recapping? 2 questions..

[Svart]

ok, question one:

Power to my channel blades comes in through the connectors and immediately is filtered through ceramic disc caps and 'lytic caps. however, it is not bypassed at the ICs at all. they are 5532, tl082, tl084 and the like, which from my experience are fairly stable even without proper decoupling. I'm thinking i should change the discs to polyester. I haven't tested this to see if there is any improvement, but for 20$ worth of caps I'm sure it can't be a waste! should i also think about decoupling near the ICs too? there are also 68pf disc caps in some of the circuits, mainly caps that have to do with feedback.. should i change those out for better monolithic caps? might as well if i'm going to do this other stuff..? (all of the EQ caps came as polyester "box" caps) I am also changing 'lytic caps out over time too since i have found a few cheapo caps that have leaked or popped or caused scratchy noises in faders.

question 2: and extension of the first question really...
In theory , i thought it was better to decouple nearest the IC. is there a reason why the designers would have done it as the power entered the board? each blade has a central star ground also, this is where most of the decoupling caps center.

thanks!

 

[Svart]

please?

:guinness:

 

[Jonathan Hayward]

well, in all my capacitor noodling on the Amek I could only get improvement by replacing old lytics and/or bypassing the big ones.
Try it out on a channel though, there's nothing to lose (except a bit of pcb track)

 

[Svart]

yeah that's what i figure too. it would cost me around 60-80$ to replace all the ceramic discs in the channels with polyester caps and GOOD ceramic monos. guess that could be considered somewhat of a waste if nothing is improved, but i would be highly suprised if NOTHING good happened from this... :wink:

 

[Steve Jones]

Usually there are decoupling caps on every CMOS logic IC as every time a logic gate switches it throws a short circuit momentarily across the power.

 

[Svart]

well i **** you not, most of the ICs don't have power decoupling caps anywhere near them! looks like i might have to just do a few channels and give it a try.

 

[Bjorn Zetterlund]

The decoupling at the point where power enters the module is partly to isolate it from other channels; most of the time you will also find resistors of 10 Ohms or so in series- which helps filter, limits switch on surge, and also will usually be rated so as to burn up in case of a short or a failed op-amp on that channel.

Whilst most data sheets recommend individual decoupling for every chip, it isn't always necessary for stability. A few TL072's on one channel are probably not gonna be that bothered. I doubt that individual decoupling will improve the sound, unless you really are having oscillations or other funny business going on.

I've never heard of feedback caps being replaced when recapping a desk; usually it's just the electrolytics. Changing the coupling caps can make a difference, but it is by no means a guaranteed 'wow that is so much better' experience. Usual ones to try are ELNA stargets (expensive-ish audio caps) and Panasonic FC's. Why not just do a channel or two and see if you notice anything better/different?

Bjorn

 

[Svart]

yeah, gonna try a pair of channels i use for super clear/clean room mics and see what happens. Not really having much problem with oscillations either. Just trying to give this console the best chance it can get at being better than stock!