Jump to content

Another Lithium battery thread


Dr Bob

Featured Posts

21 minutes ago, peterboat said:

I know you are all looking at ways of controlling the alternator, but surely my cheap 6 squids board from Amazon, conected to a relay would do the job of turning off the alternator? Set it for 14 volts at the battery, when thats achieved it disconnects the alternator, use something in the boat batteries drop below 14 volts, it starts charging again simples, and very cheap as the relay wouldnt be taking much load

Yes, but I was hoping to operate each day without a relay turning off the Lithiums as it will eventually fail. We discussed earlier in the thread about how to 'stop' an alternator and the safe way if using a hybrid system is to just activate the lithium bank disconnect. Again I dont want to have to do this daily but think my AtoB will shut down before I get to 100%.

Nicks point about crusing all day and depleting the batteries - that to me is not a problem as I will only disconnect as I near 100% and then really want to get it down to 80% for safety....which a few hours boating will help.

Link to comment
Share on other sites

On ‎07‎/‎01‎/‎2019 at 17:10, Tom and Bex said:

 Yes should do, will try to get them over to you later. 

 

Did you recompile the Arduino code? I am getting a compile error "


'average_EPC_utilization' was not declared in this scope"

 

My god the Arduino IDE is caveman-like compared to Microchip's IDE. Truly crap!

 

Did I mention I hate Arduinos?

 

Edited by nicknorman
Link to comment
Share on other sites

10 hours ago, Dr Bob said:

 From your data and my experience of the AtoB, I may just get away with using the AtoB. If not I will turn the Li's off manually short term and get a proper alternator controller.

Worth trying, but wouldn't be so sure. My guess is it would spend a long time in bulk, and therefore a long time in absorption.

 

8 hours ago, MoominPapa said:

I have briefly considered if it has the making of a product, and come to the conclusion that it might, but there'l be an awful lot more engineering to do. I don't mind editing source code and re-compiling to set parameters. A buyer wouldn't want to do that. In fact you'd have to make the thing entirely independent of a PC, or face supporting PC software too (on a Macintosh, or Windows! "Just run Ubuntu" wouldn't hack it.) 

 

My plan at the moment s to use a "tractor regulator" set to 14.0v and sensing from the disconnect contactor to control the alternator charging. That's rather A127 specific, so you'd need to include some elements of the Arduino regulator to make the thing alternator generic.

 

However if you want source and some hand-drawn diagrams, you're welcome :)

 

Cheers,

 

MP.

Sounds like an interesting project. Could make a useful upgrade in the future, so if you don't mind sharing that would be great. Still way out of my depth, but a good learning experience for me! 

 

6 hours ago, MoominPapa said:

I'm hoping that this voltage number will be good enough to be able to look up in a table giving SOC versus voltage and get a useful SOC number which can be used to counteract long-term drift in the SOC derived from coulomb counting, I don't anticipate the batteries being at 100% or 0% very often to reset the boundary conditions on the integration.

 

MP.

The problem with voltage based soc is how flat the discharge curve is. Be interesting to see how this works out. On our setup, we found the Victron BMV ah counter would drift if charge efficiency factor was set to 99% and give slightly high soc readings after several weeks, however with CEF set to 98% we have slightly low readings after a few weeks. That's one reason for us charging to 100% every 6 weeks or so - to re-synchronise the bmv.

 

4 hours ago, Dr Bob said:

It looks like to control LiFePo's you will need the WS500 regulator which looks like it will be available next month. No pricing data but the WS200 is £250 ish - so the WS500 likely to be more. It certainly looks like it is perfect for the job if a bit expensive and I'd need to take the alternator apart (I assume) to fit. Looks like a plan B. Balmar seem to have something similar (very quick look) but with less functionality for £500 ish.

I'm glad we got ours when we did now! From memory it was only around $100 including shipping.

 

3 hours ago, peterboat said:

I know you are all looking at ways of controlling the alternator, but surely my cheap 6 squids board from Amazon, conected to a relay would do the job of turning off the alternator? Set it for 14 volts at the battery, when thats achieved it disconnects the alternator, use something in the boat batteries drop below 14 volts, it starts charging again simples, and very cheap as the relay wouldnt be taking much load

Something like this could work quite well for lithium/lead hybrid installs. Set to isolate the lithium bank at 14v, and have it sensing from the lead acid. That way loads would still be provided by alternator whilst engine running, and the lithiums would take over when engine stopped as voltage would then drop below 14v. No discharge of the lead acid, and depending on alternator output vs lithium bank size, cuts charging around 80%. 

