Lithium Batteries (for the rest of us)

[chala]

Originally Posted by transmitterdan

Yes it is.

Good that it is for some.

The question was:

Originally Posted by georgec

How does the cell logger read the voltage of individual cells when they are in parallel

[vRUN]

Originally Posted by goboatingnow

Its not a question of voltage, Good practice is that no long term impressed voltage is left on a Li cell after charging.

Sry Dave: What is an "impressed" voltage ?
Something like: "Charger has a higher voltage (potential) than battery but equalization current is very low ?"

Originally Posted by goboatingnow

The correct way to share a Li battery and a load, is often referred to as Power Path Switching. For longest life, the Li battery should supply the load via a Diode sharing path ( either ordinary or MOSFET ideal) Once the power supply/charger load current is within the limits, No voltage is impressed on the Li battery. If the load is greater then the charger, both charger and battery share the load, with the battery slowly discharging.

Physics: Any 2 different charges Q1, Q2 that are connected will equalize potential U at their point of connection. You may hinder that with a very large resistance in its direct path (2 diodes in reverse, no connect i.e. capacitor) but as soon as current flows, ohms law kicks in (U~I, U=R*I).
The higher Q will have a higher current to support equalization, it diminishes faster. The total amount of current is determined by the load resistance R. Both Qs share that current load according to their internal resistance size/capacity. (Q=C*U).

But Dave, I am afraid I don't get you: Where are your 2 charges Q? Could you draw a schematic? (Please!) Any xS2p battery does equalize (balance) all the time, internally - no load. (Mini cycle?) Where do you put those Diodes (except before the load to protect the battery from being "reverse charged" by "an non ohmic load".
-confused-

PS A lot of this research is coming from GSM devices that use Li to handle the large current pulses ( 2-3A) that occur in class 10 devices. This causes lots of mini cycles and has lead to reduced life times , often unexpectedly so.

Would you have a citation for "mini cycles" ready? You know I looked at the charger IC base (including iphone charger) and did not notice these mentioned - Thanks.

Regards

[vRUN]

Originally Posted by Lagoon4us

Are $12:00 Junsie's accurate?

I do not own a Junsie (so should probably refrain from commenting but...).

The short look inside "revealed" an ATMEL uP (Microprocessor) is used internally. The Atmel voltage reference is described here. (page 5)
Since you probably do not want to read all that stuff let me summarize as:
Even under ideal conditions I expect +20 to -40mV difference
(and don't you engineers bash me for being inaccurate )
My imax B6 charger ("junsie that can charge") does work in that region, so does my (Atmel based) LiPo Voltage Checker and Balancer (all are "copies"?).

3.85 V (may be 3.87 or 3.81) (and 3.80 or 3.90 readings wouldn't shock me )

[Lagoon4us]

Cheers

[transmitterdan]

Originally Posted by chala

Good that it is for some.

The question was:

The voltages of cells in parallel must be equal to one another. They will distribute charge according to their need. There is no need to monitor the individual voltages. What would be the purpose?

Sorry if I'm being dense and not understanding your point.

[goboatingnow]

Originally Posted by transmitterdan

This isn't a technical answer If a charger is programmed to switch to a "float" voltage a couple tenths below the resting voltage of a fully charged bank I fail to see how this can be a bad thing. There will be no current from the charger and thus no "plating" or other ill effects can occur. In this case how can the bank even know the charger is powered on? From its point of view the charger is "off". It is infinitely easier to coerce a traditional battery charger to float at some lower voltage than to completely shut off shut and on based on SOC.

If you would like a complete technical breakdown of how charge is stored in Li you would realise the reason behind the comments.

I am merely relaying the approach that is taken on other areas of Li charging

The fact is small currents are flowing etc and even without that an electric field has been established

The main issue is why ate you floating a battery that has tiny self discharge.

The only reason a conventional charger floats is the user isn't there to switch it off.

The fact is using " conventional " chargers is a compromise and especially in load sharing sometimes not a great compromise.

Dave

[goboatingnow]

Originally Posted by vRUN

Sry Dave: What is an "impressed" voltage ?
Something like: "Charger has a higher voltage (potential) than battery but equalization current is very low ?"

