Any reason not to put preamp filter caps on main board?

[Hardtailed]

I've seen it done 3 different ways:
- Chassis mounted cap cans
- Filter caps on separate board
- Filters on main board

It seems though the filters on main board is not very common.

Is there something inherently bad about this layout I'm working on? (see attachment)
It has 2 8uF and 1 30uF caps right on the board, close to where they are needed. This also allows to ground them with the rest of their local circuits to diminish ground loops. I've also tried to leave more free space on each side for better "decoupling" with rest of circuit.

It seems to me that this should be better than running long wires to remotely located caps. But perhaps it really is a bad idea?

PS: If you,re wondering, the left is the cathode bias resistor and cap, the 100ohm ground reference resistors for the heater which are hooked on top of the cathode resistor for DC elevation, and the 4 screen resistors (power tubes are EL84)

[Enzo]

No, there is no electrical reason not to put the caps on the board.

Consider some old fender with the caps up under the dome. Now look at the eyelet board. Where could they have put the caps on the eyelet board? Caps today are much smaller than they were 30-40 years ago, so it is a lot easier to cram them onthe the part board.

[Hardtailed]

No, there is no electrical reason not to put the caps on the board.

Consider some old fender with the caps up under the dome. Now look at the eyelet board. Where could they have put the caps on the eyelet board? Caps today are much smaller than they were 30-40 years ago, so it is a lot easier to cram them onthe the part board.

Thanks, that's what I thought

[yankeerob]

I do on smaller amps but use large chassis mounted radial cans on bigger amps for convenience and tidiness...

[Merlinb]

Having the caps on the main board is the best option as it makes for the optimum ground scheme, and least susceptibility to noise problems! The other 'popular' options are quite old fashioned, and can caps are practically medieval.

[J M Fahey]

For what it's worth, I do put all caps (and the rest of the amp) on a single PCB, with excellent results. They are radials and mounted perpendicular to the board.
The layout ends being clean and uncluttered ... and no hum.
All parts are "normal", no exotic components.
After all, Marshall, Fender, and everybody else basically used what was commercially available at the time.
And Axial caps have no intrinsic advantage (or disadvantage) over radials.
Simply the first type is somewhat easier to use in PTP wiring, and the second one allows for more compact PCBs.
When I started into this noble profession, (late 60's), axials were the norm (yellow Siemens-Halske and grey Japanese); in the late 70's and early 80's I was offered both types to choose from (orange Siemens, grey Japanese), with radials being *slightly* cheaper for same electrical specs; nowadays radials rule.(Black EPCOS , black China, light blue Taiwan).
Of course I notice the Black ones sound much more open and airy than the blue ones, although the SSS (Sound Stage Separation) and the VTBR (Violin to Brass ratio) ... ... no, I'm kidding

[daz]

I built the same amp both ways, the last one being with all chassis mounted multi cans. Both amps sounded pretty much identical both as far as tone and noise.

[Alex/Tubewonder]

Having the caps on the main board is the best option as it makes for the optimum ground scheme, and least susceptibility to noise problems! The other 'popular' options are quite old fashioned, and can caps are practically medieval.

I definitely second this. A dog house on top of the chassis is probably the single worst place for supply decoupling capacitors.
Here's how I do it when in the mood for PTP assembly, this system makes it easy to design a PCB later with almost identical component/wiring topology:

[Merlinb]

Here's how I do it

Beautiful layout, even the heaters are done properly! *applause*

[J M Fahey]

Congratulations !!!
Beautiful wiring and layout !!

[tubeswell]

Absolutely immaculate! (Funky sound too)

[yankeerob]

can caps are practically medieval.

Aye - but some guys and gals are into the medieval look and get a thrill when things look 'like they used to'... I'm not a hobby builder any more and the first thing the guys at Manson's noticed about our amp was the three big 50+50uF's sticking out of the chassis...

Fact of the matter is - as far as sound goes - there's no difference at all. When an amp requires a 100uF first stage filter and 4 subsequent 50uF filter stages - three 50+50uF cans is a compact and easy way to go if there are other things you can accomplish with the 'real estate' you have available...

There's nothing right or wrong about what you choose... as long as it works.

BTW - nice work Al... beautifully done.

[Merlinb]

When an amp requires a 100uF first stage filter and 4 subsequent 50uF filter stages - three 50+50uF cans is a compact and easy way to go if there are other things you can accomplish with the 'real estate' you have available...

Ah, but the number one argument against can caps is that it puts ripple current noise on to the audio ground. However, in your case I suspect you are using the first can, which is the reservoir, as a 'single cap' (with the two sections in parallel), so its wiring to the audio ground is not objectionable.

And it must be remembered that lots of 'bad practices' seem to work out ok most of the time, but I do not see that as an argument for saying the practices themselves are ok. It is the method which works the greatest percentage of the time which is the correct one, all others are inferior substitutes.

(Also, I have found guitarists seem to have very low expectation of their equipment. They often describe things as being "dead quiet" when I would describe them as "lots of hum". Go figure.)

[J M Fahey]

Aye - but some guys and gals are into the medieval look and get a thrill when things look 'like they used to'...

That's the point.
That's why they buy "Tweed Champs" filled with digital effects and PC CPUs wrapped in tolex labeled "Valvetronix", thinking they bought "Brian May's amp".
I do not like to give away good ideas for free to monster commercial makers, but the "cap vault" over the chassis is an excellent place to hide the hard disk, he he.

[yankeerob]

Ah, but the number one argument against can caps is that it puts ripple current noise on to the audio ground. However, in your case I suspect you are using the first can, which is the reservoir, as a 'single cap' (with the two sections in parallel), so its wiring to the audio ground is not objectionable.

I fully comprehend what you're saying... and furthermore - there's a lot to be said for the quality of which ever type you use.

I personally design all of amps with the entire power supply at the power transformer end which - as good design would have it - is furthest away from the input. Further to that - I am not one of those who will accept something's 'dead quiet' unless it is... I'm saying it's all according to design and the space taken up by axials at the audio end might be useful for other things - that's what I'm always concerned about when I design an amp layout - getting other bits in - if I need to - without it getting too crowded.

I find these discussions are usually geared towards the ideal world scenarios of one-off or at best, very low volume building - trouble starts when you start trying to make them in numbers and at a price that is in reach of more players... If you were to jump in my shoes for a minute - the matter of how many you can you build - consistently - in a given period of time has to become a consideration. I have - however - established a solid reputation for quality - both in build and in sound - at a very good price break. Now people are asking for send/return loops etc - which are possible but require some 'clever boxing' where layout design is concerned. That is why I have chosen to have a central power supply completely separate from the audio section.

And it must be remembered that lots of 'bad practices' seem to work out ok most of the time, but I do not see that as an argument for saying the practices themselves are ok. It is the method which works the greatest percentage of the time which is the correct one, all others are inferior substitutes.

Amen to that. Like said earler - consistency is king and if your customers feel that the amp they're buying from you is exactly the same as next and the one before it then they will be confident their money's being spent well. If they're worried about getting the one in ten that isn't quite up to the mark they might not part with it at all... I can't afford to be in the latter situation.

(Also, I have found guitarists seem to have very low expectation of their equipment. They often describe things as being "dead quiet" when I would describe them as "lots of hum". Go figure.)

It's not their fault - look at all the rubbish they've had to put up with over the last 20 yrs - and it has been that long... or more that mainstream mftr's have gotten away with producing cheap and nasty stuff in the name of good VFM when it is - as you are well aware - anything but. I do take your point and do take full advantage of our ability to produce products that amaze those very same guys where this point is concerned