the response of this setup:
(Though first I would put the curves for the Jamicon only, 2 inches away from the pins)
What you can see, that the net impedance, not really ESL, but the resulting equivalent inductance of the original capacitor is significantly increased!
What will happen now with our classical bypass arrangement?
(Though first I would put the curves for the Jamicon only, 2 inches away from the pins)
What you can see, that the net impedance, not really ESL, but the resulting equivalent inductance of the original capacitor is significantly increased!
What will happen now with our classical bypass arrangement?
Attachments
These pics I have made so as to contribute to the topic, have a more clear view about the things what you were talking about.
But what if someone would like to get really nice, smooth phase, low impedance, as much as possible resistive bypass impedance?
Jamicon 10000 uF, 2inch away, 470 uF phil., + 680 nF stacked film with 0,33 ohm in series, locally on pins:
But what if someone would like to get really nice, smooth phase, low impedance, as much as possible resistive bypass impedance?
Jamicon 10000 uF, 2inch away, 470 uF phil., + 680 nF stacked film with 0,33 ohm in series, locally on pins:
Attachments
Pinkmouse, thanks a lot for the fast moderation!
Jorge, I don't think that bypassing in itself would be a dangerous thing. These curves only saying that bypassing with FILM capacitors, because of their high Q & lowish ESR, is dangerous! but, as you see from the last graph, they can be tamed, by adding series resistance.
An other point is: It is not as much the high ESL of big cans that is causing problems, but the parasitic trace inductance! It means, film capacitors are always risky, not only with electhrolytics!
By the way, for the same reason, also OS CON-s can cause problems, although, otherways, they have very good parameters
example:
Jorge, I don't think that bypassing in itself would be a dangerous thing. These curves only saying that bypassing with FILM capacitors, because of their high Q & lowish ESR, is dangerous! but, as you see from the last graph, they can be tamed, by adding series resistance.
An other point is: It is not as much the high ESL of big cans that is causing problems, but the parasitic trace inductance! It means, film capacitors are always risky, not only with electhrolytics!
By the way, for the same reason, also OS CON-s can cause problems, although, otherways, they have very good parameters
example:
Attachments
oops, previous picture is an OS COn, sanyo SH 150 uF, 6.3v
I would like also to add, before going to bed, that unfortunately I do not think that these measurements would shed light on the real nature of Carlos's snubber modification. What I want to say, these are just measurements, I did not try them in the amplifier setup, so I don't know, what sonic differences they cause.
There is one thing, what I have never understood about using big caps - one can play with the bypass setup, and it is even possible to reach quite reasonable results. But what can You do with the enourmously raised level of current peaks, high current sharp impulses crashing into that big cap, radiating EMI all around?
Good night, george
I would like also to add, before going to bed, that unfortunately I do not think that these measurements would shed light on the real nature of Carlos's snubber modification. What I want to say, these are just measurements, I did not try them in the amplifier setup, so I don't know, what sonic differences they cause.
There is one thing, what I have never understood about using big caps - one can play with the bypass setup, and it is even possible to reach quite reasonable results. But what can You do with the enourmously raised level of current peaks, high current sharp impulses crashing into that big cap, radiating EMI all around?
Good night, george
for your measurements. 
Joseph K said:
It wasn't me, another mod got there just before I did! Seriously, though, for posts with high info content like yours, moderation is a formality. We don't do it to annoy new members, just to keep a lid on timewasters.
Anyway, back to your results. I would find it interesting to see such measurements on real amp pcb traces. Do you have anything around that you could just stick on your jig as a typical example?
We don't have a smooth impedance curve as shown but how much does this affect on an amp with a bandwidth of 100 kHz and a power bandwidth of 90 kHz?
These measurements shows also that Carlos' 1 ohms may be too little to eliminate this resonance.
Anyway the topic is interesting but unknown how important it is.
I'll appreciate Joseph's technical attitude to the question and not boasting that he "get's it". Gold star for this.
These measurements shows also that Carlos' 1 ohms may be too little to eliminate this resonance.
Anyway the topic is interesting but unknown how important it is.
I'll appreciate Joseph's technical attitude to the question and not boasting that he "get's it". Gold star for this.
Kuei Yang Wang said:
No Thorsten,
I DONOT believe that someone who hears audible differences between cables is a sucker with wooden ears. My point is, that those cable peddlers etc would like you to believe so, and thus influence you into believeing you hear what they say you should hear, making you feel you're one of those golden ears. Independently whether there IS an audible difference or not.
That is my point, not more, not less.
Kuei Yang Wang said:
Thorsten, I understand you tactic to use these adjectives to put images up into people's mind that have nothing to do with the discussion at hand. You throw terms around like "pseudo" or "religious" etcetera. You're good at it, but it is not my game. I think it is cheap and unfair. So, you'll have to play this one without me, I'm afraid.
Jan Didden
jeff mai said:
You are still involved in the results, but if the test is set up well, you have no way to know which of the tested devices is the one YOU made. So, if the end result is that you can only identify YOUR stuff 50-50, there would be no audible difference. Yes I know there is more to it with significant deviations etc, but as a first approximation, that's the gist of it.
If you can identify YOUR stuff let's say 3 times out of four, without having ANY clue (other than the sound) which set is playing, or without even knowing that the set is changed between changes, if you get me, well, then I would feel strongly that there is a clear audible difference.
On the other hand, if you go to this store for the umpteemth time to drool at these great looking cables, and after a lot of nailgnashing you lay out 2 month's pay to buy them, and you get home, try them and report they sound great, hey, gimme a break!
Jan Didden
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.