Physics: Any 2 different charges Q1, Q2 that are connected will equalize potential U at their point of connection. You may hinder that with a very large resistance in its direct path (2 diodes in reverse, no connect i.e. capacitor) but as soon as current flows, ohms law kicks in (U~I, U=R*I).
The higher Q will have a higher current to support equalization, it diminishes faster. The total amount of current is determined by the load resistance R. Both Qs share that current load according to their internal resistance size/capacity. (Q=C*U).

But Dave, I am afraid I don't get you: Where are your 2 charges Q? Could you draw a schematic? (Please!) Any xS2p battery does equalize (balance) all the time, internally - no load. (Mini cycle?) Where do you put those Diodes (except before the load to protect the battery from being "reverse charged" by "an non ohmic load".
-confused-

Would you have a citation for "mini cycles" ready? You know I looked at the charger IC base (including iphone charger) and did not notice these mentioned - Thanks.

Regards

Switched power path means in effect isolating the battery when an external power source is applied. Hence the diodes isolate each source from interacting , ie source A will not try to charge Source B ( which happens to be a battery

The battery therefor is charged independently of the power source.

Have a google on power path lots of ICs implement the metholodogy.

There is a few good research papers on mini cycles. When I'm back in the Lab. Ill reference a few.

By impressed voltage I mean where the charge source remains connected with a regulated voltage. Ie for example float mode. An electric field is therefor present again read into Li structures

Dave

[chala]

Originally Posted by transmitterdan

There is no need to monitor the individual voltages. What would be the purpose?

May be to detect a non-performing cell.
The question remain:

Originally Posted by georgec

How does the cell logger read the voltage of individual cells when they are in parallel

[georgec]

Originally Posted by chala

Is that a clear answer ???

Yes it is, except for the measuring part. With 12 cells I would place sets of 3 in parallel and those banks in series. Wouldn't the measuring be across the paralleled banks so as to read the cell voltage?

[goboatingnow]

The purpose of measuring the series voltage of the battery strings is to ensure that the designated cutoff voltage is not exceeded in the fairly rare case where a cell ( or parallel set of cells) does not uniformly balance the charge replenishment evenly. ( or where changes in effective resistance cause unexpected HVCs)

It's not there to detect non performance persay. This requires more analysis.

Dave

[vRUN]

Originally Posted by georgec

Yes it is, except for the measuring part. With 12 cells I would place sets of 3 in parallel and those banks in series. Wouldn't the measuring be across the paralleled banks so as to read the cell voltage?

12cells in the arrangement you choose are called
4S3P (four in series, three paralleled) ... but you knew that ;-)

Without adding resistors and stuff and only being able to measure voltages (which it sure is) you can access:

(A)---[3xcell]--(U1)--[3xcell]--(U2)--[3xcell]--(U3)--[3xcell]--(B)

A-B Voltage will be your battery voltage, most likely somewhat 12V-ish
A-U1 will be your 1st 3P-Cell voltage
U1-U2 will be your 2nd 3P-Cell voltage
U2-U3 will be your 3rd 3P-Cell voltage
U3-B will be your 4th 3P-Cell voltage

You will not be able to access a single cell (those will self-balance to one single voltage, dead cells will mostly get warm, heat up)


@chala: If you want to measure single cells you could add resistors and use Kirchhoffs law ...The current entering any junction is equal to the current leaving that junction. i2 + i3 = i1 + i4

[dlentz]

Originally Posted by goboatingnow

If you would like a complete technical breakdown of how charge is stored in Li you would realise the reason behind the comments. I thought I did, LI doesn't change physics or Ohm's law.

I am merely relaying the approach that is taken on other areas of Li charging

The fact is small currents are flowing etc and even without that an electric field has been established With less than zero voltage potential, how could any current flow?

The main issue is why ate you floating a battery that has tiny self discharge.

The only reason a conventional charger floats is the user isn't there to switch it off. Yes, true.

The fact is using " conventional " chargers is a compromise and especially in load sharing sometimes not a great compromise.
Probably, but not entirely convienced.