 

This setup, with a similar low voltage disconnect set to say 12.8, and cell monitoring with alarms such as ISDT BC-8S that we've used, could make for a fairly cheep but safe lithium install @Dr Bob.

 

 

10 hours ago, Dr Bob said:

 From your data and my experience of the AtoB, I may just get away with using the AtoB. If not I will turn the Li's off manually short term and get a proper alternator controller.

Worth trying, but wouldn't be so sure. My guess is it would spend a long time in bulk, and therefore a long time in absorption.

 

8 hours ago, MoominPapa said:

I have briefly considered if it has the making of a product, and come to the conclusion that it might, but there'l be an awful lot more engineering to do. I don't mind editing source code and re-compiling to set parameters. A buyer wouldn't want to do that. In fact you'd have to make the thing entirely independent of a PC, or face supporting PC software too (on a Macintosh, or Windows! "Just run Ubuntu" wouldn't hack it.) 

 

My plan at the moment s to use a "tractor regulator" set to 14.0v and sensing from the disconnect contactor to control the alternator charging. That's rather A127 specific, so you'd need to include some elements of the Arduino regulator to make the thing alternator generic.

 

However if you want source and some hand-drawn diagrams, you're welcome :)

 

Cheers,

 

MP.

Sounds like an interesting project. Could make a useful upgrade in the future, so if you don't mind sharing that would be great. Still way out of my depth, but a good learning experience for me! 

 

6 hours ago, MoominPapa said:

I'm hoping that this voltage number will be good enough to be able to look up in a table giving SOC versus voltage and get a useful SOC number which can be used to counteract long-term drift in the SOC derived from coulomb counting, I don't anticipate the batteries being at 100% or 0% very often to reset the boundary conditions on the integration.

 

MP.

The problem with voltage based soc is how flat the discharge curve is. Be interesting to see how this works out. On our setup, we found the Victron BMV ah counter would drift if charge efficiency factor was set to 99% and give slightly high soc readings after several weeks, however with CEF set to 98% we have slightly low readings after a few weeks. That's one reason for us charging to 100% every 6 weeks or so - to re-synchronise the bmv.

 

4 hours ago, Dr Bob said:

It looks like to control LiFePo's you will need the WS500 regulator which looks like it will be available next month. No pricing data but the WS200 is £250 ish - so the WS500 likely to be more. It certainly looks like it is perfect for the job if a bit expensive and I'd need to take the alternator apart (I assume) to fit. Looks like a plan B. Balmar seem to have something similar (very quick look) but with less functionality for £500 ish.

I'm glad we got ours when we did now! From memory it was only around $100 including shipping.

 

3 hours ago, peterboat said:

I know you are all looking at ways of controlling the alternator, but surely my cheap 6 squids board from Amazon, conected to a relay would do the job of turning off the alternator? Set it for 14 volts at the battery, when thats achieved it disconnects the alternator, use something in the boat batteries drop below 14 volts, it starts charging again simples, and very cheap as the relay wouldnt be taking much load

Something like this could work quite well for lithium/lead hybrid installs. Set to isolate the lithium bank at 14v, and have it sensing from the lead acid. That way loads would still be provided by alternator whilst engine running, and the lithiums would take over when engine stopped as voltage would then drop below 14v. No discharge of the lead acid, and depending on alternator output vs lithium bank size, cuts charging around 80%. 

 

This setup, with a similar low voltage disconnect set to say 12.8, and cell monitoring with alarms such as ISDT BC-8S that we've used, could make for a fairly cheep but safe lithium install @Dr Bob.

 

 

  • Greenie 1
Link to comment
Share on other sites

3 hours ago, nicknorman said:

Did you recompile the Arduino code? I am getting a compile error "


'average_EPC_utilization' was not declared in this scope"

 

My god the Arduino IDE is caveman-like compared to Microchip's IDE. Truly crap!

 

Did I mention I hate Arduinos?

 

Think that's the original code he sent me. I did modify it to use feature out as a battery combiner. Can't remember getting any errors, will get the laptop out tomorrow and have a play (need to modify end of charge anyway, would like tail current a bit higher to keep further away from knee on charge curve).

Link to comment
Share on other sites

2 hours ago, Tom and Bex said:

Something like this could work quite well for lithium/lead hybrid installs. Set to isolate the lithium bank at 14v, and have it sensing from the lead acid. That way loads would still be provided by alternator whilst engine running, and the lithiums would take over when engine stopped as voltage would then drop below 14v. No discharge of the lead acid, and depending on alternator output vs lithium bank size, cuts charging around 80%. 

