Capacitor distortion

[jasonhanjk]

I often see electrolytics used in the feedback path of a power amp (to ground) which also is poor practice. Also sometimes see electrolytics as an input capacitor on amps without any bias voltage. Poor engineering.

Thanks dude. I didn't know electrolytic need to have some dc bias.
I have a circuit bias at 0V, will use non polar ecap from today onwards.

 

[Bob from Florida]

New deflection. Because you mentioned it, respect must be earned. So far, you failed miserably.

Whatever.
Not like anything you or I say about this subject will make anyones life better.

 

[solderdude]

Thanks dude. I didn't know electrolytic need to have some dc bias.
I have a circuit bias at 0V, will use non polar ecap from today onwards.

Yep, and the DC bias also has to respect the + and -. Especially with tantalum. These little buggers can't stand reverse voltages.

Indeed if one has to use electrolytics as coupling caps and there is no voltage difference between either side and you need a high value capacitor and small size then you can still use electrolytics but must use 2 of them (so you need caps with double the intended value) in anti-series.
So like this:

Does not matter if the + are tied together or the - of both caps.
I would not recommend this with tantalums.

When there is very AC little voltage and you need a high capacitance in a small space you can also but 2 capacitors (with a high as possible voltage rating) in anti-parallel.
In such case the total capacitance doubles. Could be used for line-level audio signals while using 25V to 63V caps.

Would also be fun for @pma to measure this (along side with anti-series) to see how good this works. Should have lower distortion.

 

[restorer-john]

Would also be fun for @pma to measure this (along side with anti-series) to see how good this works. Should have lower distortion.

I'm gonna say the opposite- more distortion. You have twice the ESR, (2 inductors, 2 resistors and 2 capacitors) all in series. And twice the capacitance value for each cap.

As long as PMA measures the same types and equal overall capacitance this time- I'm interested to see what he comes up with too.

 

[solderdude]

ESR is not important with input coupling caps or output coupling caps of line-level signals.
It is for decoupling caps, especially in SMPS. Paralleling (or compounding) helps greatly.

And yes, the same value should be compared.
In anti-parallel 2x 47uF = almost 100uF and in anti-series 2 x 220uF is slightly higher than 100uF.

@pma might be in for these type of experiments. I am too lazy for that.

 

[somebodyelse]

I'm gonna say the opposite- more distortion. You have twice the ESR, (2 inductors, 2 resistors and 2 capacitors) all in series. And twice the capacitance value for each cap.

As long as PMA measures the same types and equal overall capacitance this time- I'm interested to see what he comes up with too.

IIRC Bateman's measurements showed reduced distortion from back to back electrolytics.

 

[restorer-john]

ESR is not important with input coupling caps or output coupling caps of line-level signals.

ESR is absolutely important. A rising ESR is probably the best indicator of a failing capacitor, regardless of the part of the circuit it is in. Rising or high ESR is usually coupled with the capacitance value dropping and in the case of audio, mismatches between channels due to RC differences. I use an ESR meter all the time for determining problems. That and a thermal camera for SMPS caps will show you more capacitor issues than any amount of visual/removal and testing.

 

[pma]

Effects of el. capacitor load on distortion

In my posts #1 and #14 the 10uF/100V and 4.7uF/6.3V capacitors were loaded with input impedance of the E1DA Cosmos ADC, which is approx. 830 ohm at the input divider setting used. This input impedance results in some LF roll-off, which makes -0.4dB for the 10uF cap and -1.3dB for the 4.7uF cap, both values taken from real life measurement. The measured plots are shown below, in a comparison with the loopback response.

This roll-off would not have much effect on audibility, but has an effect on capacitor distortion by increasing ac voltage across the capacitor.

