DIY High Voltage Battery Reconditioner - Grid Charger

Nice work... Very clean and clear presentation compared to the chaos we have on the thread over at:
----> Build Hybrid Battery Maintenance Gear For Under $100 | PriusChat

Have you given thought to a resistor based discharge beyond the light bulbs? I still use the lightbulb method, mostly because sourcing an affordable solid state resistor is problematic. One guy on here used an electrical brake from an old elevator.

 

Yea, ideally you'd want to rig up a system to discharge into your home powerwall, or even just for powering your hobby chargers for module work, but as always managing this gets expensive real quick and most people just need the light bulb trick.

What's good about the light bulb method is the intensity of the light with a quick glance tells you your discharge status. What's bad about it is the resistance of the filament changes depending on its temperature so you aren't discharging with a stable resistance value.

Over in another thread we're talking about this SKYRC discharger that allows you to get a discharge chart for each module on your computer. That's a game changer for me when it comes to getting more accurate discharge data: SkyRC BD250 Battery Discharger Analyzer I'll probably buy one of these in the next few months.

 

Doh... Just found this thread after purchasing the Prolong setup.

That you found an ideal power supply for $30 is rad...

 

Impressive build and excellent pdf documentation. A big thank you for sharing this!

 

Nice work. Thanks for detailing this setup. Being studying every bit of the schematics on page 4 ever since I download the paper.
Any updates yet on your automation plan?
For my own setup, I'd be using a different brand of power supply, with the following specs:

MOSO X6-150M43 LED driver
Input: 100-240V, 50/60Hz, 2.0A Max.PF:0.95
Output: 80-143V, 0.15-1.50A
Max: 170V, Max power: 150W.

I'm connecting these two power supplies serially, to obtain, ~300VDC @ 1.5A.

https://www.mosoleddriver.com/linear-industrial-lighting-led-driver/57314486.html

 

Thank you for your reply. I have already gone far, with my prototype using the power supplies I have.
1. I used a mobile phone charger, as the power source for the volt/amp meter
2. Sourced 4 diodes 1N4007(must not be 4), from an old phone charger also, and a 3M ohm resistor
5. A piece of flat plywood, with piece of connectors.
6. I have tested the circuit, and works perfectly. Used a 240V bulb, as the "battery", just to see if it works.
7. Continuing with the DC side later tomorrow, again.
Thanks so very much.
Your pdf was very helpful.

 

I am going for the gentler, lower current approach. I just received my Mouser order of 5 @ MeanWell APC-8-250, and an eBay order that has a battery pack chassis, so I can try my ideas with a few modules on the bench. The cooling solution is still pending. I am going to mount this on an oak board with rubber feet, and an all-hole metal cage for tourist protection. I vaguely hope for the steampunk look.

 

I think

The oak board idea is pretty cool. I did same thing too, to try out my ideas, of the gridcharger I saw here too. That helped me prototyped everything on a plywood board, before transferring my charger circuits into a stainless steel sheet metal I made out of a scrap printing machine I scavenged from a junkyard. I got all the printers harness wires, and connectors, and some parts elsewhere to assemble it.
Would post my own grid charger once I'm done here too.

 

It's kind of making me think of the old Southwest Technical Products look.

 

The first test was the temperature on the modules during charging. I placed 6 decommissioned cells between pressure plates, with 1 reverse connected and 1 out-of-circuit (attached photograph is different configuration). This kept the voltage within the limits of the constant current charger at all times. I re-patched the connections so that each cell went to maximum charge, and afterward went to a deep discharge. Being in Texas, the temperature in the garage was pretty warm at night, but the module temperatures stayed not far above ambient.

This design pushes the energy from the discharging module into the other modules, does not have a load resistor, and results in each module being charged to saturation for a long period. I don't know if that is a problem. It does let the system run at constant current before the end condition, so that I can directly calculate amps (or coulombs) in and out. I can measure voltage on each module.

This is, of course, a research phase. That is because I am a scientist, my car battery is expensive, and I am using an abundance of caution.

The next task is to build a resistive load, so that the constant current supply, the module 6-pack, and the resistor are all in series, and the voltage supplied by the power supply is within limits all the time.. This will dump the discharge battery energy into a light bulb, and the elapsed time will directly correspond to the amps discharged. In this design the light bulb will have constant brightness. I think this will let me know the capacity of the rejuvenated modules, once I know how to interpret the voltage by amps data. I don't have a data logger at hand, so I will be taking voltage measurements from time to time. I will look for a negative spike in the 2nd derivative as the endpoint.

Looking forward, I am building the connector to the traction battery. At present, I have a shopping cart at Mouser Electronics with 16 ga stranded wire (5' red, 6' black) and 20 ga stranded wire, 1', for a fuse-able link.

Although I would like to use the same connector that tgtech used for his J1 & J4, because the community is better off if we can design interchangeability, I want to use a 2 pin connector and his choice is 3 pin. I welcome discussion on this. Maybe he will change his choice

 

Would your two pin connector be able to handle the current "draw" capacity your grid charger or AC circuit be pumping out?
Do you have schematics or diagram of your set-up?

 

For whatever reason, the photo did not post earlier.

To be clear, this is a 0.250 A charger. Although it is a "grid charger" it is not high amperage.

 

The next task is to build a resistive load, so that the constant current supply, the module 6-pack, and the resistor are all in series, and the voltage supplied by the power supply is within limits all the time.. This will dump the discharge battery energy into a light bulb, and the elapsed time will directly correspond to the amps discharged. In this design the light bulb will have constant brightness. I think this will let me know the capacity of the rejuvenated modules, once I know how to interpret the voltage by amps data. I don't have a data logger at hand, so I will be taking voltage measurements from time to time. I will look for a negative spike in the 2nd derivative as the endpoint.

[/QUOTE]
a.mcintosh, Could you please explain how to apply your "2nd Derivative" math to this application? I'm sorry, I barely made it through high school math many years ago! Trying to catch up on my math skills, and it's very helpful to have real-world examples to work with! Thanks!

 

Just in case a.mcintosh isn't watching this thread, I'll guess he meant when you see a relatively sudden change (which would include a reduction) in the rate of change in voltage.

 

I recently found your post as I attempt to restore hybrid batteries on (4) of my gen 2's. You have an excellent build, thank you for posting!

To draw the hybrid battery down in stages the wire wound variable resistor which uses center tap connectors looks promising to me. My plan is to get the 300w 500ohm unit with (3) center taps and clamp the tap at the correct value. They will also make a "custom".

https://www.aliexpress.us/item/3256802277471615.html?spm=a2g0o.cart.0.0.782638daU1zsV2&mp=1&gatewayAdapt=glo2usa