Modifying a charger cradle to run off of DC

[Sonia's Driver]

I need to modify a NiMH radio battery charger cradle to run on DC. It currently runs on 120v AC.

I had been planning on just identifying the DC rail, probing the voltage and then just soldering in a buck or boost converter, but it works differently than I expected and I need some help to understand how it works. Before I removed it from the case, I was expecting it to be wired so that the transformer steps down the voltage, the bridge rectifies the lower AC voltage, it's filtered, and then the PWM IC switches it down to final voltage. But it doesn't look like it works that way.

Pics:

The two rectangular holes marked "+" and "-" connect to the where the battery contacts on the cradle. They are at 8.5vDC with open circuit.
The SMD near the AC hot input looks like a MB10F bridge rectifier, the DIP-8 IC on the back looks like a LY2117 PWM chip (they're all over Alibaba and I think it's a knockoff of an IR2117). The other IC on the back has the identifying marks "PS 817C 1723", which might be an optocoupler.

So, if I'm following the traces correctly, the mains AC is first rectified, then it's fed into the transformer, and then into the PWM chip, so I'm pretty confused.
I just need to know where to wire in the buck/boost converter...

Please help! Thanks.

 

[rjenkinsgb]

Measure the voltage across C5 with both a totally flat battery and a fully charged battery connected.

If that is constant (or within a fraction of a volt), you can feed DC in at that point. If not, the charge current may be controlled via the switch mode feedback circuit (the opto isolator) so a fixed DC voltage may not be practical.

For the voltage testing, I'd add flying leads to the capacitor connections and fasten them to your meter probes so you do not have to touch the PCB while it has mains power connected - the voltages are lethal.

 

[alec_t]

it works differently than I expected and I need some help to understand how it works.

Welcome to ETO!
It seems what you have is a switch-mode power supply (SMPS). The rectified mains voltage is smoothed by a fat capacitor (C1). Then an oscillator circuit chops up the DC at high frequency (typically hundreds of kilohertz). The chopped DC drives the transformer primary. The stepped down transformer secondary voltage is rectified and smoothed to give the supply output voltage. This voltage (or a representation of the output current) drives the input of a feed-back opto-isolator (U2), whose output controls the PWM duty cycle in U1 to regulate the supply voltage to give the required charger output.

 

[Nigel Goodwin]

I suspect (like rjenkinsgb) that the current regulation 'may' well be done via the opto-isolator, thus constant current rather than constant voltage.

I would suggest trying the tests given in post #2.

 

[Sonia's Driver]

Thanks for the explanation! alec_t I think I understand how the power supply works now, (but not why the transformer has two pins connected to the PWM chip in addition to the + and - from the rectifier...) I'm sure that would explain why there was no voltage between the two battery-side transformer outputs when I probed it; I didn't have the battery connected, so the opto-isolator was restricting the voltage down to ~0.

I agree that rjenkinsgb's test is a good idea, but I can't do it for a week until I can retrieve thin enough wire from my workshop to solder to C5. However, it seems like the more straightforward approach to modifying the circuit would be to just inject the DC at the MB10F's outputs. Aside from the difficulty of boosting from 12v to 120vdc, is there any reason why I wouldn't want to just do that?

 

[Visitor]

An easier method might be to cut the traces from at least one of the contacts of the 120v charger, and connect a dedicated DC NiMH charger module in its place. Looks like they can be had for around 3 bucks on ebay.

 

[Diver300]

However, it seems like the more straightforward approach to modifying the circuit would be to just inject the DC at the MB10F's outputs. Aside from the difficulty of boosting from 12v to 120vdc, is there any reason why I wouldn't want to just do that?

If you've got DC at around 150 V (although you might get away with a lower voltage) you can just run the charger from DC without modification.

 

[Sonia's Driver]

An easier method might be to cut the traces from at least one of the contacts of the 120v charger, and connect a dedicated DC NiMH charger module in its place. Looks like they can be had for around 3 bucks on ebay.

Excellent idea! I found one that'll work for only $4, with shipping.
I don't even need to cut the traces; I can just wire the cradle's contacts to the new board and throw the old one away!