DIY portable, useful LiFePO4

[markopolo]

Here's a link with good descriptions of what I observed when charging parallel cells: https://www.impactbattery.com/blog/t...s-in-parallel/

The methods of connection can be described as:

1. not so good
2. better
3. best

Most folks with parallel lithium cells probably use connection method #2. The result will be unequal use of the cells.

Here's an example of an installation of prismatic cells that would rank between #2. better and #3. best - in my opinion. Have to give it a 2.5+ on the very simplified ranking scale: https://www.entropypool.de/2015/08/0...mplementation/

It must be noted though that if parallel connecting assembled cell packs with each assembled cell pack (aka battery) having a separate BMS then connecting each battery to the system positive and negative busbars separately might be the only option. Otherwise, one BMS tripping could bring the whole system down.

[booster]

Interesting stuff. This is the kind of balance issues that are very common on many of the lead acid systems we see, like etreks. The cable lengths and connection points to maintain the balance spoke of here is different than the balance from the balancer that many put into the etreks to address the 1/2 voltage output tap they used.


I don't know if it is significant or not or feasible, but in the lithium pack, would they be better off to swap the negative cable and BMS positons on the last string to get more even connection lengths, of maybe just put the negative on the same connection as the BMS?


I do like the idea of using the woven strap with a kink in them between cells. I have always been worried about solid buss bars connection battery posts with no room for movement and the load taken by the posts.

[markopolo]

I marked up the photo from the Entropy site to show the most balanced way to do it IMO.


The small boards you see on top of the cells are for monitoring.


image.jpg


To me, it's all about cell distance / cell path to the nearest charge or discharge point.


series_parallel.jpg


It's more critical with more cells. I'll have 40 cells, not 16 for example. Some of the Renogy batteries for example have 120 cells.

[booster]

If you really wanted even closer than your markup, you could use two feeds to your 4 wire input with two cables feeding common connection pairs of your 4 feed wires. The big thing would be to get the feeds all the same, either by the pairing or 4 separate ones of equal length to a common tie point. I think you would want to take off the jumpers between the posts at that point for each end of the string.

[JohnnyFry]

Marco, I just received my 150 A/H Prismic cells yesterday and am still waiting on the BMS with temperature cut off and Bluetooth. All 4 cells showed 3.33 volts on checking so they arrived in a balanced condition. As I understand it he BMS performs balancing on a continuous basis from then on. The new BMS shows the individual cell voltages as well as temperature, Amp Hour status, current in/out and cycle count. Can't wait to get it all hooked up and running!

[markopolo]

That's exciting Johnny!

If it's a Jiabaida (JBD) BMS then it will have default settings. Balancing will occur at a set voltage (adjustable). The extent of balancing is adjustable also, example 50mV or 30mV or 10mV difference between cells etc. I'm fairly certain it's only in the mA range though and might be 50mA per cell. You can choose Charge Balance (charging) or Static Balance (idle).

I decided that I will add an additional set of balance leads during pack assembly and use a separate 5A capacitor based balancer if/when needed.

Voltage accuracy has become critical for me. I doubt I'll ever buy a meter now with less than 19,999 counts and worse than 0.1% accuracy.

Some meters round values & some just drop the digit they don't display so seeing at least the 3rd digit to the right of the decimal is invaluable to me.

https://www.youtube.com/watch?v=NAMwiSblf_c

[Bud]

Originally Posted by markopolo

Here's a link with good descriptions of what I observed when charging parallel cells: https://www.impactbattery.com/blog/t...s-in-parallel/

The methods of connection can be described as:

1. not so good
2. better
3. best

Most folks with parallel lithium cells probably use connection method #2. The result will be unequal use of the cells.

Here's an example of an installation of prismatic cells that would rank between #2. better and #3. best - in my opinion. Have to give it a 2.5+ on the very simplified ranking scale: https://www.entropypool.de/2015/08/0...mplementation/

It must be noted though that if parallel connecting assembled cell packs with each assembled cell pack (aka battery) having a separate BMS then connecting each battery to the system positive and negative busbars separately might be the only option. Otherwise, one BMS tripping could bring the whole system down.

After reading the "Unbalanced Charging" link, I don't understand.

