Gen II Prius Individual Battery Module Replacement

When you got the second Hitec X4 did you get a second PC power supply?

 

How do you program the 7250 mAhr charge/discharge on the Hitec X4? Do you need to be in manual mode or auto mode?

 

How do you set the mAhr? The menu for working with NiMH on my X4 doesn't include a step for changing mAhr. I can chose Auto/Manual, charge, discharge and number of cycles. Thanks

 

In my unit (a Supermate DC6) there is a "safety" timer that you set to limit the charging cycle. If you are charging at 5 amps, setting the limit at 1.5 hours would result in a 7.5 amp-hour limit or 7500 milliamp-hours.

JeffD

 

I ran 3 cycles on 4 modules without setting the target charge and got the following results:


Cycle 1 Cycle 2 Cycle 3
Cell # Dis Chg Dis Chg Dis Chg
1 901 5000 1399 3896 1399 3406
3 1399 3829 1399 1358 1399 2124
5 1399 4144 1399 2916 1399 3039
7 1399 4263 1399 3160 1399 3351

It looks like there's another setting for capacity shutoff that I need to enable. The way I understand it, what I did was tell it charge until the charger determines the battery is full when it would stop. From what I can tell, the charger determined the batteries were full well before 7250 mAh so I'm not sure enabling capacity shutoff will do anything.

I'm charging at 5 A...

 

I think I am just a few days ahead of you in my balancing. I had remarkably similar (disheartening) results after my first attempted cycle:

Module DCHG1 CHG1 DSCH2 CHG2 DSHC3 CHG3
1 1399 3615 1348 2553 1351 2674
15 1399 3289 1346 2674 1352 2430

I am using the Supermate DC6 that JeffD, BritPrius, etc. use but it seems many of these charger models have similar (and not entirely intuitive) user interfaces. Hence, I will tell you what I found in case you have a similar problem.

First, note that 1400 mAhr at a 0.7A discharge works out to 2 hrs (120 min). (0.7A is slightly over the max current for a charger that can handle a 5W discharge. I presume this is limited by the heat dissipation of the charger.) On my DC6, 120 min was the default setting for the charging safety timer. In fact, this is the value I wanted:
7.25 Ahr/5A = 1.45 hr = 87 min
Consider that the delta V will periodically interrupt the 5A charging in order to take a 'resting' voltage of the battery. (I have yet to actually scope this to get the period or length of the rest time. I do see this 'rest period' as I manually observe the charge current.) The instruction manual for the DC6 recommends setting the safety timer at 140% of the calculated time needed for the capacity sought. Doing this leads to:
(87 min)*1.4 = ~120 min

I think on my charger that the cutoff time was not just the setting for the charging as I could not find a similar setting for the discharge side. When using the automated "DSCH->CHG" cycling, the cutoff time appears to be applied to the DSCH period as well as the CHG period. In any case, I increased the value to 720 min and reran. I got much saner (though not stellar) results:

Module DCHG1 CHG1 DSCH2 CHG2 DSHC3 CHG3
1 5591 7251 6104 7251 6226 7250 (yes!)
15 4462 6606 5198 7099 5431 7250 (so so ... this is the middle of the pack)

On the DC6, the safety timer setting is somewhere in the "user set program" menu. (The capacity cutoff is also in this menu.)

I am curious if there is a reason why you are charging modules in relatively close proximity to each other. (I am assuming you started charging at roughly the same time but maybe you have staggered the starts.) Perhaps I am being anal, but I have spaced my 4 chargers evenly across the pack (e.g. modules 1,8.15,22) to avoid unnecessary heat buildup and have staggered the start times (somewhat) to avoid charging too many modules simultaneously.

Good luck and I look forward to following your progress!
Marc

 

That's a good point. I'm planning to get a second (4-channel) charger, so I didn't want to space charging out too much. I was think that skipping one module would be sufficient, but maybe I can skip two and still run two chargers simultaneously.

I'll have to look at the default discharge time to see how if it was timing out instead of getting to 6V. That would explain a lot. Thanks for point that out! I might have an old heat sink that I could add to the bottom of the charger.

 

The long discharge time is what makes this process so laborious (~6.5 Ahr @ 0.7A discharge = 9+ hrs). It seems a shame that this passive process takes so much time but I suppose it's not an issue for the fairly low-capacity RC battery world.

My chargers get only slightly warm during the discharge. As you suggest, adding heat sinks and/or putting some air flow across them would presumably allow them to discharge more without damage. However, the firmware in my chargers limit the discharge current according to the number of cells connected.

 

I did a quick search but didn't find what I was hoping to find. Is there a written description of what cycling the battery modules does to make them regain their capacity? I'm seeing a gradual increase in discharge capacity as I go through cycles, but I'm not sure why this is happening. I know its a good thing, I just would like to understand a little bit about why it happens.

