Gen II Prius Individual Battery Module Replacement

Everything is still working well and my discharge results look pretty good on the first 4 modules after 3 cycles, over 5,200 mAh.

Quick question, is it acceptable to discharge at 1A? I read that a lot of people are using 0.7A,so I want to make sure 1A isn't going to hurt. If not, can I go higher than 1A for discharge?

 

0.7A is used because the toy chargers are limited to 5W discharge power.

I discharge at 20A.

 

So does that mean I must use 0.7A for discharging to reach the desired level of discharge? What are the implications of using 1A? Please pardon my ignorance.

 

No problem. I thought you would connect the dots.

0.7A is used because the equipment is very limited.

I use 20A.

The car will discharge at 120A

1A is much less than 20A or 120A, so the only implications of discharging at 1A are that it will take a little less time than 0.7A, and you need to use 1A for all discharges for consistency.

However, in reviewing your choice of chargers, I believe your charger is limited to 10W/2A. 10W/7.2V = 1.3A.

I recommend you set to 2A and accept what it gives you, 1.3A at start and 1.7A at end.

 

The hobby chargers usually just have 1A discharge setting and in practice it usually comes out to around 0.7A on the readout.

 

So everything is looking good so far. A couple of questions though:

I thought 3 modules needed replacement, but I was able to do additional cycles on 2 of the 3 to get them up to >5k mAh discharge. Should I just go ahead and replace these anyway, or could I use them with no worries?

Also, what is the minimum gauge wire that I should use to do the final voltage balancing? I have 24awg untinned lashing wire, but I'm afraid that may be too thin. I'm trying to avoid the lengthy process of building the connectors from speaker wire. Any recommendations?

 

May be fine. If you're just going by capacity @ 5-10W, you can't be sure.

"Voltage balancing" = you're going to hook all the modules in parallel to "balance" their voltage? If so, please don't. If you leave it for about a week, you might get there. NiMH reacts very quickly to voltage changes to the point that very little current flows past an initial surge.

Better to charge empty modules to a given peak voltage or discharge them to a given min voltage.

Good luck,

Steve

 

I'm only following the example from other people I've seen in this thread, which has always been hooking the pack up in parallel to balance voltages. I want to do everything right so I can get the maximum life from this battery. If discharging to a predetermined voltage will yield better results, then I'm all for it. I should do this with the X4, or is there a better way? Could I just use the 55W halogen bulb I used to test the voltage drops initially?

 

Not by me. There are many other references on the site concerning the ineffectiveness of parallel charging/harnessing. It creates an illusion of balance by imparting a "surface" charge. They will eventually equalize, but it's going to take a while to get there at the trickle the tiny voltage disparity allows.

Given that you have a quad charger, you could easily balance in less time than it would take to effectively achieve balance by harnessing a whole pack.

If your modules are all at full charge, discharge @1.2A to some arbitrary voltage well above 7.2V. You only need to bleed off a couple hundred mAh. Start with your lowest capacity module and discharge 200mAh and see where that puts you. Discharge all modules to that voltage at the same current. You're not just balancing based on voltage in this case, but to internal resistance as well.

If they are in the discharged state, charge at max current (6A?) until a given voltage is attained. I would recommend it be some arbitrary voltage achieved AFTER you have put at least 1000mAh into the module.

Yesterday, I assembled a pack using an analogous matching technique. During an in-car discharge test as described in my sig, the DeltaV at the end of the test @ 40% SoC was 0.17V. In the 60% SoC range, it was in the 0.07V range.

 

Awesome, Keith. Thanks for the advice.

I'll do the discharge method with the X4 as you suggested. I actually prefer this method because I'm already setup for it, nothing else will be required.

 

Even better. I forgot about that.

Man... talk about cross-posting.

You will get better results with the 55W, and it's literally as easy as discharging to an arbitrary voltage. You might get it down to just a minute or two per module.

 

Make sure to hook up both the high and low beams on the headlight (55w is just the low beam) at once so you can draw maximum current when doing a voltage drop test. Run the test for 2-3 minutes each then make a list and compare the voltage drop. Ones in good shape will have a lower voltage drop while the ones that are not so good, vice versa.

 

Had picked up car around 145k miles and found out later someone had cobbled together a working battery..no serials # inline ..half cells have red dots.. missing or broken underside screws.

Pulled the HV pack a couple of weeks ago at 194k miles - have already had out twice this year for single cell replacement. When first pulled 3 showed below 7v , rechecked voltages (not a load test) next morning , 5 modules were below 7v . There were two more at or below 7.2 v and now they show under 7v also. Dropping like flies ..ack!

Got a 6 cells on way from eBay and trying to get 7th locally. Was going to do the described reconditioning and try to balance , have purchased a REAKTOR 250W/10A and power adapter . To try out and set up like the B6 redux video I grabbed a old bad cell removed previously and it was DSC and CRC very low 260mAh / 1261mAh and swelled up even though it was charged in a vice bracketed by 2x4 .