I have 2 marine batteries connected with 4 gauge cable the Wrong Way, about 8 feet of cable. When on the Battery Tender the voltage meter is about 13.1 volts and with shore power dumb charger about 13.5-6. If I disconnect both batteries and test the voltage immediately and hours later, the battery voltages will be different?

and/or when I use the batteries is it that one of the batteries is supplying more amps so I'm simply using up the total amp hours of that battery sooner?

Bud

[booster]

Originally Posted by markopolo

That's exciting Johnny!

If it's a Jiabaida (JBD) BMS then it will have default settings. Balancing will occur at a set voltage (adjustable). The extent of balancing is adjustable also, example 50mV or 30mV or 10mV difference between cells etc. I'm fairly certain it's only in the mA range though and might be 50mA per cell. You can choose Charge Balance (charging) or Static Balance (idle).

I decided that I will add an additional set of balance leads during pack assembly and use a separate 5A capacitor based balancer if/when needed.

Voltage accuracy has become critical for me. I doubt I'll ever buy a meter now with less than 19,999 counts and worse than 0.1% accuracy.

Some meters round values & some just drop the digit they don't display so seeing at least the 3rd digit to the right of the decimal is invaluable to me.

Those are some mighty nice and spendy meters in that video. I would really like to get a 87v max, but really gulp at spending that much for the rare times I would need it. The ability to read true voltage from variable speed drives would be really handy, I think.


I was surprised at some reviews on Amazon about Fluke customer service being horrible, as they used to be the pillar of support of tradespersons. Things like lifetime warranties that aren't repair warranties but only replacement if still in production are essentially useless as pointed out in some reviews. The apparently were bought out a while ago and now are perhaps morphing into an all for profit margin mode, which is sad if true.


What would your top choice of a 20 or 50K count meter be?

[markopolo]

Originally Posted by Bud

After reading the "Unbalanced Charging" link, I don't understand.

I have 2 marine batteries connected with 4 gauge cable the Wrong Way, about 8 feet of cable. When on the Battery Tender the voltage meter is about 13.1 volts and with shore power dumb charger about 13.5-6. If I disconnect both batteries and test the voltage immediately and hours later, the battery voltages will be different?

and/or when I use the batteries is it that one of the batteries is supplying more amps so I'm simply using up the total amp hours of that battery sooner?

Bud

It depends on how long the batteries were on the chargers. Eventually, they will be equal or pretty close to equal. Float charging lead acid batteries will let you get away with less than perfect wiring.



As for the second question, the battery nearer the load will work harder. It might be a 60/40 split at first but eventually that would reverse as the nearer battery gets used up. I've put an ammeter on that and it's quite visible. If the discharge never gets to where it's pulling more from the farther battery then the nearer battery won't last as long as the more distant battery.

[markopolo]

Originally Posted by booster

Those are some mighty nice and spendy meters in that video. I would really like to get a 87v max, but really gulp at spending that much for the rare times I would need it. The ability to read true voltage from variable speed drives would be really handy, I think.

I was surprised at some reviews on Amazon about Fluke customer service being horrible, as they used to be the pillar of support of tradespersons. Things like lifetime warranties that aren't repair warranties but only replacement if still in production are essentially useless as pointed out in some reviews. The apparently were bought out a while ago and now are perhaps morphing into an all for profit margin mode, which is sad if true.

What would your top choice of a 20 or 50K count meter be?

The 87v max would be nice to have. My DMM budget would be like 10% of that.

I'm interested in DC voltage accuracy so I'm happy with meters that might not be suitable for other uses. A lot of the cheaper multimeters are just not suitable for industrial or professional use. The CAT ratings are very suspect from what I've read. It's not worth the risk using a cheap meter around high voltage stuff. For checking RV related DC voltage there are some reasonably priced units with 0.1% accuracy or better.

Low cost / relatively low cost stuff for checking DC voltage:

For just a simple voltmeter then it might be impossible to beat a $4 BY536V "5 bit" meter.

Measurement accuracy: 0.01%（+/-2）*
*In the worst case, the error is no more than 10mV. In general, the error is about 2mV.

That meter would be useful for 24VDC systems also. The display would be 24.000V. In comparison, a 20K count meter would display 24V as 24.00V - the third digit after the decimal would be lost.

