does it makes sense to use electrolytic can as bypass cap ?

Oct 24, 2007 15 Replies

i have a circuit board that appears to have had some mods done as components do not match or blend with other components.



my concern for now is two polarized electrolytic caps that appear to be used as "bypass caps" ?? i thought tantalum and ceramic were the best choice for bypass? is this a problem ?



one is a electrolytic 1uF/50v polarized cap between 32v line to Vcc2 on chip and 0v ref


32v dc ---- 36 ohm --|-- 1uF/50v (polar) (-) ---| | | Vcc2
0v ref


and the other is an electrolytic 22 uF / 6.3v (polar) cap betweem



5v line to VCC on chip an 0v ref


any way i just recently read to use ceramic but i am wondering if there is a reason one would want to use electrolytic ? or was this just a dangerous amateur who did this ?


*OTHER Question *

there is a 36 Ohm 1/4 watt resistor ( orng blu blk gold ) between (inline with) the 32V supply volts to a Vcc2 pin on an IC, what is reason to use this setup and could it be another mistake made by previous repairer ?



thanks for any useful help, robb


It creates an RC low-pass filter to smooth out the power even more than just a capacitor can do. I do something similar on chips that are especially sensitive to supply ripple.

an

more

that

thanks for reply,

so is there any magic with the 36 Ohm value anf 32.6 Volts? could i use 30 Ohm or 47 Ohm resistor or other value ?

thanks again, robb

It depends what you are trying to bypass ?

The electro's types are basic caps and good for local rail stability but do not work well in preventing RF from traveling around. Using ceramic type caps solves the issue of high frequency traveling around. it has to do with material used and construction forms. In most cases all you need is the electrolytic to help stabilize local rail voltage on a particular component how ever, if your circuit have functions of generating RF or would be sensitive to stray RF then you need to use caps for by pass that will work to suppress it.

You may find combinations of a low value non inductive type cap being coupled side by side with an electrolytic. This is performing 2 different jobs, not summing them.

"I\'m never wrong, once i thought i was, but was mistaken" Real Programmers Do things like this. http://webpages.charter.net/jamie_5

32 Volts is getting a bit out of the wheelhouse for tantalum, so electrolytic is probably a fair choice.

This would be a good application for tantalum, but electrolytic is cheaper.

Neither one of these sound like a "bypass" application, rather they are bulk storage so that supply regulators don't have to respond to every tiny little current pulse the IC's ask for and/or to isolate other circuits from noise so generated.

Ceramic capacitor 1 uF or greater is likely to be a big sucker. Aluminum electrolytics in that value range are compact, cheap, adequate, readily available.

They may have wanted a bit of time delay for Vcc2 to come up after Vcc1. Or just to decouple it from other circuits. What kind of IC is it? Could be perfectly normal.

Tantalum capacitors are also electrolytic.

That is true.

Other tantalum constructions exist as well as electrolytics other than aluminum, but I didn't feel the need to nit-pick, rather contented myself to common usage within the apparent scope of the OP's question...

snipped-for-privacy@where.> > one is a electrolytic 1uF/50v polarized cap between 32v line

to

0v ref
[trim]

an

Thanks Bill for help,

it is a VFD display *driver* IC, 32 bit shift reg with latches. VCC1 5-12 V to power the IC and the VCC2 is the controlled voltage 0-60V (< 40 mA) which drives an the VFD display.

So i can replace the 36 ohm (it is fried) with anything close and probably should choose one on the higher side (i.e. 47 Ohm over the 22 Ohm) ?

thanks again for help, robb

Really? As far as I\'ve been able to find out the only types of capacitors that use tantalum in their construction are electrolytic. Am I missing something?

Small electrolytics are fine as they dont have a lot of inductance.

However in my designs I always use tants and a ceramic to bypass IC's.

formatting link
\\pcbcad28.htm

Figure out why it fried, fix the problem, replace with same value unless you have a compelling reason to change it.

snipped-for-privacy@where.> > it is a VFD display *driver* IC, 32 bit shift reg with

latches.

and

over

think there was a shorted connection between 32 v and ground and i think that is what fried the resistor ?? a bad solder joint bridged to ground track

i do not have a 36 Ohm but i have 22 and 47 and so i figured i should err on higher side (the 47 Ohm)

thanks for the help, robb

Rob

What about two 22 ohm Rs in parallel, then in series with one other (that is assuming you have a number)?

Robb,

In changing the resistor from 36 ohm to 47 ohm, just consider that your Vcc2 will decrease by about another 1/2 volt (in addition to the 1-1/2 volt already dropped by the 36 ohm one). On a 32 volt supply, it totals to less than 10%, so you might be just fine. It's hard to say from here just how low your VCC2 can go without affecting operation or reliability.

Considering power dissipation in the resistor, 40 ma through 47 ohms will dissipate just 75 mw, so a 1/4 watt resistor will still be fine.

and

over

that your Vcc2

1-1/2 volt

totals to less

just how low

ohms will

fine.

Thanks Ninja,

i found a 39 Ohm resistor among my meager supply so i am a bit closer to original.

the application was the supply power to a VFD display so i suspect a lower voltage will just dim it up a bit maybe give the VFD a bit more life .

Thanks for the help, robb

other (that

Thanks for help Suzy,

fortuneately i found a closer equivalent a 39 ohm so that should be close enough and the application is the supply power to a VFD display so lower voltage just means dimmer and maybe longer VFD life ?

thanks again, robb

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required