Cheap LiFePO4 BMS?

[nicknorman]

13 minutes ago, peterboat said:

Years of working on cars has meant knowing exactly how canbus works, fixed a jags non starting before Christmas with canbus problems [wire nearly broken through causing high resistance] I sold the garage four and a half years ago and still I end up working on the sodding things for them!

The thing with CANBUS is that it only specifies the physical and electrical layers. In vehicles, there are other standardised specifications for the application layers that are readily available. These tell you what the bunch of binary data whizzing around the bus actually means. With a proprietary and secret variant like Masterbus, there is no such information available, so I have to reverse engineer it.

[peterboat]

10 minutes ago, nicknorman said:

The thing with CANBUS is that it only specifies the physical and electrical layers. In vehicles, there are other standardised specifications for the application layers that are readily available. These tell you what the bunch of binary data whizzing around the bus actually means. With a proprietary and secret variant like Masterbus, there is no such information available, so I have to reverse engineer it.

You are right, strange thing is its rarely ever the actual canbus at fault, its the power feeds and earth to the nodes that go wrong, but its easy to check with a canbus reader after the diagnostics have pointed you in the right direction.

Richard at Finesse is building boats with canbus operations, cant wait for the BSS inspectors getting a look at them ?

[nicknorman]

29 minutes ago, peterboat said:

You are right, strange thing is its rarely ever the actual canbus at fault, its the power feeds and earth to the nodes that go wrong, but its easy to check with a canbus reader after the diagnostics have pointed you in the right direction.

Richard at Finesse is building boats with canbus operations, cant wait for the BSS inspectors getting a look at them ?

Yes the physical layer is very robust. Our boat, built in 2010, has 2 separate CANbus systems, the aforementioned Masterbus, and the Empirbus DC distribution and control system (all lights, pumps, etc). No visible fusing except for a master 100A fuse. Fortunately our BSS chappie was receptive to the description and explanation and it wasn’t a problem. So not a particularly new concept.

Edited December 31, 2019 by nicknorman

[peterboat]

1 minute ago, nicknorman said:

Our boat, built in 2010, has 2 separate CANbus systems, the aforementioned Masterbus, and the Empirbus DC distribution and control system (all lights, pumps, etc). No visible fusing except for a master 100A fuse. Fortunately our BSS chappie was receptive to the description and explanation and it wasn’t a problem. So not a particularly new concept.

I remember reading that yours was canbus before, Its a much better idea than the usual spaghetti mess that you see on Narrowboats, If I was building a new boat it would be what I would do

[Richard10002]

Here's an interview with the CEO of a seemingly respected Lithium Battery manufacturer. You would assume he knows what he is talking about, and seems to dispel some of the fears there seem to be regarding LiFePo4 batteries. 

 

 

 

 

[peterboat]

7 hours ago, Richard10002 said:

Here's an interview with the CEO of a seemingly respected Lithium Battery manufacturer. You would assume he knows what he is talking about, and seems to dispel some of the fears there seem to be regarding LiFePo4 batteries. 

 

 

 

 

Really good video Richard very helpful indeed, My truck sits outside with solar so in winter its a very low amps going in, so looking at this viseo its not going to hurt the batteries. Clearly our indoor batteries will be ok as they are going to be warm enough to charge quicker. I will watch it again tonight.

[Richard10002]

3 hours ago, peterboat said:

Really good video Richard very helpful indeed, My truck sits outside with solar so in winter its a very low amps going in, so looking at this viseo its not going to hurt the batteries. Clearly our indoor batteries will be ok as they are going to be warm enough to charge quicker. I will watch it again tonight.

Yes.... even though the chances are that my indoor jobs would never reach as low as 0C when charging, the fact that I charge at about 0.12C suggests that it wont matter much if it does.

 

In addition, he also suggests that it is OK to hold them at 3.65V, (14.6V), for a while, or even float at 13.6V for longer than a while, is comforting. It means that I could leave them floating on the genny for a whole day in order to give the internals a chance to balance the cell that seems a bit out.

 

He seems to say that balance is important, and not allowing them to discharge completely is very important.

 

I'm still going to fit a temperature controlled relay and a high voltage controlled relay in the charging cable, perhaps more out of interest that necessary protection.

Edited January 7, 2020 by Richard10002

[Dr Bob]

On 07/01/2020 at 13:33, Richard10002 said:

 

 

In addition, he also suggests that it is OK to hold them at 3.65V, (14.6V), for a while, or even float at 13.6V for longer than a while, is comforting. It means that I could leave them floating on the genny for a whole day in order to give the internals a chance to balance the cell that seems a bit out.