Next up for a DMM would have to the Aneng AN870 / Zotek ZT219 / Richmeters RM219 (all the same as far as I can tell).
DC Accuracy is listed as ±(0.05%+3) in the manual but sometimes shown as ±(0.5%+3) on some websites so that's confusing. Regardless, it seems like a pretty nice low cost meter.

https://www.youtube.com/watch?v=nRjlL98q9hg

I have a Dawson Tools DDM645 - https://www.jameco.com/z/DDM645-Daws...r_2208462.html

It would be the same as a Mastech MY65.
DC accuracy in the ranges I'm interested in is listed as: ±(0.1% + 3 counts)

I just got it but it's an older design. It's big though so forget about putting it in a pocket. I know where the DC voltage calibration potentiometer is located on the board so I can fine tune it / re-calibrate it.

Next up would be the Uni-T UT61E. Note the E in the model name. The UT61 series varies in accuracy specs. https://www.uni-trend.com/html/produ...ies/UT61E.html

DC accuracy is listed as: ±(0.1%+2)

I think it is an older design as well but a good budget priced multimeter.
I've ordered an Aneng AN870. It'll be interesting to compare it to the Dawson/Mastech unit.

If you have a good meter then a good voltage reference would be nice to have. You might be able to find out how to calibrate your meter by search online.

A good source for a quality voltage reference is here: Voltagestandard.com

I have a couple of budget voltage references and have ordered a couple more. Like $30 total. (one wasn't working so it was free). The two I have haven't been measured by precision equipment so accuracy is a range based on the chip until I can get more accurate readings. The two on order use the same or maybe one version better chip but more importantly have been measured and come with a sheet with the measurements. Note: some vendors send out the same sheet to customers. If you order one of those then Google your numbers to see if other folks have the exact same numbers. If so, the numbers are fake.

I'm getting more dialed in on voltage precision. My guess is that the Dawson unit is only a few millivolts off at worst. Here's the result of the test of the Dawson on the two voltage references I have:

AD584JH voltage reference.jpg

[Bud]

Originally Posted by markopolo

It depends on how long the batteries were on the chargers. Eventually, they will be equal or pretty close to equal. Float charging lead acid batteries will let you get away with less than perfect wiring.

As for the second question, the battery nearer the load will work harder. It might be a 60/40 split at first but eventually that would reverse as the nearer battery gets used up. I've put an ammeter on that and it's quite visible. If the discharge never gets to where it's pulling more from the farther battery then the nearer battery won't last as long as the more distant battery.

Thank you Marko for getting the first answer right! Just kidding, good to hear as the batteries get fully charged often, after most uses.

I don't understand the second answer. "but eventually that would reverse as the nearer battery gets used up. If the discharge never gets to where it's pulling more from the farther battery then the nearer battery won't last as long as the more distant battery."

I can understand the reversal if the nearer battery is dead for example. But what determines when that happens, at what SOC the nearer battery is at when the reversal happens? It does not seem in my case to be near a 0 or a 10% state of charge or the batteries would be dead nearly 4 years later.

I suppose no one has measured/determined what happens. Just wire it right in the first place which I did Not (lazy).

Interesting stuff Marko, thank you. I just need to decide what to do when these batteries give up. Serious choices:

- 2 new marine batteries

- new charger and agm or lithium

- new B with lithium/ac or somewhat new B with all new electrics - lithium/ac

Bud

[markopolo]

A long time ago, I ran a short duration test on two 200Ah banks separated by 20' of wire. It would have been a fairly large load like 50A to 80A, I don't remember exactly.

The bank that was 20' of wire closer to the load supplied most of the current by a fair margin. The test was long enough to make some guesses that eventually the load would be shared equally then, at some point, more current would be supplied by the more distant battery. More capacity would have been removed from the nearer battery early on so my assumption was that more current would come from the farther battery later on.

I don't recall you having problems with your current setup so ... if it ain't broke, don't fix it!

[Bud]

Originally Posted by markopolo

A long time ago, I ran a short duration test on two 200Ah banks separated by 20' of wire. It would have been a fairly large load like 50A to 80A, I don't remember exactly.

The bank that was 20' of wire closer to the load supplied most of the current by a fair margin. The test was long enough to make some guesses that eventually the load would be shared equally then, at some point, more current would be supplied by the more distant battery. More capacity would have been removed from the nearer battery early on so my assumption was that more current would come from the farther battery later on.

