Another Lithium battery thread

[Dr Bob]

6 minutes ago, MoominPapa said:

You don't need to switch the alternator off. Everything is connected to the lead acid battery, including the alternator and all the loads. When the Lithiums are full, the contactor isolates them from the lead acid side, so the lithiums stop charging. The LA is still loading to alternator, so it doesn't go bang, and continues to float the LA and support the loads. The BMS has to be able to sense the LA voltage, so it knows when the engine has stopped and the LA is no longer at float, then it closes the Lithium contactor again.

 

MP.

 

Ah, that doesnt seem to fit with what the author Eric says. He is adamant that BMS only isolates the lithium bank in the event of an over or under voltage event and the 'finsihed charging' event happens when the lithiums are full because the charge device reduces its charge itself. Isolation is just a failsafe if something goes wrong. At least, that's the way I read it......and I could easily be totally wrong!

[Dr Bob]

12 minutes ago, WotEver said:

And if I understand matters correctly it is actually detrimental to leave Lithiums at 100%, is that not correct?  So other than the occasional lift to 100% for top balancing (although Peter suggests this is rarely required) you’re actually better off never taking them to 100%. Yes?

 

Well that is what I had gleaned over the past 6 months and it's best being around the 50% mark.

However, in the blog post referenced, Eric the author is pushing the point that LiFePo's NEED to be taken to 100% every month or two otherwise they will loose capacity. I have never seen this written anywhere else. It looks like he has real experience of the systems and reading the posts at the end of the blog no one seemed to be questioning this point. He talks about a 'memory' effect which I thought had been dismissed in the past and the 100% was nothing to do with balancing. He did say taking them to 100% made them more sensitive to balancing.

If there is no problem in not going over 80% and only going to 100% to balance the system, then aiming for an upper limit of 13.8v on the solar MPPT and then restricting the alternator and IP22 to 13.8V seems a good idea. We are then back to the question of how to do this.

 

eta

he does say you should not stay at 100% - which I wouldnt do as our inverter is always on and the fridge is taking a load.

Edited January 6, 2019 by Dr Bob

[MoominPapa]

14 minutes ago, Dr Bob said:

Ah, that doesnt seem to fit with what the author Eric says. He is adamant that BMS only isolates the lithium bank in the event of an over or under voltage event and the 'finsihed charging' event happens when the lithiums are full because the charge device reduces its charge itself. Isolation is just a failsafe if something goes wrong. At least, that's the way I read it......and I could easily be totally wrong!

I couldn't find that on a quick check - can you point to a particular place, or offer a quote?

[Alway Swilby]

When we were having our boat built the builder suggested that we might like to think about lithium domestic batteries. The brand they were suggesting were Relion which are marketed as drop in replacements for LA. In the end we went for 3 x 24v 50ah in parralell. The batteries have thier own built in BMS and are connected to the alternator, the Victron inverter/charger and the solar MPPT controler just as if they were normal LA batteries. We also have a 24v bow thruster LA bank connected via a VSR to the same charging circuit so while charging the LA bank is also in circuit.  (There is a separate issue with the VSR in that because the lithium bank voltage never drops below 26v even when the engine has stopped it doesn't disconnect the LA bank but we are looking into a different VSR that will be programmable to disconnect at higher voltages.)

The shunt for the Victron BMV is only on the lithium bank. When first starting the engine in the morning about 50 to 60 amps go into the batteries (we have a 100 amp 24v alternator). This current drops fairly evenly over the course of a couple of hours down to about 10 amps by which time the BMV is reporting about 90-95% SOC (assuming it has been at about 50-60% when starting). The voltage by then is about 27.5v. If we are cruising and continue running the engine eventually the current drops to 2 or 3 amps but doesn't drop below  this figure. The voltage might creep up to 27.6 but not more than this.

This all seems to work fine and trouble free. We did have trouble getting hold of the batteries but that is another story. Eventually the batteries were installed in April 2018 and haven't blown up yet or died yet!