 

This setup, with a similar low voltage disconnect set to say 12.8, and cell monitoring with alarms such as ISDT BC-8S that we've used, could make for a fairly cheep but safe lithium install @Dr Bob.

 

 

Thanks for all of that T&B. Have a greenie.

Yes, great info. I like the idea of the 12.8V and 14.0v cutoffs plus the cell monitoring. A cheap solution and worth a try. I hadnt thought of sensing from the lead acid but now that seems obvious so it connects again.

 @peterboat can you give me info on your 6 squid controller from Amazon? That sounds realy useful.

Link to comment
Share on other sites

5 hours ago, nicknorman said:

Well 2 things, firstly what is the hysteresis of your sick squid board? If it shuts off at 14v, the battery voltage will rapidly drop to say 13.6v and then if the relay closes again, you will just have a chattering relay.

 

And secondly for people who cruise their boats a lot, once the battery was fully charged and the relay opened, for the rest of the cruising day the battery would be depleted by the boat's loads. Well unless you have a LA battery as well, but that seems a fudge to me.

Nothing, for most boats. The alternator remains operating until it stops (or nearly stops) spinning

But its cheap, and LifePo4s dont drop that fast, and if its needed the LA gives you that protection, also some alternators stop charging when the key is turned off.

Link to comment
Share on other sites

46 minutes ago, Dr Bob said:

Thanks for all of that T&B. Have a greenie.

Yes, great info. I like the idea of the 12.8V and 14.0v cutoffs plus the cell monitoring. A cheap solution and worth a try. I hadnt thought of sensing from the lead acid but now that seems obvious so it connects again.

 @peterboat can you give me info on your 6 squid controller from Amazon? That sounds realy useful.

No probs Bob I will have a look for it tomorrow if thats ok?

  • Greenie 1
Link to comment
Share on other sites

3 hours ago, Tom and Bex said:

The problem with voltage based soc is how flat the discharge curve is. Be interesting to see how this works out. On our setup, we found the Victron BMV ah counter would drift if charge efficiency factor was set to 99% and give slightly high soc readings after several weeks, however with CEF set to 98% we have slightly low readings after a few weeks. That's one reason for us charging to 100% every 6 weeks or so - to re-synchronise the bmv.

 

The plan is to use the voltage to "nudge" the SOC: most of the tracking of changes will be done integrating the current, but if the voltage is consistently giving a lower or higher answer, the SOC can be trimmed up or down as needed. I think I need to get the real batteries in place and record a few complete discharge cycles first, to see if I'm generating data which has useful information in it. There's plenty of storage in the computer to log stuff.

 

One problem I may have with coulomb counting is that I'm using a hall-effect current  sensor, not a shunt. It's pretty accurate, but can suffer a small persistent offset due to induced magnetism after large current excursions. The spec says this is 300mA max (in plus/minus 200A) but it does represent some drift. Time will tell.

 

480Ah bank is arriving tomorrow!

 

MP.

 

  • Greenie 2
Link to comment
Share on other sites

26 minutes ago, MoominPapa said:

 

One problem I may have with coulomb counting is that I'm using a hall-effect current  sensor, not a shunt. It's pretty accurate, but can suffer a small persistent offset due to induced magnetism after large current excursions. The spec says this is 300mA max (in plus/minus 200A) but it does represent some drift. Time will tell.

 

MP.

 

Are you using an open loop or closed loop (nulling) hall effect sensor? I've never used them but I would have thought a closed loop sensor whereby an opposite magnetic field is generated to create a null (which is all that the hall effect sensor then detects) might not suffer from magnetisation issues since the net magnetic field is zero.

 

https://uk.rs-online.com/web/p/current-transducers/1883783/

 

Edited by nicknorman
Link to comment
Share on other sites

9 hours ago, nicknorman said:

Are you using an open loop or closed loop (nulling) hall effect sensor? I've never used them but I would have thought a closed loop sensor whereby an opposite magnetic field is generated to create a null (which is all that the hall effect sensor then detects) might not suffer from magnetisation issues since the net magnetic field is zero.

 

https://uk.rs-online.com/web/p/current-transducers/1883783/

 

I'm using this

 

https://uk.farnell.com/allegro-microsystems/acs758ecb-200b-pff-t/ic-current-sensor-200a-3cb/dp/1791391

 

which I guess is open loop, since there's not room in the package for an opposing-field electromagnet. 

 

The integrated package devices seem to be more accurate, but the ones you link to would certainly by more convenient to use. Interestingly, it seems that all the closed-loop ones a single current direction, and to get a sensor which reads in both directions, important for battery monitoring, you need to go to open loop. I don'r understand why that should be.