I don't recall you having problems with your current setup so ... if it ain't broke, don't fix it!

Thanks Marko.

Bud

[booster]

Originally Posted by Bud

Thank you Marko for getting the first answer right! Just kidding, good to hear as the batteries get fully charged often, after most uses.

I don't understand the second answer. "but eventually that would reverse as the nearer battery gets used up. If the discharge never gets to where it's pulling more from the farther battery then the nearer battery won't last as long as the more distant battery."

I can understand the reversal if the nearer battery is dead for example. But what determines when that happens, at what SOC the nearer battery is at when the reversal happens? It does not seem in my case to be near a 0 or a 10% state of charge or the batteries would be dead nearly 4 years later.

I suppose no one has measured/determined what happens. Just wire it right in the first place which I did Not (lazy).

Interesting stuff Marko, thank you. I just need to decide what to do when these batteries give up. Serious choices:

- 2 new marine batteries

- new charger and agm or lithium

- new B with lithium/ac or somewhat new B with all new electrics - lithium/ac

Bud

I think that you have to remember that most of us are using smart chargers, so you have a limited time at absorption before going to float. The scenario that Marko mentioned is correct, eventually the second battery will get near fully charged, but in general that would apply to if they are put on fixed voltage charger. With a multistage charger it gets a lot more complicated because you quickly get into choosing the lesser of two evils, or basically dam*ed if do, dam*ed if you don't thing.


It the second battery is charging slower than the first, the absorption stage can never finish at the correct time for both of them as they need a different amount of time to get to full. If you increase the absorption time to get the second battery full, the first battery will see absorption voltage on top of a full battery which is not a good thing for battery life. If you keep the absorption time right for the first battery, the second one will be left substantially lower than full and also at lower voltage so it may take days or longer to get even near. Being in a partially discharged state is not good for batteries and can shorten their life due to capacity walkdown. If the batteries charge at the same rate due to good cabling, neither of the above can happen if the absorption time is set right or even better amps to the batteries controlled.


This all said, in the real world for most of the RVs on the road, it probably would have, at most, a minimal life impact on batteries and a minor impact on usable capacity while camping. This is because nearly every charger made either over or under changes the batteries depending on the algorithm and depth of discharge. We have very accurate charging, so for us cable lengths were more important, IMO, as they could mess up the good accuracy of the charger. Most campers don't have that kind of a charging system (on all 3 sources) so their benefits would be lesser than ours. Our charging paths on the two set of 6v GC2 batteries are probably within 2" of each other or so, as I was very careful to do it that way.


A lot of this stuff like cable matching is tough to do within the restraints of an upgrade to an existing system. It gets much easier if you are starting from scratch like we did, as we removed the entire old system and moved the batteries to a centralized location.

[Bud]

Originally Posted by booster

I think that you have to remember that most of us are using smart chargers, so you have a limited time at absorption before going to float. The scenario that Marko mentioned is correct, eventually the second battery will get near fully charged, but in general that would apply to if they are put on fixed voltage charger. With a multistage charger it gets a lot more complicated because you quickly get into choosing the lesser of two evils, or basically dam*ed if do, dam*ed if you don't thing.


It the second battery is charging slower than the first, the absorption stage can never finish at the correct time for both of them as they need a different amount of time to get to full. If you increase the absorption time to get the second battery full, the first battery will see absorption voltage on top of a full battery which is not a good thing for battery life. If you keep the absorption time right for the first battery, the second one will be left substantially lower than full and also at lower voltage so it may take days or longer to get even near. Being in a partially discharged state is not good for batteries and can shorten their life due to capacity walkdown. If the batteries charge at the same rate due to good cabling, neither of the above can happen if the absorption time is set right or even better amps to the batteries controlled.


This all said, in the real world for most of the RVs on the road, it probably would have, at most, a minimal life impact on batteries and a minor impact on usable capacity while camping. This is because nearly every charger made either over or under changes the batteries depending on the algorithm and depth of discharge. We have very accurate charging, so for us cable lengths were more important, IMO, as they could mess up the good accuracy of the charger. Most campers don't have that kind of a charging system (on all 3 sources) so their benefits would be lesser than ours. Our charging paths on the two set of 6v GC2 batteries are probably within 2" of each other or so, as I was very careful to do it that way.