 

 

Edited January 6, 2019 by Alway Swilby

[MoominPapa]

8 minutes ago, Dr Bob said:

Well that is what I had gleaned over the past 6 months and it's best being around the 50% mark.

However, in the blog post referenced, Eric the author is pushing the point that LiFePo's NEED to be taken to 100% every month or two otherwise they will loose capacity. I have never seen this written anywhere else. It looks like he has real experience of the systems and reading the posts at the end of the blog no one seemed to be questioning this point. He talks about a 'memory' effect which I thought had been dismissed in the past and the 100% was nothing to do with balancing. He did say taking them to 100% made them more sensitive to balancing.

If there is no problem in not going over 80% and only going to 100% to balance the system, then aiming for an upper limit of 13.8v on the solar MPPT and then restricting the alternator and IP22 to 13.8V seems a good idea. We are then back to the question of how to do this.

 

eta

he does say you should not stay at 100% - which I wouldnt do as our inverter is always on and the fridge is taking a load.

I think the problem here may not be loss of capacity, but loss of knowledge of capacity. The voltage/state-of-charge graph for lithiums is very flat - they have pretty much  the same voltage at 25% charge as they do at 75% charge. BMS therefore mainly uses "coulomb counting" to determine state of charge. Count the amp-hours in, then count them out again. As lithiums are very energy-efficient, this works well, but like anything based on integrating noisy signals, it drifts over time. The best was to reset the Ah counter is to either fully charge or fully discharge the battery, as the voltage/State-of-charge graph does get a lot steeper as the cell approaches these two extremes.

 

MP.

 

[peterboat]

I can name two electric car makers that dont charge their banks up to 100% Tesla although it does have an override to do this and Vauxhall with its Ampera. Both these car makers have long lived battery banks, on forums I am on its suggested that  GM hasnt had failures at all, so for me I will stick to the 20-80% that James, JohnV and myself have decided is the best method of ensuring long life of the battery bank.

If I was going to do what you are planning Dr Bob I would put a Gell in parallel with the Lithium's as they best suit the LifePo4s charging regime, my Midnite solar controller is on gell and that limits at 13.8v dropping to 13.6v.

My disconnect is a cheap panel off amazon and the solenoids are the same for domestic. For electric drive as I have a solenoid and a reverse solenoid designed to operate at 69 volts from Allbright who make these things to order. both of these are at 10% over the maximum amps the motor controller runs at

[dmr]

37 minutes ago, Dr Bob said:

Well that is what I had gleaned over the past 6 months and it's best being around the 50% mark.

However, in the blog post referenced, Eric the author is pushing the point that LiFePo's NEED to be taken to 100% every month or two otherwise they will loose capacity. I have never seen this written anywhere else. It looks like he has real experience of the systems and reading the posts at the end of the blog no one seemed to be questioning this point. He talks about a 'memory' effect which I thought had been dismissed in the past and the 100% was nothing to do with balancing. He did say taking them to 100% made them more sensitive to balancing.

If there is no problem in not going over 80% and only going to 100% to balance the system, then aiming for an upper limit of 13.8v on the solar MPPT and then restricting the alternator and IP22 to 13.8V seems a good idea. We are then back to the question of how to do this.

 

eta

he does say you should not stay at 100% - which I wouldnt do as our inverter is always on and the fridge is taking a load.

There still appears to be some uncertainty and conflicting views about how to handle and charge lithiums. This is nothing new, there was loads of debates and theories (remember the memory effect?) when little lithiums first became popular, but I am surprised that they are not fully understood by now. The big lithiums going into boats are not really new technology, as far as I can see they are just made up of lots and lots of familiar little lithiums.

 

A big problem with making a DIY system is defining exactly what that system needs to actually do to keep the batteries happy, especially under unusual conditions....which "expert" do you believe?. An advantage with a ready made system, such as Victron, is that hopefully Victron do understand how to handle the batteries. 

