Question

[Tryinmybest]

Hello

just wondering some basic stuff

for any given schematic that states only the uF value for a capacitor, how does one go about figuring out the volt, tolerances, etc aspect of the capacitor?

thanks

 

[Svart]

for signal or for power? for power, find the next value higher than the rail and then pick the NEXT higher value. You'll never have a problem.

example: 24v rail. find the 25v caps and then choose 35v. plenty of safety.

Signal is a bit harder, you need to know the RMS value of the signal and do the same. most audio will be low enough to use 6.3v caps except when using the caps to block phantom power which is 48v. find 50v caps then use 63v caps.

 

[alk509]

Context. If the cap has 300V across it, you don't use a 10V cap. If you need EXACTLY 2.2uF, you don't use a 20% cap, et cetera.

Best,
Al.

 

[Tryinmybest]

thank you guys!

svart, thats a cool way of doing it. thanks for that, makes sense

 

[Infernal_Death]

[quote author="Svart"]
example: 24v rail. find the 25v caps and then choose 35v. plenty of safety.
[/quote]

Hmm what happens if you feed a voltage doubler on this 24v rail for supplying phantom power ? Wouldn't be wise to use 35V caps there. I think it always depends on the context of the used circuit. Just use the next higher value of the voltage you get at any given point. The harder part is to know the voltages on all points of the circuit.

Flo

 

[Tryinmybest]

oh, i see what you're saying infernal. so you mean dc voltage induced from phantom power creating "higher" values needed?

so heres a question... lets say you mic a source and thats phantomed powered and then you feed the output to this unit that is in ques as far as the V values go. now, is that introducing or inducing a DC voltage at the input? if so, in this case would one then pick the proper cap volt values from there? so if its a 15v rail plus 48v phantom should one pick then the next value up from 63v and then pick the next higher value as our cap?

 

[Svart]

well I consider the voltage doubler a specialized circuit and special rules apply, but the same way of picking caps should be done, find the highest voltage that the cap(s) will see then find a value of cap higher than that and then pick the next higher value.

 

[Tryinmybest]

so svart, out of curiosity, what is the threshold of over shooting the value? say you need a 30v and you use a 100v cap. is this where its tolerance comes into play? how much more can one over shoot?

 

[alk509]

[quote author="Svart"]find the highest voltage that the cap(s) will see then find a value of cap higher than that and then pick the next higher value.[/quote]

That's an excellent rule of thumb. But like ID said, the hard part is figuring out the voltage across the cap, since it's really specific to each circuit...

Peace,
Al.

 

[Tryinmybest]

okay so heres my assessment and the parts i'm going to purchase for this project, if i understood everything correctly

the project:
http://home3.netcarrier.com/~lxh2/tapesat.html

parts im getting:
For the 3 opamps
http://www.njr.co.jp/pdf/ae/ae04023.pdf

As for the resistors, I have 1/4 watt -/+5% tolerance

For the 1n34A, i'm ordering these updated variations:
http://www.nteinc.com/specs/100to199/pdf/nte110mp.pdf

and for capacitors C2, C3
http://www.mouser.com/catalog/specsheets/pf.pdf

power supply, i was thinking of a wal-wart adapter -/+15 Volts, 50-100ma

Any comments?

 

[pstamler]

[quote author="Tryinmybest"]parts im getting:
For the 3 opamps
http://www.njr.co.jp/pdf/ae/ae04023.pdf [/quote]

Looks like a reasonable choice.

As for the resistors, I have 1/4 watt -/+5% tolerance

Should be okay, except for the resistors which control the pre-emphasis and de-emphasis curves, which are R2, R8 and R9. Those should be tighter tolerance, especially if you're building two channels for stereo. Of course, you can select resistors from 5% tolerance units using a DMM.