 

 

It's an interesting video but this guy is trying to sell batteries! Of course he is going to try and push the boundaries. The fact is that LiFePo4s should not be charged when they are 100%. If you notice he does not say "It is fine to keep up at 14.6V". When pushed on high voltage charging, he says "its OK" but listen to the way he says OK (around the 14 mins point in the video). It's more ...yea, you can get away with it. He doesnt mention anything about charge current but if his company are selling chargers that push up to 14.6V then you will be at the CV point when it reaches 14.6V and the charger needs to be left on at 14.6V until the tail current drops. He does say you can leave them at 14.6V overnight(!!!!!) which sounds really bad. Maybe his chargers cut back to zero current (is that possible?).

He pushes the point that the cells need to be balanced. At 14.6V, a balanced set of 4 cell will be at 3.65V. If they are 300mV out of balance (quite possible) then one cell could be 3.50V and another 3.80V. Holding a cell at 3.80V with current going in will knacker that cell pdq. He is assuming that holding at 14.6V is with balanced cells.

 

On his point about a float voltage of 13.6V, he is assuming there is a current draw.  I am happy holding mine at 13.6V but only IF there is current going out as well as in...ie solar charging with something on in the boat drawing power...ie the inverter powering the TV etc on a sunny afternoon. That means there is likely to be little net current going in.

 

I will go back to my original statement. He is trying to sell batteries......'of course there is no problem charging to 100% and keeping them there forever' and get an advantage over other battery sellers.

 

Have a look at the Tesla site. Tesla are selling cars that use Li batteries. The new Model 3 long range can do 320 miles of a full charge. The factory default for charge is 80%. Yes you can increase this to 90% or 100% but it is not recommended as the norm. Their Turbo chargers charge to 80% then back down big time. They spend a lot of time telling you not to go to 100% as a routine. This will be due to both battery cycle life and safety. Tesla know what they are doing as they are trying to sell cars (not batteries) and they warrant the batteries for 8 years - so they have to get the battery advice right ....unlike someone selling batteries.

 

So Richard, whilst you may be ok to keep your cells up in the knee for 24 hours to initially balance then, the advice still has to be keep below 100% as the norm for longer life and less need to balance. Go to the cruiser forum and read all the stuff on there. They are not making it up. Overcharging Li's is bad bad practice.

Edited January 8, 2020 by Dr Bob

[peterboat]

28 minutes ago, Dr Bob said:

It's an interesting video but this guy is trying to sell batteries! Of course he is going to try and push the boundaries. The fact is that LiFePo4s should not be charged when they are 100%. If you notice he does not say "It is fine to keep up at 14.6V". When pushed on high voltage charging, he says "its OK" but listen to the way he says OK (around the 14 mins point in the video). It's more ...yea, you can get away with it. He doesnt mention anything about charge current but if his company are selling chargers that push up to 14.6V then you will be at the CV point when it reaches 14.6V and the charger needs to be left on at 14.6V until the tail current drops. He does say you can leave them at 14.6V overnight(!!!!!) which sounds really bad. Maybe his chargers cut back to zero current (is that possible?).

He pushes the point that the cells need to be balanced. At 14.6V, a balanced set of 4 cell will be at 3.65V. If they are 300mV out of balance (quite possible) then one cell could be 3.50V and another 3.80V. Holding a cell at 3.80V with current going in will knacker that cell pdq. He is assuming that holding at 14.6V is with balanced cells.

 

On his point about a float voltage of 13.6V, he is assuming there is a current draw.  I am happy holding mine at 13.6V but only IF there is current going out as well as in...ie solar charging with something on in the boat drawing power...ie the inverter powering the TV etc on a sunny afternoon. That means there is likely to be little net current going in.

 

I will go back to my original statement. He is trying to sell batteries......'of course there is no problem charging to 100% and keeping them there forever' and get an advantage over other battery sellers.

 

Have a look at the Tesla site. Tesla are selling cars that use Li batteries. The new Model 3 long range can do 320 miles of a full charge. The factory default for charge is 80%. Yes you can increase this to 90% or 100% but it is not recommended as the norm. Their Turbo chargers charge to 80% then back down big time. They spend a lot of time telling you not to go to 100% as a routine. This will be due to both battery cycle life and safety. Tesla know what they are doing as they are trying to sell cars (not batteries) and they warrant the batteries for 8 years - so they have to get the battery advice right ....unlike someone selling batteries.

 

So Richard, whilst you may be ok to keep your cells up in the knee for 24 hours to initially balance then, the advice still has to be keep below 100% as the norm for longer life and less need to balance. Go to the cruiser forum and read all the stuff on there. They are not making it up. Overcharging Li's is bad bad practice.

