Are your Capacitors Installed Backwards

[Bryan]

Interesting video.
https://youtu.be/BnR_DLd1PDI

[Thor-Arne]

I'm aware of this, but I'm not sure how significant it is.

Wouldn't the manufacturers mark this properly if it was important?

[smjcuk]

That's why I like electrolytics. They're honest. At least if you install them backwards they complain loudly!

[akis]

In addition I read recently that MLCCs lose their capacitance as the voltage increases sometimes to a fraction of nominal.

[dom0]

In addition I read recently that MLCCs lose their capacitance as the voltage increases sometimes to a fraction of nominal.

This is (afaik) a property of all ceramic capacitors, but mainly affects those with high dielectric constant dielectrics.

[Fungus]

I'm aware of this, but I'm not sure how significant it is.

Wouldn't the manufacturers mark this properly if it was important?

I don't think it's as important as the audiophools would have you believe (or capacitors would be marked, as you say).

The signal he's picking up sure looks scary (woooh!) but it's really in the millivolt range and you wouldn't be using big honking power-supply capacitors on things with millivolt signals and high impedance inputs. You'd be using itty-bitty ceramics for that, right?

[paulie]

In addition to concern for installing non-polarized caps backwards also make sure those resistors are not reversed too. Specially important for audio applications use oxygen free copper on the PCB.

[dom0]

Fungus: I didn't watch the video (1 hour!), but what you write sounds like the vid is about the orientation of the outer film/foil layer (which is susceptible to external electric fields). Is this correct?

If so, this effect rarely matters, except for very big film capacitors with extremely low applied levels (think sub-nV²/Hz amplifiers) driven by low impedances. As electric fields are easily shielded mitigation is simple in cases where it matters.

[TimFox]

Years ago, I learned that paper (then) capacitors had the "outside foil" marked for safety purposes.  That lead should be connected to the node with lower DC voltage, since the insulation outside the outside foil was minimal.

[Fungus]

Fungus: I didn't watch the video (1 hour!), but what you write sounds like the vid is about the orientation of the outer film/foil layer (which is susceptible to external electric fields). Is this correct?

Yes.

He's an audio guy and in the video he builds a special device to detect the "correct" capacitor orientation.

He starts out with a really old paper covered capacitor which has "outer film" clearly marked on it then goes on to show that modern film capacitors are misleading because don't have this marking. They obviously need a special device to detect the "correct" orientation, which he goes on to build (I fast-forwarded most of the "build" part of the video).

If so, this effect rarely matters, except for very big film capacitors with extremely low applied levels (think sub-nV²/Hz amplifiers) driven by low impedances. As electric fields are easily shielded mitigation is simple in cases where it matters.

Yep. I'm not an expert but about half way through the video I started to suspect it wouldn't make a real difference in places where that size of capacitor is used in real life (ie. power supplies, etc.)

Still: His device will look good next to the one which detects proper speaker cable orientation.

[Cliff Matthews]

Thanks! I enjoyed your video. I have lots of those orange-dipped caps.

[Kjelt]

Interesting indeed.
Hey is it just me or does it bother others that the guy calls the oscilloscope settings in mHz per division?
I always use time per division not frequency, maybe that was different on the older scopes?

[Len]

Fungus: I didn't watch the video (1 hour!), but what you write sounds like the vid is about the orientation of the outer film/foil layer (which is susceptible to external electric fields). Is this correct?

Yes.

He's an audio guy and in the video he builds a special device to detect the "correct" capacitor orientation.

He starts out with a really old paper covered capacitor which has "outer film" clearly marked on it then goes on to show that modern film capacitors are misleading because don't have this marking. They obviously need a special device to detect the "correct" orientation, which he goes on to build (I fast-forwarded most of the "build" part of the video).

If so, this effect rarely matters, except for very big film capacitors with extremely low applied levels (think sub-nV²/Hz amplifiers) driven by low impedances. As electric fields are easily shielded mitigation is simple in cases where it matters.

