On 2012-09-22, at 17:29, Gerrit Heitsch wrote: >>>> BTW, Gerrit - what do you think was the point in providing two +5Vs in the 64? One called +5 and the other CAN+5? Looking at 252312-right, they dropped this idea on the narrow boards. But I am wondering why they introduced it in the first place. >>> >>> There are 3 reasons I can imagine: >>> >>> 1) Power, 5V, 2A might have needed a different plug >> >> I think this DIN should handle 2A w/o problems but In fact I am not sure about the actual specs. > > Hm... It would be borderline... If the contacts are just a bit worn, you start having a measurable voltage drop on the connection. But there are two pins allocated for +5. Just by adding second pin to the plug they could have doubled the current there. >>> 2) Keeping the power for the VIC and clock generation seperate from the rest of the system to prevent RF noise from leaking. Remember, the first boards had the discrete PLL inside the can. >>> >>> 3) Providing the VIC with its own power source that doesn't contain all the noise from the rest of the system which might otherwise become visible on the screen. >>> >>> I think it's a mix between 2) and 3). With the creation of the 8701, 2) became less important. >> >> I see. I am mostly concerned about [3)]. Shall probably have to check it by changing the supply point for the VIC. Hopefully I won't overload the PSU with this. > > As I wrote, the regulator in the PSU is a 78S05 and therefore good for 2A. Without any expansion, the old C64 draw about 1A on +5V. VIC itself can draw up to 200mA and the clock generator should be less than 100mA (since nothing gets hot). That's below the 1.5A the PSU is supposed to be able to supply. Yes, I am just quite precious about this particular PSU. It is the one with two fuses and a switch on the case. A jewel ;-) And since it is over 30 years old and worked so well for so long... you know. I think I need to fetch one of those newer bricks from the basement for such tests. > For a short (!) test, I don't see any problems. > > Just remove the 7805 and run a connection between +5V and the output pad of the 7805. You still need the 9VAC for the voltage doubler to supply the 7812 so that VIC and SID get their +12V. First I simply bent out pin 40 out of the socket and supplied it with +5 from the PWR switch. I noticed (really barely) a degradation to the picture quality but when supplying the same pin with CAN+5, there was virtually no difference. Meaning also a slight (if at all) degradation of the quality. So I decided to do better test and keep the two decoupling capacitors in the loop. So I cut the trace and installed a pinhead like here: https://dl.dropbox.com/u/58002657/vic_pwr_supply.jpg and after switching back and forth the supply between +5 and CAN+5 I don't see any difference. So as the last iteration I removed pin 2 of 7805 and supplied the pad with +5. https://dl.dropbox.com/u/58002657/whole_can%2B5_replaced.jpg Again - no noticeable difference at least on the screen. Maybe on a scope someone could notice something but on the screen I see no noticeable difference. After doing this I tend to believe that it must have been something around 2) or something about utilising the PSU's current capacity better. -- SD! Message was sent through the cbm-hackers mailing listReceived on 2012-09-22 16:00:37
Archive generated by hypermail 2.2.0.