Questions:
What is a good discharge capacity target? If I continue cycling beyond 3 cycles, will the capacity continue to increase or will it hit the limit of diminishing returns?

How do I measure internal resistance? Just the resistance of each module? What is a good value for each module/whole pack?

Why is it necessary to pair up modules of like capacity? Aren't they all connected in series, so how important is this (they are all side-by-side pairs)?

I've noticed some of the eBay sellers say they have already cycled their modules 3 times. Does anyone have any experience buying these, or should I just expect to cycle the cells after I get them?

Thanks

 

Each module has 6 battery cells in series. As a battery is cycled over and over in your Prius, the cells get out of balance. That means that some cells have a higher State of Charge (SOC) than the others in that module. The module amp-hour capacity is then limited by the weakest cell in that module. Rebalancing is a process where we repeatedly slightly over charge the module so that the weak cells can catch up while the stronger ones dissipate the excess energy as heat

Most good modules can recover to their rated capacity, 6.5 amp-hours.

I found that going beyond 3 cycles is not justified.

A good module will have a resistance of 10-15 milliohms and your Prius ECU will report the resistance of 2 modules in series (their resistances add). You can measure module resistance of a charged module by pulling a known current and measuring the voltage drop.

No, all of the modules are in series and the weakest one will limit overall battery capacity. All of the modules should have the same capacity and SOC when reinstalled.

A reliable seller (there are few) that says that the modules have been balance should guarantee a given capacity (I did when I sold my 27 rebalanced modules). One discharge-charge cycle will confirm if the module is ready to be used.

 

Thanks for the clarifications. I'm trying to think of a good load source so I can determine the internal resistance.

The third discharge for my first 4 modules gave the following: 4991, 5245, 4872, 4994 mA-h. These values were 10-35% better than the first discharge, but still nowhere near 6500 mA-h. Does this indicate my reconditioned pack will have limited life (substantially less than the 150,000 miles it has already given me)? Would there be any benefit to going for a higher charge to say 7500 mA-h or higher?

 

I am curious about the charge setting also. For a good battery module that still has the capability of achieving a 6.5 AHr capacity, charging with 7250 AHr implies a charging efficiency of 90%. From experience in other applications, NiMH charging efficiency is typically around 90% when charging at a 1C rate in the mid-part of the SOC. The efficiency gets worse as the charging rate drops, e.g. a NiMH trickle charge at 0.1C typically achieves only 70% efficiency in the mid-part of the charging. Also, the efficiency decreases as you get closer to a full SOC. (In the limit, the efficiency goes to 0 as would happen if we charged beyond the point where the delta-V stops the charging.)

Most of us charge with 5A (0.77C) so I would expect a charging efficiency slightly below 90% for the mid-point of the charging curve. Considering that we attempt to charge to near capacity, the efficiency could be significantly below 90%. All this implies a charging capacity cutoff of over 7250 mAhr might be better. I have seen in other posts where jdenenberg suggested a cutoff of 7500 mAhr (implying 87% charge efficiency).

I would love to see some data/statistics to determine if we could safely increase the charging limit and if the charging rate of 5A is optimal. It seems the following data could be useful:
- Charging efficiency in the middle part of the SOC for different charge rates (e.g. 0.1C, 0.5C, 1C). (The Prius module spec I have seen recommends charging in the range 1.3 to 3.25A. Why do they recommend 0.2C to 0.5C? This runs counter to charging efficiency recommendations in other contexts where 1C or higher is optimal.)
- Charging efficiency at different points on the SOC curve, e.g. when charged to 10%, 50%, 90%, etc.
- Absolute charging limit for typical modules (if there is such a thing), i.e. if charged all the way to the delta-V cutoff
- Internal resistance vs charging cycles (or age or mileage)

I imagine someone who does this type of rebuilding for a living (Dorman/Reinvolt, 3prongpaul) probably has some of this data. I imagine they would not want to share it which is perfectly understandable. If anyone is willing to share this data, please do so. Alternatively, a thorough explanation/derivation of someone's charging capacity limit and charging rate would be interesting. Otherwise, if there is interest in this data I might be willing to sacrifice a couple of modules and run some experiments provided experts (e.g. BritPrius, JeffD or the professionals (e.g. 3prongpaul, seilerts) provide some guidance.

 

A 12v light bulb is a useful load, but you will need to measure the current as well as the voltage

Are you driving in a hot climate or very hilly terrain? If so that would account for the somewhat degraded status of your modules. Your Prius will operate happily with an HV battery with 3 amp-hours capacity so you can assume that your modules, after balancing, will give you a reasonable number of miles (maybe 100k) before you will have to revisit this issue.

Note that the Gen3 modules are improved versions over those that were originally in a Gen2 Prius (lower series resistance and therefore less heat generation and higher peak current capability) so you should consider a full set of them next time.

