Conrad,
I have only the listed brands in my original post. I have not had others. I did try various values of each brand with or without bypassing caps (MKP 0.01uF to 0.1uF).
The cap at the adjustment pin is the most sensitive.
The less the resistence added to the regulator output, the less the influence given by the caps but more severe the ringing would be. I don't have a scope to check it.
I think I will probably eventually give it up and build another regulator. My worry is that I tried one shunt regulator and it did not sound as good as the plain LM317/337 in terms of clarity, details and transient response despite of having possibly less cap colourations.
I think the Jung regulator, which is well praised, doesn't like a large or low ESR cap at the output neither, and if not careful, ringing can occur. So how does it perform in terms of transient response? Without a large cap at the output, when a sudden draw of current occurs, the transient response has to suffer.
Perhaps a shunt regulator does not have the same problem and does not rely on a large cap at the output? But how come the shunt reg I had did not sound that good?
Another thing I could do is to use the PSU schematic of a high end CD player I have. It is a series regulator with one MOSFET, two transistors, a zenar, and some resistors and capacitors. An very simple one. Unfortunately, I can't post the schematic here.
Regards,
Bill
I have only the listed brands in my original post. I have not had others. I did try various values of each brand with or without bypassing caps (MKP 0.01uF to 0.1uF).
The cap at the adjustment pin is the most sensitive.
The less the resistence added to the regulator output, the less the influence given by the caps but more severe the ringing would be. I don't have a scope to check it.
I think I will probably eventually give it up and build another regulator. My worry is that I tried one shunt regulator and it did not sound as good as the plain LM317/337 in terms of clarity, details and transient response despite of having possibly less cap colourations.
I think the Jung regulator, which is well praised, doesn't like a large or low ESR cap at the output neither, and if not careful, ringing can occur. So how does it perform in terms of transient response? Without a large cap at the output, when a sudden draw of current occurs, the transient response has to suffer.
Perhaps a shunt regulator does not have the same problem and does not rely on a large cap at the output? But how come the shunt reg I had did not sound that good?
Another thing I could do is to use the PSU schematic of a high end CD player I have. It is a series regulator with one MOSFET, two transistors, a zenar, and some resistors and capacitors. An very simple one. Unfortunately, I can't post the schematic here.
Regards,
Bill
One thing I am yet to do is to use larger MKPs.
I think I will try using a 20uF MKP for bypassing the adj pin (ditch the electrolytic) and another 20uF to shunt the main cap.
I will see if this smooths out the impedance or not.
I think I will try using a 20uF MKP for bypassing the adj pin (ditch the electrolytic) and another 20uF to shunt the main cap.
I will see if this smooths out the impedance or not.
Hi,HiFiNutNut said:
Maybe the shunt was also unstable.
What did you have before the shunt regulator?
and did you adjust the CCS to flow sufficient current, but not excessive, to the load and the shunt?
Last weekend I started trying parallelling different values of caps from Panasonic and Rubycon ZL. This approach looks promising. The ideal is to flatten the impedance curve since various caps have various impedance at various frequencies.
I will order a couple of dozens of caps next weekend to try them out.
I will order a couple of dozens of caps next weekend to try them out.
Peranders,
Interesting thread.
I have long been thinking about buying the Jung reg from you but afraid of having the same issue - cap sound. From my understanding, the supereg does not like large caps at the output neither or it can oscilate.
Here is my story with the Panasonic. Last weekend, I put on a pair of 4,700uF 25V Panasonic FC in parallel with a pair of 2,200uF Rubycon ZL on the +/-15V rails. I was so happy that the sound was much much better - the high frequency hzzzzz was gone. There was detail, clarity, ... That ran for just over an hour. I thought I had finally solved the problem that had not been solved for many weeks' efforts. The rest would be to fine tune the passive XO.
Then I found that the Panasonic were connected in the wrong polarity! They were not exploded! It just shows they are really good quality stuff. So I put them in the right polarity, then the frequency response inregularities came back. So my hard work continues...
Are caps supposed to sound better in the wrong polarity?
Regards,
Bill
Interesting thread.
I have long been thinking about buying the Jung reg from you but afraid of having the same issue - cap sound. From my understanding, the supereg does not like large caps at the output neither or it can oscilate.
Here is my story with the Panasonic. Last weekend, I put on a pair of 4,700uF 25V Panasonic FC in parallel with a pair of 2,200uF Rubycon ZL on the +/-15V rails. I was so happy that the sound was much much better - the high frequency hzzzzz was gone. There was detail, clarity, ... That ran for just over an hour. I thought I had finally solved the problem that had not been solved for many weeks' efforts. The rest would be to fine tune the passive XO.
Then I found that the Panasonic were connected in the wrong polarity! They were not exploded! It just shows they are really good quality stuff. So I put them in the right polarity, then the frequency response inregularities came back. So my hard work continues...
Are caps supposed to sound better in the wrong polarity?
Regards,
Bill
That's bizarre and I don't pretend to understand all that might be going on. However, reversed caps usually leak like a sieve, so maybe your power supply would benefit from more loading. Choose some power resistors to draw maybe 100mA and try loading the supply. Does the sound change? Try the resistors in various physical locations, including exactly where the reversed caps were installed. IMO, cap changes as you've described never result in large changes in sound. Either one is exaggerating the effects, or there is some subtle and undiscovered design problem.
