i've been busy swapping out my motor on my 2014 STI(woo hoo 569 whp, i know way opposite of my hch ii lol) that i didn't have time for my daily but,

here's an update, because i live in an apartment with 3 floors and living on the 3rd, i decided to go a route that you suggested not to. lol

i did this 3 weeks ago to get my inspection done and it's been holding. (did 3 negative recals total on the first two days after battery swap but nothing after that and it's been 3 weeks). i drive the car in traffic heavy roads during rush hour back and forth. i live in westchester ny so i drive down the 95s, bruckner, to the 87n and then local over a bridge over to the FDR south and then over the queensboro bridge over to my office in LIC in brooklyn, so it's really bad for the battery but it's been holding.

first the batteries - the SD rate of these batteries were horrible, i mean horrible. after the 4th day after a full 3 cycle charge(discharge to 1v/cell at stick level and then full charged to 1.45v/cell at subpack level) , it would not pass the load test. so i did a full discharge cycle to 0v per subpack. another reason why i did this was because at stick level their discharge rates were different, one stick may have been fully depleted while the other had charge. i wanted to even out the playing field so each subpack was discharged to 0v for at least 12 hours. i know you may cringe but i remember somewhere that Nickel batteries have SD and the worse they were maintained, the higher the SD were and a full cap discharge and then pulse charges may break up the crystalline formation which will slow down the SD rates. so what i did was i fully discharged all the subpacks, let it sit at 0v for about 3 hours, and then let the cells bounce back to about 5v at stick level and then pulse charged at 1 min charge 30 second rest until the whole subpack was at 15v and then full charge at 5 amps @ subpack level to 17.5 volts. after all that was done i did a 4 day load test and the results were spectacular. the sd rate was much lower and after a 20 second load test, they still held above 5.4 volts at stick level on load.

Second the pack itself.

i got 3 80-90 cfm fans and stuck them on the case with zip ties. i didn't use hot clue just because i hate it lol.

i cut 3 holes on the bottom of the case where the fans are mounted. a little background to why i did this is because when i put the 130+ cfm fans on first, it was WAY TOO LOUD in the car with the fans running, it was just painful to drive, second when i tested the fans, it wasn't pushing as much air because first, there wasn't much space to pull air from because of the channels on the pack itself. basically there was a giant plastic bar in the way for the fans to really pull air. so i cut out 3 inch holes on the pack itself to move some more air. and then i tested and it did start pulling more air but it was still too loud, but it was a little quieter due to the holes being cut. second, i got 80 cfm fans with more fan blades to quiet down the whole operation. it worked and with the whole pack mounted with the metal shroud and the seat back in, it is just a hum now. the fans come on via accessory power and i put a fuse tap on with a 10amp fuse on an empty slot of the fuse panel that was accessory key on. great thing about this is, i dont' know if the car pwm fan stays on after the car is turned off. so i can turn the engine off and leave it in accessory for 3 minutes after the car is off so i don't waste gas but still move some air in the batteries after use. i am maybe thinking about rigging my turbo timer i have laying around to accessory on even with the key out so the fans push air after i leave the car for about 5-10 minutes.

Third the car itself.