I always set my safety limit to 7500 mA-hours and the charging cuttoff was done by the Delta-V (The module voltage under charge is compared to a momentary resting voltage) function of my SuperMate DC6. This raised all 27 of my good modules (195k miles of mostly highway use in the NorthEast) to over 6.5 amp-hours of capacity.

JeffD

 

I guess you figured out that the instruction manual is not that great for the x4. I think the capacity setting is in the setup menu that you save. The auto/man is switched while the A is flashing by pushing the increase and decrease buttons together.I ran three of the Hitec x4 chargers on automatic mode set at 5A charge, .7A discharge, 7250 mAh capacity cutoff, and 660 minute safety timer cutoff. Two of the x4chargers were DC powered using a car and a boat battery with a trickle charger for each one. The other x4 had its own AC power supply. The car battery did go down on me once or twice by not watching the charge level since I had to go to work. The deep cycle battery did fine with just the trickle charger on it. I charged every other module and set a fan to blow on the battery pack.
I was able to get all of my mods between 5500 and 6200 mAh of discharge and 7250mAh charge. There were 2 or 3 that I had to run several cycles on. Either I had some kind of operator error or the delta V caused a shutoff and I got some really low numbers. I was going to pitch those mods but decided to try again and they behaved. I thought about trying another cycle or 2 to get them at the 6500mAh mark but decided I didn't want to put they extra time in it or take a chance on messing one up. I decided to group my mods in pairs with the closest capacities and to put the lowest ones to the oustide and the best ones in the center. My hope is that the ones near the ends will more easily dissipate the heat that will result from more frequent charging. I was about to install the battery pack today when the sunshine melted into rain and I don't have a shop. Guess I'll have to wait some more.

 

These are the type of questions I'd like to find answers to, but probably will be happy to just get it back on the road. Maybe someone could offer to do this locally for all of the cars that will be needing reconditioned batteries! It could be an opportunity. The local wrecking yard is charging $1000! for a pack pulled out of a wrecked car. That seems to be their standard price, regardless of age/miles on the car. No warranty!

I ended up buying 2 of the X4 chargers and and running both from a computer power supply that's rated for 550 watts at 12V+. Instead of cutting the wires, I riffled through a scrap bin and found a motherboard adaptor so I could just plug right in to the power supply wiring harness. I cut the wires coming out of the adaptor and tied those in to supply power to the chargers.

I started the next 8 modules this morning before I left for work. Maybe the next I'll run batch to 7500 mA-h as JeffD does. Would mind getting everything up to/above 5500-6000 mA-h.

I'm not sure my wife shares my view, but its kind of an interesting project!

 

The last couple nights its been getting down to at/below freezing, so I'm wondering how much cold temperatures would limit modules during the discharge cycle. Last night I recorded 1st cycle discharges when I stopped the chargers during the 2nd cycle discharge so I could change the charge limit to 7500 mA-h. The 1st cycle discharges weren't that great, 2900-4000 mA-h. I wasn't there during the cycle, but I wonder if the cold limited the discharge rate and the timer timed out as opposed to the modules getting to 6 V. When I restarted the chargers for 7500, it seemed like the voltage on some modules wasn't dropping between last night and this morning.

Would it help to wrap a blanket around the pack? Thanks

 

Cold will affect the modules, but take care as overheating can destroy a module. If you stopped a discharge cycle and then restarted one, the total discharge in the restarted cycle will be low since you had already lowered the SOC befor the new cycle.

JeffD

 

The first cycle was undisturbed. I stopped the second discharge and will disregard it when analyzing the data. I'll have to wait to see what the final charge looks like.

 

Good point.

Per my thinking, if one starts this process with a discharge, that first discharge will merely drain the battery from whatever state of charge the Prius left in it (minus, depending on timing and module health, a little self-discharge). The target SOC when in the vehicle is in the mid-part of the curve, i.e. 40-80%. It will not be near a full (100%) charge. Given that range of expected SOC, I don't think that first discharge figure if very indicative. Correct?

 

ok need everyones expert advice. Heres whats going on. Battery goes from full charge to dead and back very fast. Also from a stop it drives like a manual in 2nd gear. Very sluggish. I let the car idle for about 20min before removing the battery so I could warm up the inside of it. Im in Park City Ut and its very cold out. Would the extra idling cause a bad cell to read as normal? What else can I do? Assuming its a battery issue I removed the battery and volt metered every cell

1 8.14
2 8.14
3 8.14
4 8.13
5 8.13
6 8.13
7 8.13
8 8.13
9 8.13
10 8.12
11 8.13
12 8.13
13 8.13
14 8.12
15 8.13
16 8.13
17 8.13
18 8.13
19 8.12
20 8.13
21 8.13
22 8.12
23 8.12
24 8.14
25 8.13
26 8.13
27 8.14
28 8.15

So Im at a loss. now what? Please help.