My next step was too charge one of the eBay received cells and it actually shut off charging because it hit 7501 mAh ..Reset for 2 auto cycle and it finished 7268 mAh ...but swollen slightly .

Left today with the REAKTOR running on cell1 , and will hopefully see no swelling as it is still mounted in the stack. Am curious what capacity it will show when complete.

I still think this current pack is not worth the trouble of doing and focus on a whole batt purchase , and shove my current pack back in the vehicle with the 7 cells replaced . Will grab a bulb and load test before as it is quick , but the real #s seem to be coming from the REAKTOR - but afraid in going through the cycling will cause more cell swelling .

Kenny .

 

The above is a perfect example of what simple module replacement gets you. Even a "cobbled together" pack can be made to function very well within the limited operational range the car uses if efforts are made to balance the modules. Simply slapping them in and hoping for the best is a recipe for repeated efforts.

IIRC, Reaktor 250W is limited to 20W internal discharge, 2.5A-3.3A (beginning @8V and ending @6V)

Swelling is normal and unavoidable above 80% SoC. Set NiMH sensitivity to 3mV and NiMH forming charge to 6.5Ah and 6 cells. Set capacity limit to 8450mAh. Charge at 6.5A max.

NiMH voltage change should halt about 10 hours after the last charge/discharge operation. Continued drop after this time suggests high self discharge. If the balance of your modules halt their drop, they are candidates for reconditioning.

Steve

 

Published new 3 minute video yesterday. To show how fast (~30 seconds) and easy it is to scan and check the hybrid battery modules manufacture dates.

You all know by now that the warranty could be void if the battery is not genuine (e.g. swapped for the older pack or one individual module was replaced).
You perhaps already had that experience that second hand batteries may not be what they are claimed to be by the sellers.
You've perhaps heard already how fast the battery in Prius can be swapped to another one. Takes about 20 minutes in second gen Prius as I was told by my friends in US / AUS who are doing them full time.

In 30 seconds it can be revealed if the battery is manufactured at about the same time as the car or whether there are any individual modules have been changed in an attempt to fix failed pack.

I've designed this tool first of all to help garages who use my Battery Analysers and are dealing with the huge number of packs, also wreckers, second hand battery dealers and compliance companies to make informed decisions.

For the DIY person buying professional scanner might not be justified investment but if you are buying the pack from those who do it full time – surely they can afford that - ask them for the manufacture dates and the load test results. If they can't produce those - buy from those who can...

Some facts:
- one pack in that video was made in late 2003. Owner said 1 module failed. Load tested - 1 dead, 5 modules - near the end of life (<1.5 - 2AH) the rest ~ 2.7-3.0AH (@ 4A load). That was genuine pack not worth rebuilding...
- another pack I've recently scanned claimed to be 2009 by wrecker - in fact was from 2003 :-( Clueless seller. Knowing them being honest - assume the pack was swapped prior to the car being dumped in the wreck yard (already happened few times in here) ...
- old leftovers from first Gen - 8 NP1s - one NP1 module was made in 2011 (at the end of that video) was taken from the wreck in 2015. That was replacement pack by the dealership and was looking promising. From 38 modules 36 were complete waste (<1AH), one survived and one was blown. The car was sitting for too long, modules were completely discharged and did not take charge (even "burp" charge). That supports another point - seemed to be "fresh" genuine 2011 pack could be complete waste :-(

 

Well, that's a nifty gadget for sure; however, you've established that it's not justifiable for a DIY person... in a decidedly DIY thread. For the DIY user, can't one just simply decode with the first four digits to establish manufacture dates?

ABCD

AB = two digit day of month
C = 1-9, X, Y, Z for month
D = Year, with "A" starting in 1999.

In my experience, unmolested packs always have those four digits the same, but I've only seen a dozen or so. Toyota's return guidelines permit up to 3 months variation, so your first example isn't a smoking gun at all. 9/27 to 12/16 is within a 3 month period.

I give you an "A" for "niftiness".

How much $?

 

Pulled the bus bars on the pack, and week later they have settled down no other cells below 7.2v .

Currently REAKTOR set to 3A current limit , 2A discharge down to 6.3v - Forming was set (but unused) 3A 7.2V (6S) , NiMH sensitivity is on default . Capacity cut off 7500 mAh ... had followed Ultaniums B6 redux video suggestions including cycling 3 times - would this be better reduced to a single pass or two ?

Kenny

 

Kenny,
Thank you for this story.
So often people assume that it will be be a walk in the park when it comes to fixing their battery pack.
Hard for them to believe me when I say it can be a real challenge.

 

Ultanium bought the Reaktor 300W on my recommendation, but I never watched his video.

Bottom line, 3A is too low of a charge current to reliably trip delta-V cut-off. You really need to be at 1C charge rate (6.5A, hence my recommendation) to get reliable delta V cut-off. No point in limiting discharge to 6.3V. Straight down to 6.0V @ 3A You won't pull that for the whole test, but you'll be at that level for the important last portion of it.

Cycles are effective, but you have to make your own call.

 

AB= 21st, C= February, and H = 2006