-X0000-A0010612pdfEnglishENG_DS_V23130-X0000-A001_0612.pdf1-1414939-4REC

£60-70 quid new. Cheaper from car breaker yards - so I am told.

[rusty69]

12 hours ago, Dr Bob said:

but would be simpler if we had a decent alternator controller.

What are the pro's and con's of using a battery charger powered by Alt/LA/inverter to charge the LiFeP04's as suggested earlier?

 

Answers must be words of one or two syllables,sentences less than 10 words, and a minimum of techno lingo. Definitely no mention of thresholds, balancing or cell monitoring to be included.

 

Thankyou

Edited November 17, 2019 by rusty69

[rusty69]

On 19/10/2019 at 13:04, Richard10002 said:

Sounds a bit complicated for the ordinary Joe at the moment. I can live with killing a set of LA batteries within a couple of years, but am not prepared to take the chance with lithiums yet  

What a difference a month makes.

[peterboat]

7 minutes ago, rusty69 said:

What a difference a month makes.

He was on the cusp, just like he is on the cusp of becoming a leaver?

[rusty69]

4 minutes ago, peterboat said:

He was on the cusp, just like he is on the cusp of becoming a leaver?

Me too. I am currently on cuspers last stand.

[Dr Bob]

1 hour ago, rusty69 said:

Me too. I am currently on cuspers last stand.

Bo**ux. You are 2nd behind Smelly wanting to leave.

[Dr Bob]

2 hours ago, rusty69 said:

What are the pro's and con's of using a battery charger powered by Alt/LA/inverter to charge the LiFeP04's as suggested earlier?

 

Answers must be words of one or two syllables,sentences less than 10 words, and a minimum of techno lingo. Definitely no mention of thresholds, balancing or cell monitoring to be included.

 

Thankyou

The big pro has to be that you dont overstress you alternator.

Say your Li's are half full, you alternator will be working flat out so may burn itself out if you do this for any length of time and it wont turn off when your Li's get full.

If you use the Alt/LA/inverter, then the Li will take the power from the inverter/charger at the rate it wants, the LA's will power the inverter and the alternator will charge the LA's as if they are in bulk. When the Li's get to full,  the inverter/charger will back off - IF you can set it to the voltage you need - Richard can as he has a programmable Sterling - so then the alternator will cut back in current as the LA's demand....but still 14.4V to keep the LA's topped up.

Negatives? There could be an issue if the inverter/charger puts a lot more current into the Li bank than the alternator is putting into the LA's. In that case the LAs will drop in SoC as the Li's increase in SoC. The beauty of using LAs with Li's is that the LAs are alway near full as you dont drop the voltage below 12.7V.  If they drop to 50% then you are going to need 14.4V to get the LAs back to full which you may not have if you then stop the engine. Yes, the Li's will charge the LAs back up to full but only at 13.3V or less. The thing is though that most(?) inverter/chargers are programmable so you should be able to set the charge current similar to the LA charge current. A lot here will depend on the capacity of the LA vs capacity of the Li.

This is going to open up a huge can of worms - as Nick criticised my system back in March - saying I needed 14.4V to keep the LA's in good shape. 6 months on, my LAs are in the same shape as life before Li's despite only being charged at 13.7V max - as I doubt they were ever discharged below 95%. There is a logical argument for this which is outside your rules for a response ? so I wont bother outlining it.....but you will need 14.4V to charge LAs if they go down to 50% SoC.

What I am not sure about is what happens if your LA bank is 90% charged and you take 50A out with the inverter, what current is you alternator controller give? I've been using an AtoB with my LAs so cant talk about 'normal'charging of LAs. Maybe sulphation is the big downside if the LAs go down to 50% or another downside is lower charging rates from the alternator than if the alternator was directly connected to the Li bank.

Need also to think about the wiring so you isolate the Li's from the alternator yet want to have your domestics are fed from the Li's.

Someone needs to try it and keep a close eye on SoC of the LA's.

 

[rusty69]

15 minutes ago, Dr Bob said:

The big pro has to be that you dont overstress you alternator.

Say your Li's are half full, you alternator will be working flat out so may burn itself out if you do this for any length of time and it wont turn off when your Li's get full.That sounds like a big plus.

