The 330nF capacitor in the C64-OPSU

[286]

I've been reading up on the optimized psu and only one question remains. Is there a particular reason that the 330nF capacitor is exactly 330nF? As I understand the purpose of the 2200uF cap is to remove large ripple and the 330nF cap filters noise.

The reason I'm asking is that I have a bunch of caps at home but no extra 330nF since that isn't as usual as the series "1, 2.2, 4.7, 10, 22, 47...".

I figure that one could use another small value cap for noise filtering with the same result (220nf or 470nF). Am I wrong?

(Ordering a 330nF wouldn't really be a problem. It's more that I'm genuinely interested in the design.)

[nILS]

I figure that one could use another small value cap for noise filtering with the same result (220nf or 470nF). Am I wrong?

Sorta. Changing the capcitance changes the filter's behaviour. There's lotsa material on DC filtering on the net, makes for a nice reading evening ;)

[286]

I found a bunch of interesting reading material. Sadly I didn't save the links but here are two that at least show the basics (In case anyone else is interested in stuff like this).

http://www.electronicrepairguide.com/filter-capacitor-function.html <- This is about as much as I knew when I started but it could be interesting for people wanting to learn the basics.

http://www.allaboutcircuits.com/vol_2/chpt_8/2.html <- Another quite nice explanation of filtering.

I also borrowed some lecture notes from a friend and those teached me a lot. But drawing any real conclusions is difficult. Changing the 330nF to 470nF will change the cutoff frequency slightly. Without actually testing on the circuit it's almost impossible to calculate if this a good or bad thing. It may add or remove noise all dependent of the rest of the circuit.

What really messes things up is that apparently different capacitors with the same capacitance but of different type have different frequency filtering characteristics. The distances in the circuit is also an important factor that is difficult to use in a calculation.

The last thing that bothers me is that the C64OPSU design is made for both 50Hz and 60Hz inbound AC. This would "pollute" the circuit with different noise and ripple and therefore different capacitor combinations would be optimal.

But the 330nF is there for a reason and it was probably selected using an oscilloscope or something rather than Thorsten just thinking... "what can I use this big bag of 330nF caps for"  :D

Still I think that it could be possible to get good results with a 220nF or 470nF cap instead. The uncertainties of all other factors make me believe that the difference between 220uF, 330uF, 470uF is smaller than the difference between two different builds of the same circuit using different C64-bricks in different countries with different AC-frequencies.

I'm ordering a 330nF cap anyway. ;)

[strophlex]

What about the 2200uF one? It should be OK to replace it with some cap of higher capacitance as long as the voltage rating is 25V or higher? And by the way, does the voltage rating have to be as high as 25V?

[286]

What about the 2200uF one? It should be OK to replace it with some cap of higher capacitance as long as the voltage rating is 25V or higher? And by the way, does the voltage rating have to be as high as 25V?

I can't say anything about replacing it.

The reason for it being rated 25V or higher is that there is sort of a golden rule when in comes to electrolytic capacitors. You always want to make sure that your cap can take twice the supply voltage.