Hybrid battery balancing

I'm not sure if I made myself clear - the battery is removed from the car,
sitting on a table in one piece as it came from the car with all wiring
disconnected/ modules are disconnected from one another and I am doing
a module by module reconditioning not a pack level reconditioning/balancing.

I am using the CQ3 charger with slightly reduced settings (post #117) to what TMR-JWAP had
in post #76 of this thread.The modules (info from the Toyota technician) are the same
as the Prius C and there are 20 modules in the pack.

I am writing because it was my first charging and I had a bit of a panic attack during the
charging phase when I saw what I then thought of as being high battery voltages and that
the charging was not stopping via delta V test - I thought the voltage would rise to a level
at which the modules would be damaged so I stopped the charging at 4400mAh
on all modules.

From dolj's post I presume the voltages are normal which has given me confidence
that I may be doing the right thing and that the modules may be OK but I would appreciate
if someone experienced could confirm whether I should continue with my "more conservative settings"
or use TMR-JWAP's and whether I should add a an automatic capacity rule lower than 6000mAh
- I want to avoid stressing the modules and ending up with garbage at the end.

 

lower amp charging is universally considered safer. The thing is that even the very low amp charge cycling, is that there is still no guaranty you will not have garbage results from bad modules even with the lower slower cycling.
If I read correctly TMR-JWAP posted that he sets Delta V to default because the Delta V cutoff doesn't appear to work as it should for him.

If the cycling doesn't help the weaker module, you have to be prepared to locate replacements.

 

Well, AFAIK the whole point of cycling the modules is to "gently" stress them in order to recover capacity. Discharging and charging within "normal" limits seems to accomplish very little. The general consensus is that if a few cycles do not show decent capacity during the discharge phase, then you go to a lower voltage on the discharge and higher mAh on the charge. I can't find my notes from my refurb, but some modules needed 5-6 cycles - some I had to go down to 5V and charged 7500 mAh (no delta V)- with a fan pushing air through the bottom duct continuously. I replaced 5 or 6 modules total- seems to be doing decently over 1 year and 15,000+miles later.

 

I fully appreciate and agree with the need to "gently" stress the modules and

the methods to cycle to wake up the modules to see whether they can be revived but there's a few points
which are not clear to me after actually doing the first cycle :

Let's start with the charge cycle as that is where I have most concern now:
Does everyone allow the charger to go to voltages of the order I have seen (8.49V - 8.67V) (dolj confirmed
that they were normal - would like to hear what others have to say).

From TMR-JWAP's post above which looks like I didn't read closely enough/ associate closely enough to the charge/discharge
curve (attached) it may be that at the voltages I was seeing, I was just pumping charge into heat and I
should have stopped the charge at or below eg. 6*1.4V = 8.4V . I would have expected the charger to stop the charging
via the Delta V test before these voltages but the charger stuck to the voltage and kept pumping.
I am also now wondering whether my choice of 15mV for Delta V (from reading it should be 3mV-5mV/cell),
instead of making the test more sensitive than TMR-JWAP's 20mV - maybe the charger is not seeing the jump
in voltage and thinks it should continue charging. I am especially concerned about this because I am reading
how everyone is not able to reach charging capacity but in my case, the charger is just pumping away.

I don't have any temperature sensors and would really like to know how TMR-JWAP senses the "heat generation"
i.e. what type of sensor/where they are placed to get early heat detection as the modules are cooled
with air and I suppose it is difficult to measure unless it is very sensitive.

For the discharge cycle, I would like to know if my idea of going down to 6.6V instead of 6.0V makes sense
(at least for one cycle) to ensure most unbalanced cells are not reverse charged.

 

chances are the weak module you are worrying about is already at it life expectancy, but you can't tell from 1 cycle without tons of experience from cycling other modules. Second and third cycles will typically tell more, and 4th and 5th cycles should show the most a module has left to offer.
Weaker modules will charge to higher voltages. as well as show other signs of degradation before and during discharge. Before the discharge cycle they typically loose charge quicker than the stronger modules (self discharge) and during the discharge cycle (the discharge slope is much steeper) than the slope of stronger cells.
Without knowing how TMR-JWAP gets his modules temperatures, (I'd also be interested to know ) you can use and infered thermometer, remove the fan for a minute at take a few reading to see if the temps increase in the module. or just use you hand to see if the module is getting hotter.

Finally, limiting the discharge cutoff to 6.6 volts for the first cycle is OK. some might not agree but in my mind it's just nit picking about how another person chooses to get there discharge data. Further,

there is a process called deep discharge where the operator reduces the discharge voltage on successive cycles. Again, some have positive results and some others don't.

 

Sorry if I'm beating a dead horse asking all sorts of questions but it beats me making a lot of mistakes.
You didn't say anything about the voltages I quoted (8.49V to 8.67V) so I assume you also consider these voltages
to be OK for the charging cycle.
What do you consider as normal voltages for a newer module and "higher voltages"
of older cells ?
Does the charging cycle end for you via Delta V check from the charger or seem not to end
(voltage not changing or sometimes stepping down slightly and then back up) with charge continuing / should I
just be using a lower capacity cutoff ?
It is sounding to me like I'm the only one the charge cycle seems to be continuing or am I coming to the
wrong conclusion i.e. it will eventually drop voltage and Delta V check stop the charge ?