 

Sharing knowledge on forums like this is obviously a good start.

 

...............Dave

[mrsmelly]

2 hours ago, Dr Bob said:

Well yes.....unless you are a muppet!

Following the links, it seems to be a diy self build and the links on the blog where it is described are a bit (lot) over the top for me. Now if it was an off the shelf item then no problem. I am only at the very basic level programming a Rasp Pie and this seems well beyond my skill level. Are there any off the shelf units with that functionality?

As these devices that cleverly shut off the charge are an electrical gadget does that mean when,  not if that particular gadget goes wrong which at some point it will,  do the very very very expensive batteries become damaged?

[Dr Bob]

1 hour ago, MoominPapa said:

I couldn't find that on a quick check - can you point to a particular place, or offer a quote?

No the reference is buried deep in the text so not at all easy to find!!!.....look at response 11 although if you read all his words  carefully, that is the way he is setting it up. Here are 2 quotes from his response 11 (3/4 of the way down the blog)

First of all, the battery disconnection scheme is a safety system that should never operate, not a way of controlling charging. Relying on HVD to terminate the charge of the lithium would be both bad design and poor in terms of performance, because the lithium would get no absorption time and this can represent around 30 minutes when charging with a lot of current, like what you get when running two engines and alternators in parallel. The consequence of this is that you still have to control your voltage properly, from all sources. This can require changes and investment.

......snip.......

The reason why the manual system has been working is because it has been managed manually AND all the charging voltages were adjusted. The owner has to manage it this way because it can’t be automated: he used drop-ins and the BMS is of no help. With a proper BMS, there are additional things you can do, but implementing charge termination means creating a way for the BMS to get the charge controllers to cut out and this is also a little engineering project. Regardless of what you do, once you are charging lithium cells you have to make all your charging voltages and charging profiles acceptable for the lithium. Battery charging requires voltage regulation, you can’t do it by pulling the plug.

 

I think that is why the isolation device is only a safety system and not the normal way of stopping a charge.

[Johny London]

So we have batteries with internal bms (for cell balancing) or single cells with nothing. Plus we then have a BMS over all the batteries. And we set all out various charge controllers to appropriate settings (though the alternator seems to be the most tricky). And lastly we have "emergency" cut offs for high, low and possibly temp.

 

Those "drop in" Relions are looking more and more like a bargain. I wonder if Minimalists are still going ok - anyone know? 

[Dr Bob]

10 minutes ago, mrsmelly said:

As these devices that cleverly shut off the charge are an electrical gadget does that mean when,  not if that particular gadget goes wrong which at some point it will,  do the very very very expensive batteries become damaged?

The way Eric's document is laid out is that the charge devices throttle themsleves back when the Li battery gets full. If however something fails and starts to overcharge, then the over or undervoltage device cuts in and isolates the battery. Therefore there are two levels of protection. I would anticipate even in the design of the most simple system, you would have a separate audible alarm for high and low voltage ( Tom & Bex have an audible alarm ) so a 3rd level of protection. Two things would have to go wrong at the same time to wreck the batteries.

You asked earlier about cost. I'm looking to get used Lithium's and want to pay less than a grand for them rather than the £4-5K if you went to someone like Victron for an idiot proof system that likely has 4 or 5 levels of protection. To me the biggest risk is on the charging side and I am usually on the boat when charging so maybe manageable.

[nicknorman]

1 hour ago, MoominPapa said:

You don't need to switch the alternator off. Everything is connected to the lead acid battery, including the alternator and all the loads. When the Lithiums are full, the contactor isolates them from the lead acid side, so the lithiums stop charging. The LA is still loading to alternator, so it doesn't go bang, and continues to float the LA and support the loads. The BMS has to be able to sense the LA voltage, so it knows when the engine has stopped and the LA is no longer at float, then it closes the Lithium contactor again.