the car with the bad battery pack i was getting neg recal every 5-10 minutes. my drive is about 1.5 hours long each way so i was getting about avg 10 neg recals on my way to work. that took a toll on my mpg. i was getting about 26-28 mpgs on my fe meter but i would get only about 260 miles with 1 bar left on my gas gauge. now with the new pack i am getting about 36-38 mpgs. the auto stop start kicks in a lot but still has battery to do so. Never goes below 5 bars, and trickle charge kicks in cruising at 5 bars and stop when it reaches 6. i never got more than 2 auto stop starts during commute as it was always on neg recal. i filled up last week and when i filled up i got about 350 miles. which is wayyyyy better than what i got before. that's about 90 miles / 2 weeks. so it's about 2300 miles more i can get out of the car in a year which is about 900 dollars i can save a year in gas @2.55 a gallon. Again this is just 3 weeks of usage. another thing i did rig to lessen my hassle is something that you were against. i made a harness to come out to my trunk that connects directly to the battery. every 3 months or every 2 neg recals i will do a full discharge on the pack to close to 0v or 12 hours of discharge, and then let it sit for 2 hours to have the cells bounce back, disconnect my 12v battery for 30 seconds, start the car let it charge itself to full and repeat 2X more times. something i would like to disclose if this works or not is, the guy i sold my hch i to does it like this and he has only had to do this once when battery started to act up. i sold the car about 4 months ago. that car was my test mule but there was such a good deal on my current hchii that i had to give that one up. i don't know if this method would work but hopefully i can get at least 2 months out between discharges. i know this is just speculation but if this works, then there is no need for a trickle charger. also another benefit that this may bring is that a higher charge rate in may be able to lessen the SD rates on these cells as the higher charge rate my break up the crystalline formation in these nickel cells. i will check back the next time it starts acting up. another thing to maybe help seal the back panel and lower the noise produced by these fans, i was thinking of maybe using dynamat to seal the whole panel and seam seal the edges so all the air is being sucked out and there is not parasitic leak for the cooling system. i know it will definitely help with the sound deadening.

Fourth - a backup plan.

the orange harness connector i believe is where all the voltages are read, so if these are direct connections to the subpacks, do you think it may be possible to directly feed power to balance charge these subpacks? im pretty good mechanically and electrically so i think maybe if i splice into this harness and create a piggy back so it reads each 14.4v (subpack) i can maybe put a 500mah charger to the unbalanced pack or discharger to the unbalanced pack so i dont' deteriorate the WHOLE pack and concentrate on the problematic subpack. it's a shame that honda didn't go the step further and maybe redirect their DC to DC Converter power into this harness so it can balance out whichever cells were lower on voltage. that could've been a remedy to make the car perform up to par.

anyway i know that was a long read but if it works, hey it's easier to do this than for me to take the whole pack out of the car and walk up 3 floors to discharge and charge and then back down to reinstall and to note, i don't know if this method will work but as of now i have nothing to lose so why not?

 

The Honda case inlet and exit are weird like that due to the staggered sticks and flow dynamics. Altering the inlet or outlet may disrupt cooling and create hot spots.


It's not a shame that Honda didn't go the step further. It's an impossibility as the (-) of one pair is the (+) for the next. Monitoring it is one thing, applying current with the right polarity and being able to swap it is an entirely different matter - something no NiMH HEV manufacturer has deemed worthwhile.


It may be possible to do manually; however, the wiring is VERY thin, and you will need to make sure you have the polarity correct. It doesn't take much to handle 500mA, but those wires may not be enough.


As you've discovered, SD is a big problem. What you describe as "pulse" charging is not pulse charging. Pulse charging involves the application of high voltage current to create a very high current "pulse" that lasts for fractions of a second repeated many times a second. You just stopped and started your charging. The 5A charge to full is all that's needed.


Knowing that SD is the bigger issue, you've drawn the incorrect conclusion. You don't need periodic discharges, you need periodic charges.


There is no point in a waiting period after deep discharging. There is no usable energy in the cells, and you don't want the voltage to bounce to the point that the car thinks there's enough juice so that it pulls 30A from the pack when it tries to spin the ICE. Better to start off the 12V and have the car attempt to force charge and actually impart some capacity into the pack. It may take multiple 12V resets.