Yep. I'm not an expert but about half way through the video I started to suspect it wouldn't make a real difference in places where that size of capacitor is used in real life (ie. power supplies, etc.)

Still: His device will look good next to the one which detects proper speaker cable orientation.

Paul's a "radio guy" more than an "audio guy". The capacitors he's talking about are for repairing old tube radios or amplifiers with paper foil capacitors. (Not modern audiophile equipment!) He’s got a lot of experience with tube RF & audio circuits, so if he says the capacitor orientation makes a difference I'm inclined to believe him.

Also, note that he’s talking about tube amplifier circuits not power supplies. Did you watch any of the video?

[Muxr]

I watched some of his other videos too. He does some great work with his homebrew PCBs. It's a good series on how to apply clean fixes and workarounds to the old gear.

[German_EE]

Capacitors installed 'the wrong way round' can have an effect in high gain audio circuits. Some radio amateurs avoid the traditional superhet receivers and build units called Direct Conversion Receivers where all of the receiver gain is in the audio stages. This can mean up to 40dB of audio gain so hum pickup from various sources can be a major problem. Making sure that the outside layer of a capacitor is wired to the low impedance source helps.

[Fungus]

Also, note that he’s talking about tube amplifier circuits not power supplies. Did you watch any of the video?

Where did I say he's talking about power supplies?

[Thor-Arne]

I'm still not convinced that this is significant, at least in most applications.

Perhaps in some high gain amplifiers, but I suspect that the rest of the circuit (pcb traces, resistors, inductors) will have a larger effect than just having the capacitors "installed backwards".

I have no idea about the effect in tube circuits.

For the older paper-type capacitors in HV applications I can understand it, but that is a insulation issue rather than a noise pickup issue.

Still, if this was significant I'm sure the capacitors would be marked.

[Len]

Where did I say he's talking about power supplies?

You said that's the only place that such capacitors are used "in real life", which is clearly not the case.

[SeanB]

Kind of important in valve circuits, as you have a very close source of hum, the cathode heater. The extra shielding was very useful to reduce this, along with having the heater wires wound as a tight twisted pair right to the base pins.

[Thor-Arne]

As I said, I have no idea about tube (valve) circuits, decided to not mess with it many years ago because of the HV involved.
However, I vaguely remember to have seen caps with a extra metal can around as shielding.

[Kjelt]

As I suspected there is also a capacitance limit to which you can test this.
With a 10n MKP it is visible, but with a 2u2 MKP nothing.

[vk6zgo]

Clever EEs back in the day would orient capacitors so as to use the grounded outer foil as a shield between
input & outputs of high gain valve (vacuum tube) circuits.

If you wired the caps the other way round,the shielding was not as effective,& the stage might oscillate.
This was mainly done with early  MW TRF radios,& was not common with later Superhet designs,except maybe in those with HF RF stages.

Replacing those older caps with modern much smaller caps might give you problems with instability,even if you did mount them with the outer foil connected to ground.

[Pentium100]

I had to replace leaky paper caps in a tape recorder. The capacitors had a metal case that was not connected to either pin of the capacitor, but was used as a shield (with a wire wrapped around it and connected to ground. Naturally, the new caps have a plastic case and the circuit started to oscillate. I had to wrap the new caps in aluminium foil and connect that to ground.

Also, good to know about the caps, now I will make sure to install them properly when I am building a tube amp.

[Dave Turner]

It is specifically mentioned that his interest is in repairing real old school kit where the capacitor orientation was important.

My thoughts are why not just use a double pole double throw toggle switch in a box oriented so that the toggle points to the relevant alligator lead?

[mikerj]

It is specifically mentioned that his interest is in repairing real old school kit where the capacitor orientation was important.

My thoughts are why not just use a double pole double throw toggle switch in a box oriented so that the toggle points to the relevant alligator lead?

I was wondering this as well, a battery driven circuit seems like overkill.