john curl said:
I've modded out a few of your designs, John..and even then I've found that doing work on the PS for the servo improves sonics. Interestingly enough. Of course, the levels of 'change' noted are not even remotely economically feasible (as a return for $ in the manufacturing chain) until the near the 'no compromise' execution/price level. Since it's a servo, the short cut is one that does not work for 'audio' signal circuits...which is the subbing of standard elect. caps in the servo PS with OS-Con types. Oscons are cheap these days, due to the extreme use in high end computer gear. Extreme Gaming video cards and motherboards are now so common that they've driven the price of OS-Con caps into the ground. So use them in the servos where they won't be heard - But their effect will be.
Thank you Mr. Curl, for your enlightment 😀
Mr. Curl, your name is mentioned in this article
http://waltjung.org/PDFs/Picking_Capacitors_1.pdf
http://waltjung.org/PDFs/Picking_Capacitors_2.pdf
How good (or how bad?) the performance of series capacitor which center is biased (fig D, E, F)?
Mr. Curl, your name is mentioned in this article
http://waltjung.org/PDFs/Picking_Capacitors_1.pdf
http://waltjung.org/PDFs/Picking_Capacitors_2.pdf
How good (or how bad?) the performance of series capacitor which center is biased (fig D, E, F)?
Attachments
john curl said:
Hi,
Cyril Bateman in his series on capacitors in Electronics World was in disagreement with Jung's tests. Available here :
http://uk.geocities.com/cyrilb2@btinternet.com/downloads_4.html
john curl said:
Hello John
Have you ever tried copper foil Polyphenylene Sulphide capacitors in the servo (Panasonic make such a part)
Regards
Arthur
Bateman did NOT understand the differential null test. He could not understand WHY we could not get a deep null. He criticized it without consulting us. He wanted to reinvent cap testing, and he left us out. Typical, but in poor taste.
PPS caps look very interesting. I have measured samples and found them interesting, and low in non-linear distortion. Did not test for DA however.
Have you explored the low end?
cutoff of 1hz vs. 0.1hz vs. 0.01hz?
or just assumed you can't hear that low so it doesn't matter?
cutoff of 1hz vs. 0.1hz vs. 0.01hz?
or just assumed you can't hear that low so it doesn't matter?
With DA, extending the bandwidth reduces, but does not eliminate the problem. Personally, I think that almost any -6dB/octave reduction in bandwidth below 5Hz is probably OK, IF the roll-off mechanism is about perfect. However, servos should operate well below this frequency, because they are NOT perfect, and with caps, most people can't afford the best ones, either.
However, IF you are willing to pay 30 Euros/ cap, or more, then maybe a 5 Hz bandwidth might be OK.
However, IF you are willing to pay 30 Euros/ cap, or more, then maybe a 5 Hz bandwidth might be OK.
I believe that DA itself it is lesser evil than a non-linear DA, that is the lesser evil than non-linearities of semiconductor devices. However, combination of DA and non-linear transistor input impedance can cause more complex non-linearities than in case of a cap with less DA in the same topology.
We don't use mylar anymore, because there is metalized polypropylene, which measures much better for DA, as well as polystyrene and Teflon for the best projects.
I heard the problem with mylar, 30 years ago, when I designed it into one of my Symmetry electronic crossovers. I really thought that stacked foil mylar was a great solution at the time. Unfortunately, we could hear it in the circuit. After that, it was not long before metalized polypropylene became available, and of course, servos became popular.
IF my associates and I had not HEARD the sound of the mylar coupling cap in the circuit, I might be still using it today. I don't think much of electrolytic coupling caps either for quality audio equipment, but I realize that it might be necessary in auto amps, and other battery powered equipment. I would try to use the best cap that I could find, IF I had to design something that required an electrolytic coupling cap.
However, PPS caps might be practical, I don't know for sure if they are better than mylar, but they could well be.
What I am concerned with here, is the dilution of design quality by lowering standards, just because it is convenient for some.
I heard the problem with mylar, 30 years ago, when I designed it into one of my Symmetry electronic crossovers. I really thought that stacked foil mylar was a great solution at the time. Unfortunately, we could hear it in the circuit. After that, it was not long before metalized polypropylene became available, and of course, servos became popular.
IF my associates and I had not HEARD the sound of the mylar coupling cap in the circuit, I might be still using it today. I don't think much of electrolytic coupling caps either for quality audio equipment, but I realize that it might be necessary in auto amps, and other battery powered equipment. I would try to use the best cap that I could find, IF I had to design something that required an electrolytic coupling cap.
However, PPS caps might be practical, I don't know for sure if they are better than mylar, but they could well be.
What I am concerned with here, is the dilution of design quality by lowering standards, just because it is convenient for some.
john curl said:
John,
It looks like everything you discovered/heard/decided/etc... happened 30 to 40 years ago? I have two questions for you:
1. Is there anything new that you figured out since, say, 1990?
2. Did you, or do you have any plans to, double check your old results and findings against new devices (trannies, opamps, capacitors, resistors, etc...) built in the last, say, 10 years?
PMA said:
Looks very good. Similar with what I got in YAP for +/-40V swing into 4ohm (without Zoebel, output coil and input RF filter).
- Status
- Not open for further replies.