|
Post by peterboat on Apr 5, 2021 11:07:01 GMT
The batteries own internal BMS has a SOC meter, its what for me dictates its state of charge. If kris does as you say and slowly charges the batteries on an individual basis, whilst the battery is connected to the valence software, he will when he hits 14.6 volts and all cells balanced have a 100% SOC. I did it using a 180 watt solar panel and solar charger. I did it in stages starting at 13.8 for all the batteries then 13.9 etc until I hit 14.6 its because I was putting the batteries in series that I did top balancing the once. I then found that when the batteries were discharged to 13.8/9 the internal BMS says the batteries are at 80%, which is why I bulk charge to this voltage, absorb at 13.6 and float currently 13.4 volts as I am not at the boat. I have stuck to these voltages for years with no ill effects and John with the same batteries uses similar voltages I’d agree that the internal SoC is probably the best indicator of SoC. Although I suppose we don’t actually know how accurate it is. But something isn’t adding up. You say that 13.8v is the zero current voltage after discharging to 80% SoC. Unfortunately I can’t find any graphs of zero current voltage vs SoC for valence batteries. However, there is Valence’s own graph for voltage at various rates of discharge, all of the curves are fairly fast discharges, there’s one at C/5 and one at C/3. These give about 13.25v and 13.15v at 80%. Of course, one would expect the voltage to be lower under discharge, than at zero current. Then I looked for a graph of standard LiFePO4 discharge voltage, found one at C/4, which gave 13.25 at 80% SoC which is commensurate with my findings. If we take the mid point of the Valence C/5 and C/3 to give an estimate for C/4 it would be 13.2v. So slightly lower than for standard LiFePO4, and this might be explained the the longer wiring runs inside the battery (interconnecting all the small cells) and the probably presence of some kind of shunt within the Valence battery. But anyway, with these figures being very close to the standard LiFePO4 it seems to me unlikely that the magnesium has any significant effect on the voltage vs SoC relationship. And for standard LiFePO4 the zero current voltage at 80% SoC is about 13.3v, which is a long way away from your 13.8v. As I said, something doesn’t add up. And looking at Valence’s own graph, surely it is unfeasible that the voltage would drop from 13.8v at 80% SoC no load, to about 13.3v at C/8? Li batteries just aren’t like that! In summary I would agree that your charging voltages are entirely ok, but I suggest that actually you are charging close to 100% or certainly to well over 80%. How close depends on how long you hold the 13.8v after reaching it, how long you hold the 13.6 after reaching it. Interesting stuff on here Nick, its a forum but plenty of valence stuff on it
|
|
|
Post by Telemachus on Apr 5, 2021 11:21:32 GMT
I’d agree that the internal SoC is probably the best indicator of SoC. Although I suppose we don’t actually know how accurate it is. But something isn’t adding up. You say that 13.8v is the zero current voltage after discharging to 80% SoC. Unfortunately I can’t find any graphs of zero current voltage vs SoC for valence batteries. However, there is Valence’s own graph for voltage at various rates of discharge, all of the curves are fairly fast discharges, there’s one at C/5 and one at C/3. These give about 13.25v and 13.15v at 80%. Of course, one would expect the voltage to be lower under discharge, than at zero current. Then I looked for a graph of standard LiFePO4 discharge voltage, found one at C/4, which gave 13.25 at 80% SoC which is commensurate with my findings. If we take the mid point of the Valence C/5 and C/3 to give an estimate for C/4 it would be 13.2v. So slightly lower than for standard LiFePO4, and this might be explained the the longer wiring runs inside the battery (interconnecting all the small cells) and the probably presence of some kind of shunt within the Valence battery. But anyway, with these figures being very close to the standard LiFePO4 it seems to me unlikely that the magnesium has any significant effect on the voltage vs SoC relationship. And for standard LiFePO4 the zero current voltage at 80% SoC is about 13.3v, which is a long way away from your 13.8v. As I said, something doesn’t add up. And looking at Valence’s own graph, surely it is unfeasible that the voltage would drop from 13.8v at 80% SoC no load, to about 13.3v at C/8? Li batteries just aren’t like that! In summary I would agree that your charging voltages are entirely ok, but I suggest that actually you are charging close to 100% or certainly to well over 80%. How close depends on how long you hold the 13.8v after reaching it, how long you hold the 13.6 after reaching it. Interesting stuff on here Nick, its a forum but plenty of valence stuff on it Did you mean to post a link?