 

MP.

 

IMO this is a fairly inelegant way to do things. You are adding in a LA battery simply to cater for poor system design. Much better (IMO) to ditch the LA and have a smart alternator controller which, when the Lithiums are at whatever SoC you choose (say 80%), reduces the output voltage to one which causes no further current to flow into the Lithiums (I am loath to call it float, but that is what it is!). That way the boat’s systems and services are still powered by the alternator, and there is no need for clunky relays etc. Well there is of course, but these are only for emergency isolation if an upper or lower charge limit is exceeded. Ie the ultimate protection under fault conditions, not the routine charging control.

[MoominPapa]

12 minutes ago, Dr Bob said:

No the reference is buried deep in the text so not at all easy to find!!!.....look at response 11 although if you read all his words  carefully, that is the way he is setting it up. Here are 2 quotes from his response 11 (3/4 of the way down the blog)

First of all, the battery disconnection scheme is a safety system that should never operate, not a way of controlling charging. Relying on HVD to terminate the charge of the lithium would be both bad design and poor in terms of performance, because the lithium would get no absorption time and this can represent around 30 minutes when charging with a lot of current, like what you get when running two engines and alternators in parallel. The consequence of this is that you still have to control your voltage properly, from all sources. This can require changes and investment.

......snip.......

The reason why the manual system has been working is because it has been managed manually AND all the charging voltages were adjusted. The owner has to manage it this way because it can’t be automated: he used drop-ins and the BMS is of no help. With a proper BMS, there are additional things you can do, but implementing charge termination means creating a way for the BMS to get the charge controllers to cut out and this is also a little engineering project. Regardless of what you do, once you are charging lithium cells you have to make all your charging voltages and charging profiles acceptable for the lithium. Battery charging requires voltage regulation, you can’t do it by pulling the plug.

 

I think that is why the isolation device is only a safety system and not the normal way of stopping a charge.

First para. Agreed, but I think what he means is that you shouldn't rely on detecting the conditions for HVD for normal charge termination. NOT that the mechanism has to be different. It's fine to detect normal "charge full" and use that signal to open the Lithium-LA paralleling relay. The action is the same, but the conditions that cause it are different.

 

Second para. The crucial thing here is that charge terminal depends on charge current as well as voltage - this means that, for instance, a conventional alternator controller, which doesn't know the current, can't do the business. IThe Arduino controller that Tom and Bex use does measure battery current).

 

MP.

 

 

[nicknorman]

21 minutes ago, mrsmelly said:

As these devices that cleverly shut off the charge are an electrical gadget does that mean when,  not if that particular gadget goes wrong which at some point it will,  do the very very very expensive batteries become damaged?

This is why one has a normal control system to regulate day to day charging, and a separate “emergency” system to isolate the battery when things go wrong. The latter should never come on to play except when the main system goes faulty.

[Dr Bob]

37 minutes ago, dmr said:

There still appears to be some uncertainty and conflicting views about how to handle and charge lithiums. This is nothing new, there was loads of debates and theories (remember the memory effect?) when little lithiums first became popular, but I am surprised that they are not fully understood by now. The big lithiums going into boats are not really new technology, as far as I can see they are just made up of lots and lots of familiar little lithiums.

 

A big problem with making a DIY system is defining exactly what that system needs to actually do to keep the batteries happy, especially under unusual conditions....which "expert" do you believe?. An advantage with a ready made system, such as Victron, is that hopefully Victron do understand how to handle the batteries. 

 

Sharing knowledge on forums like this is obviously a good start.

 

...............Dave

I agree. For me, paying £5K to go down the Victron route is a non starter. I'd sooner buy a another big lens for the camera. However building a simple system for £1K (as in T&B's post) is attractive.