To verify if there is any capacitor distortion if the capacitors are loaded with high impedance, thus no roll-off in audio band is added, my Audio Buffer was inserted into the measuring loop, behind the capacitor, ensuring that load impedance is now 100 kohm. My Audio Buffer is described here:

Audio Buffer in english

Measurements into 100 kohm load

The same set of measurements as in post #14 was performed, using the Audio Buffer in the loop, as a capacitor load. Same capacitors were used and measurements with them compared to the loop with Audio Buffer (without capacitors).

We can say that:

- overall distortion has increased a bit due to the Audio Buffer inserted into the loop
- all 3 measurements overlap and we can see that the capacitors now do not add any distortion
- low frequency distortion of electrolytic capacitors has disappeared by using high impedance load (100k input of the Audio Buffer) and the LF distortion is now lower than the distortion with capacitors loaded directly by E1DA input
- there is still some slight rise in distortion in all 3 plots below 70Hz, most probably due to input electrolytic capacitors in E1DA, that are loaded with quite low impedance of the input opamp of the E1DA ADC. This might be a space for further improvement of E1DA
- my measurements also demonstrate why some of us (me, @restorer-john ) complain on too low input impedance of SOTA ADC and preamps

Edit (Jan 13, 2023): as a comparison, I am adding distortion plot of 100nF X7R capacitor loaded with 100 kohm, measured under same conditions as the graph above

 

[solderdude]

A rising ESR is probably the best indicator of a failing capacitor

Of course it is I did not claim otherwise. A rising ESR is not an issue for coupling caps in line inputs and outputs. Loss of capacitance can be if the value was chosen a bit 'tight' anyway. Output coupling caps (speaker) is a different matter. Decoupling... for sure.

I use an ESR meter all the time for determining problems. That and a thermal camera for SMPS caps will show you more capacitor issues than any amount of visual/removal and testing.

Yep.

- low frequency distortion of electrolytic capacitors has disappeared by using high impedance load (100k input of the Audio Buffer) and the LF distortion is now lower than the distortion with capacitors loaded directly by E1DA input

Of course it is as with coupling caps it is all about the AC current passing through them. Lower the current and performance increases.

- there is still some slight rise in distortion in all 3 plots below 70Hz, most probably due to input electrolytic capacitors in E1DA, that are loaded with quite low impedance of the input opamp of the E1DA ADC. This might be a space for further improvement of E1DA

Good catch.
He is probably forced to use electrolytics there due to the low input imp when AC coupling is essential.
There appear to be differences in that area between the various capacitor lines and manufacturers.

- my measurements also demonstrate why some of us (me, @restorer-john ) complain on too low input impedance of SOTA ADC and preamps

Also not all sources like to deliver current and when output capacitors are being used and these are not designed to drive 600ohm or 1k or even 2k loads roll-off can be an issue.
Most (certainly not all) modern DACs all use buffer amps, often with additional output resistors, that have no issues driving 1k loads. At most at the expensive of a slight output voltage loss. When mixing older gear or 'audiophile' gear that is not buffered one can run into issues.

For the (rather pointless) SINAD race low input resistance is essential.

 

[restorer-john]

Of course it is I did not claim otherwise. A rising ESR is not an issue for coupling caps in line inputs and outputs. Loss of capacitance can be if the value was chosen a bit 'tight' anyway. Output coupling caps (speaker) is a different matter. Decoupling... for sure.



Yep.



Of course it is as with coupling caps it is all about the AC current passing through them. Lower the current and performance increases.



Good catch.
He is probably forced to use electrolytics there due to the low input imp when AC coupling is essential.
There appear to be differences in that area between the various capacitor lines and manufacturers.



Also not all sources like to deliver current and when output capacitors are being used and these are not designed to drive 600ohm or 1k or even 2k loads roll-off can be an issue.
Most (certainly not all) modern DACs all use buffer amps, often with additional output resistors, that have no issues driving 1k loads. At most at the expensive of a slight output voltage loss. When mixing older gear or 'audiophile' gear that is not buffered one can run into issues.

For the (rather pointless) SINAD race low input resistance is essential.

