Switching Boost Converter only works while probing with o'scope?
I'm trying to improve a 4S LiPo (14.4V nominal) to 19V/3.5A boost converter based on the MC34063 PWM switch controller. I initially built it once with a Schottky diode for the catch diode, and a IRLB8701 N-channel for the switching transistor.
Now I'm trying to do synchronous switching based on the IRF2183 MOSFET driver, and two IRLB8701 N-channels for both ends of the switching. (One assmption is that the high-side 8701 built-in body diode will provide some amount of forward conduction even before the built-in cross-conduction delay of the 2183 kicks in and pulls the gate of that thing higher.)
The problem I'm seeing is that when turning on, the circuit doesn't switch -- it just sits there at 13.8V output with 14.3V input.
As soon as I probe pin DRV_C (or the connected resistor R3 on the connected side) of U1 with my o'scope, it starts switching like a champ, and outputs a buttery-smooth 19V into my 250 mA test load (a 100 Ohm 5W resistor)
Why would this be?
My guess is that perhaps there is no initial charge to boost the gate drive of the high-side N-channel -- but the forward body diode of that device should (and does) provide enough conduction to output voltage, and the voltage sensor of the 34063 should sense "low voltage" no matter what, so I feel this guess is poor.
Here's the schematic:
And here's the actual board, in case it helps someone (I know this is kind-of hard to read as-is; red is top layer; blue is bottom layer.)
Also, I'm somewhat considering hacking a 22 pF capacitor and 10 MOhm resistor between GND and DRV_C, and hoping that makes it work without the scope, and then calling it good :-/ Better proposals highly appreciated :-)
I'm trying to improve a 4S LiPo (14.4V nominal) to 19V/3.5A boost converter based on the MC34063 PWM switch controller. I initially built it once with a Schottky diode for the catch diode, and a IRLB8701 N-channel for the switching transistor.
Now I'm trying to do synchronous switching based on the IRF2183 MOSFET driver, and two IRLB8701 N-channels for both ends of the switching. (One assmption is that the high-side 8701 built-in body diode will provide some amount of forward conduction even before the built-in cross-conduction delay of the 2183 kicks in and pulls the gate of that thing higher.)
The problem I'm seeing is that when turning on, the circuit doesn't switch -- it just sits there at 13.8V output with 14.3V input.
As soon as I probe pin DRV_C (or the connected resistor R3 on the connected side) of U1 with my o'scope, it starts switching like a champ, and outputs a buttery-smooth 19V into my 250 mA test load (a 100 Ohm 5W resistor)
Why would this be?
My guess is that perhaps there is no initial charge to boost the gate drive of the high-side N-channel -- but the forward body diode of that device should (and does) provide enough conduction to output voltage, and the voltage sensor of the 34063 should sense "low voltage" no matter what, so I feel this guess is poor.
Here's the schematic:
And here's the actual board, in case it helps someone (I know this is kind-of hard to read as-is; red is top layer; blue is bottom layer.)
Also, I'm somewhat considering hacking a 22 pF capacitor and 10 MOhm resistor between GND and DRV_C, and hoping that makes it work without the scope, and then calling it good :-/ Better proposals highly appreciated :-)
No comments:
Post a Comment