Thanks for the tip. Had another negative (bottom end) recal yesterday on a different road, complete with the forced regen. It appears that recal triggers more readily (battery drains faster?) when the car is driven hard in the cold morning right after startup. I suspect it has something to do with the wide temperature swing here in Colorado, night = 50F and day = 85F.

My Hybrid Automotive charger & discharger (simple light bulb type) is due to arrive today according to USPS, so I'll try installing the harness tonight.

Is it harmful to grid-charge for just ~8 hours and drive away? I understand it won't achieve true 100% balance if I don't let it finish but is it harmful to the battery if I stop charging in the middle?

 

There is no harm in partially charging. 8 hours may actually be enough -IF- you first top off the battery (either while driving it or Rev to 3000 in park to get charging going, and then run it charge to 8 bars).

The grid charger is doing small currents, so that's why it takes a long time.

And you definitely WANT the small currents when you get to 100% full because high currents would create dangerous heat.

 

Like Hunter said, no harm, but I would disagree that it may be enough. In an unbalanced pack, the weakest cells can be at a significantly low state of charge when the car thinks the pack is full, AND charging above 80% SoC becomes significantly less than 100% efficient.

It will take at least 5 hrs to get the strongest cells topped off. Given the description and frequency of recals, I would estimate your weakest cells could be as low as 30% when the car thinks they are full, and they will take on the order of at least 12 hours. I have personally seen cells in a severely unbalanced pack at 0% SoC, but this car threw an IMA light and would do recals every 5-10 minutes under non-cruise driving.

As you can see, there is a lot of uncertainty. As a general rule, if you hold peak voltage for 5+ hours, you can be confident you are topping off the pack and achieving top balancing.

The fastest way to top balance:

Make sure you have your Navi or radio security code.
Disconnect your 12V for 30 seconds.
reconnect 12V
Start and idle car until you show 100% SoC.
Grid charge to peak voltage and maintain peak voltage for 5 hours.
Total grid charge time not to exceed 24 hours.

The HA grid charger only reports voltage to the ones place. When possible, I like to get resolution to the tenths so peak voltage becomes more obvious. To my knowledge, there is no way to hook a more precise voltmeter inline with the charger. In any case, if you hold a given peak voltage for 5 hours, even to the nearest volt, you should be close enough to not really matter.

Note that in 75°F conditions, you should expect a peak voltage in the 188-194V range.

Since your recals aren't that frequent, you can likely expect very positive results.

While I believe HA recommends multiple progressively deeper discharges, I do not subscribe to this. Every cell but a few will get reversed in the process. In the multiple cycle deep discharges, the weakest cells get reversed multiple times. I prefer to just do it all once.

I recommend a single deep discharge to less than 1V at less than 1A current when at 120V, so a 100W bulb is fine as that will give you 0.83A. Note that the single bulb discharger puts a hefty strain on the bulb because of the initial voltage of 180+. At that point, the bulb is at 150% rated power, will shine like the sun and possibly burn out.

If at all possible, I recommend the addition of an ammeter in series at the discharger. By recording time, volts and amps, you can compute the total capacity of your pack. You only record values from the top down to 120V as capacity below that is not useable. If you don't want to record data points, you can set a video to record or download an app that will take time-lapse pictures.

Note that when you do a deep discharge, this eliminates any voltage depression and can cause your next grid charge peak voltage to be a little higher, typically 1-2V.

If you're in a hurry, after the deep discharge, hit it with the grid charger for about 5 minutes to get your pack voltage up to near 120V or so. Disconnect your grid charger and do another 12V reset. Start the car and let it idle until it shows 100% SoC. This should only take about 15-20 minutes and get you to 80% SoC. Then it's only another 5-8 hours grid charging to get back to peak voltage.

Good luck!

Steve

 

Thanks! if I have time to actually install the charger tonight, I'll probably just drive it downhill (on that 4 mile hill next to my house) to get it showing 8 bars, then grid charge at the 350mA. Depending on my battery's real condition it might be balanced by the time I drive it to work tomorrow morning.

From the description of the deep-discharge process, everything I have read assume that discharge should begin at 100% SOC (not 8 bars on dash but "real" 100% SOC). Why is this? Why can't I drive around, get to 50% SOC, and then do the discharge? Is this because of the risk of reversal would be greater if I drain at the high rate of discharge that typically occurs with "ASSIST"?

Also, I wonder if I should get a european 220V light bulb for that first phase of discharge at 180V.

