Okay... one more data point...
Drove home and recorded the whole drive with Torque pro and charted it Excel...

• SoC is blue line. Value is on left axis. As you can see it wandered between 60 and 70.
• Amps is the current. Positive current is assist/discharge, negative current is regen/charge. As you can see, it's all over the place with each stop and start. That segment between 16:42 and 16:46 was freeway cruise once traffic cleared up. 17:05 and on was me sitting in the driveway talking on the phone.
• Fan is the fan speed X10, so it shows on the chart. Value on left axis. It peaked at 30 (speed 3). as you can see it was off for about 3/4 of the drive.
• Ambient is air temperature, left axis.
• T_in is the battery inlet temperature, left axis. Pretty stable the whole drive between about 80 and 88°F
• TM is the Temperature, maximum of the 3 temperature probes - almost always the middle one, left axis. Started at 82°F and ended at 108°F
• DeltaV is MaxV - MinV on the blocks, value on RIGHT axis. usually between .1 and .3, but it occasionally spikes higher under heavy assist or regen. Yes, my battery kinda sucks, and it looks like it improves as the dive progresses - indicative of imbalance and capacity deterioration. It's built from the garbage from 3 packs - part of my initial inventory two years ago. Zero matching on any parameter, zero balancing. Mis-matched in EVERY parameter - capacity, IR and SD. I would NEVER put this in a customer car. It is a "lead-the-fleet" battery, and I wanted to see how long a sh!tty pack would last... it's been two years now! Die already, dangit!

I was actually being a little conservative. My driving style commands about 42 mpg. I got 48 on this drive.

 

Thanks for sharing, it's good to be able to look at all that data on one graph so you can see exactly whats going on. Ya your DeltaV is pretty high im surprised it hasn't coded out, i've seen them throw codes for a >0.2 variance (at least thats what I was able to see when I graphed it). I ordered that EC570-90, some leads and a pack of diodes, should be fun to put it together and try it out this weekend, I hope that unit works! Still haven't heard back from the customer, I figured they would probably have me do the job so I waived the diagnostic fee, all I did was pull codes and check the aux battery and a few other things, I normally just let people apply the diagnostic fee towards labor, especially if I think they will have me do the job and it only took me 15 min to diagnose, but I did drive for about 30 min each way, but at least im in a Prius, hardly costs $1.50. My mileage sits around 42.7 because of how I drive, a little less now because I got some pretty aggressive snow tires on, which I would highly recommend to anyone who lives where it snows significantly. Also, got a 4-channel PicoScope yesterday with a pressure transducer kit, terminal probe kit, amp clamps, and their first look pulse diagnostic sensor kit (for the ICE engine but I thought it would be nice to also be able to look at the ICE and see if theres any issues, this sensor goes in the tailpipe or dipstick tube and feels exhaust pulses and can tell if theres a burnt valve, worn piston rings or other mechanical issues while the engine is running)

 

If you had a pack code out at > 0.2V delta, then something else was wrong besides just deltaV. IR was high, or deltaV was MUCH higher at some other point where current was low that you may not have captured. I've seen many codes result from pack that were very well behaved in the upper range, but a cell craps the bed at 60% SoC and deltaV goes to >1.2V with consistency. THAT will definitely code a car.


When you dig into the various repair manuals, you can find diagnostic criteria that 0.3V is the implied upper limit, but it won't code there. There are also higher allowances for transient deltaV. Yet another factor is data skew. All recorded data is read sequentially over the course of 0.6-0.9 seconds. The min/max voltages are read at different times, and they require all 14 voltages to be read with high/low selected from the list, and it can really skew the results during high current or current transition phases. Even a difference of 10-20ms between readings can have surprising results when current is changing.


The ELM327 reader and Torque Pro/Hybrid Assistant combo are also less capable than Techstream.


With Techstream when I subject my pack to forced charging at a sustained 70A, the deltaV is right around 0.35V. When I subject my pack to a discharge test at about 20A, the deltaV starts at around 0.15V at 80% SoC and finishes at about 0.28V at 40% SoC. Constant or near-constant current also helps eliminate the data skew I mentioned.


Regardless of all that, I KNOW this pack is meh. 3800-5000mAh capacity, Techstream reported IR from .027-.030 and about 25% variation in SD.


It's just mind boggling that it's lasted two years in that state, particularly, given the fact that there's two Phoenix summers in there, and I abuse the crap out of it... and my car is black and get's hotter than ***** in the summer even with windows cracked and a sunshade.


Anyway, be ready to test the EC570-90. With all used-things via ebay, they are either great or they are crap. I recently bought one for $25 with $20 shipping that powers up, looks like it's going to work, but then it just won't raise past a certain voltage or deliver more than 60mA current. If it doesn't work, you'll want to send it back. Keep an eye open for those things.


Sounds like you're armed for bear on the ICE side. I don't go much past the batteries, inverter coolant pumps, combo meters and 3 way valves. In fact, I'm trying to distance myself from all of that to focus on just batteries.


One thing I've noticed in over 100 Prius jobs... the vast majority of them neglect everything but oil changes. TB and MAF sensors almost always need a cleaning, and most people don't even think about changing the spark plugs. That plus an air filter makes for a very quick "tune-up" that can keep the ICE running like a top. About $35-40 in material cost and what... 30 minutes of work?