 

If you use the Alt/LA/inverter, then the Li will take the power from the inverter/charger at the rate it wants, the LA's will power the inverter and the alternator will charge the LA's as if they are in bulk. When the Li's get to full,  the inverter/charger will back off - IF you can set it to the voltage you need - Richard can as he has a programmable Sterling - so then the alternator will cut back in current as the LA's demand....but still 14.4V to keep the LA's topped up.

Negatives? There could be an issue if the inverter/charger puts a lot more current into the Li bank than the alternator is putting into the LA's. In that case the LAs will drop in SoC as the Li's increase in SoC.How about a lower output charger than the alternator can put out, say a 30A Victron on a 50A alternator.

 

 The beauty of using LAs with Li's is that the LAs are alway near full as you dont drop the voltage below 12.7V.  If they drop to 50% then you are going to need 14.4V to get the LAs back to full which you may not have if you then stop the engine. Yes, the Li's will charge the LAs back up to full but only at 13.3V or less. The thing is though that most(?) inverter/chargers are programmable so you should be able to set the charge current similar to the LA charge current. A lot here will depend on the capacity of the LA vs capacity of the Li.We currently have 4 x 250W of solar.Perhaps 3 could go on the Lithium bank and one on the LA bank.

 

This is going to open up a huge can of worms - as Nick criticised my system back in March - saying I needed 14.4V to keep the LA's in good shape. 6 months on, my LAs are in the same shape as life before Li's despite only being charged at 13.7V max - as I doubt they were ever discharged below 95%. There is a logical argument for this which is outside your rules for a response ? so I wont bother outlining it.....but you will need 14.4V to charge LAs if they go down to 50% SoC.

What I am not sure about is what happens if your LA bank is 90% charged and you take 50A out with the inverter, what current is you alternator controller give? I've been using an AtoB with my LAs so cant talk about 'normal'charging of LAs. Maybe sulphation is the big downside if the LAs go down to 50% or another downside is lower charging rates from the alternator than if the alternator was directly connected to the Li bank.

Need also to think about the wiring so you isolate the Li's from the alternator yet want to have your domestics are fed from the Li's.Your BEP motorised switch controlled by a relay should do this somehow connected to the ignition circuit.

Someone needs to try it and keep a close eye on SoC of the LA's

 

Many thanks.

 

So, it sounds like a possibility. The plus side being that if you already have a suitable charger and inverter,and a relatively small output alternator you don't have to worry about the alternator output so much.

 

I need to give it more thought.

 

 

[nicknorman]

4 hours ago, Dr Bob said:

This is the tyco relay that I think MP and Tom are using. I dont know much about them as I havent got one. MP can advise.

https://www.te.com/commerce/DocumentDelivery/DDEController?Action=showdoc&DocId=Data+SheetV23130-X0000-A0010612pdfEnglishENG_DS_V23130-X0000-A001_0612.pdf1-1414939-4REC

£60-70 quid new. Cheaper from car breaker yards - so I am told.

Thanks, this answers the question I was asking - the answer being that, depending on the type of battery isolator, the system either needs to output an on/off signal (and I think it’s effectively on to turn the isolator off and vice versa, but doesn’t matter at this stage), or one output going momentarily on to motor the tyco one way, a different output going momentarily on to motor it the other way. Anyway the relevant point is that hardware-wise, there needs to be 2 outputs.

[MoominPapa]

19 minutes ago, nicknorman said:

Thanks, this answers the question I was asking - the answer being that, depending on the type of battery isolator, the system either needs to output an on/off signal (and I think it’s effectively on to turn the isolator off and vice versa, but doesn’t matter at this stage), or one output going momentarily on to motor the tyco one way, a different output going momentarily on to motor it the other way. Anyway the relevant point is that hardware-wise, there needs to be 2 outputs.

Shout if you want access to the circuit of my tycho failsafe driver, which takes two pulse inputs from the microcontroller, one for open and one for close, and fails safe by opening the contactor if the microcontroller doesn't pulse the "close" signal every couple of seconds. It also ensures that if the controller output gets stuck "on" the tycho coil doesn't get permanently energised, which will burn it out.

 

MP.

 

[nicknorman]

The tyco relays look pretty good. But question for MP or T&B - the spec sheet says it’s important to pay attention to load current direction. Rather implying that it is only designed to pass current one way. Which isn’t how a battery isolator could be - obviously it need to pass both charging current and discharging current. I my own case, up to 175A charging current and maybe 200A discharging current (electric kettle). Does anyone know why current direction could be important?