Thanks, that gives me confidence that my logic makes sense.

Hope others can chime in and give their view.

 

Yes,
Now I get what Dolj was referring to when mentioning the 20% to 80% SOC and .34 amp float charge. As well as your concerns about continuing the 1.6 amp charge above that voltage level in the modules.

A lot more than I could have imagined from reading some of the three threads I've seen your posts in about the yaris.

 

The final discharge and charge voltage is going to vary depending on the current setpoint you're using, the wire gauge you're using, and the internal resistance of the module. There's no perfect answer.

In my case, with my settings, if I have a final charge voltage that's over 9.0, I consider it a trigger for further evaluation/examining.
Let's assume all 28 modules being cycled are in the same range of 9-9.1 volts. Is that a problem? Not necessarily. Do they all test with similar capacity? Are any swelling? Are they all holding the charge?

If I have 28 modules and 26 finish at 8.69 volts and 2 of them finish at >9 volts, then it's more likely those modules have some issues. It's also likely those modules may show some swelling compared to the others. When in doubt, throw it out. There's no need to build a battery that has a marginal module. Why go through all the work and then slap it together with a questionable module?

Temperatures are easy. My test rig is the case bottom from a Gen 2 with an OEM duct and OEM fan blowing air into it. When I attach an assembly of 28 compressed modules on the case, I also attach the bottom plastic flow blocker, which creates an air supply plenum under the modules. Cooling air is now forced upward between all 28 modules. It's basically just like OEM but in the opposite direction.

If I have any concerns about a module I can monitor it's temperature either using my fluke temperature gun or extech Type K thermocouple reader. Every module has a 'well' in the top that works pretty well for inserting a thermocouple tip. You can even pack the 'well' with bits of paper towel to seal the tip in it. It's pretty rare I worry about temperature because there's plenty of other indications of a module being bad.

Every time I rig up a new set of 28 modules, I ALWAYS perform the discharge first so I know which modules are likely to be problems.
During the charge cycle, I can run my hand across the top of the battery and feel the air flow. It's easy to feel warm air or not warm air. If it's only halfway through the charge cycle and the air is extremely warm.

 

Hat off to you/thank you very much TMR-JWAP !!!- I now feel confident to continue and hopefuly
I don't ruin my daughter's battery

 

If you haven't already taken a look at the Wiki on HV battery repair I started, take a look. It references lots of other videos and work by earlier/better folks but will be hopefully useful in adding large amount of info you'll need to know to do this project. I think it's in my signature/footer right below all my posts...

 

In addition to mentioning "read up as much as possible", as I just did in my earlier reply to you, if I were doing it again, I'd put little water/party balloons over all the vent valves in the modules. Yes, you want to be working on them individually in your garage, basement etc. No you don't want NiMH venting to your space!

 

Yes, you really want a fan pushing air on your precious modules. I took the ducting from the car and taped it up to the modules kept bolted into their battery case's rods/cage. I connected a blower (really just a large 12v computer fan to a 12v power source (heck, I bet you have one!) and that blew air through the ducting and over the modules (just like in the car!) all the time they were charging...

 

Have a CQ3 coming in the mail to repair my hybrid battery pack on a 2008 Prius. Is 14awg an okay wire gauge for the alligator clip wires? If not what gauge should they be?

I ask because ordered the alligator clips and banana plugs already, and just need to get the wire to make the leads. I'm thinking 14 awg should handle the current easily, but not too sure about the measurement aspect. Specifically wanting to make sure the leads will have a voltage drop good enough to not cause a measurement error.

Want to get this right. I miss driving my Prius. Especially at the gas station.

 

14 is plenty of wire. I use 16g speaker wire, but my cycle regimen for modules uses 2a for charging and 1.5a for discharge. I use heavier cable for load testing at 60a.

 

Thanks! I'm planning to use your charge settings so that's good to know!

 

So I completed 5 discharge > charge cycles on the cells. It went faster because Amazon lost my CQ3 in shipping, gave me a refund which I used to buy another CQ3 on Amazon, and they both showed up the same day. So I got lucky and ended up being able to do 8 cells at a time. By the 5th cycle all the cells had a 5.5ish amp hour discharge.

None of the cells seemed bad, so being cautious I did a 2 minute load test with a headlight on each cell. None of the cells seemed to have that bad of a voltage drop, and all but one where pretty closer to each other. The outlier had much less of a voltage drop. Wondering, could the P0A80 just be corrosion on the terminals? They were pretty corroded Here's what they looked like:


I've cleaned the corrosion with vinegar, salt and boiling water. The corrosion certainly added a lot of resistance. Would it be okay to put Ox-gard on them when I finish rebuilding the pack after balancing finishes?

