Originally Posted by Banfi T.
The PSU's capacitors are not really for smoothing, rather for energy storage to bypass any power fluctuation common in domestic power networks. The headamp has 4700uF puffers and stabilzers quiet enough to power MC headamps (for vinyl playback I mean) I doubt that any ripple rejection is gained by using that many capacitor in the external PSU. On the other hand you can now disconnet and reconnect the mains cable without hearing it
Anyway, the stablizers are on the same board as the headamp, the four smaller heatsinks are for them. (The regulators are also non-feedback discrete designes.)
ah... i see... interesting concept, but most likely lots of stress for the transformer. do you use regular fuses on the primary and/or secondary side?
As for the p2p...Untill recently I did work with pcbs, but I had to realize that for prototypes the protoboards are much more suitable. The first headamps I built this way were done in a traditinal way, I used the legs of the components to make connections. Many times even the components themselfs were at the downside of the panel. While this certainly worked for smiple circuits this can be rather messy with more complicated projects. I ended up using more and more wires which were crossing other tracks and I did not liked this. On the other hand, if you use the leg of a part, so that you bend it and solder it to an other part it is not too easy to remove that part if you want to. (With prototypes it happens all the time). So, instead this I figured that I put every component in as they were in a real PCB, so I cut down their legs. Then I use the legs as connections, but this time I build up the track like if it were a real PCB. In this way I could do a much cleaner board with less cross wires and the parts can be easily changed. The following picture illustrates the headamp + regulator board having one side done in the old way and the other in the new. I think the picture speaks for itself. (Note, that the side done in the new way is more complete having the output stage also almost completely connected)
i understand exactly what you mean and switched myself short time ago from simply connecting parts to pre-wiring some sort of traces before i solder in the main components. but i guess, it'd be better to reduce the solderpoints to the necessary minimum for optimal current-flow... well... just a guess.