[nicknorman]

9 minutes ago, MoominPapa said:

Shout if you want access to the circuit of my tycho failsafe driver, which takes two pulse inputs from the microcontroller, one for open and one for close, and fails safe by opening the contactor if the microcontroller doesn't pulse the "close" signal every couple of seconds. It also ensures that if the controller output gets stuck "on" the tycho coil doesn't get permanently energised, which will burn it out.

 

MP.

 

Always interested to see other people’s circuits! I wasn’t thinking that such a fail safe would be necessary really, the PICs have a watchdog timer, brownout reset and as such I wouldn’t expect it to crash permanently. I suppose it depends on the relative reliabilities of any protection/fail safe driver vs the controller. And also some kind of “risk assessment” as to whether the probability of an over or under voltage situation combined with a microprocessor failure is realistic possibility. The tycho burning out isn’t in itself too catastrophic, I would have thought. But keeping an open mind at this stage!

 

In a system without LA batteries, having the contact open as a fail safe might be a bad idea!

 

Any thoughts on the tyco current direction?

Edited November 17, 2019 by nicknorman

[1st ade]

6 minutes ago, nicknorman said:

Any thoughts on the tyco current direction?

I don't know if that's what happens here but contactors on trams used to have a permanent magnet to extinguish any arc. As the contacts opened there is the possibility of an arc; the magnetic field "pushed" the arc away from the housing. This made the arc longer so it was stretched out and extinguished quicker.

 

Conceivably the physical design is that in the case of reverse current either the arc doesn't extinguish or it hits something it shouldn't. This would affect the max breaking current.

[Dr Bob]

1 hour ago, rusty69 said:

How about a lower output charger than the alternator can put out, say a 30A Victron on a 50A alternator.

 

 

 

Yes, but my victron IP22 30A has a LiFePO4 setting but doesnt cut the current and voltage as it gets to 100%. Best have a charger that will go to float (ie set to low 13V's) at whatever termination voltage you put in and one where you can set the current. I think Richard's Sterling does this. My IP22 would be no good to shut down itself but would work if you terminated via a auto disconnect ie BEP or tyco but the relay cutting the charger rather than isolating the batteries.

 

1 hour ago, rusty69 said:

We currently have 4 x 250W of solar.Perhaps 3 could go on the Lithium bank and one on the LA bank.

 

I would wire all 4 into the Li as it will be the battery to use when not charging via the engine.

 

1 hour ago, rusty69 said:

Your BEP motorised switch controlled by a relay should do this somehow connected to the ignition circuit.

 

 

 

See response above on shutting off the charger. Yes you do then need to think about how you isolate your Li's from the LAs and I am not sure what the optimum is as I have not thought it through. (Acute brain drop occurs if you try and think too much!) Maybe Richard can explain his proposed system? -it sounds like worth trying but there is a lot of detail to iron out. What you dont want is two BEPs as they are £150 squid each. The Tyco's are much cheaper but then your going to have to talk to MP for his circuit.

If you are thinking of trying Li's then it is worth experimenting a bit.

If you buy them now and get them installed, say before Christmas, then you will get b**ger all solar but you can charge for an hour in the morning and an hour in the afternoon and not burn out your alternator as it is cold and you will never be near 100% SoC. By Spring, by the time it warms up and you need to control your alternator temp there will be more sun and you will be moving to a more automated system. When I put mine in, I was planning to manually isolate the alternator each day and just use solar when I didnt have time. Within a month I had worked out a solution and so spend no time at all (apart from the word I am not allowed to say but it what you do when you spin lots of plates). Nick will have his alternator controller out by Chrismas!

[MoominPapa]

33 minutes ago, nicknorman said:

Always interested to see other people’s circuits! I wasn’t thinking that such a fail safe would be necessary really, the PICs have a watchdog timer, brownout reset and as such I wouldn’t expect it to crash permanently. I suppose it depends on the relative reliabilities of any protection/fail safe driver vs the controller. And also some kind of “risk assessment” as to whether the probability of an over or under voltage situation combined with a microprocessor failure is realistic possibility. The tycho burning out isn’t in itself too catastrophic, I would have thought. But keeping an open mind at this stage!

