can i replace polyester capacitors with tantalum

[shadowfire]

I've been wiring up guitars for years and the last few months I wanted to make a few pedals. So this ones a quick question I couldn't find on google. Pretty much I ordered about 30+ different components not realizing that there is polarization in tantalum capacitors and the biggest reason to get tantalum is its size compared to an electrolytic for guitar pedals and so forth.

Anyways this is a simple question
I've got a Maxon OD808 inspired project on the go and the capacitor I'm concerned about in this circuit is the 220nf (0.22uf) capacitor in it. In this instance do I have to put the cathode to ground or it doesn't matter on pedals? Just need some quick assurance

or is it a better idea to just replace the capacitor with something else. At this point it's a first build so I'm not too concerned about getting the best tone possible out of this. Just want to know the do's and don'ts about dealing with polarized capacitors that I don't know.

[R.G.]

Anyways this is a simple question
I've got a Maxon OD808 inspired project on the go and the capacitor I'm concerned about in this circuit is the 220nf (0.22uf) capacitor in it. In this instance do I have to put the cathode to ground or it doesn't matter on pedals? Just need some quick assurance

Polarity always matters to electrolytics unless they specifically are the bipolar or non-polar versions. If you have polarized electros, whether aluminum or tantalum, make sure the + lead is always 0V or more positive with respect to the - lead.Tantalum is slightly more tolerant of reversal than aluminum, but that just means it fails after a longer time. I don't know the exact circuit that's "Maxon OD808 inspired" so I can't say about this cap without seeing the schematic.

or is it a better idea to just replace the capacitor with something else. At this point it's a first build so I'm not too concerned about getting the best tone possible out of this. Just want to know the do's and don'ts about dealing with polarized capacitors that I don't know.

See above. They're marked with polarity for a reason.

Another reason to use film instead of small valued aluminum or tantalum (or even niobium these days) is that electrolytic caps drift as they age. One rule that doesn't get a lot of play among beginner circuit designers is to never use an electro cap for a filter where you want the cap to set the frequency rolloff. They will drift off even if you get one that has the right value to start with.

[Transmogrifox]

If you're doing something based on the OD-808, then it probably looks like schematic I linked below.  In this particular schematic the 0.22uF caps are tantalum and polarity is shown.  This will give you ideas for how to do this.

It's probably best to breadboard your circuit first, measure across all polarized caps and confirm they are inserted the right way.

For whatever it's worth, multi-layer ceramic chip capacitors are available readily up to 10 uF (there are some up to 100 uF but very expensive).  Those would be surface-mount, but the SMT 1206 size is easy to handle and it happens to fit nicely on the pad-per-hole perf boards.  I have also seen relatively small-sized leaded ceramic caps up to 100 nF.

[amptramp]

Tantalum capacitors have low leakage compared to aluminum electrolytics but you are using them in a tone control (if the above schematic is correct) and variations with temperature and aging can be expected.  If you build a second one, the tolerances are enough that it would probably not sound the same, but if you only build one, that is not a problem.  You have enough resistance in series that the typical failure mode of defects in the tantalum pentoxide dielectric transforming to conductive tantalum dioxide cannot occur because the resistance limits the current enough to keep the internal temperature down.

This is OK for a one-off but not production since you would have the Fuzz Face problem of no two units sounding or behaving alike.

[Transmogrifox]

You have enough resistance in series that the typical failure mode of defects in the tantalum pentoxide dielectric transforming to conductive tantalum dioxide cannot occur because the resistance limits the current enough to keep the internal temperature down.

Please correct me if I'm wrong, but I think it only takes reverse voltage to cause the chemical change.  The high resistance in the circuit limits reverse voltage against the reverse leakage currents and has a slower rate of degradation.  In this instance the failure is not sudden and catastrophic.

Without current limiting, as the part deteriorates it has higher leakage and then dissipates more power, more power means more heat, and that is when your nose is the first to tell you something is wrong.  The part can be degraded over time with a smaller reverse bias even if the circuit does not provide enough power to bring out the smoke.

I would guess the first sign of a problem would be leakage current causing voltage drop at "BIAS", potentially enough to cause nasty clipping in the tonestack.  There is a good chance you would see high leakage on a reverse biased tant cap even if it wasn't damaged.  You would know if it was damaged when you swap the polarity and it still has excessive leakage.

I have only ever used tantalum caps for power supply filtering while I don't find the lower-voltage poly film caps to have excessive sizes for values <470nF.

On the scientific side of things this is an interesting technical article.  Has some neat microscopic photos and thermal images.
http://www.kemet.com/Lists/TechnicalArticles/Attachments/199/2014%20EDFA%20Tantalum%20Cap%20Failure%20Analysis%20Review%20by%20Javaid%20Qazi.pdf

[shadowfire]

The only big differences in my OD808 is it's going to be 18v instead of 9 using a voltage doubling circuit (all caps are 50v minimum) and more the less with a bit of work turned it into an onboard preamp with a few push pulls. Plus I am not a fan of 2 volume, 2 tone I feel they can be doing far more useful functions with that "Real estate". I usually do passive mods all the time but I figured I'd try this.

so back to my original post.
It's been a while since i looked at the legit schematic as I was mainly focusing on the veroboard website that I saw this on for the maxon OD808 and the user who posted the diagram was right it is a polarized tantalum cap, alright that solves that question.