|
|
|
Post by peterboat on Apr 5, 2021 15:57:22 GMT
Interesting stuff on here Nick, its a forum but plenty of valence stuff on it Did you mean to post a link? LOL I did but got involved with something else and forgot
|
|
|
Post by Telemachus on Apr 5, 2021 16:18:52 GMT
Did you mean to post a link? LOL I did but got involved with something else and forgot
So the first thing I come across on that thread is this picture of the chap’s BMS display after he fully charged it. Note, 99.61% Soc, current virtually zero, voltage 13.438v. Sorry but saying 13.8v is 80% SoC is definitely wrong by a long way.
|
|
|
Post by kris on Apr 5, 2021 16:32:41 GMT
My bmv is reading 90%soc at 13.2v which can’t be right? To windy to put my new solar on today maybe tmw. I think I’m going to have to buy a fridge and maybe a dc water heating element to keep a load on these batteries when I get my new solar.
|
|
|
Post by Telemachus on Apr 5, 2021 16:40:20 GMT
My bmv is reading 90%soc at 13.2v which can’t be right? To windy to put my new solar on today maybe tmw. I think I’m going to have to buy a fridge and maybe a dc water heating element to keep a load on these batteries when I get my new solar. Well it’s not outrageously far out. Is that with zero current? I would expect about 13.35 to 13.4 at 90%. Have you set the correct capacity in the BMV and set the charge efficiency to 99 or maybe 100%?
|
|
|
Post by peterboat on Apr 5, 2021 16:48:33 GMT
LOL I did but got involved with something else and forgot
So the first thing I come across on that thread is this picture of the chap’s BMS display after he fully charged it. Note, 99.61% Soc, current virtually zero, voltage 13.438v. Sorry but saying 13.8v is 80% SoC is definitely wrong by a long way. Depends whether he took it to 14.6 volts if you read thats when it resets SOC
|
|
|
Post by kris on Apr 5, 2021 16:49:11 GMT
My bmv is reading 90%soc at 13.2v which can’t be right? To windy to put my new solar on today maybe tmw. I think I’m going to have to buy a fridge and maybe a dc water heating element to keep a load on these batteries when I get my new solar. Well it’s not outrageously far out. Is that with zero current? I would expect about 13.35 to 13.4 at 90%. Have you set the correct capacity in the BMV and set the charge efficiency to 99 or maybe 100%? No that’s under load only about 3amps. Yes capacity and charge efficiency are set.
|
|
|
Post by Telemachus on Apr 5, 2021 16:51:47 GMT
It started out at 0% SoC so to get it up to 100% (well, 99.6%) surely it has to have been fully charged?
|
|
|
Post by Telemachus on Apr 5, 2021 16:58:21 GMT
Well it’s not outrageously far out. Is that with zero current? I would expect about 13.35 to 13.4 at 90%. Have you set the correct capacity in the BMV and set the charge efficiency to 99 or maybe 100%? No that’s under load only about 3amps. Yes capacity and charge efficiency are set. One problem is that we are talking about quite small differences in voltage, but exactly where it’s being measured and thus what voltage drop 3A might create, are unknown (to me). Ditto the accuracy of the measuring device. Anyway, you could surely turn off the load and recheck the voltage after a while. I have found that Li takes several hours to fully recover to its rested voltage after even a small discharge. But it seems to me that the figures you quote for voltage and SoC are not unreasonably far out. It’s really difficult to determine SoC from voltage other than at the extreme ends, hence the need for the BMV.