 

Here's a link to the spreadsheet if anyone want to look at the results yourself:

Basically the spreadsheet documents (from left to right):
Took an initial voltage of each cell
Discharged and Charged each cell 5 times, red highlights on low discharges, green on good seeming cells
Stats about the discharges
Cell voltages after discharging and charging
Battery load testing voltages
Deltas (differences) after a 2 minute load test
The last column will be cell voltages after balancing the cells for 48 hours


Won't know till it's done, but hoping to get her back on the road.

 

TMR-JWAP: Trying to use your settings the CQ3 charger (quoted below)
but when I start the program (D->C)
the CQ3 charger is going directly into delay "DLY" after 5 seconds.
The delay is set for 20 min.,
and after 20 min.,
the program starts the "charge" cycle instead of the "discharge" cycle.
I am new at this, I've not had the charger long, but wonder if there is something wrong with the charger or something that I am doing wrong.
Thank you for your time and efforts!

 

Have you measured the starting voltage and if so, how does it compare to the discharge value you have set? What you describe is seen when you set discharge for 7.2 volts and the module is already below that voltage (as an example).

That or you could have a bad charger module (whatever 1-4 is called). In my last charger purchase, I have one that consistently runs at a lower discharge amperage than all the others. Not huge enough to hassle for a refund, but enough to delay the discharge cycle.

 

Some comments now that I read backwards on your settings. I don't see your NIMH discharge value, which may be your problem. Go to the D/C option, then go left once. You should see NIMH discharge. It will already be set in terms of amps from the D/C screen, but you need to set the NIMH discharge voltage. No lower than 5.8v (minimum safe level I have found, have not gone lower). Higher is OK, but you won't be exercising the bottom voltage of the cell.

Safe is good, I burned about 4 modules in my experimentation. Learned much.

On the PROGRAM SELECT---USER SET menu:
Precharge Time ----OFF
Wait Time---CHG>DCHG----20min

NiMH sensitivity D. Peak 20mv/cell <---- you might want to leave this at default for the first cycle, just to catch modules that hit DeltaV and stop. This will allow you to keep some level of DeltaV protection while still gently overcharging. If it doesn't hit at default, there's no need to change it. If it does hit at default, you need to decrease the sensitivity by raising from default. If you don't care about any deltaV protection and don't mind burning the occasional module, 20 usually shuts it off (but not always).

NiCD sensitivity D. Peak Default
Temp Cutoff......OFF <------anyone have a source for temp sensors?
Safety Timer......OFF

Capacity Cut-off......7200 (I vary this anywhere from 5000 to 8200 depending on the modules) <--- If you track the initial charge value and then resulting discharge value, you can calculate the module efficiency for that cycle and adjust this value to hit 6000. i.e. first charge 5000, subsequent discharge 4000 = 80% efficiency. 6000/.8 = 7500 as an initial target. I don't use 6500 as a base as I know most all batteries being reconditioned will be below optimal from normal use.

Input Power low Cutoff.....11.0v

Key Beep....ON <------- This drives me nuts, personal preference to turn it off

Buzzer.....ON
Back Light.....50%
Battery end volt LiPo...4.2v/c

Then PROGRAM SELECT----NiMH Battery
NiMH CHARGE Man
CURRENT 2.0A

NiMH CHARGE Aut <-----this is the value used in a D/C or C/D cycle, it varies the charge amps as the battery charges (auto does this)

CURRENT 2.0A <----- this is a nice safe amperage, will take longer but you only get just below 2A on the discharge, so nothing lost.
As a side note, heat is the enemy, so I charge the pack standing on end with a fan blowing air through the pack, not just over the top. Works hands-free at this amperage most of the time. Get some small water balloons and put them on the exhaust nipples. If they start inflating, you are stressing the battery. Leave them on, the battery will re-absorb the gas (the baloons do not slowly deflate, they deflate suddenly.

NiMH DISCHARGE 1.5A 5.8V <----- I set this to 5, which you will never get on this charger. 2A max, but since this is the least stressful part of the cycle as well as the longest, I just max it out. I still only get 1.8-1.9A even when I set it higher.


NiMH CYCLE D->C
C=2.0A D=1.5A

I am using the same chargers, 4 of them as it takes forever cycle with just one and with airflow through the pack, no issues with doing 14 at a time.

 

One more thing, I never damaged any cell on the discharge, only on the charge (overcharge). No signs of stress at all - no gassing (balloons), no modules getting fat, nothing. Only on charge.

Therefore, on the C/D cycle, the 20 min delay lets the cells cool a bit and makes sense. On a D/C cycle I reduce delay to 5 and saw no impact. I was also able to overcome stubborn DeltaV (even at 20) by repeated discharges (waiting for the cell to bounce back up between discharges). Multiple discharges may be too risky, but if I can't get capacity above 4100 because of deltaV, what do I have to lose?

When you figure 3-5 (avg 4) cycles per module * 12 hours per cycle * 28 modules, 20 min per cycle really adds up. Going higher amps will speed half the cycle but needs watching. I've gone as high as 1C (6.5a) in testing and saw no real return since I had to watch everything closely during charge. 2A worked just fine, no capacity improvement at higher amps and I didn't have to watch it.