 

 

 

MP.

 

Link to comment
Share on other sites

9 hours ago, MoominPapa said:

 

The plan is to use the voltage to "nudge" the SOC: most of the tracking of changes will be done integrating the current, but if the voltage is consistently giving a lower or higher answer, the SOC can be trimmed up or down as needed. I think I need to get the real batteries in place and record a few complete discharge cycles first, to see if I'm generating data which has useful information in it. There's plenty of storage in the computer to log stuff.

 

One problem I may have with coulomb counting is that I'm using a hall-effect current  sensor, not a shunt. It's pretty accurate, but can suffer a small persistent offset due to induced magnetism after large current excursions. The spec says this is 300mA max (in plus/minus 200A) but it does represent some drift. Time will tell.

 

480Ah bank is arriving tomorrow!

 

MP.

 

What batteries are you getting and where from? Apologies if you mentioned this earlier in the thread.

Link to comment
Share on other sites

1 hour ago, MoominPapa said:

I'm using this

 

https://uk.farnell.com/allegro-microsystems/acs758ecb-200b-pff-t/ic-current-sensor-200a-3cb/dp/1791391

 

which I guess is open loop, since there's not room in the package for an opposing-field electromagnet. 

 

The integrated package devices seem to be more accurate, but the ones you link to would certainly by more convenient to use. Interestingly, it seems that all the closed-loop ones a single current direction, and to get a sensor which reads in both directions, important for battery monitoring, you need to go to open loop. I don'r understand why that should be.

 

MP.

 

As far as I can tell ( because the datasheet is a bit sparse) these sensors are bidirectional. Quoted current range is +-200A and it mentions being suitable for AC. I suppose the question not directly answered in the data sheet is whether the output indicates the current flow direction. But since the nulling coil’s current must be bidirectional it seems likely that the output is too.

 

It is of course a lot more expensive than the sensor you are using.

 

edit:definitely bidirectional, you can see from the schematic that the output can swing both ways. Downside is that it needs a + and - power supply.

 

12898B9D-0DE3-48BF-AB62-C8FFD0B90FF7.png.669e85a132209955f20aad608bb2c99e.png

Edited by nicknorman
Link to comment
Share on other sites

measuring up to 200A is fine for charging but if you have a big inverter that plus a pump start can easily exceed 300 A. What happens on the hall effect sensors I suspect they just read 200.  One way would be two wires and two sensors on the load side.  Counting amps in and out from earlier posts seem to require an efficiency factor around 0.985 which you easily factor in for your SOC, and tweak with experience.

Link to comment
Share on other sites

4 hours ago, Detling said:

measuring up to 200A is fine for charging but if you have a big inverter that plus a pump start can easily exceed 300 A. What happens on the hall effect sensors I suspect they just read 200.  One way would be two wires and two sensors on the load side.  Counting amps in and out from earlier posts seem to require an efficiency factor around 0.985 which you easily factor in for your SOC, and tweak with experience.

 

The inverter is 1.8Kw, so can't exceed 200A except perhaps transiently. The current sensing element in the hall effect sensor ir just a meaty copper bar, so I doubt any ill effects from overload.

 

On a related subject, the current sensor and Lithium battery will never see starter-motor currents, despite being normally paralleled with the start battery. The lower internal resistance of the Lithiums means they're likely to source a large proportion of the starter current if the stay paralleled, and I want to fuse the lithiums with a fairly conservative fuse, which would likely blow. To avoid this I'm interlocking the starter circuit with the paralleling contactor. The contactor will be opened when the start button is pressed and before the starter is energised.

 

MP.

 

8 hours ago, nicknorman said:

As far as I can tell ( because the datasheet is a bit sparse) these sensors are bidirectional. Quoted current range is +-200A and it mentions being suitable for AC. I suppose the question not directly answered in the data sheet is whether the output indicates the current flow direction. But since the nulling coil’s current must be bidirectional it seems likely that the output is too.

 

It is of course a lot more expensive than the sensor you are using.

 

edit:definitely bidirectional, you can see from the schematic that the output can swing both ways. Downside is that it needs a + and - power supply.

 

12898B9D-0DE3-48BF-AB62-C8FFD0B90FF7.png.669e85a132209955f20aad608bb2c99e.png

Interesting. These would certainly be easier to integrate than the solution I'm using.

 

MP.

 

Link to comment
Share on other sites

1 hour ago, MoominPapa said:

Interesting. These would certainly be easier to integrate than the solution I'm using.