A lot of this stuff like cable matching is tough to do within the restraints of an upgrade to an existing system. It gets much easier if you are starting from scratch like we did, as we removed the entire old system and moved the batteries to a centralized location.

Thanks booster, understood I think. Wow, so I've kind of been saved from my own dumb cabling by a dumb charger!

[booster]

Originally Posted by Bud

Thanks booster, understood I think. Wow, so I've kind of been saved from my own dumb cabling by a dumb charger!

In some instances, that is absolutely possible, as long as you keep good track of it so it doesn't run way to long and cook things.

[Bud]

Originally Posted by booster

In some instances, that is absolutely possible, as long as you keep good track of it so it doesn't run way to long and cook things.

I've sure thought about that. At home I never used shore power because of the 13.5-6 from the charger, just let it sit for 30 days until the next LOUD ONAN exercise. Early on I did not know I needed a full charge more oftern.
Near 4 years ago for the first time I started being nice to lead acid batteries and used the Battery Tender 13.0-1. I sometimes drive off with a full charge for many hours at much higher than 13.5-6. Still, so far so good.

Marko, booster thanks again.

[JohnnyFry]

I have finally received my "Smart" BMS (XIAOXIANG) from Battery Hookup and have connected it to my four 150 A/H Prizmic cells from Ali Express. The cells arrived in balanced condition and roughly 80% charged. After connecting the pack in place of the FLA 232 6 volt battery pair it started the Onan 2.8 generator without hesitation and with more vigor than the old FLA batteries, so far so good. I turned on the TrippLite to bring the charge up and if did so without incident. I have not finished installing the Renogy DC/DC charger so I have not tested that aspect.

I am having one issue: When the battery is brought up to charge by the TrippLite and further charging is cut off by the BMS, I get flickering lights. Apparently the TrippLite is unhappy not having a battery as a voltagel reference. When I bridge in the FLA starting battery the problem disappears. I am not sure what to try next and am looking for suggestions. So far I have considered the following options: 1. get a small 12V motorcycle battery and connect it in parallel across the 12V buss. This would stay in cricuit when the Li BMS takes the Li battery off line when fully charged or 2. install a large (60,000 mFd) capacitor across the output of the TrippLite. Any thoughts Marco, Booster, Bud?

[booster]

Originally Posted by JohnnyFry

I have finally received my "Smart" BMS (XIAOXIANG) from Battery Hookup and have connected it to my four 150 A/H Prizmic cells from Ali Express. The cells arrived in balanced condition and roughly 80% charged. After connecting the pack in place of the FLA 232 6 volt battery pair it started the Onan 2.8 generator without hesitation and with more vigor than the old FLA batteries, so far so good. I turned on the TrippLite to bring the charge up and if did so without incident. I have not finished installing the Renogy DC/DC charger so I have not tested that aspect.

I am having one issue: When the battery is brought up to charge by the TrippLite and further charging is cut off by the BMS, I get flickering lights. Apparently the TrippLite is unhappy not having a battery as a voltagel reference. When I bridge in the FLA starting battery the problem disappears. I am not sure what to try next and am looking for suggestions. So far I have considered the following options: 1. get a small 12V motorcycle battery and connect it in parallel across the 12V buss. This would stay in cricuit when the Li BMS takes the Li battery off line when fully charged or 2. install a large (60,000 mFd) capacitor across the output of the TrippLite. Any thoughts Marco, Booster, Bud?

The Tripplite, AFAIK, needs a battery in the system to reference properly, so that would certainly need to be addressed. If you are going to be using the BMS to control charging, which I am not certain if it is a good idea or not so good compared to having a separate control system so the BMS is a fail safe only, you my want to get away from the "not so smart" Tripplite and go to a fixed voltage unit like a Progressive Dynamics, which many/most of can hold voltage without a battery online. You can also use a time delay relay and other relays on the charger that brings in the batteries first before the charger is allowed to run and cuts out when the batteries disconnect.

[Luv2Go]

The Tripplite does have two battery types, wet or gel. The gel setting has the lowest bulk voltage 14.2V and float of 13.6V, that may be enough to keep the BMS from opening. The wet settings are 14.4V and 13.3V respectively.