Sweet Honey O/D cap question

[stax]

Hi just a quick question about polarised vs non polarised caps.
A verified schematic of the SHOD shows the two 1uf caps coming from legs 1 and 7 off the IC as being polarised but on the other layout one is a box type and the other a ceramic disc, any idea why this would be and which is preferable?
Cheers
Stax

[IvIark]

I will always use poly or multilayer ceramics instead of electrolytic caps.  Electros can last for years but they still have the shortest useful life out of all the components in an average build, and that along with their inherent deficiencies such as leakage, poor tolerances etc, then I'd prefer to use what I consider to be higher quality cap types, which are still available at a reasonable cost.  So most of my layouts that have 1u caps will have been replaced by polys or MLCs.  Anything above 1u I tend to stick with electros or tantalums because they are easier to find at a reasonable cost than polys or MLCs (although I do have MLCs up to 10u and so may still personally opt for something else for personal builds), so I leave electros in place because that is what the majority will use.

[stax]

Thanks for the reply, as a novice I somehow assumed the electrolytic caps were better (don't know why).
Looking forward to starting this when the components turn up tomorrow, I have tried to build this before with a different layout and had a few problems (my fault), this layout looks less cramped so  I'll give it a go.

[GoranP]

My feeling is that electrolytes are the bottom of the capacitor barrel, as far as useful characteristics go.
Their only redeeming quality is when it comes to values >100u where they are still less money (and sometimes space) consuming than the ceramic/metal film variants.

[jaredcohen]

IvIark and Goran, thank you for helping out all of us who have questions. To follow up on the OP's question, I understand what you both are saying about cost and availability of poly or ceramic caps and the quality of electrolytic.

My only question then, is that the original schematic has the 1uf's as polarized. Does the fact that the film/ceramic caps aren't polarized affect the circuit at all? (I'm learning as I go, and polarized caps are pretty confusing for me still... thank you for your patience with my stupid basic electronics questions).

[induction]

You can always replace polarized caps by nonpolarized, assuming you can find the values/voltages/sizes that fit. Polarization is a limitation of many electrolytic (and tantalum) caps, not a feature. There is no electronic benefit to polarization in caps (unless you like watching them explode).

In more detail: Caps are useful for blocking DC and passing AC. Polarized caps only block DC when correctly biased. When reverse-biased, they pass DC and the resulting current destroys the dielectric and kills the cap. The higher the reverse voltage, the more current flows, the more damage to the cap. When they die, they behave like a jumper, not an open-circuit.

It is generally ok to use polarized caps when there is no chance of reverse-biasing them. For example, they are fine for power filtering. They can be used as coupling caps only if there is a DC offset between the stages being coupled. For example, the output of an op-amp (in a single-supply circuit like most stompboxes) usually has AC signal riding on top of a 4.5 V DC offset. If this stage feeds the output, we want to remove that offset before feeding the signal to the output jack.  Here you can use an electrolytic cap with the anode connected to the op-amp side. Even though the AC signal swings both up and down, there will always be a higher voltage on the op-amp side. You should not use an electrolytic cap between stages with the same DC offset (including zero), because half the time the signal will reverse bias the cap. Even short reversals will reduce the lifetime of the cap.

None of these problems occur with unpolarized caps. They are fine for all applications that fit their capacitance and voltage specs. Incidentally, you can make a nonpolarized cap from two polarized caps in series with either anodes or cathodes connected together. (Caps in series add the same way as resistors in parallel: C = C1*C2/(C1+C2), so use two caps of double the value you need. For example, if you need C=1uF, set C1=C2=2uF.) In this arrangement, when the signal swings positive, the first cap will be forward-biased, and the second cap will be reverse-biased. Reverse-biasing is not harmful in this situation because the DC current is blocked by the other cap.

[jaredcohen]

That's an incredibly helpful explanation Induction. Thank you so much. I can't express how appreciative I am for the help all you guys give.  I enrolled in a basics electronics course at the local college and I'm learning way more from message boards!

[GoranP]

Thanks Induction!

As usual, straight to the point and beyond reproach...