Dave

Dave, I'm not trying to be argumentative here, but I don't get it.
I'm running a Sterling Ultra 60 on my 700 amp Winston bank. I've been away several times (6) for extended times (2-4 weeks). I have the charger set at 13.8 volts for bulk & absorption and float at 13.2 V. Once my bank reaches 13.8 v the charge drops back until the charger "floats", at which time there is zero amps applied to the bank. The bank never falls to 13.2 v (when left with only the fridge on.) so little to no additional charging takes place. The highest pack voltage my logger has ever seen is 13.9v (probably supplied by my 150 A DC generator, set at 14.1v for charge to speed up charging, but never ran unattended.) and the highest cell was 3.49 v. With a maximum of 50mv delta V on the cells.

What I don't understand in my setup is, oddly enough, when the fridge cycles the charger picks it up and supplies the current. This doesn't kick the charger into another bulk or absorption cycle and it doesn't put any current into the battery (zero as per Link monitor). The charger is still at 13.2 V and the battery is usually around 13.3-13.4 at this point.

Maybe a clue is the remote monitor for the charger. It labels the charge cycles as "Conditioning" when charging and "Ready" when in float. So maybe "float" is a misnomer in this particular instance. When in the "ready mode" the charger is set at a lower V potential than the pack, so I see no way to move electrons into the pack at that point. But why/how could the fridge cycle from the charger and not the pack?

What am I missing?
Dwain

[vRUN]

Originally Posted by goboatingnow

The fact is using " conventional " chargers is a compromise and especially in load sharing sometimes not a great compromise.

Dave

Thanks for the "power path" explanation. Big words for something seemingly obvious. (and yes I did google but found only things like the microchip "power path reference design" - which I <ahem> thought "bs")

Now I get it: "System Power Path Management allows end-users to charge their batteries without interruption." <eek>

Together with your above comment I can "dig" quite a few discussions (&vocabulary) I see in this thread. Li batteries are so very much closer to an immediate storage medium than a Pb battery. All those "intelligent" Pb chargers in parallel with a Li battery plus on top a Solar charger connected to the varying (ship) load do make for some really nice and unexpected ideosyncrasies.

Good thing my Pb charger is charging Pb batteries...
it will stay that way

[CAELESTIS]

Okay, a question for the experts (after me reading most of the entire thread but not being an electrical expert myself).

Current: 8 LA Interstate 6v batteries wired as 24v--550amps @ 24v, Outback charger/inverter with Mate, 140amp @24v alternator, Balmar charge reg, and additional Linkpro battery monitor.

If I switch to Lithium, what else do I need to change besides the batteries and charge parameters? What brand battery appears to be the best (American preferred)?

Space constraints are height at about 11.5in. Can batteries be mounted on their sides?

Thanks!

[vRUN]

Originally Posted by dlentz

Dave, I'm not trying to be argumentative here, but I don't get it.
I'm running a Sterling Ultra 60 on my 700 amp Winston bank.
...the charger set at 13.8 volts for bulk & absorption and float at 13.2 V.
...
... when the fridge cycles the charger picks it up and supplies the current. This doesn't kick the charger into another bulk or absorption cycle and
it doesn't put any current into the battery (zero as per Link monitor). The charger is still at 13.2 V and the battery is usually around 13.3-13.4 at this point.

Maybe a clue is the remote monitor for the charger. It labels the charge cycles as "Conditioning" when charging and "Ready" when in float. ...
Dwain

Dwain, (I am not Dave but)

"...when the fridge cycles the charger picks it up and supplies the current.
This doesn't kick the charger into another bulk or absorption cycle
and it doesn't put any current into the battery (zero as per Link monitor).
The charger is still at 13.2 V and the battery ... 13.3-13.4 at this point.
"

Easy explanation for your observations ???

The battery bank is at 13.3 Volt most likely via a (silicon) diode connection to the house (fridge)
and house is at charger voltage i.e. at 13.2V to the fridge.

HOUSE: 13.3V +(-0.7V) = 12.6V battery vs. 13.2 V charger

... the battery won't supply, it can't be charged from house against the diode...
... charger does it all...

Your "intelligent!" charger did notice that no current was flowing into the battery --> no bulk cycle.
Your "intelligent! Pb charger" also noticed that the battery voltage was too low
while current was needed which it had to supply all alone
--> so the "defunct Pb battery" is "conditioned".

Seems all very (ILL-) Logical to me ...
...but what do I know about your system, especially the "Sterling Ultra 60"

At least maybe we need a schematic