[peterboat]

6 minutes ago, Johny London said:

So we have batteries with internal bms (for cell balancing) or single cells with nothing. Plus we then have a BMS over all the batteries. And we set all out various charge controllers to appropriate settings (though the alternator seems to be the most tricky). And lastly we have "emergency" cut offs for high, low and possibly temp.

 

Those "drop in" Relions are looking more and more like a bargain. I wonder if Minimalists are still going ok - anyone know? 

 

2 minutes ago, nicknorman said:

IMO this is a fairly inelegant way to do things. You are adding in a LA battery simply to cater for poor system design. Much better (IMO) to ditch the LA and have a smart alternator controller which, when the Lithiums are at whatever SoC you choose (say 80%), reduces the output voltage to one which causes no further current to flow into the Lithiums (I am loath to call it float, but that is what it is!). That way the boat’s systems and services are still powered by the alternator, and there is no need for clunky relays etc. Well there is of course, but these are only for emergency isolation if an upper or lower charge limit is exceeded. Ie the ultimate protection under fault conditions, not the routine charging control.

 

5 minutes ago, Dr Bob said:

The way Eric's document is laid out is that the charge devices throttle themsleves back when the Li battery gets full. If however something fails and starts to overcharge, then the over or undervoltage device cuts in and isolates the battery. Therefore there are two levels of protection. I would anticipate even in the design of the most simple system, you would have a separate audible alarm for high and low voltage ( Tom & Bex have an audible alarm ) so a 3rd level of protection. Two things would have to go wrong at the same time to wreck the batteries.

You asked earlier about cost. I'm looking to get used Lithium's and want to pay less than a grand for them rather than the £4-5K if you went to someone like Victron for an idiot proof system that likely has 4 or 5 levels of protection. To me the biggest risk is on the charging side and I am usually on the boat when charging so maybe manageable.

Yes us minimalists are doing fine, JohnV has LAs in his system and an alternator plus solar with no issues on charging at all. James is pure solar/whispergen same as me, so we have no alternator problems, both solar and whispergen have set voltages to avoid overcharging, and the whispergen has an automatic start if under voltage is sensed so maybe we do have technology to help us

[Dr Bob]

3 minutes ago, MoominPapa said:

First para. Agreed, but I think what he means is that you shouldn't rely on detecting the conditions for HVD for normal charge termination. NOT that the mechanism has to be different. It's fine to detect normal "charge full" and use that signal to open the Lithium-LA paralleling relay. The action is the same, but the conditions that cause it are different.

 

Second para. The crucial thing here is that charge terminal depends on charge current as well as voltage - this means that, for instance, a conventional alternator controller, which doesn't know the current, can't do the business. IThe Arduino controller that Tom and Bex use does measure battery current).

 

MP.

 

 

I agree with you but I'd sooner have a 'smart' alternator controller that reduced voltage rather than a clunky relay that operated every time the batteries were full.

Are there any other alternator controllers commercially available that work like the Arduino controller?

[MoominPapa]

3 minutes ago, nicknorman said:

IMO this is a fairly inelegant way to do things. You are adding in a LA battery simply to cater for poor system design. Much better (IMO) to ditch the LA and have a smart alternator controller which, when the Lithiums are at whatever SoC you choose (say 80%), reduces the output voltage to one which causes no further current to flow into the Lithiums (I am loath to call it float, but that is what it is!). That way the boat’s systems and services are still powered by the alternator, and there is no need for clunky relays etc. Well there is of course, but these are only for emergency isolation if an upper or lower charge limit is exceeded. Ie the ultimate protection under fault conditions, not the routine charging control.

 

The LA battery is the engine start battery, so it's not an extra component. Doing it this way does have some significant advantages.

 

1) It doesn't require BMS control of all the charge sources.

2) It allows the engine start battery to power the loads when the Lithiums do discharge termination. Much better than a sudden lights-out.

3) It allows the lithiums to be stored whilst we have mains over the winter at 40% to 50% charge. We simply discharge to that state, then do a manual disconnect and turn on the battery charger. The lithiums stay at 40% and we live on the engine start battery floated by the charger.

