# Can I use electrolytic capacitors in an oscillator?

I'm trying to learn about oscillators so I thought I'd put together something like one of the colpitts oscillator circuits I've seen and play around with it. I'm pulling scavanged parts from my junk box and realized As far as matched pairs, I have almost entirely electrolytic capacitors. Is it even possible to use these to build an oscillator since they are polarized?

If I can and orientation matters, how would I orient the cathode and anode? Are there other things I should consider?

Please excuse what I'm sure is a simple question.

• If you're trying to build a Colpitts at such crazy low frequencies, how are you getting an inductor that goshawful huge?! Nov 10, 2016 at 4:14
• It is not my intention to target a frequency, just use the parts I have to make a Colpitts. Are you saying that electrolytics would only be useful in constructing an extremely low frequency oscillator? Nov 10, 2016 at 4:17
• Generally, if you're operating at frequencies that demand the use of capacitors that large in value, you use RC topologies (such as the Wien-bridge that propelled Hewlett and Packard to fame) instead of LC, primarily due to the ghastly size/weight and nonlinearity of large value inductors. So yes, electrolytics are only needed if you're going to make an extremely low frequency oscillator (like lower end of the audio band, or infrasonic), and in those cases, you don't use a LC topology like the Colpitts because the inductor's even worse than the caps at that point! Nov 10, 2016 at 4:20
• I'm understanding you to mean the typical values of electrolytic caps are impractical for an LC oscillator because the corresponding inductor that would oscillate in the circuit would be ridiculous. (Guess I shouldn't strictly trust the calculators.) I'll check out the Wein-bridge and see if maybe I can put that together with what I have. The history sounds interesting at the least. Thank you! Nov 10, 2016 at 4:29

Yes, you can use electrlytics in a Colpitts.

A Colpitts oscillator has places where you can use polarised capacitors, and for the places that experience bipolar operation, you can use two in series with a Mohm bias resistor to the centre point to a high voltage to keep them both properly biassed.

Use the choke from a flourescent fitting as the inductor, and you might get operation way down in the low audio. I don't normally recommend mains transformers as inductors, and certainly they don't ever make good power inductors, but here they would be a good source of large low frequency inductance. Excellent learning opportunity. Stability, tolerance, Q, suitable values, well, there's a reason electrolytics don't get used much.

You will probably find it won't work with a small inductor and big capacitor. The sqrt(L/C) gives you an impedance. Any given design of sustaining amplifier will only work well over a range of tank impedances. Too low an impedance and the active device gets excessively loaded, and the losses in the components drop the Q values too low to resonate. Keep the impedance up with a large inductor. That's part of the learning, but unfortunately an oscillator that just 'doesn't go' can be a bit of an enthusiasm damper.

I will agree with the comments that if you just want a quickly working low frequency oscillator, then go RC. If you want to learn, then find a big inductor to go with the big caps.

• Thank you! The idea of creating an oscillator at audio frequencies is interesting. It would be fun to actually hear it. I would like to start by getting a working one first because of the "enthusiasm damper" effect you mentioned. Maybe this is a follow-up project. Your insight is very helpful. Nov 10, 2016 at 14:49

Following Neil's answer, yes you can do that, but you usually won't.

As said, it's necessary not to build an LC oscillator with vastly differently sized reactive and capacitive elements, and finding an inductor in the same order of magnitude will be a bit expensive (if you can't scavenge one from something else).

Also: If you want an oscillator, you're often actually interested in an exact, stable frequency oscillation.

Now, most electrolytic capacitors are actually sold with a 20% value tolerance. That's not a great start to hit an exact frequency. You say you've got matched pairs – but are these really matched, or do they just carry the same specification?

Also, electrolytics are usually mainly used as relatively long-term, large value "energy storage" in the power supply of loads. As such, they're optimized for high capacity density, but not for low equivalent series resistance, so instead of this typical Colpitts

simulate this circuit – Schematic created using CircuitLab

You'd have to consider the circuit including the parasitic series resistances; and also, electrolytics capacitors aren't totally flat over all frequencies, so if you don't operate at low frequencies, subtract a few percents from the nominal capacity value:

simulate this circuit

Suddenly, your oscillator is dampened, and the the fact that there's voltage drops over R1par and R2par means that the transistor isn't quite working at the same Uce, which also means the amount of energy stored per oscillation will change. This makes stable operation a bit tricky.

Regarding Polarization/Orientation: as usual, make sure the caps are oriented the same way as the biasing.

• Thanks for the additional information Marcus. Sounds like a tricky thing to get working. I'll probably look for a more reliable option for my first oscillator and come back to this idea when I'm a bit more informed and can handle the troubleshooting better. Nov 10, 2016 at 14:52
• 100 Ohms ESR ? Are you sure? I would expect maybe 1 Ohm in a general-purpose 470 uF elyt. A "wet" low-impedance Al elyt or tantal of that capacity would be about 100-200 milliOhms. And a modern solid-poly would be as low as 6-10 milliOhms, at the specced frequency (admittedly worse in the lower audio / near DC range). Also, I don't think the capacitors in the Colpitts need to be precisely matched to get an oscillation... = it might just as well work :-)
– frr
Nov 10, 2016 at 14:57
• On the back of an envelope, with a 1 mH inductor, the circuit would have a resonant frequency of about 300 Hz and each component (L, C) would have an impedance of about 2 Ohms. That's not much, but at the resonant frequency these cancel out, so the parallel LC circuit behaves as quite a high impedance. In the schematic suggested by Mr. Müller I would like to stress that the power supply must present preferably a very low impedance. Use an elyt of maybe 10 mF or more (as much as you have) to prevent the PSU from skewing your resonant properties.
– frr
Nov 10, 2016 at 15:06
• The two Ohms (or one Ohm combined?) of resonant impedance will compare to the ESR of your capacitors and of the inductor, to give you some overall "quality" figure of the resonator. As others have suggested, by using a larger inductor you should get a lower resonant frequency and hence higher impedances of the lumped components (L,C), which should/might result in a higher Q (check the resistance of your inductor if you use something large, like a 100W mains transformer).
– frr
Nov 10, 2016 at 15:12