C64 Optimized PSU -- routing the power

[wadesey]

I've read through the countless PSU questions posted over the past 10 years here, and unfortunately most of them refer to broken links that seemed to answer everyone's questions. Most notably these two:

 http://www.danielpri...h-diy/sid_2.htm

http://www.midibox.o...hp?topic=4455.0

 

I have a decent amount of experience constructing kits and fixing things, but I've always been a little shaky understanding power supplies and distribution...

 

So my question is: what is the best way to route power from the 5v and 14v headers on the PSU board to multiple destinations? For example, the schematic shows +5v being split and sent to J2 of the SID module(s) and J2 of the CORE module(s). What's the best way to physically make that mult from a single header pin on the PSU board to two or more destinations on other boards? Should I install separate headers on the PSU board for each destination and not one, as represented in the schematic?

 

I realize this is probably super obvious and/or I'm thinking about this the wrong way, but I've been literally researching for days trying to figure out PSU questions and haven't been able to conquer this one... Thanks!

[jjonas]

What's the best way to physically make that mult from a single header pin on the PSU board to two or more destinations on other boards? Should I install separate headers on the PSU board for each destination and not one, as represented in the schematic?

 

Hi,

 

if you're just asking about whether to have one header for several destinations, or a separate header for each destination (which is the level I'm able to give any advice on), I would have several headers, one for each destination. For example if you have four 5VDC destinations, you could - instead of feeding all destinations from one "master" 5VDC header on the PSU board - have four 5VDC-GND headers on the PSU board (i.e. four times two pins) that are fed by the "master" 5VDC (and GND) source on the PSU board.

 

I don't know how many destinations you'd have, but the more you have, the easier it becomes to troubleshoot power supply problems if each destination can simply be unplugged (without need for de- and resoldering, for example), to test one module at a time.

[wadesey]

Yes, that's exactly what I was asking -- thanks for confirming!

[wadesey]

Another kind of related power issue that hopefully someone can answer...

 

I successfully made the C64 optimized power supply and get +9v and +5v at the correct IC pins on the SID module -- so the +5v output of the power supply is confirmed at +5v. However, at some point in the power section of the CORE module the +5v drops to +3.97v. Will that be a problem?

 

As instructed, I used a jumper across the in-out where the 7805 regulator would go and changed C5 to 1000uF instead of 2200uF. Did I misunderstand and a 2200uF capacitor should still be used, causing the voltage drop?

[Imp]

Did you connect the 5V to pin J2 of the core? It seems  you used j1 instead, so some voltage drop occurs on the rectifier.

[wadesey]

Yep, I had the properties of J1 and J2 confused. Everything measure correctly now. Thanks!

[skunks]

Wadesey, the first page seems to be archived, but without images, so it isn't very helpful:

https://web.archive.org/web/20121114031919/http://www.danielprice.org.uk/synth-diy/sid_2.htm

 

In my experiments I found that power connection topology really matters. Of course, ground loops must be avoided, but even without them optimal Vs and Vd connection scheme is not obvious for me.

 

If you are building single core-SID pair, probably you supply both modules with say 9v AC and 12v AC. Then you connect Ground and  +5v from core module to SID. That's the only point grounds are connected, so no ground loops.

 

The second idea is to rectify, filter and regulate AC to DC inside PSU box. Grounds will be connected there as well (and not between modules).

 

As far as I understood 1st idea is better then 2nd. But filtering capacitors in SID/Core modules should not be removed. Am I right?