[Daryl]

Quote from Steve at SAS Audio Labs

Pretty amazing spec for a non-polar electrolytic cap. 

At the link, examine figure 7, and notice the very very poor specs of most electrolytic caps, different brands. All are showing their inductance under 10khz, and some, for amplifiers, below 1khz. One shows effects at around 300hz. Is yours a special cap(?), and what brand.

http://waltjung.org/PDFs/Picking_Capacitors_1.pdf

Cheers.

Quote from Daryl

Hi Steve,

It's just an inexpensive Bennic from Madisound and over 15 years old.

I bought a mouse today for my measurement computer and I will fire it up when I get the time.

The capacitors in that article were power supply caps and not non-polars.

I will measure some power supply capacitors when I can and start a new thread.

I would not recommend that article at all.

Another quote from Steve.

Any ideas why it measures radically different than the other, well known brands?

Bipolars are essentially two electrolytics back to back, so I fail to see your point. The DA should be much higher with a NP than polar.

Are you kidding? Wow. Tell that to Dr. Karl Weber, a chemist, who discusses the subject and what happens at the molecular level. His explanation almost, if not perfectly, correlates to the sound we appear to be contributing to the insulation material in a capacitor.
http://www.jacmusic.com/auricap/htm/on_capacitor_dielectric_material.htm

The RCA Radiotron designers handbook (26 engineers) also has a nice description of different capacitors, and their problems and superiority. And it was written some 40-60 years ago. So the knowledge is not new.

[Daryl]

Hi again Steve,

Like you I would think that a power supply type capacitor would be designed for the lowest possible impedance over the widest bandwidth possible and not being burdened with the need for non-polar status would make the task easier.

They don't give a lot of details about the Bennics and I have always assumed they were nothing speacial.

To me the measurements in Fig. 7 (Picking Capacitors linked above) looked like poor performance.

The only clue I could find was that all of the capacitors were standard electrolytics and not the non-polar type you would use in a speaker crossover and I have not tried to measure that type but like you thought they were better.

So I thought why not dig through my junkbox and see what I've got to put on the grill.

You never use polarized capacitors reverse biased (they will blow up in your face) so the polarized types will need to be biased.

I decided to use them in identical pairs connected in series back to back which will self bias themselves.

I also decided to measure the non-polars (NPE's) in identical series pairs as well since I happened to have two of each in my junkbox.

This will double the impedance that I measure and give me greater separation from the systems residual noise and give me an average of two samples of each type.

The results will be scaled to reflect the impedance of a single capacitor.

I found three indentical pairs of standard electrolytics...

47uf 25v Nc radial lead
470uf 25v Nc radial lead
4700uf 35v Nichicon axial lead

The 47uf and 470uf appeared to be marked Nc.

Not shure because the diagonal slash of the 'N' was round instead of straight but I think it was Nc and I think that might be short for Nichicon (I think I bought them all from Rippoffshack at some point).

I also found two identical pairs of NPE's...

50uf 100v Bennic axial lead
200uf 100v Bennic axial lead

Heres the measurement results...


The standard electrolytics measure similar to what is portrayed in fig. 7 of the Picking Capacitors article.

And HOLY COW the Bennic NPE's blow them out of the water!

You can see it best on the 47uf and 50uf units since they are about the same value.

I wonder if other NPE brands measure similar to the Bennics.

If anyone would like to send me a cap or two to measure in an overstuffed envelope contact me.

[Steve]

Hi Daryl,

     The bennics appear to be somewhat different than the panasonic and mallory electrolytic caps. It appears to me that the inductance is rearing its head around 50- 60khz for the 47uf bennic and around 10khz for the 200uf.

I also wonder if your gadget is of the same quality as a professional instrument?

Of course voltage rating and dimensions, internal connections make a difference. I wonder if they are high frequency types? Any chance you can test those?

Thanks Daryl.
Steve

[art]

Anyone wanting caps to be measured for ESR, SRF, and all that can contact me. I can measure out to 200 MHz with a VNA.