battery and charger talk

[dnmun]

i saw there are some pictures on my disk from working on some chargers and disassembling some ping packs so i thought i would post them up since nobody ever takes them apart so they don't know what's inside.

the larger kingpan and EMC chargers use a bigger switching transistor format. like the TO-247 package instead of the To-220 package on the smaller chargers. this is 200V 15A. the collector is the tab and middle leg and the emitter is the left leg and base is the right. i just measured the npn transistors on corbins charger and it is 585mV forward bias from base to emitter in the circuit. collector base is about 575 which you would expect because the collector is more heavily doped. since the pn junction in the middle of the transistor doesn't conduct current until the base injects some electrons into the depletion region, then it will measure open circuit on your diode tester. both ways collector-emitter.

on the Sanken C4140 i took out it measures 250mV or 250mR of resistance, so dead, and the other FJP13009 i have here reads 20mR.



this is what it looks like soldered into the pcb. leave a little short space above the shoulder on the leg so you can bend the legs to allow the transistor be flush against the case and compress the heat sink insulator uniformly in order to spread the heat load out and keep the tab corner from digging into the insulator and cutting through the silicone and shorting the collector to the case.



this is what happens when the choke overheats and burns through the trace underneath and short out the hi voltage output of the transformer through the choke to the input of the 12V regulator. the trace runs from the top of those 4 diodes that form the rectifier for the circuit current in the back end. the trace runs right under the choke so when the choke wears through the shellac, actually burns it off, then the hi voltage is carried onto that rectified circuit and you can see where the little 35V cap to the left of the burn that is attached to that trace also is removed because it blew, and the two caps on the output, the 35V and the 100V blew. also the little 35V cap next to that dual op amp also blew ans was removed.



i decided to make my own choke by rewinding the core with larger wire to handle more current without overheating. when i installed it i made sure to add a mica insulator underneath the choke and shellac it in place and mounted the choke higher away from the surface and glued it in place too.



and i added some more conductor from the output of the schottky diodes to the end of the charger to increase current capability.

View attachment 3

and also i have this picture of where you have to solder the two leads together when you remove the thyristor daughterboard that kinpan has started adding to these chargers to make them 'universal voltage' and can result in damage to the charger when used on 2 wire 240V circuits. which is what happened to cargo tom. but removing the thyristor daughter board by unsoldering the one leg on the right and cutting the other two short allows you to solder the two little stumps of wire together to make the circuit complete to convert it to 120V only. which i do to every kingpan charger i open that have it. not all of them do, just the small ones.


this is the 16S 20Ah ping pack i got from ambrose for parts. it always is the top pouch of the top cell that puffs it seems but this is the bottom, but still on the end where the distortion caused by the tabs being pulled against the subconnector as the pouches expand with age..

 

[dnmun]

taking the ping pack from ambrose apart.
View attachment 6

both sections cleaned up and good cells remain,



this is the other 8S section after removing shrink wrap. all cells look good.
View attachment 4

and more pouches from wineboy and dogman in piles after having been cleaned up of all the puffers




then assembled the pouches into packs i could charge up and balance to test for capacity



you can see how i cut each cell apart to separate the four pouches from each to to be able to find the low pouch out of the 4 when i had low capacity on one cell. notice that the tabs are unsoldered by washing the solder off the end with two soldering irons, then i cut the subconnector on the ends of the slots to free the tabs from the subconnector. this is the least damaging way to take the ping packs apart and the subconnectors are never used again because i solder the tabs to each other. note that there is still the little piece of pcb under the tabs so i go back a second time and unsolder the tabs from the little piece of pcb and fold the tabs out of the way to remove it so they are all free. then i straighten the tabs in some heavy pliers to make the tabs flat again and sticking straight up.

 

[dnmun]

then i got to where i was able to attempt to build the two 15Ah 23S ping packs with all the assembled pouches before i knew the capacity of all of them. this is how each looked when i assembled them to test each side for capacity.


i assembled the two separate 23S 15Ah packs into one big pack after testing capacity on each side so i had some idea in advance of which were low and which had the full (or close to) 15AhView attachment 4

this is the pack again after big sections removed to test capacity of individual cells again.

this is how i cut the subconnectors from each side of the slot.

i soon discovered i did not have enuff good pouches to make it to 23S of 15Ah so i had to drop down to 22S by taking apart the cells with bad (low capacity) pouches and conserving those that still had full capacity or close to 5Ah anyway. so i had to measure the capacity of each pouch which did by fully charging them to 3.65V on the big single cell charger and then discharging them in series through a big cement resistor so that i could measure capacity of each pouch exactly. you can see the notes i would record of voltage right after full charge, and then the next day, to find the ones that dropped in resting the fastest. part of the selection, but not the only or exclusive one since i found some pouches that dropped fast still had full capacity, which i had already demonstrated previously.


my single cell charger charging 12-13 pouches at a time.

 

[dnmun]

this is how i supported the motor when i pulled the tranny on my honda wagon to work on the clutch.

i used some short tire chain replacement sections attached to the rafter on my carport with a 1/4" bolt.


it turned out that the clutch was ok but the clutch fork return spring had worn out, then the ends of the spring broke off at the wear point on the throwout bearing, slipped past the throwout bearing on top of the fork fingers and then the clutch fork was jammed too far out so i had no freeplay.

the previous shade tree mechanic had broken one of the pressure plate bolts off from using an impact wrench to tighten them i think so he had just kinda drilled it out and stuck another shorter bolt in that had not threads cut so i had to drill out the remainder of the bolt, remove the flywheel and then retap the threads for the bolt from the backside with a metric tap, and then used one of the other original pressure plate bolts in the threads and used the other bolt in one of the other threads that had never been damaged. i think it will last. just had to do it twice because the new clutch i put in was warped so i took it apart again and replaced the old clutch which had only about 15k miles anyway.