 

i don't think it will create hot spots, it may relieve more air when the air is being pulled. i did the modification at the exit of air of that pack, so air should still travel through and through all cells before hitting the fans.

im not saying they had to do it through that harness per say, im saying that they could've designed a balancing mechanism on stick or subpack level. i sure they have the resources.

hum... ok maybe i should make a something that can pulse charge a few times per second. maybe even try programming my powerlab 8 to do that.

my conclusion to SD may be incorrect but it's working for now and it's much a less of a hassle so let's see how this works out.

the car won't think it has enough juice as it started via the starter, basically it's like the car sat for a long time and SD itself dead which im sure the car would not even hit the IMA if it sees the voltages. with 2hrs to have the cells bounce back it wasn't even 110v so i think it's ok

on my HCH I i tried this right after i disconnected the discharger meaning it was like at 45v after 5 minutes off the discharger and disconnected the battery and the car just shot me an ima light. so when that happened the way i got the battery back up was to use the car for about 10 minutes driving, and then parked it, did a 12v reset and then the car started to charge the IMA with all the lights out. didn't hurt the pack i think because that pack is still running strong.

again im not going against anything you are saying, im just throwing ideas out there to people with ailing batteries that are tired of taking the pack in and out, im a lazy bastard so im just throwing some ideas out there that may work, although not fully proven i've been using this car since my last discharge for more than 3 weeks now so let's see how it turns out

 

You are wrong. Down stream affects upstream in the form of deliberate stagnation points and pressure to help direct flow. You may be flowing more through the pack, but you've likely changed the flow dynamics significantly, and you run a genuine risk of cooking cells. Only 3 of the cells have temperature sensors on them, and they represent what encompasses the hottest and coldest locations in the pack. You've potentially changed that.


Again, extremely difficult and would defeat the purpose of selecting NiMH - cheap and easy battery management. The relatively low cost and management demands of NiMH is what allowed hybrids to emerge when they did. Otherwise, you wouldn't have seen hybrids for another 5-10 years when Lithium became cost effective.


The PowerLab won't. You're wasting your time. Just charge at 5.5A and call it good. True pulse charging involves and oscillator and thousands of cycles per second.


110V after 2 hours? All the more reason to not bother with a wait. Nothing is gained by letting the battery sit in a discharged state as it climbs.


I recommend repeated 12V resets and idle charging, no driving until the IMA light is out. This gives you a relatively controlled 4-5A charge rate.

 

i understand where the concern lies but there are already hotspots that are NOT being monitored on the pack. as you know and i do as well from opening the pack there are 3 probes on 3 cells in the entire pack as you stated. but even in the stock form, that's only 3 probes. there may be cells that have high IR that are NOT monitored that do get hot. im sure you know this because of how many packs you've been through. when charging a pack there are cells in sticks that get hot. that's inevitable at this stage to actually monitor all cells unless we have ptc strips but they stopped that after the hch i packs. if you look at the construction at the exhaust of the pack there are 2 slits on the top and the bottom and the center is totally blocked off. i just made a hole in that center piece. the volume of air may have increased due to these holes which subsequently affected the velocity that the air travels. Now in a perfect vacuum im sure what i've done will affect the overall performance but due to the enclosure not being perfectly sealed i do not see it greatly affecting the overall performance of what it was compared to stock.

i don't know about extremely difficult because i can think off the top of my head with some relays and programming it can charge up a stick/subpack whenever it needs to by routing power only to a pack let's say during a cruise period.

ok if it involves oscillating wheels, then isn't that more simple? i can get a round pcb solder a piece of thin flexible stainless on one end, feed power on another stationary piece of flex stainless and then run the wheel with a motor and change the speed of the wheel with a pot? maybe i should try it out lol <- edit::i read that as oscillating wheel not oscillator lol... but again wouldn't my theory in general work?

 

You're missing the point. There's a reason they put those three sensors on those three cells. You changed something thinking you made it better. You likely haven't. You've added puny muffin fans to a leaf blower while altering the designed flow characteristics. You've likely moved the hot spots to cells that have no monitoring whatsoever.

If you understood how healthy NiMH works, you wouldn't be carrying on about relays and power routing.

Again, pulse charging is a waste of time.