I was more interested in the charging in the cold,  for my truck spends hours in the winter charging from solar at under 1 amp its also above freezing so I am reassured by this video.  The batteries he is selling however are well respected and very tough 

[Dr Bob]

1 minute ago, peterboat said:

I was more interested in the charging in the cold,  for my truck spends hours in the winter charging from solar at under 1 amp its also above freezing so I am reassured by this video.  The batteries he is selling however are well respected and very tough 

Yes, he seems to be towing the line on charging when cold. His technical explanation of electrons moving but Li ions not moving as fast is in line with all the current wisdom so Lithium plating can happen (and dendrite formation) but this is not as bad....or non existent.... when the current is very low (so your truck at 1 amp should be fine). He does comment that LA's are as bad at low temperature ...which they are in terms of capacity restriction etc.... but it is still possible to wreck Li's when charging a bit below freezing. For us in boats, it seems easy just to put them inside and if you leave the boat unheated for a while, then just heat them up a bit before you put charge in.

It was interested that he said 25°F (ie - 3°C ish) was the cut off. I think my Thunderskys are meant to be better than that as they have Yttrium in them. We left our boat for 2 weeks over Christmas and the cells went down to 5°C in that period. Since we have been back they have settled at 12°C (under the bed) whilst the boat basks in 25°C heat.

 

For your truck, if we get a cold spell and you end up at -4°C one morning, are you just discharging to run it or is there a charger putting charge into the batteries?

[peterboat]

2 minutes ago, Dr Bob said:

Yes, he seems to be towing the line on charging when cold. His technical explanation of electrons moving but Li ions not moving as fast is in line with all the current wisdom so Lithium plating can happen (and dendrite formation) but this is not as bad....or non existent.... when the current is very low (so your truck at 1 amp should be fine). He does comment that LA's are as bad at low temperature ...which they are in terms of capacity restriction etc.... but it is still possible to wreck Li's when charging a bit below freezing. For us in boats, it seems easy just to put them inside and if you leave the boat unheated for a while, then just heat them up a bit before you put charge in.

It was interested that he said 25°F (ie - 3°C ish) was the cut off. I think my Thunderskys are meant to be better than that as they have Yttrium in them. We left our boat for 2 weeks over Christmas and the cells went down to 5°C in that period. Since we have been back they have settled at 12°C (under the bed) whilst the boat basks in 25°C heat.

 

For your truck, if we get a cold spell and you end up at -4°C one morning, are you just discharging to run it or is there a charger putting charge into the batteries?

Just solar I do 30 miles a week in it and solar seems to cover it mostly if I know its going to be very cold I turn off solar 

[nicknorman]

I sent my battery monitor PCB design off to manufacture in China today. Double sided 100mm x 100mm with the usual solder masks, silk screen, plated through holes etc. and 24hr production time. Thought I would try JLCPCB and it turns out they had a special offer on for new customers. So the total cost for manufacturing 5 PCBs and shipping to U.K. via air mail (7-10 days) was... ...

 

 

£1.95
 

 

Ridiculous, and much cheaper than veroboard!

[Tom and Bex]

On 17/12/2019 at 09:55, nicknorman said:

It’s quite interesting that there is a “80%” cult on here. The members are categorically saying “oooh don’t go above 80% / don’t charge to 3.7v per cell” etc. But where is the evidence that this is a bad thing? Is it just an old wives’ (or spouses’, perhaps!) tale thing? We all know that if the voltage gets too high, the cells will be damaged. But the zeitgeist seems to be that not only must the max voltage not be exceeded, one must also leave a (randomly decided) large safety margin. Where is the evidence e for this strategy?

 

Surely one either has effective protection systems, or not. If one has, why leave a large safety margin? If not ... well one should do!

We're not obsessed with 80%. I consider we have fairly robust protection, and just charge as and when. On our recent 10 days cruise, we switched our alternator controller off every other day, so ended up charging to 100% every 2 days or so. From what I've read, charging to 100% is ok, so long as it's not held there which on a boat in constant use it won't be. Our charging is set to 14.0v, but batteries seem to not get there until over 80% so not seen the need to go any higher anyway. 

 

 

On 17/12/2019 at 12:54, MoominPapa said:

Your battery monitoring takes 500mA? Blimey! Base load on NB Melaleuca is 800mA and that's the BMS AND the internet router AND the back boiler pump AND the inverter standby AND the digital alarm clock AND three laptops in sleep mode AND the radio memory protect.

 

MP.