 

I still recommend a 12V reset. It guarantees max SoC. 8 bars during normal driving doesn't necessarily mean max SoC. If you can manage a full 12 hour charge, you'll probably get pretty close to true 100%.

Car max SoC is true 80%, but it is ultimately limited by the strongest subpack. Regardless of the actual state of the battery if one subpack of the 11 hits the upper voltage limit during charge, the car will stop charging and register max SoC. So 8 bars just tells you that one of 11 subpacks is at ~80% SoC.

Best to top balance rather than run the weakest cells in full reversal for the full duration of the discharge (will take a long time). I have tested individual sticks that put out as little as 45mAh as-removed from the pack (when the car thinks it's fully charged). The strong sticks put out 4,000mAh. The stick had a single cell that was completely discharged. Top balancing absolutely minimizes the duration of polarity reversal and ensures it occurs at the lowest voltage and lowest current. It's a "better safe than sorry" situation.

Your drive around until you get to 50% SoC would be viable AFTER a 100% top balance. The timing required to make that happen might be a bit tricky. The lower current draw of the bulb discharge is a smoother process and minimizes the impact to cells that may have increased internal resistance or voltage depression. However, that's mostly academic as the "grid-charge-only" used to be the primary fix, and deep discharging was explored as a follow-up measure later.

I'm guessing you're a really impatient person looking to shave time off everywhere. Please take your time on the grid charger harness install and don't rush. That's some high voltage **** in there.

The idea of a Euro spec bulb for 220V isn't a bad one. I'd look for a 200W.

The original discharger developed for this purpose was a Lowe's 600W halogen lamp with the 2X 300W bulbs re-wired in series. Due to it's higher current, the cycles with progressively deeper discharges was utilized.

Good luck!

Steve

 

Hehehe. I like to get things started, so just wanted to know if installing the charger tonight is worthwhile if it I can't start the charge-discharge-charge until I have 50+ hours free. I estimate ~20 hour balance assuming there are some very un-charged cells at 80%SOC, then ~12h drain, then ~17h recharge (6Ah/ 350mA) purely on grid charger. I guess I will just install the harness tonight, check charger operation for a few hours, then do the entire process over this weekend. I think I can park the car on Thursday night and keep it there until Monday...

 

Heh. I get it.

This is one of those times where patience pays off. Limited fiddling about with shorter times won't hurt anything. As long as you commit to a single full grid charge and discharge, you'll get there. Furthermore, once you get there, don't over-fiddle with it. Once you restore improved function to the pack, it's best to leave it along until you notice a significant performance hit and/or recalibrations. If you want to do it as a maintenance item, I recommend quarterly.

Lastly, I recommend the "shortcut" I mentioned on the post-discharge charge (use the car at idle). The high current charge the car provides can help rejuvenate inactive electrolyte improving capacity and reducing internal resistance. ~15 minutes of idling saves 10+ hours off the recharge time.

Steve

 

I think I read something from bumblebee about gridcharging at 6A to "break through," is that the line of thinking here? Seems most others recommend a slow gridcharge after deep discharge. Also, the shortcut charge would have to occur after slow gridcharging to about 120V right? I think you already said this just confirming that the car is not designed to charge from 0V.

About instrumentation and measuring current: in my old job I use to do just that sort of thing, oscilloscopes, sign wave generators, National Instruments data logger, etc. The main thing I did was doing nano-voltage measurements over a spot welded 0.003" diameter wire. I never thought much of it back then, but now that I think about it that nanovoltmeter cost twice as much as my HCHII. haha. The unitek spot welder cost more than my HCHII as well. These days I am content with my $20 voltmeter and turn-key solutions--thus the HybridAutomotive unit instead of building my own.

 

That is exactly it except I believe the car will charge at about 20A. I routinely do subpack charges with 15-20A and have no data to suggest this harms it in any way. Besides, full braking regen will dump about 50A into the pack.

The car might charge an extremely low pack voltage, but I know it will throw a low voltage input for the subpack taps, and you start out with an IMA light. I prefer to not run the risk of having the car freak out. It only takes a few minutes to get it back up to 120-132V.

Heh... your standards are still too high... get the $4-5 Harbor Freight voltmeter...

I have about a dozen. Finally sucked it up and bought a decent $25 one with autoranging and such. Campbell's soup can seem like ambrosia when you're used to dirty-dishwater-stew.

 

If you had this Fluke you'd never need another one...

But that's like 4 months worth of fuel. Or a new set of tires. or something else useful.