Feedback loops caps - film vs ceramic

Given the choice, what would sound better in feedback loop positions around opamps?  I guess this is to avoid stray capacitance and for bandwidth limiting.

A film and foil PP cap, or standard C0g MLCC?  I'm talking through-hole parts btw.

Wima makes some low value pf foil caps in their FKP range, but I wondering if a c0g radial would be superior if it offers lower inductance etc.

Much appreciated

I've always used the Kemet Goldmax MLCC here.

I have to say though, I only see red in most of the boutique builds I've seen around so I'd be interested in what the thoughts are on this too..... I'm sure it's been mentioned here before but who knows......

Is there something particular about the Kemet Goldmax you prefer over other MLCC types?

jrasia said:

Is there something particular about the Kemet Goldmax you prefer over other MLCC types?

Click to expand...

I think I read something where Foote said he used them, at least for the bypassing when I was looking for recommendations.....could've been someone else though..
Kemet is reputable. I see them used in a lot of gear,especially in the feedback loops, , although I do stick with the 200v ones ..... I think I can hear more clarity or cleanliness I guess when I replace the lentils with them in circuits....Could be because I read I should...... or because the higher voltages I use keeps them behaving???/ I really don't know....

I do like the Wima's and use them for certain things.....Be interesting to hear if there's a reason to use them in these spots....... or not....

NPO/COG dielectric should be very neutral (not have any sound of its own).

JR

I'll second the suggestion for an NP0/C0G multilayer ceramic. Films can work well, but their physical size makes their strays more annoying for a feedback capacitor. You really just want a small, wideband cap in that position, placed as close to the device(s) as possible so that the feedback path is as short and compact as possible. Component leads are not your friend here, so a surface mount MLCC will always be preferable.

Plus, with an 0805 or 1206 size cap, you can usually fit it into existing PCB traces very close to the devices, if you're doing a retrofit. These caps will generally fit between the pins of a 0.1" DIP package, so for a dual op amp, you can fit it right between the device pins for the inverting input and the output, making it basically as small as possible. This is a useful trick if you're replacing an older slower op amp in a rangy PCB layout - you can put a small feedback cap right at the chip pins and effectively shunt away the rest of a less than ideal PCB layout, something very common with PCBs from the early 70s that were designed for slow op amps.

Thanks Guys.  Love this place!

Thats a great idea about using smd caps on dip pins Monte.  Is stray capacitance really that fussy about lead length?

I am modding console strips with new high speed opamps (lme49720, opa1612), but the current feedback caps are only a few millimeters away from the pins already in most places.  Worth worring about?

Just be aware of your power supply being taxed too far.

It's not as hard to do as I had originally thought. You'd be surprised  (well I was  :-[) how much a couple of chips will take away.......