4) As you say, the system needs relays anyway. The more complex system need two. This system uses one. The relays are the most expensive single component apart from the cells.

 

 MP.

 

4 minutes ago, Dr Bob said:

I agree with you but I'd sooner have a 'smart' alternator controller that reduced voltage rather than a clunky relay that operated every time the batteries were full.

Are there any other alternator controllers commercially available that work like the Arduino controller?

You're going to need the clunky relay for emergency disconnect anyway. By definition when the batteries are full it's not going to be breaking a large current.

 

MP.

 

[nicknorman]

9 minutes ago, MoominPapa said:

 

The LA battery is the engine start battery, so it's not an extra component. Doing it this way does have some significant advantages.

 

1) It doesn't require BMS control of all the charge sources.

2) It allows the engine start battery to power the loads when the Lithiums do discharge termination. Much better than a sudden lights-out.

3) It allows the lithiums to be stored whilst we have mains over the winter at 40% to 50% charge. We simply discharge to that state, then do a manual disconnect and turn on the battery charger. The lithiums stay at 40% and we live on the engine start battery floated by the charger.

4) As you say, the system needs relays anyway. The more complex system need two. This system uses one. The relays are the most expensive single component apart from the cells.

 

 MP.

 

Ok well I can see the point to some extent, however if the relay is in continuous use to avoid overcharging, what happens when the contacts weld up etc? Surely there should be an “ultimate protection” system that isn’t normally in operation. If you just have the one relay controlling both normal and emergency operation, you don’t really have backup/emergency protection.

 

And in our case we already have two alternators, one small one for the starter battery and a large one for the domestics. So it would require a fairly significant reconfigure. Also, what about alternator roasting? If you have a system whereby the alternator is going to be at full output for a few hours, how long is it going to last? Certainly with our Iskra 175A alternator I think I would want some sort of current limiting when it got hot.

Edited January 6, 2019 by nicknorman

[Dr Bob]

52 minutes ago, peterboat said:

 

 

Yes us minimalists are doing fine, JohnV has LAs in his system and an alternator plus solar with no issues on charging at all. James is pure solar/whispergen same as me, so we have no alternator problems, both solar and whispergen have set voltages to avoid overcharging, and the whispergen has an automatic start if under voltage is sensed so maybe we do have technology to help us

Peter, have you got a link to JohnV's system? He's the only one of the bunch that I've not read their stories. Interested in how he solves his alternator charging problem.

[MoominPapa]

53 minutes ago, nicknorman said:

Ok well I can see the point to some extent, however if the relay is in continuous use to avoid overcharging, what happens when the contacts weld up etc? Surely there should be an “ultimate protection” system that isn’t normally in operation. If you just have the one relay controlling both normal and emergency operation, you don’t really have backup/emergency protection.

How much redundancy do you want, or need? If the relay is only used under error conditions, you don't find that the contacts are welded up until you really need it. If it's used routinely, then you find out the problem before it's needed for ultimate protection. The BMS measures the current through the relay, so it can alarm if the relay is commanded to open and the current doesn't fall to zero. 

 

53 minutes ago, nicknorman said:

 

And in our case we already have two alternators, one small one for the starter battery and a large one for the domestics. So it would require a fairly significant reconfigure. Also, what about alternator roasting? If you have a system whereby the alternator is going to be at full output for a few hours, how long is it going to last? Certainly with our Iskra 175A alternator I think I would want some sort of current limiting when it got hot.

 

We have a single 70A A127 in a well ventilated engine room. It can be run at maximum output for hours on end charging the current LA bank, and I've not seen any problems, so I don't anticpate any.

 

MP.