 

I wish we had a base load as low as that! Ours seems to be about 1-1.2A, and that's without the back boiler pump that draws about that on its own! And that base load does not include our monitoring and protection (BMV, BMM8v2, ISDT BC-8S, battery balancer, motorised switch, and bi-stable relay driver), all of which are connected prior to the shunt, but I suspect (hope!) are quite low. 

 

On 20/12/2019 at 17:52, Dr Bob said:

I agree with Peter. Just using voltage is the easiest way to track your battery performance. MPs system seems to have a very good correlation to SoC but that is very advanced for those of us just bolting bits of kit together. I rely on my BMV for battery monitoring and Ahr counting and whilst the Ahr counting is great for use on a day to day basis (so you can see you have used 120Ahrs overnight since the sun last shone) it is not good as an accurate Ahr out reading as it gets well out of sync  with the real position over 3 months. Yes, I could try and fine tune it over the 3 months by changing the settings but why bother when the voltage is the best indicator of battery SoC. I could chose to resync the BMV when the voltage gets to a point where I get to on my normal charge routine ie say 13.7v charging at 30A (which may be 90% ish) so it would say 100% when it is 90% but why? If you know your voltage limits then that is all you need.

We find our BMV gives fairly accurate soc even after 3 months. Last time when I watched it approach 100% after 3 months of not being near, the BMV reached 100% and I estimate the batteries took another 20-25Ah, so well within 10%. I did have to slightly tweak the settings the first couple times to get that though, but still worth it for reliable, easy to understand soc.

 

Mostly I don't really get hung up on what the batteries are doing, and just charge for 2-3 hrs every 2 days in winter. Not had any recent alarms so must be doing something right! The soc % on the isdt facing me as I go into the bedroom is a good general guide, if it's down in the 30's I know I need to charge soon, if in the 70's the batteries are fairly well charged! That's about all I need to know!

[MoominPapa]

21 minutes ago, Tom and Bex said:

I wish we had a base load as low as that! Ours seems to be about 1-1.2A, and that's without the back boiler pump that draws about that on its own!

 

My number doesn't include the boiler pump either. It's a Bolin and pulls about 0.3A, I think. Pity they're not made any longer.

Our inverter is at least 15 years old, but the standby current in energy-save mode is still remarkably good, even by modern standards, It's a Xantrex Prosine.

 

MP.

 

ETA. I see I claimed the 800mA included the pump. Maybe I was exaggerating just a little.

ETAM. I did measure the BMS consumption before it was installed. I think it was about 80mA

Edited January 11, 2020 by MoominPapa

[Dr Bob]

3 minutes ago, MoominPapa said:

It's a Xantrex Prosine.

 

 

That sounds like a rather painful medical procedure!

[MtB]

17 minutes ago, Dr Bob said:

That sounds like a rather painful medical procedure!

 

I thought it sounded like a car built by Citroen.....

 

 

[system 4-50]

7 hours ago, nicknorman said:

I sent my battery monitor PCB design off to manufacture in China today. Double sided 100mm x 100mm with the usual solder masks, silk screen, plated through holes etc. and 24hr production time. Thought I would try JLCPCB and it turns out they had a special offer on for new customers. So the total cost for manufacturing 5 PCBs and shipping to U.K. via air mail (7-10 days) was... ...

 

 

£1.95
 

 

Ridiculous, and much cheaper than veroboard!

Can you give a link or tell me that I should trawl back thru the thread?

[Sea Dog]

11 hours ago, nicknorman said:

I sent my battery monitor PCB design off to manufacture in China today. Double sided 100mm x 100mm with the usual solder masks, silk screen, plated through holes etc. and 24hr production time. Thought I would try JLCPCB and it turns out they had a special offer on for new customers. So the total cost for manufacturing 5 PCBs and shipping to U.K. via air mail (7-10 days) was... ...

 

 

£1.95
 

 

Ridiculous, and much cheaper than veroboard!

It can only be virtual slave labour or a loss leader designed to kill off the competition.  Since the postage alone has to be more, my money is on the latter, probably with some element of the former!

 

Nice work though, Nick. Fingers crossed for the outcome.

 

[nicknorman]

4 hours ago, system 4-50 said:

Can you give a link or tell me that I should trawl back thru the thread?

This is for my own design of battery monitor linked to smart alternator controller. Still in its infancy. So the £1.95 is just for the bare PCBs to my own design. Other folk on here have put stuff together using veroboard etc, so the point is just that if one can be bothered to design a pcb (using free software) one can get a really good and neat product for minimal financial outlay (but yes of course, many hours of design work). https://jlcpcb.com
 

 

[nicknorman]

8 minutes ago, Sea Dog said:

It can only be virtual slave labour or a loss leader designed to kill off the competition.  Since the postage alone has to be more, my money is on the latter, probably with some element of the former!