Conrad,
What you said does make sense to me.
The circuit itself already draws 83mA current.
I think caps usually do not sound as different as what I described - it may well be due to the system. My system is very revealing - I have recently got the guys who run the Sydney Audio Club to come and listen to my system, and they all said it is one of the best systems they have ever auditioned. They have been audiophiles in the audio games for dozens of years. Perhaps it is for this reason, cap comparison on this sytem make more sense.
Really, the reversed caps sounded better possibly because of the lack of the higher frequency exaggeration.
Regards,
Bill
What you said does make sense to me.
The circuit itself already draws 83mA current.
I think caps usually do not sound as different as what I described - it may well be due to the system. My system is very revealing - I have recently got the guys who run the Sydney Audio Club to come and listen to my system, and they all said it is one of the best systems they have ever auditioned. They have been audiophiles in the audio games for dozens of years. Perhaps it is for this reason, cap comparison on this sytem make more sense.
Really, the reversed caps sounded better possibly because of the lack of the higher frequency exaggeration.
Regards,
Bill
What is the purpose of a super regulator?HiFiNutNut said:
Using a super regulator is a super large cap made by electronics so why do you want to have large caps also?
How high voltage rating did the cap have and how much reverse voltage did you apply?HiFiNutNut said:
Peranders,
25V rating on +/-15V rails.
Yes a supereg is a large cap. However, you still need some small caps here and there. I think they will give the colourations.
I have no question whatsoever that the supereg design is great. I only have doubts on capacitors.
Regards,
Bill
25V rating on +/-15V rails.
Yes a supereg is a large cap. However, you still need some small caps here and there. I think they will give the colourations.
I have no question whatsoever that the supereg design is great. I only have doubts on capacitors.
Regards,
Bill
Comparing modern electrolytic capacitors
"Sometimes you worry too much. Most colouration will your speakers contribute with."
There is nothing wrong with trying to obtain the most neutral response from your system before the loudspeakers.
The majority of preamplifier and amplifier designs seem to suffer from mid range colouration due to the inability of normal bypass capacitors to have a very low impedance in this critical area of the spectrum. FWIW, I also disagree with Douglas Self's normal use of a 10 ohm resistor and a 1,000uF capacitor inserted in the -VE rail of his amplifier's supply to the front end in order to improve PSRR, for the same reason. It is far better to supply the front end from additional voltage regulators or gyrators.
SandyK
"Sometimes you worry too much. Most colouration will your speakers contribute with."
There is nothing wrong with trying to obtain the most neutral response from your system before the loudspeakers.
The majority of preamplifier and amplifier designs seem to suffer from mid range colouration due to the inability of normal bypass capacitors to have a very low impedance in this critical area of the spectrum. FWIW, I also disagree with Douglas Self's normal use of a 10 ohm resistor and a 1,000uF capacitor inserted in the -VE rail of his amplifier's supply to the front end in order to improve PSRR, for the same reason. It is far better to supply the front end from additional voltage regulators or gyrators.
SandyK
sandyK said:
I've also used this circuit in various applications and can say it works very well indeed.
Its not expensive to try, only downside is it takes a little room so if space is tight it may cause a problem.
I used it for op-amp and discrete analogue stages
Conrad,
I don't know why do you say that?
Look at this graph, it's not a quote, I can repeat it any time [the 1ohm ref line is at -28dB, the bottom of the curve at 100kHz is -57db for the simpler cap, that is 35mohm at 100kHz, and will rise to 1 ohm at 11MHz!!! all this at 50V rated voltage.
Other than 1ohm at 3kHz..
Ciao, George
http://www.diyhifi.org/forums/viewtopic.php?p=14283#p14283
I don't know why do you say that?
Look at this graph, it's not a quote, I can repeat it any time [the 1ohm ref line is at -28dB, the bottom of the curve at 100kHz is -57db for the simpler cap, that is 35mohm at 100kHz, and will rise to 1 ohm at 11MHz!!! all this at 50V rated voltage.
Other than 1ohm at 3kHz..
Ciao, George
http://www.diyhifi.org/forums/viewtopic.php?p=14283#p14283
Oops, forgot that maybe one should registrate to see the pics.. Though it definitely worth the pain..🙂
Here it is, a different meas., same cap, this time from this same forum:
http://www.diyaudio.com/forums/showthread.php?postid=549338#post549338
or the pic re-attached:
Here it is, a different meas., same cap, this time from this same forum:
http://www.diyaudio.com/forums/showthread.php?postid=549338#post549338
or the pic re-attached:
Attachments
Anxiously awaiting my return email from the registration, but the pic helps until then. Can you give me the details of the instrument and set-up for the test?
Here's a question- if the HF performance of electrolytics is so good, why can't you bypass a high speed op-amp with a pair of big electrolytics (even Oscons) and get stability? Or, maybe *you* can, but it certainly didn't work for me!
Here's a question- if the HF performance of electrolytics is so good, why can't you bypass a high speed op-amp with a pair of big electrolytics (even Oscons) and get stability? Or, maybe *you* can, but it certainly didn't work for me!
- Status
- Not open for further replies.