MCU controlled lead-acid battery charger

1 post / 0 new

Hello,

I made a design for a 24V lead-acid battery charger using one of your IC's (the TNY280GN).
The needed specs are:
- can be switched on and off by the MCU by supplying 200Hz PWM 80% at enable input of sheet (meaning the chip is 20% on)
- charge current 0.4 A
- fed from 220V
- should be off when no PWM supplied, to get the longest life possible (at least 15 years needed)
- no regulated output voltage needed, only a more or less fixed charging current

I put a 680K resistor between the enable and bypass pins, to make the chip go into undervoltage mode when there is no PWM present. Otherwise it will try to autorestart every few seconds, which is bad for controlability and lifetime.

I used the attached wurth transformer using the full primary winding.
The whole thing seems to work and run relatively cool on the test PCB.

I have a few questions though:

1) Is the chip used within its design specifications the way I am using it?

2) even with the 680K resistor and no PWM from MCU (so in undervoltage mode) there still seems to be some (parasitic?) switching going on, because the unloaded voltage at C19 is not zero (but about 8V) and I can see the switching on the scope.
Is there a way to stop this (because bad for lifetime and EMC), maybe by using another resistor value between enable and bypass?

3) the charger is working (getting 80% PWM from MCU) for about 1% of its lifetime. 99% of the time the enable input is low. Do you think it will last for more than 15 years used like this? (220V is always present)

4) when I turn on the PWM before supplying 220V, sometimes the charger doesn't start. Is there a way to avoid this?

Thank you,

Bart Van Thielen

첨부 파일Size
charger.pdf32.87 KB
wurth trafo tinyswitch.pdf152.67 KB