 

Nice work though, Nick. Fingers crossed for the outcome.

 

Cheap Chinese labour, not too worried about environment, economies of scale and investment in modern automated equipment I suspect. There are 3 or 4 such companies in China all offering a very similar international service. I had previously used Elecrow but I just missed the last order date before Chinese New Year (factory closed for a week) whereas JLCPCB is not closing for that holiday and I guess they have spare capacity as a result since the whole Far East pretty much shuts down for that holiday. So the 1/2 price offer on the PCB manufacture and the massive subsidy of nearly 100% on shipping costs is just a subsidy to try to win customers. So almost certainly a loss leader, just like many other special offers etc

 

And yes, let’s hope it works. I still have the software to write which will be the most onerous part of the project. However I have added some redundancy (eg provision for direct battery temperature and current monitoring, real time clock, in case my attempts to reverse engineer the Masterbus fails.)

 

I have also selected components with power down modes and my strategy will be to have stuff powered down /sleeping when charging is not taking place, waking up for a few mS every few seconds to check cell voltages. I am thus hoping for an average power consumption of under 1mA. When charging is taking place the interfaces will all be powered and so it will perhaps take 20mA or so, but that doesn’t matter.

[rusty69]

9 hours ago, Dr Bob said:

That sounds like a rather painful medical procedure!

I got one last month on the cheap. It was quite painful as a neutral-earth bond was required.This involved a bit of dismantling and much gnashing and wailing. Still, it's all healed now!

[Dr Bob]

1 hour ago, rusty69 said:

I got one last month on the cheap. It was quite painful as a neutral-earth bond was required.This involved a bit of dismantling and much gnashing and wailing. Still, it's all healed now!

Oh, that's good. Did you get the Mk3 or Mk4? The Mk4 comes with a upgrade this month to silver and and extra 50Gbs.

[rusty69]

3 minutes ago, Dr Bob said:

Oh, that's good. Did you get the Mk3 or Mk4? The Mk4 comes with a upgrade this month to silver and and extra 50Gbs.

I don't know that. I got a yellow one to match my new yellow drill!

[peterboat]

13 hours ago, Tom and Bex said:

We're not obsessed with 80%. I consider we have fairly robust protection, and just charge as and when. On our recent 10 days cruise, we switched our alternator controller off every other day, so ended up charging to 100% every 2 days or so. From what I've read, charging to 100% is ok, so long as it's not held there which on a boat in constant use it won't be. Our charging is set to 14.0v, but batteries seem to not get there until over 80% so not seen the need to go any higher anyway. 

 

 

I wish we had a base load as low as that! Ours seems to be about 1-1.2A, and that's without the back boiler pump that draws about that on its own! And that base load does not include our monitoring and protection (BMV, BMM8v2, ISDT BC-8S, battery balancer, motorised switch, and bi-stable relay driver), all of which are connected prior to the shunt, but I suspect (hope!) are quite low. 

 

We find our BMV gives fairly accurate soc even after 3 months. Last time when I watched it approach 100% after 3 months of not being near, the BMV reached 100% and I estimate the batteries took another 20-25Ah, so well within 10%. I did have to slightly tweak the settings the first couple times to get that though, but still worth it for reliable, easy to understand soc.

 

Mostly I don't really get hung up on what the batteries are doing, and just charge for 2-3 hrs every 2 days in winter. Not had any recent alarms so must be doing something right! The soc % on the isdt facing me as I go into the bedroom is a good general guide, if it's down in the 30's I know I need to charge soon, if in the 70's the batteries are fairly well charged! That's about all I need to know!

The problem is I believe companies like Tesla, GM and all the other electric car companies that stick to the 80% rule.  They want a claim free long life of the battery pack.  It's not like it really matters in real life operation, my solar system under normal day to day use just works.  If something goes wrong and it goes to low the whispergen starts up and brings the level of charge to a normal level.

[nicknorman]

Just a brief update:

1:  PCB manufactured 24 hrs after I sent the files off, now winging its way slowly back to UK

2: I have sussed the Masterbus (canbus), so now I can access various data including battery temperature, current, SoC, AH removed etc from the Mastershunt, and can view the parameters on the Combi such as whether shore power (or travelpower) is present, what the maximum charging current is set to, what the bulk, absorb and float voltages are. And more importantly, I can change these voltage and current settings. So my gadget will not only control the alternator but also control the Combi, should shore power or travel power be on. I'm chuffed!

Edited January 16, 2020 by nicknorman