Where should I put the capacitor ?
Between the TX and the rectifiers it just loads the TX and does not filter anything to gnd... my TX are not center tapped and not connected to gnd.
Last edited:
Yes, the secondary leads connected directly to the rectifier.
So you mean placing the cap parallel to the secondary windings right ?
Will do
Sorry for not getting to this, the forum just decided to stop sending me post notifications or this thread! So I had no idea anything happened here until I accidentally clicked the tab in my browser. I feel like throttling someone...
If your cap is under 2.2u, an RC may be okay, but I've found it's actually better to use as large a film cap as you can, and omit the resistor. Unless you use a strong carbon comp resistor, it is likely to blow out at some point (the resistor is extremely vulnerable to line transients!). This is fundamentally different from an RC snubber, and is a different approach.
Merlin, R5 and C4 must NOT be changed! C1 can be bypassed if you want. A 100nF cap at the output should be okay.
If your cap is under 2.2u, an RC may be okay, but I've found it's actually better to use as large a film cap as you can, and omit the resistor. Unless you use a strong carbon comp resistor, it is likely to blow out at some point (the resistor is extremely vulnerable to line transients!). This is fundamentally different from an RC snubber, and is a different approach.
Merlin, R5 and C4 must NOT be changed! C1 can be bypassed if you want. A 100nF cap at the output should be okay.
I used 2 u caps without the resistors and the situation got even worst, so I decided to remove the kmult from the psu and place it inside the riaa amp case, just before the regs.
After some judicious grounding mods, I was able to reduce hum to almost nothing... so the kmult is not responsible by the hum.
It does something good to the sound also... Cleans up the highs so things become more "at ease" and lows go really deep down.
Thank you again for your good design.
After some judicious grounding mods, I was able to reduce hum to almost nothing... so the kmult is not responsible by the hum.
It does something good to the sound also... Cleans up the highs so things become more "at ease" and lows go really deep down.
Thank you again for your good design.
Concerning the secondary winding RC, I've found that in some cases an RC sounds better, but these were cases where the circuit would never sound right, there were sibilance issues and the balance was always off. If these are taken care of the plain C always seems to sound better. This is through a few speaker amp prototypes fed from a computer source, so your small-signal circuits may have entirely different considerations.
That is what is supposed to do. I.e. Relaxing the background. Does it also mild out some subjectively put "enthusiasm" in the playback? In the manner some mains filters do, if you get my drift. Or its only a plus across the performance range? When its not used to fight an urging EMI or hum & buzz problem I mean, but only as an add on enhancement.
I just started listening test yesterday, but the results seem comparable to what I found when using it in the riaa psu. The effect is far from what I know EMI filters do to the sound.... with the kmult there is no shrinking of the soundstage.... I feel the sound becomes more "at ease".... detail is there but does not force itself as before.
In this case (headphone amp) the psu is quite basic and uses a low cost low spec smoothing cap, so the effect of the kmult is quite audible.
In this case (headphone amp) the psu is quite basic and uses a low cost low spec smoothing cap, so the effect of the kmult is quite audible.
In the folded, I am using 4 pole smoothing caps and the shunts are quite well tuned so although there is some relaxation of the sound, I slightly loose in soundstage size and "punch factor". I believe that although these v12r shunts have a very good psrr, they can easily reflect the character of the preceding caps and TX and in this case the kmult somehow "erases" the benefits of the Jensen four poles and r-core partners.
I am now confirming my initial thoughts.... We can use the kmult with great results if we use "regular" smoothing caps and tx.
In the headphone amp this is obvious and in the paradise psu, the effects are even more spectacular.
I am now confirming my initial thoughts.... We can use the kmult with great results if we use "regular" smoothing caps and tx.
In the headphone amp this is obvious and in the paradise psu, the effects are even more spectacular.
OK, since builds differ here is what I would propose then. An independent KM made as an in-line add on wherever there is a raw PSU umbilical. In a mini box, metallic and grounded if possible. Using in/out connectors too. That way comparisons are made easier or overall system synergy can be checked and modified in no pain.
Attachments
That is what I did.
That is why I can easily plug it between the several psu's and amps.
Kmult built as proposed in keantoken's site is quite flexible regarding operating voltages.... I always draw more than 300mA through them in my builds (the headphone amp is powered by two BiB's at 300mA)....
Care must be taken while handling it because it can easily be damaged by discharging caps.... I have built 6 kmult's because I already burned two of them
That is why I can easily plug it between the several psu's and amps.
Kmult built as proposed in keantoken's site is quite flexible regarding operating voltages.... I always draw more than 300mA through them in my builds (the headphone amp is powered by two BiB's at 300mA)....
Care must be taken while handling it because it can easily be damaged by discharging caps.... I have built 6 kmult's because I already burned two of them