For the 1n34A, i'm ordering these updated variations:
http://www.nteinc.com/specs/100to199/pdf/nte110mp.pdf

Well, maybe. You'll probably get a different compression curve from these than from 1N34A's -- you may like it better, or not as well. Don't forget there's also a pair of 1N914s in the circuit too.

and for capacitors C2, C3
http://www.mouser.com/catalog/specsheets/pf.pdf

Again, since these control turnover frequencies, you'll want a tight tolerance; probably the only way to get one, though, is to select from several caps using a digital capacitance meter.

power supply, i was thinking of a wal-wart adapter -/+15 Volts, 50-100ma

Any comments?

If the wall-wart is unregulated, you may have problems with the opamps oscillating. You might want to go for a 20V supply and use a couple of fixed 15V regulators (7815 and 7915 for positive and negative respectively). In any case, put a 1000µF electrolytic cap on each rail (min. 25V) and I'd suggest a pair of .047µF film caps on each IC package, close to the pins.

Peace,
Paul

 

[alk509]

They have bipolar wallwarts??? I've never seen one! :shock:

Peace,
Al.

 

[Infernal_Death]

[quote author="Tryinmybest"]so svart, out of curiosity, what is the threshold of over shooting the value? say you need a 30v and you use a 100v cap. is this where its tolerance comes into play? how much more can one over shoot?[/quote]

Ok i am not 100% sure but i think there is no problem in overshooting but the actual size. Caps with higher voltage rating can be considerably bigger than those with smaller voltage rating. But could be that i am wrong and there are some other factors.

Svart i agree with what you said. This rule is quite good, i just thought that it's really important to know the actual voltage on the cap, not just looking at the generic power rails given in a circuit.

Flo

 

[skipwave]

One of the forum members did a modified version of that tape sat circuit, with a much more detailed schemo.

Lemme see if I can find it here..... :?

 

[Tryinmybest]

Paul thank you! Very helpful run down of my parts list. I'm learning alot here.

ok infernal, I'm going to do my best to just try for the closest needed cap's. Problem is that sometimes schematics/parts lists are vague.

So, Skipwave, I'd love to see that more thorough and detailed schemo if you find it. that would be awesome! :thumb:

 

[aurt]

[quote author="Svart"]Signal is a bit harder, you need to know the RMS value of the signal and do the same. most audio will be low enough to use 6.3v caps except when using the caps to block phantom power which is 48v. find 50v caps then use 63v caps.[/quote]

Given that my days are currently being spent re-capping a V-series Neve, I take exception to 6.3 Volts being sufficient for coupling caps anwhere besides a virtual earth node. I shoot for the same voltage that I would use as a bypass cap were I locally filtering. Most op-amps are not rail to rail output, but if I remember correctly, electrolytics aren't so fond of the reverse voltage they are subjected to for half the audio cycle, so a bit of a safety buffer never hurts. What with everybody trying to maximize their digital headroom, signals are hot these days.

And yeah, re-capping the desk is a bitch, but I just got a very cool desoldering station that makes it sooooo much easier. I have no idea how I've lived without one.

 

[skipwave]

Here we go: http://doorstopelec.typepad.com/

Not much more too it, I guess. Just some extra resistors around the diode pairs.

 

[Tryinmybest]

ok thanks skipwave. I guess i'll just see what happens. i mean ive already built a simple mic pre like 2 years ago and i wanted to get back into building stuff. im just gona attempt this and understand it best i can. if it fails oh well

 

[sodderboy]

Aurt, you SO have my sympathies! Make sure you have good ventilation, and I hope you are using 105 degree C caps? They will last longer
Mike

 

[aurt]

Definitely 105s, uprating the voltages where I can as well. I don't want to have to do this again very soon :wink:. The current lot have been in there six years, so, not that bad a run, really.

It's a bummer that very little progress has been made in miniaturizing axial electrolytics. It's also a bummer panasonic stopped making their axial 105s. I'm kludging in radials of a more appropriate rating where I can, but what a pain.