 

 

[nicknorman]

13 minutes ago, MoominPapa said:

How much redundancy do you want, or need? If the relay is only used under error conditions, you don't find that the contacts are welded up until you really need it. If it's used routinely, then you find out the problem before it's needed for ultimate protection. The BMS measures the current through the relay, so it can alarm if the relay is commanded to open and the current doesn't fall to zero. 

 

 

We have a single 70A A127 in a well ventilated engine room. It can be run at maximum output for hours on end charging the current LA bank, and I've not seen any problems, so I don't anticpate any.

 

MP.

 

 

Contacts weld up from opening and closing and making an arc. They don’t weld up when the contacts remain closed.

 

Your earlier argument had merit. This one doesn’t really, although I take the point about the alarm with mismatch between current and relay commanded state. However alarms are only any good if there is someone there to hear it and additionally to be in a position to do something about it.

 

But anyway, you still have a simplex system. One BMS and one relay to control both routine charging and emergency conditions.

 

my 175A Iskra is not that much bigger than your 70A A127 and so I think is more “highly strung”. It is fairly hard to have a 175A alternator working hard for a long period, but quite normal to have a 70A alternator working flat out for a long period. I suspect this is taken into consideration when thermal loads are being considered at the design stage.

[dmr]

Reply to various post above....

We have the 100amp Iskra alternator which is a smaller frame than the 175. We've run this for about 5 years now with only one replacement (slip rings worn out) so it is able to run flat out for longish periods, but I am amazed it does this, it runs much too hot and smells quite bad some days.

 

I've put a fair bit of thought into alternator controllers over the last few years but not got motivated enough to make anything myself, my feeling is that all this relay-disconnect stuff is Not the way to go, and a well thought out Arduino (or equivalent) bespoke alternator controller Is the way to go (with the inbult regulator removed). I really like my Adverc but its all done in hardware, with a bit of processor power and knowledge it should be possible to make a really good controller.

 

..............Dave 

[MoominPapa]

1 hour ago, nicknorman said:

Contacts weld up from opening and closing and making an arc. They don’t weld up when the contacts remain closed.

 

Your earlier argument had merit. This one doesn’t really, although I take the point about the alarm with mismatch between current and relay commanded state. However alarms are only any good if there is someone there to hear it and additionally to be in a position to do something about it.

 

But anyway, you still have a simplex system. One BMS and one relay to control both routine charging and emergency conditions.

 

my 175A Iskra is not that much bigger than your 70A A127 and so I think is more “highly strung”. It is fairly hard to have a 175A alternator working hard for a long period, but quite normal to have a 70A alternator working flat out for a long period. I suspect this is taken into consideration when thermal loads are being considered at the design stage.

The relay I'm using is specced to 70,000 operations at 100A, so I'm happy it will be reliable.

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

 

The BMS electronics has a fail-safe (relay open) watchdog feature if the CPU fails or crashes.

 

I suspect you're right that your alternator will need a BMS capable of dialing back the output for thermal management.  I know from experience that mine doesn't.

 

MP.

 

 

 

 

[nicknorman]

10 minutes ago, MoominPapa said:

The relay I'm using is specced to 70,000 operations at 100A, so I'm happy it will be reliable.

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

 

The BMS electronics has a fail-safe (relay open) watchdog feature if the CPU fails or crashes.

 

I suspect you're right that your alternator will need a BMS capable of dialing back the output for thermal management.  I know from experience that mine doesn't.

 

MP.

 

I am fairly cynical when it comes to relying on specifications for critical systems. I flew helicopters which were designed to very high standards and with a critical fault rate of 10^-9 but they still went wrong often!

 

As I understand it the relay has to be powered on or powered off. Thus the BMS electronics aren’t really fail safe. For example a simple wiring disconnect between the BMS and the relay would cause the relay to remain closed when it should be open. And whilst that might cause the BMS to alarm, a more complex failure of the BMS electronics could cause a failure of both the relay to open, and the alarm. A watchdog is only about a software crash due to bad code, not about hardware failures.