Member Christo did a series of measurements on Zeners and LED's a few years back if you search the forum you should find the results. Easy to get into Excel and plot some graphs.
I was very surprised at how quiet the Zeners were as well - an LM317 without sizeable Vadj bypassing is -2 orders of magnitude noise ire in some cases.
Weed and electronics
It's awfully hard to hold a disagreement here. Normally the procedure is like partial differentials, you look at a single factor, with all other variables held constant. Here we have this kind of river-rapids of ideas, where it's barely worth disagreeing, it's just chips in the stream.
The thread on 'weed and music' is much more comfortable to be in. Oddly, there seems to be less confusion about what is real and what is not than in some other parts of the forum.
Maybe a thread on weed and electronics?
It's awfully hard to hold a disagreement here. Normally the procedure is like partial differentials, you look at a single factor, with all other variables held constant. Here we have this kind of river-rapids of ideas, where it's barely worth disagreeing, it's just chips in the stream.
The thread on 'weed and music' is much more comfortable to be in. Oddly, there seems to be less confusion about what is real and what is not than in some other parts of the forum.
Maybe a thread on weed and electronics?
Just for fun, I compared two different connections of polarized caps, the back-to-back series caps and paralleled caps (parallel vs antiparallel). I was curious because one of the posted articles claimed that the antiparallel caps give lower distortion than series back-to-back. I used some '60s vintage Sprague 150D tantalums, 1u/35VDC. The load was a 3k3 resistor, and for each connection, I set the excitation signal to the corner frequency in order to see distortion behavior and to compare apples to apples. As a check, at midband, the distortion was about the same as residual. The caps weren't matched and appear to have come from two different manufacturing lots. 3VAC excitation.
Series back-to-back: -80dB 2nd, -100dB 3rd, -120dB 4th
Parallel: -72 dB 2nd, -88dB 3rd, -105dB 4th
Antiparallel: -73dB 2nd; -88dB 3rd, -108dB 4th
These data contradict the claim in that paper.
Series back-to-back: -80dB 2nd, -100dB 3rd, -120dB 4th
Parallel: -72 dB 2nd, -88dB 3rd, -105dB 4th
Antiparallel: -73dB 2nd; -88dB 3rd, -108dB 4th
These data contradict the claim in that paper.
Testing capacitor distortion in a test jig is one thing, testing in a real world circuit stage is another.
Source and load impedances over frequency are typically not perfectly constant.
More useful would be to test distortions of typical stages with variety of capacitor types.
THD, IMD and multitone IMD would be more appropriate testing methodologies.
Dan.
Source and load impedances over frequency are typically not perfectly constant.
More useful would be to test distortions of typical stages with variety of capacitor types.
THD, IMD and multitone IMD would be more appropriate testing methodologies.
Dan.
I just repeated my 32 year old test with the same tant cap values and the same test conditions. Inverse parallel was about 15dB BETTER than the same parts in series. 20db better compared with a single 4.7uf tantalum cap. Perhaps my test conditions encouraged this result, but I don't see how. What SY measured, I cannot answer for.
Yes, I find this thread exhausting. I feel guilty for reading it so much, surely there are better uses of my time? But I still can't keep up, I have to skip over blocks of pages of posts every now and then. Where do these guys find the time? I am doubly impressed by those who actually make measurements and post them, I haven't got enough hours in a day.
The funny part is that my impression is that most of the "regular" contributors to this thread hold each other in very high regard, but snipe at each other for sport. It is easy to get drawn into that dynamic if one hangs around here too much.
Weed and electronics? Sadly, I consumed my earthly allotment of weed many years ago (tobacco too). Now I'm working on my whisky ration.
I don't feel that weed has anything left to teach me, perhaps I'm wrong about that. For me it never mixed with logical thinking or technical matters, but I have known some very good techs who worked stoned all the time. I once went back to work after a dinner which included some aromatic after-dinner treats, and was working on a tape deck tune-up, which should have taken about half an hour, maybe 45 minutes, but an hour later I was still doing the first steps for the 5th time. 300Hz test tones get boring after a while.
Sorry my mistake. Should have put question mark in my post. I thought it was relevant to Dielectric Absorption. I was searching google if there is any electron microscopy of capacitor's working. Just simple curiosity.
Regards
In case it seems like the now very familiar game here of someone is right and someone is wrong game -- or someone is smarter and someone is not so ego game.......Others can see the whole context of the 'non-linear' part of caps as THD "Picking Capacitors part 1 Audio Feb 1980.
http://waltjung.org/PDFs/Picking_Capacitors_1.pdf
and the rest here: http://www.elpee.info/Documenten/PickingCapacitors2.pdf
-RNM
Last edited:
Zeners are quite noisy at 6.2 volts, and the reason for a buried zener is for bette long term drift, 6.2 happens to be close to zero tempco, that is why the usual voltage for high stabiity is aound the 6.2 region.
Walt Jung is no slouch, 5 years after my tests, Walt and JC came up with a completely different way to test caps.
Read his conclusion about this test he designed with JC about capacitors and DA.
"The residual from the units is dominated mostly by DA errors."
Read the whole context here:
http://www.waltjung.org/PDFs/A_RealTime_Signal_Test_For_Capacitor_Quality.pdf
THx-RNMarsh
Read his conclusion about this test he designed with JC about capacitors and DA.
"The residual from the units is dominated mostly by DA errors."
Read the whole context here:
http://www.waltjung.org/PDFs/A_RealTime_Signal_Test_For_Capacitor_Quality.pdf
THx-RNMarsh
Last edited:
- Status
- Not open for further replies.