McIntosh MC60 capacitor question

[tuberadios69]

After reading through Terry DeWicks great Tips and Hints, I decided to replace all the Bumble Bee caps in my Mac MC60's due to his warnings. I was surprised to find that the .25uf @ 600 volt caps on the phenolic(?) board (straddling the 10uf @ 450 v bias electrolytic) had been replaced at some point with .47uf @ 600 volt caps (they look like brown colored "orange drop" caps). These are the caps in the circuit closest to the output tubes according to the schematic so I presume we call them coupling caps(?). Is there some kind of advantage to this change to .47uf or do you think someone simply goofed in the past? By the way, the selenium rectifier has been replaced by a silicon diode (2.5 amp @1000PIV) and a 5.2K resistor replaced the old 1.8K, although this new resistor is 2 watt instead of 1/2 watt).
So, 1) should I replace the .47 caps with the schematic specified .25 or .22 uf caps or leave as is;
2) is there any problem with the 5.2K resistor (Terry specifies 5.1K) being 2 watt instead of 1/2 watt?
Thanks everyone! Steve

 

[Rodeodave]

A cap that blocks DC from one part of the circuit from another part is generally called a coupling cap.

The increase in capacity lowers the -3dB point of the bass roll off, but increasing coupling capacity beyond the design target it is not necessary and can also do harm, namely be causing blocking distortion (http://www.aikenamps.com/BlockingDistortion.html). I would revert to the original value. We are talking about C9/C10, right?

The increased resistor value is necessary to compensate for the lower voltage drop of the new silicone diode. You would have to measure the actual bias voltage to determine if the value of 5k2 is correct, my guess is it's close enough (like 2% off or so from 5k1).
2W is perfectly fine. One could argue that it's even better since the physically larger 2W part will have a lower temperature rise than a 1/2W at the given currents and therefore have a lower drift in value, given that they have comparable temperature coefficients. Plus larger parts generally have lower noise figures since fringe effects in their resistive elements have less influence.

For replacements caps people like the Illinois Capacitor MPW series, or Mallory 150s, or Sprague 716p (copper leads!), or even Russian PIO caps. Just make sure they're labeled non-inductive and you should be fine. Measure which lead is the outer foil (requires oscilloscope) and conserve proper polarity when changing the caps.

 

[tuberadios69]

Hi Dave, thanks for the quick reply! Yes, C9 and C10 are the caps I was questioning. I'm using MIT Multicap PPFX for the .22uf (see pic) since I had two in my parts bin as well as one Hovland (.47uf) and two Kimber Kaps for the 0.047uf's (on backorder from parts express so the pic shows those last two bumble bees). I used MIT Multicaps (RTX) and Infinicap SETI 's when I rebuilt/refurbed my MC-30's several years ago and am floored at how good they sound. I have 715P's on hand but only 0.1uf so I had to order the Kimbers (less than $5 each). I'll take your advice and replace the incorrect C9/10 0.47 caps with the proper 0.22uf, maybe bypassed with some Jensen 0.033uf PIO's I bought years ago from partsconnection on closeout. Does that sound like a good idea? Thanks again for your wisdom, I appreciate it!