 

I should have been clearer. They would be physically easier to integrate. Electrically, they're a pain. +/- 12v power supplies, and an output which swings around 0v too. Not at all microcontroller friendly. The chip I'm using runs on 5v and has a voltage output that swings between 0.5v and 4.5v. Dead easy. Though I did make it a bit more complicated by using a differential input to the ADC with the other leg fixed at 2.5v with a pair of resistors. That allows the use of  the programmable gain in the ADC system to get better resolution at low currents.

 

MP.

 

Link to comment
Share on other sites

  •   1 hour ago, tree monkey said:

    So far this is the only post I have understood in this thread :)

     

     

    I'm glad to see that this post hasn't deteriorated into the usual stupid comments and banter, instead providing lots of useful insight. However, I must admit it is totally beyond not only what I understand but also beyond the lengths I'd be prepared to go to for. 

So frustrating - the holly grail of energy storage is so near yet so far!

I'm going to wait for a set of Battleborn or Relion to come up in ebay.

In the meantime, if anybody (who has mastered the dark art of lithium control) would put out a wire by wire how to, complete with components etc, I'd still be up to give it a go. 

Edited by Johny London
Link to comment
Share on other sites

2 hours ago, MoominPapa said:

On a related subject, the current sensor and Lithium battery will never see starter-motor currents, despite being normally paralleled with the start battery.

When our starter battery dies, I will likely do away with it completely and run purely from the lithiums. I've got an emergency switch to parallel them if needed, and the engine turns over and starts much quicker from the lithiums. Even s heavily discharged lithium bank will still easily start the engine. Ours are fused at 300A though which seems more than enough for our Isuzu engine!

 

13 minutes ago, Johny London said:
  •   1 hour ago, tree monkey said:

    So far this is the only post I have understood in this thread :)

     

     

    I'm glad to see that this post hasn't deteriorated into the usual stupid comments and banter, instead providing lots of useful insight. However, I must admit it is totally beyond not only what I understand but also beyond the lengths I'd be prepared to go to for. 

So frustrating - the holly grail of energy storage is so near yet so far!

I'm going to wait for a set of Battleborn or Relion to come up in ebay.

In the meantime, if anybody (who has mastered the dark art of lithium control) would put out a wire by wire how to, complete with components etc, I'd still be up to give it a go. 

I think the arrangement I described in post 128 is about the easiest and least costly diy route, albeit with some compromises. 

Link to comment
Share on other sites

21 hours ago, nicknorman said:

Are you using an open loop or closed loop (nulling) hall effect sensor? I've never used them but I would have thought a closed loop sensor whereby an opposite magnetic field is generated to create a null (which is all that the hall effect sensor then detects) might not suffer from magnetisation issues since the net magnetic field is zero.

 

https://uk.rs-online.com/web/p/current-transducers/1883783/

 

I had a quick look at the spec of both types a while ago, the closed loop ones do look a lot better ( i suspect for the reason you suggest) but appeared to have a significantly higher current consumption which is not good in a battery system. The magnetic hysteresis should be predictable so maybe it could be sorted out in the software????

 

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

Link to comment
Share on other sites

19 minutes ago, dmr said:

I had a quick look at the spec of both types a while ago, the closed loop ones do look a lot better ( i suspect for the reason you suggest) but appeared to have a significantly higher current consumption which is not good in a battery system. The magnetic hysteresis should be predictable so maybe it could be sorted out in the software????

 

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

Yes it does use a bit of juice, although when the measured current is zero it’s not massive. But the differential supply and output swing is a bit of a pain. The differential output can be dealt with by a fairly simple circuit but not sure how that would affect the critical zero point accuracy.

 

If i we’re doing it I think I would use a shunt and digitise the value locally, send it via CANBUS etc to the main control system (all done with a PIC, obviously!). Which of course is exactly what my Mastervolt Mastershunt does (so I don’t claim any originality!)

 

Incidentally I just sent off my PCB files to China, £29 for 10 boards 75mm x 50mm on 48 hr production and 2-3 days delivery. We’ll see what happens!

Link to comment
Share on other sites

44 minutes ago, Tom and Bex said:

When our starter battery dies, I will likely do away with it completely and run purely from the lithiums. I've got an emergency switch to parallel them if needed, and the engine turns over and starts much quicker from the lithiums. Even s heavily discharged lithium bank will still easily start the engine. Ours are fused at 300A though which seems more than enough for our Isuzu engine!

I have a 250A fuse, and I don't hold much for its chances starting a Lister with a flywheel the size of manhole cover. 

 

MP.

 

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.