 

Yes in regards to oil changes i've seen the same thing, people just neglect to do them, I had a customer that used the AMS Oil 25k mile oil and I was like, no please stop using this, let's go synthetic and change it at 10k. I also love putting Seafoam in the oil and gas about 300 miles before a change, i've seen that make a huge difference in fuel economy and better idling, and reduce oil consumption. Yes the MAF always needs cleaning, and spark plugs haven't been changed since the car was under warranty, neither have the COPs. I have flip-flopped between only doing hybrid stuff and trying to expand into full-service, I was looking at getting a shop last year but it was just too big of a commitment ($2k/month with minimum 3 year lease) I need a lot more customers before I can commit to that. My thinking was if you have a shop to be full-service that way I could do everything for my customers, although not as profitable as batteries, but so I could be their one stop shop. I've been talking to some of the auctions and dealers in my state and trying to be their go to guy for anything hybrid related. I thought they would like it if I could also do some trouble-shooting and tune-up stuff on the ICE, if I can get good with the scope and accessory kits I would do some work on ICE cars if I don't have a lot of other work on hevs, the mechanic that knows electrical and is good at diagnostics is always sought after. I also got a CardaQ-M and I want to get into reprogramming, for hevs and especially pure evs they need to get the latest updates, which can solve a lot of problems and increase performance, but its not cheap for the OEM subscriptions, at this point i'm just going to pay per flash. You know whats crazy is the OEM thats the most expensive for their diagnostic and reflash capabilities, its not BMW or Mercedes, its freakin Volvo, almost $8K/year! I won't be working on any of those, at least not at the OE level, I had an old 240DL i got for free years ago (found out later it was infested with bed bugs, found it on CL the guy said if I give him a ride to the airport he'll give it to me, i thought it was a little strange when he showed up in it wearing scrubs, he said he worked at the hospital and didn't have time to change, LOL, gave my friend and his Japanese gf a ride to Costco and I thought they were just little baby roaches, they took a bus home LOL) ok way off topic, sorry. I also got an AutoEnginuity Proline bundle and I like it a lot for my generic scan tool, I haven't got a chance to hook it up to a Volt yet but i'll post on here when I do. Ok I need to get off this computer and clean the garage its already noon here!

 

One question Mr. Keith, I got the EC570-90, I soldered the diodes in the charge leads as you described, when I hook it up to a Prius battery and set it to charge it says no load. I'm going to try resoldering it, hopefully that is all the problem is and its not a bad unit.

 

Once you have the desired voltage and current set, the leads attached and the battery safety plug installed, you must press and HOLD the ON/OFF button until the voltage rises above the pack resting voltage and you start to see current increasing. Once current is flowing, you can release the button and it will continue to operate as a CC or CV power supply depending on set voltage/current.

If that doesn't work, make sure your diodes are not backwards. (+) lead has stripe on battery end. (-) has stripe on power supply end.

If your diodes are not backwards, you may have a defective unit.

 

Thank you for your quick response! I tried holding down the on/off button and the voltage and amperage starts fluctuating, I held it down for about 30 seconds but once you release the button it still says no load. I double checked the diodes they are wired correctly, I got a pack of 10 from ebay, maybe they're just cheap, or perhaps I got them too hot when I soldered them, but I was pretty careful and I don't think I damaged them. I tried it on an individual module, still no luck. I even tried using leads with no diodes on a module but it did the same thing.

 

As a test, set voltage to 260V current to 1000mA and time 1.00 hours.

With everything connected and safety plug in place, pressing and holding the ON/OFF button should result in 0-1mA of current displayed and a voltage that increases somewhat erratically, but in an upward trend. Once the voltage has risen above the pack voltage, voltage holds mostly constant, and current begins a steady climb to the set point. Once voltage is achieved and current is climbing, you can release the button.

You can also check continuity of your leads. In one direction, you should get 0Ω. In the other direction, you should get continuity.

If it's not a lead issue, and the power supply doesn't behave as I described, it sounds like a bad unit. Send it back.

I'm 1 for 2. I tried to buy a second one for backup advertised as "powers on, but untested", and it behaves somewhat as you describe except it peaks at 60V and won't go higher. Won't work on a single module either.

 

BTW... 48°F ambient this morning, 64°F battery inlet temp. Battery hit 111°F.

I used Hybrid Assistant to force the fan full blast, and it was down to 104°F in about 10 minutes. I let the car manage it at that point, and the fan was on 2-3 to maintain 104°F on a steady highway drive (very little battery current in either direction).

 

So I am working that Tahoe, the 12v battery was marginal and the positive clamp was loose, it was a quick release one so I couldn't just tighten it, I replaced both terminals. Problem was still there. I logged data with Tech2win, and it shows the #10,11, and 12 blocks are bad, there was a 1.7v-3.6v difference from the average voltage of the other 17 blocks. This was my first time using tech2win and its pretty crappy compared to techstream, it doesn't tell you what the min/max block voltage number is so I had to find them by looking at the block voltage data, and the weird thing is the numbers don't match perfectly, but its still easy to tell which are bad. It also doesn't tell you IR for blocks, only for the whole pack.

Anyways, I started rebuilding and I realized I don't know which number each block is in the pack, I know the bad ones are in the middle and towards one side, but I don't know which side they start counting from. I would assume they would start from the computer side of the pack, not the fan side, but i'm not sure. I'll have to dig through the service information tomorrow but I have a feeling they won't have it in there, because they don't rebuild packs, just swap them. Any ideas?