circuit theory question about capacitors not handling AC (cathode bypass caps?)

[mikeymike]

Hello everyone, I just wanted to start off by saying hello, this is my first post here. I'm relatively inexperienced at amplifier repair/modding/circuitry but I have learned a good deal over the last decade or so from reading online. I just came across an article on premierguitar.com written by RG Keen called The Immortal Amplifier Mod that says capacitors that are polarized can't handle AC (this was in reference to a rectifier tube failing short resulting in sending AC into the filter capacitors causing them to short as well).

So I was wondering, do cathode bypass capacitors, on say a blackface preamp circuit, deal with any AC? My question revolves around the idea that cathodyne phase inverters use both the plate and the cathode for signal and thus there is AC present on the cathode (at least in a cathodyne phase inverter?).

Like I mentioned, I'm still learning so if I misunderstand how the cathodes function in a blackface preamp (or what Mr Keen was saying specifically regarding the filter caps and voltages/current present there) I would love to be enlightened and look forward to any greater understanding I can glean from yall.

[sluckey]

You're taking that statement out of context. Keen was talking about shorted rectifier tubes. He actually said, "But sometimes when the rectifier tube fails, it does not open. Sometimes it shorts and dumps the full, high voltage AC power into the filter caps. Polarized filter caps cannot stand AC, so they usually fail shorted." I underlined a key word in this statement.

 He was also speaking to a mostly non technical musician audience and offering a way to proctect the filter caps from a shorted rectifier without going into much detail.

In normal functioning amps polarized caps handle AC just fine. They pass AC and block DC just like any other capacitor. Look at some solid state circuits. You will see polarized electrolytic caps used a coupling caps. This is necessary because transistors are low impedance devices which need low values of resistors to operate. This means that the capacitors used must be BIG values (polarized electrolytics) in order to provide the necessary RC time constant to work well with AC signals.

[mikeymike]

Look at some solid state circuits. You will see polarized electrolytic caps used a coupling caps. This is necessary because transistors are low impedance devices which need low values of resistors to operate.

Is this in reference to ESR being lower on larger capacity capacitors? I watched a video a week or so ago talking about the relationship of capacitor capacity (hah) vs their ESR and a general statement was made by the youtuber that larger capacitors will have lower ESR's than smaller ones.

Regarding the article, I know that caps pass AC and don't mind it that's why I was confused about what was being said by Mr Keen. If you have any further detailing that would help enlighten me (and you have the time/desire to do so) I would love to understand this better. From what I gather you're saying the key point is that they're filter capacitors. Would it be that the filter capacitors are sending the relatively small AC ripple to ground in their role as filter capacitors but if full AC signal was sent to them it would be like shorting the AC to ground through the filter capacitors?

I am a little confused by your response regarding the shorted rectifier tube comment. I was trying to be clear in my original post and did make mention of the rectifier failing short:

(this was in reference to a rectifier tube failing short resulting in sending AC into the filter capacitors causing them to short as well).

Sorry if this is coming off as sophomoric it's not my intention at all. Just trying to grow my knowledge of tube amplifiers and would like a little help in understanding them more 

[sluckey]

Polarized caps are just that. The polarity must be observed. They cannot tolerate reverse voltage. If you don't believe it, connect a 40µF cap backwards across a 400VDC supply. It will probably explode!

So, that filter cap operates just fine in a normal functioning power supply. But if the rectifier becomes shorted you will now have AC voltage (swinging positive and negative) applied across the filter cap. Half the time the AC voltage will be the correct polarity. The other half of the time, the polarity will be reversed. This reversed polarity will smoke the filter cap, maybe even explode it.

If you need a more detailed explanation, look at some theory books. Or maybe one of the deep theory guys here will chime in.

[mikeymike]

Polarized caps are just that. The polarity must be observed. They cannot tolerate reverse voltage. If you don't believe it, connect a 40µF cap backwards across a 400VDC supply. It will probably explode!

I totally believe you, I've heard of this happening to a peer in electronics class when I was in high school haha, there was a loud pop and some laughter from the teacher and some subsequent explaining as to what he did wrong

So in the example with the filter caps, when the rectifier is functioning (flipping the lower half of the waveform back up above zero) the ripple filtering is occuring all above 0 volts so it's not techincally reverse polarity, just rising and falling above the 0 volt mark. This is okay for polarized capacitors but if it were to go under 0 then problems arise?

Is this also occuring in the cathode bypass caps? All swings are staying above 0 volts? Is there an offset applied? Is this elevated in reference to ground (I think thats the right way to term it?).

Sorry for all the questions, maybe I can do some googling

[jojokeo]

Polarized caps are just that. The polarity must be observed. They cannot tolerate reverse voltage. If you don't believe it, connect a 40µF cap backwards across a 400VDC supply. It will probably explode!

It's been a while since doing this on purpose (boredom/playing around/teaching others around the shop, etc.) but there's no probably - they DO explode! And loudly too better than firecrackers depending on the caps...

[sluckey]

So in the example with the filter caps, when the rectifier is functioning (flipping the lower half of the waveform back up above zero) the ripple filtering is occuring all above 0 volts so it's not techincally reverse polarity, just rising and falling above the 0 volt mark. This is okay for polarized capacitors but if it were to go under 0 then problems arise?

Exactly

[punkykatt]

Hmmm, I have a Princeston Reverb build that blew the line fuse, replaced the fuse, was good for a couple of gigs, fuse blew again, replaced the fuse flipped standby on fuse blew right away. Next day I put the amp on the bench installed new fuse hooked up the light bulb limiter turned the amp on flipped the standby all was normal until I touched the rectifier tube, dead short.  Put another rectifier tube in and all is fine.  What are the chances the filter caps got stressed? The amp plays fine no hum.  What would be the best way to test for stressed caps?  Thank you any info.

[mikeymike]

Hmmm, I have a Princeston Reverb build that blew the line fuse, replaced the fuse, was good for a couple of gigs, fuse blew again, replaced the fuse flipped standby on fuse blew right away. Next day I put the amp on the bench installed new fuse hooked up the light bulb limiter turned the amp on flipped the standby all was normal until I touched the rectifier tube, dead short.  Put another rectifier tube in and all is fine.  What are the chances the filter caps got stressed? The amp plays fine no hum.  What would be the best way to test for stressed caps?  Thank you any info.

Well judging from what R.G. Keen said, i'd replace the caps if I were you! (and maybe do his immortal mod haha)

[jjasilli]

Hmmm, I have a Princeston Reverb build that blew the line fuse, replaced the fuse, was good for a couple of gigs, fuse blew again, replaced the fuse flipped standby on fuse blew right away. Next day I put the amp on the bench installed new fuse hooked up the light bulb limiter turned the amp on flipped the standby all was normal until I touched the rectifier tube, dead short.  Put another rectifier tube in and all is fine.  What are the chances the filter caps got stressed? The amp plays fine no hum.  What would be the best way to test for stressed caps?  Thank you any info.

Google how to test capacitors:  one hit:  https://www.ifixit.com/Wiki/Troubleshhoting_logic_board_components

[pdf64]

It’s worth noting that generally, ecaps aren’t subjected to ac; rather there’s always an overall Vdc bias, such that at any instant, the Vac + Vdc never results in the total effective voltage across them being opposite to their intended polarity.
Like the HT current in tubes, the ac signal modulates the dc, but the the overall current never reverses (as it would for pure ac).