Orson Cart
Advanced Member level 3
- Joined
- Jun 18, 2009
- Messages
- 798
- Helped
- 347
- Reputation
- 1,146
- Reaction score
- 558
- Trophy points
- 1,373
- Location
- New Zealand
- Activity points
- 5,973
There is just a chance that the chip is designed to work better with smaller inductors (better match to the internal slope comp), but not too small, worth paralleling the 22uH to see if this improves matters, it is true that proper current mode should make the system inherently easier to stabilise, however it does appear that the chip is not quite perfect in that regard. It would be well to slow down the volt loop just to verify that the chip is stable (or not) under slower & more damped control.
We design a lot of boosters (ave current mode) and we find that a speedup network across the Vsense resistor along with suitable damping in the Vea improves performance quite a bit (our boosters 90-325VDC in, 350-420VDC out, 1.5kW).
Unless you absolutely need to have that chip - a pretty good booster can be designed with just a 555 and xtor emitter follower driving a mosfet, we have done this a lot for low power ckts (to 100W), Regards, Orson Cart.
We design a lot of boosters (ave current mode) and we find that a speedup network across the Vsense resistor along with suitable damping in the Vea improves performance quite a bit (our boosters 90-325VDC in, 350-420VDC out, 1.5kW).
Unless you absolutely need to have that chip - a pretty good booster can be designed with just a 555 and xtor emitter follower driving a mosfet, we have done this a lot for low power ckts (to 100W), Regards, Orson Cart.