On a completed system, I might agree, but on mine, as I was doing software development as I went along, it's been quite useful. The other reason for doing using it is the failsafe is powered direct from the Li batteries, and thus remains powered when the microcontroller is powered off. A simple switch in the microcontroller power line thus becomes a reliable way to remotely isolate the batteries: turn off the controller and 2 seconds later a minimal piece of hardware opens the contactor next to the batteries under the bed.

33 minutes ago, nicknorman said:

 

In a system without LA batteries, having the contact open as a fail safe might be a bad idea!

 

Any thoughts on the tyco current direction?

The datasheet implies that there is magnetic blowout, and the specs for large-current opening are only valid in one direction. If you can decipher from the datasheet which direction it is, I'd be interested, it wasn't clear to me. As this application needs to break normally modest currents in both directions, I didn't worry.

 

MP.

 

[nicknorman]

19 minutes ago, MoominPapa said:

On a completed system, I might agree, but on mine, as I was doing software development as I went along, it's been quite useful. The other reason for doing using it is the failsafe is powered direct from the Li batteries, and thus remains powered when the microcontroller is powered off. A simple switch in the microcontroller power line thus becomes a reliable way to remotely isolate the batteries: turn off the controller and 2 seconds later a minimal piece of hardware opens the contactor next to the batteries under the bed.

The datasheet implies that there is magnetic blowout, and the specs for large-current opening are only valid in one direction. If you can decipher from the datasheet which direction it is, I'd be interested, it wasn't clear to me. As this application needs to break normally modest currents in both directions, I didn't worry.

 

MP.

 

Ok thanks. Datasheet bottom right of first page has a + and - annotation, so A is - and B is +

 

I must say with RS selling this at £60 including vat, zero power consumption in either state,  it seems a no-brainer to use it. Has adequate current capability for me ie 260A at 25C. And it should in theory never need to operate!

 

Except...

 

Right next question, if there is to be a low voltage disconnect I think it needs to take into account battery discharge current. I know that Li has much lower internal resistance than LA but at say 200A discharge on a 400AH battery there must be a fair bit of voltage droop. Obviously one doesn’t want the low voltage protection to trip the batteries just because I put the 2kw electric kettle on with 40% SoC remaining. So could it be expected that there is a fairly consistent voltage drop for a given high current as a proportion of total capacity. Ie an effective internal resistance that is a function of battery capacity that could be modelled in, so that the low voltage cutoff would be x volts + (I x r) where I and r are the instantaneous discharge current and notional battery internal resistance? Does r vary a lot with either temperature or SoC?

 

So many questions!

Edited November 17, 2019 by nicknorman

[MoominPapa]

47 minutes ago, nicknorman said:

Ok thanks. Datasheet bottom right of first page has a + and - annotation, so A is - and B is +

 

I must say with RS selling this at £60 including vat, zero power consumption in either state,  it seems a no-brainer to use it. Has adequate current capability for me ie 260A at 25C. And it should in theory never need to operate!

 

Except...

 

Right next question, if there is to be a low voltage disconnect I think it needs to take into account battery discharge current. I know that Li has much lower internal resistance than LA but at say 200A discharge on a 400AH battery there must be a fair bit of voltage droop. Obviously one doesn’t want the low voltage protection to trip the batteries just because I put the 2kw electric kettle on with 40% SoC remaining. So could it be expected that there is a fairly consistent voltage drop for a given high current as a proportion of total capacity. Ie an effective internal resistance that is a function of battery capacity that could be modelled in, so that the low voltage cutoff would be x volts + (I x r) where I and r are the instantaneous discharge current and notional battery internal resistance? Does r vary a lot with either temperature or SoC?

 

So many questions!

<Consults BMS code>

 

I have two low voltage cutoffs. The first is the per-cell one, which is the last-ditch. Less than 2.9 volts on any cell for more than 5 seconds and it triggers,

 

The second is the "your battery is discharged and is you don't stop, it will get cut-off" alarm. The voltage on that is higher, because it's mainly meant to warn that the voltage has dropped enough that significant discharge of the parallel LA is happening. That trips is the battery voltage drops below 12.52 - (I/128) for more than three minutes.

 

I can't claim the factors and delays  were arrived at by careful experiment and theorising, they are numbers which seemed sensible after staring at read-outs and seem to work, so are a reasonable starting point. 

 

TL;DR for a 480Ah bank, reduce the threshold by about .75v for every 100A of discharge.

 

MP.