The current loop area from the +ve rail, cap, earth bar, cap, -ve rail, through the output transistors back to the +ve rail represents inductance.karatestu wrote: ↑Mon Feb 06, 2023 6:43 am For the input and output stages I now have a curved tinned wire going over the top of the board and connecting thd two caps. I do worry this is not optimal and that the bus bar between them should be as short as possible. You mostly see psu caps very close together with a short link between. Do you have any thoughts on that Rex ?
Probably something to minimise.
It's reasonable to expect something that's been off for a while to have discharged. In DIY equipment, it's up to you to define 'a while', but if your banks become separated (as in a fault), it's been off 'a while', you touch something and it connects again and 'BANG', the charged caps discharge into the dead short representing the discharged caps and all of sudden you have hearing issues. It's not fun.karatestu wrote: ↑Mon Feb 06, 2023 6:43 am You mention bleeders (discharge resistor to ground) in each group :grin: - do you recommend this just for safety reasons ? What resistance and wattage do you suggest ? I believe that the value shouldn't be too low as it wastes power during normal use but obviously the higher the resistance the longed it takes to discharge the caps. By the way, I always discharge them with a 60w lamp before messing about with them. With the speakers connected (4 ohm load) I always find they have all discharged apart from the output stage positive rail overnight.
Thanks, Stu
Even a 1/4w resistor dissipating 0.01w is fine, but recalc according to your definition of 'a while'.