My only other concern is say you're designing a schematic and you can use either a 1.0uf capacitor, box film style or a polarized capacitor is there ever a time where you really have to use electrolytic? The weekend is here though so I've got plenty of time to read up on these as it's not something I've given a lot of attention to prior to yesterday.

thanks for all the insightful replies.

[Transmogrifox]

...is there ever a time where you really have to use electrolytic?

1) When you can't fit the required capacitance into the available space. 
2) You're designing for a large company who makes thousands of the units and the cost savings from cheaper capacitors adds up to a significant amount.
3) You're nearly broke and you will take whatever parts you can cannibalize from recycled electronics or electronics junk you can get for almost nothing at a 2nd hand store.  Then if most of the caps you get are electrolytic it's probably what you will use.  It's a trickle-down effect from #2.

[PRR]

> 1.0uf capacitor, box film style or a polarized capacitor

1uFd is a funny spot.

Below 0.5uFd, electrolytics are less common and film caps are often practical.

Over a few uFd, film caps get huge/expensive while electrolytic is cheap.

You often see points where a 1uFd cap would do, but the designer specced a 5uFd e-cap because it was cheaper than any film and as-cheap as any e-cap.

[nachus001]

If you're doing something based on the OD-808, then it probably looks like schematic I linked below.  In this particular schematic the 0.22uF caps are tantalum and polarity is shown.  This will give you ideas for how to do this.

It's probably best to breadboard your circuit first, measure across all polarized caps and confirm they are inserted the right way.

For whatever it's worth, multi-layer ceramic chip capacitors are available readily up to 10 uF (there are some up to 100 uF but very expensive).  Those would be surface-mount, but the SMT 1206 size is easy to handle and it happens to fit nicely on the pad-per-hole perf boards.  I have also seen relatively small-sized leaded ceramic caps up to 100 nF.

Beware with the multilayer caps. They have high capacity values but they suffer from piezoelectric capacitance variations in an order of up to an 80% of its labeled value once biased. This is stated in the datsheets in a dielectric type basis. Y5V has the highest capacitance values but the worst value drift as function of the bias voltage. Conversely, the C0G/NP0 type exhibits the best  value stability as function of bias and temperature but the value span is limited to values somewhat low (200nF at most)...

[GGBB]

I have also seen relatively small-sized leaded ceramic caps up to 100 nF.

Through-hole MLCCs are available in reasonably small sizes (smaller than same-value film or electrolytic) up to 10uF or so. They are a bit pricier than film at the higher values, but huge space savers.

Beware with the multilayer caps. They have high capacity values but they suffer from piezoelectric capacitance variations in an order of up to an 80% of its labeled value once biased. This is stated in the datsheets in a dielectric type basis. Y5V has the highest capacitance values but the worst value drift as function of the bias voltage. Conversely, the C0G/NP0 type exhibits the best  value stability as function of bias and temperature but the value span is limited to values somewhat low (200nF at most)...

X7R performance in this regard is a fair bit better than Y5V and close to NP0 - you can find X7R MLCCs in pretty much any value up to 100uF. And the voltage dependence can be minimized by keeping the voltage bias to a small percentage of the cap's voltage rating, so get the highest rated ones you can use size-wise (>=50V). At bias voltages such as one would find in a 9V pedal, voltage dependence is minimal - a 100V X7R MLCC cap would vary due to bias by far less than the cap's tolerance.

[bluebunny]

In this instance the failure is not sudden and catastrophic.

I always thought that reverse-biased tants made good fire crackers?  Could be wrong...   

Having said that, I had a reversed electrolytic (the purchased board had a mistake on the silk-screen) and it made a pretty good pop for a little guy!  And it spewed some interesting guts too...   

[Ice-9]

In this instance the failure is not sudden and catastrophic.

I always thought that reverse-biased tants made good fire crackers?  Could be wrong...   

Having said that, I had a reversed electrolytic (the purchased board had a mistake on the silk-screen) and it made a pretty good pop for a little guy!  And it spewed some interesting guts too...   

Yeah when Tant's go , they make a nice pop and some smoke man !

[stallik]

Yeah when Tant's go , they make a nice pop and some smoke man !

Whooo! So they do. Perhaps I shouldn't have tried that with the family asleep

[Transmogrifox]

I always thought that reverse-biased tants made good fire crackers?  Could be wrong...   

Not wrong, just need to pay attention to the context.

That pop is Volts*Amps = Power.  More power, more pop.

In a current-limited circuit where Amps == almost nil in a low-voltage circuit, then Power is negligible and the best joy you might get is a little smoke or a crack -- or like I mentioned, just excessive leakage current that makes your op amp bias go to the rails.

The context of the statement was a very specific circuit in which I could mentally guesstimate that the power available to the reverse cap would be measured in the order of a few milliwatts.

When you put a reverse-biased tant or aluminum electrolytic cap onto something that doesn't limit power then you get entertainment value likely equal to the cost of the capacitor...

Unless it's a 400V cap connected across the wall outlet, then you might get a hefty Dr bill.  You might start questioning whether it was value for the money and pain.