Will a tantalum capacitor "behave" substantively differently from a regular electroly

EDIT.
Note to mods:

I tried to post this in General Science several times, but with no success. Could someone kindly please move it there? Many thanks.

Also, what's going on here? (With regards to posting new threads, that is.)
----------------------------------------------------------------------------------------------------------


Will a tantalum capacitor "behave" substantively differently from a regular electrolytic in a circuit?

Specifically, I'm modding an an archaic drum machine (the SoundMaster SR88) to allow for an "accent" feature (basically, a simple JFET amplifier to provide a simple boost in amplitude at selected beats) through a trigger output. The output will send an approximately 5v / 8ms burst, and I need to provide a simple envelope to allow for maybe 80 ms; so I'm using the straightforward capacitor-to-ground followed by resistor-to-ground on to power the JFET circuit method.

I'm largely appropriating a similar circuit from a Roland/Boss DR55 schematic, and it calls for a 680k ohm resistor and a .15 micro farad tantalum cap. By my inept calculations, that cap value is just too damn small. But maybe tantalum caps behave significantly differently in this regards? Low leakage and whatnots.

I'll use the tantalum if necessary, but I prefer to salvage materials from garbage I "cannibalize" and I can't find a tantalum cap even remotely close to that value.

Any thoughts? I'll post an image of the schematic if it will help.

----------------
Ok, here's the relevant bit:

Please Register or Log in to view the hidden image!

The cap in question is circled in red, and the 5v pulse is introduced at the orange line. I'm assuming the circled "T" indicates tantalum (moreover most caps of that value are not polarized), doesn't that value seem kinda small? As noted, the pulse is around 8 milliseconds and I need it to last a whole lot longer, lest I only plan on programming hardcore beats.

Oh, and ignore that CSQ bit--the signal will travel directly from the envelope circuit to the ground pin of the FET.

 

Log in or Sign up to hide all adverts.

Tantalum capacitors are lower leakage, higher Q and lower ESR. They are more accurate in terms of tolerance. However, they do not tolerate overvoltages well at all.

If you need .15uF I'd strongly recommend a ceramic. They are cheap, stable and easy to use. If you used a typical 1uF 16v electrolytic you might see around 5uA leakage; your resistor provides about 7uA pulldown current. Thus your predicted timings would be way off (almost double the current you expect.) And since they change as they dry out, the timing will change as well.

 

Log in or Sign up to hide all adverts.

Interesting. My inclination has generally been to avoid ceramics, but for a whole different set of reasons than what would be applicable here. But I shall try the ceramic--thank you!

Also, the transistor on the original schematic--a 2SK30A--is kind of archaic and hard to come by, so I've substituted a J201. Noise isn't really a concern here; after all, the "drum" circuits--a bass drum, snare, closed hi-hat, and I'll be adding a simple rimshot--are mostly comprised of noise and/or a simple twin-T style oscillator, so... I figured that any roughly comparable n-channel FET would suffice--also I've got a bunch of J201s lying about (the proper ones, not that crappy Radio Shack version which isn't even close). Do you think that should be adequate, or would you recommend another transistor?

The SR88 was designed to operate on 6v, but someone replaced the original connectors with a 9v (I picked this machine up off of fleabay for dirt cheap). Interestingly, though not surprisingly, I'm finding that voltage manipulation affects the sounds quite dramatically.

 

Log in or Sign up to hide all adverts.