How about just stop vainly attempting to be the smartest guy in the room and settle for being one of the smartest? It's not that hard.

It gets so boring watching you attempt to dig yourself out, over and over again. You get called, you attempt to deflect, and make some point nobody cares about. Come on, you can do a whole lot better.

 

[pma]

 

[BR52]

There is a simple rule of thumb, don't use Electrolytic's in the signal chain except you have high enough DC-Bias. (Tube-amp). Bipolar elco's are always build like here #43, with the penalty of no DC bias high ESR..., but they are cheap.

 

[solderdude]

How about just stop vainly attempting to be the smartest guy in the room and settle for being one of the smartest? It's not that hard.

It gets so boring watching you attempt to dig yourself out, over and over again. You get called, you attempt to deflect, and make some point nobody cares about. Come on, you can do a whole lot better.

Feeling a bit grumpy today ?

 

[pma]

There is a simple rule of thumb, don't use Electrolytic's in the signal chain except you have high enough DC-Bias. (Tube-amp). Bipolar elco's are always build like here #43, with the penalty of no DC bias high ESR..., but they are cheap.

You know, the main reason why I started this thread was the comment in post #39 in another thread:

The new DIY preamp designed during pandemic

The new DIY preamp designed during pandemic ... IC7, IC8, IC102 and IC103 are the output stage opamps. All the opamps used have very high BW and slew rate, that is 125V/us and higher. The slowest parts are the instrumentation amplifiers at balanced inputs wit slew rate of 15V/us, which is, to my...

www.audiosciencereview.com

 

[BR52]

You know, the main reason why I started this thread was the comment in post #39 in another thread:

The new DIY preamp designed during pandemic

The new DIY preamp designed during pandemic ... IC7, IC8, IC102 and IC103 are the output stage opamps. All the opamps used have very high BW and slew rate, that is 125V/us and higher. The slowest parts are the instrumentation amplifiers at balanced inputs wit slew rate of 15V/us, which is, to my...

www.audiosciencereview.com

View attachment 256840

OK now I understand. Was not really thinking you use it in the signal chain.

 

[fpitas]

It's good to discuss these things, and present real measurements. But for some reason in audiophile circles, blaming capacitors has become the equivalent of blaming the dog. It would be less distressing if any of them could design their way out of a wet paper bag.

 

[BR52]

It's good to discuss these things, and present real measurements. But for some reason in audiophile circles, blaming capacitors has become the equivalent of blaming the dog. It would be less distressing if any of them could design their way out of a wet paper bag.

It's easier to talk about things you don't really know and scratch on the surface (most audio circles). If you dig deeper, you have to spend a lot of time for it and your answers are mostly too complex.

 

[syn08]

You know, the main reason why I started this thread was the comment in post #39 in another thread:

The new DIY preamp designed during pandemic

The new DIY preamp designed during pandemic ... IC7, IC8, IC102 and IC103 are the output stage opamps. All the opamps used have very high BW and slew rate, that is 125V/us and higher. The slowest parts are the instrumentation amplifiers at balanced inputs wit slew rate of 15V/us, which is, to my...

www.audiosciencereview.com

View attachment 256840

…of which you still did not answer the main question. All you have shown is that bad electrolytics do exist, otherwise your sweeping statement “I would never use electrolytics…” is nothing but FUD. And thats before going into things like threshold of audibility.

 

[fpitas]

It's easier to talk about things you don't really know and scratch on the surface (most audio circles). If you dig deeper, you have to spend a lot of time for it and your answers are mostly too complex.

Right? And actually learning to design stuff is work!!!!

 

[syn08]

How about just stop vainly attempting to be the smartest guy in the room and settle for being one of the smartest? It's not that hard.

It gets so boring watching you attempt to dig yourself out, over and over again. You get called, you attempt to deflect, and make some point nobody cares about. Come on, you can do a whole lot better.

Come on boys. Let's get back to the topic at hand.