1 x 16uf or 2 x 8uf Filtering

[fiftynine]

Is there any reason you can't use a 22k 6w and a 16uf in place of a dedicated 22k 3w and two 8uf's as below?

Cheers.

[kagliostro]

Is there any reason you can't use a 22k 6w and a 16uf in place of a dedicated 22k 3w and two 8uf's as below?

Filtering

K

[fiftynine]

I like you already ; )

[jjasilli]

Some people prefer the low value of 8uF for the last filter stage, tonewise. 


Is there a stereo preamp? Why is the B+ line branching off?

[2deaf]

The voltage drop across the 22K will be larger with both C and E on it than with either C or E alone.

[jjasilli]

Yes, but you get the benefit of cascading filtering if the B+ rail is kept in series, without branching-off.  HT voltage can be targeted as you please with the values chosen for the dropping resistors (which also isolate the filter caps).

[PRR]

If loads C and E "hate each other", this may be poor.

In particular, if these are the 3rd and 1st stages of an amplifier chain, subsonics sneak from 3rd stage to 1st stage in-phase to cause oscillation, "motorboating".

Which is why we "usually" run the power filter chain the same way as the amplifier chain, only backward. Amplifier 1st 2nd 3rd 4th. Raw power, filters to 4th 3rd 2nd 1st. That puts two filters between the in-phase 3rd and 1st stages (or 4th and 2nd).

Then there is the "other way". Sorta-clean the raw DC and then give a separate filter to EACH stage. This may even work (though an exact design may cost more). But if you combine loads without understanding, you may get putt-putt-putt.

Ignoring the possiblity of huge sub-sonics.... If C is 22K and E is 22K, then the parallel value should be _11K_, not 22K. (Use 10K.)

6 Watts is highly unlikely. Yes, doing the job of two "3W" resistors would be 6W. However I suspect the "3W" rating is pretty conservative. In fact 22K 6W can stand a steady 360 Volts, which is probably about ALL you started with. It is more likely you only have 4mA flowing in the 22K, so 88 Volts across 22K, and 88V*4mA is 0.352 Watts steady dissipation.

You do have to be generous with this part, because at turn-on it has to fill-up the 8uFd or 16uFd, and might touch 6 Watts for a small instant. But I'd expect a 2W part to stand the surge many thousand times, and run just-warm steady-state.

[fiftynine]

Cheers.


No, nothing stereo. I just don't want to be naive finding the best arrangement for a 3 channel AC30 as below.

[HotBluePlates]

I just don't want to be naive finding the best arrangement for a 3 channel AC30 as below.

In that case, you probably should draw up the full schematic of what you want to do.

My guess: there's no "best arrangement" for a 3-channel AC-30. You already have what was original in the amp. And that amp had hum problems, due in part to insufficient filtering. But there was insufficient filtering because the filter system was chosen to keep from knocking the preamp voltage down to too-low levels, because the EL84's started with relatively low B+ and a Fender/Marshall style power supply with good filtering would the supply voltage down too much for the input stages.

Maybe some AC-30 specialists might have further advice if you describe what you're planning.

[terminalgs]

Check out the 1978 AC30 power supply layout.  three channels,  you could sub the EF86 for the 'Normal' 1/2 12AX7 stage. 


If you are adding an extra channel to a AC30 that grounds the the signal to the 1/2 the LTPI, consider using that input instead of a three way mixer somewhere else...

[fiftynine]

I gutted the Vib/Trem from an AC30/4 clone and tagged the Top Boost on to the Normal channel via 220k mixers one side of the PI. The EF86 goes in the other side of the PI as it always was. The filtering is as the filtercapsedit.jpg above. Works perfectly. Sounds beautiful.