|
|
|
Post by kris on Apr 5, 2021 17:14:11 GMT
No that’s under load only about 3amps. Yes capacity and charge efficiency are set. One problem is that we are talking about quite small differences in voltage, but exactly where it’s being measured and thus what voltage drop 3A might create, are unknown (to me). Ditto the accuracy of the measuring device. Anyway, you could surely turn off the load and recheck the voltage after a while. I have found that Li takes several hours to fully recover to its rested voltage after even a small discharge. But it seems to me that the figures you quote for voltage and SoC are not unreasonably far out. It’s really difficult to determine SoC from voltage other than at the extreme ends, hence the need for the BMV. yes I could turn my inverter off, I will at some stage. But as I say I’m not too bothered about accurate soc as long as I know voltage and have an idea of amps out and in. I’m going to have lots of opportunity to set the bmv when I’ve got this new solar fitted.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 5, 2021 17:23:25 GMT
I find it tricky to get a resting voltage reading, because there is always something happening. Either the sun comes out of a cloud and suddenly there is a bit of solar coming in, which artificially bumps up the voltage reading, or else the fridge occasionally kicks in to draw 4-6 amps of current.
But because I cruised for a few hours today, and there was a reasonable bit of solar, the batteries are now I think around 90% SoC, and when I manually isolated them using the BEP switch, the voltage was reading 13.30v
Another negative mark for my electrician- the diesel CH current draw is not being taken from the lithiums, but from the lead acid feeder battery (which I was disappointed to find out a few days ago).
|
|
|
Post by kris on Apr 5, 2021 17:43:19 GMT
I find it tricky to get a resting voltage reading, because there is always something happening. Either the sun comes out of a cloud and suddenly there is a bit of solar coming in, which artificially bumps up the voltage reading, or else the fridge occasionally kicks in to draw 4-6 amps of current. But because I cruised for a few hours today, and there was a reasonable bit of solar, the batteries are now I think around 90% SoC, and when I manually isolated them using the BEP switch, the voltage was reading 13.30v Another negative mark for my electrician- the diesel CH current draw is not being taken from the lithiums, but from the lead acid feeder battery (which I was disappointed to find out a few days ago). people did tell you about those electricians😀 At least you’ve got a system up and running. Plenty of power now as well.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 5, 2021 17:51:57 GMT
people did tell you about those electricians😀 At least you’ve got a system up and running. Plenty of power now as well. Just a thought, but one way you might get a rough idea of the SoC might be to let them run down a bit. Nick posted a graph a few pages ago showing the normal discharge curve (voltage decreasing with lower SoC). If you can assume 13v is a given SoC, and 12.9v, and so on, you can set the SoC manually at one of those lower voltages. The BMV knows your total Ah, and you can put in a rough SoC value based on the lower end of the voltage range form Nicks graph. Far from ideal, but might be a rough basis to work from as you charge them back up again.
|
|
|
Post by Telemachus on Apr 5, 2021 18:07:34 GMT
people did tell you about those electricians😀 At least you’ve got a system up and running. Plenty of power now as well. Just a thought, but one way you might get a rough idea of the SoC might be to let them run down a bit. Nick posted a graph a few pages ago showing the normal discharge curve (voltage decreasing with lower SoC). If you can assume 13v is a given SoC, and 12.9v, and so on, you can set the SoC manually at one of those lower voltages. The BMV knows your total Ah, and you can put in a rough SoC value based on the lower end of the voltage range form Nicks graph. Far from ideal, but might be a rough basis to work from as you charge them back up again. The trouble is that over the middle part of the range, which is the optimum range to keep the batteries if you don’t need their full capacity, there is virtually no change in voltage. If you look at that chart I posted you’ll see that the voltage at 70% and 40% are both quoted as 13.2v. Of course there is some voltage change, but it is down to the 100ths of a volt and the accuracy of the measuring device and the exact measuring conditions, become the predominant factors. Better to set the SoC on the BMV at quite a high SoC where the voltage change with SoC becomes significant.
|
|