Grey,
I mentioned the active loads because I have visions of converting it into an x-circuit..........🙂
As for the comp differentials acting as current sources I will wait for Mr. Curl to answer, as he is more learnad than I.
As to the Dark Side, they are stronger than you think.............
I mentioned the active loads because I have visions of converting it into an x-circuit..........🙂
As for the comp differentials acting as current sources I will wait for Mr. Curl to answer, as he is more learnad than I.
As to the Dark Side, they are stronger than you think.............
Attachments
Hi Uli,
Yes we did..............I hope to be be starting the thread up again soon.
Regards,
Jam
Yes we did..............I hope to be be starting the thread up again soon.
Regards,
Jam
...sigh...
It seems to me that Jam, et. al. are trolling for members for another site. The link given for the "X" circuit a few posts above leads elsewhere.
Jam/uli,
Being charitable, I'm willing to believe for the moment that you forgot for the moment where you were. But given the content and context of certain posts made elsewhere, the cynical side of me has to wonder where your head(s) is/are at...
Trust me, I, more than most people, have reason to question some of the things that have happened at this site, but I don't think this is an honorable way to behave, even if you have reservations about DIY.
Post the schematic here. Start a thread. Do this the right way.
Don't give into the Dark Side, Jam...there is good in you...I can feel it...
Grey
It seems to me that Jam, et. al. are trolling for members for another site. The link given for the "X" circuit a few posts above leads elsewhere.
Jam/uli,
Being charitable, I'm willing to believe for the moment that you forgot for the moment where you were. But given the content and context of certain posts made elsewhere, the cynical side of me has to wonder where your head(s) is/are at...
Trust me, I, more than most people, have reason to question some of the things that have happened at this site, but I don't think this is an honorable way to behave, even if you have reservations about DIY.
Post the schematic here. Start a thread. Do this the right way.
Don't give into the Dark Side, Jam...there is good in you...I can feel it...
Grey
Folks, back to business. First, it would be great to use a really good current source, instead of a resistor. However, I don't know how I can easily do it. Also, "Which comes first, the chicken or the egg?" In other words, if I fix the power supply, then the preamp can have a certain limited amount of 'power supply rejection' and still be useful. Now, IF I could make a more ideal current source (4), then that would be a useful addition.
Folks, I might also point out that just making a superficially similar circuit, does not make it up-to-date with all that we can do to make a better audio circuit.
Better power supply
John,
I am a power engineer by degree, also zero work experiance in that area. Started with a slide rule clipped to my belt, saw the first HP calculators. Tells you how complicated the punch card programs modeled in college were.
For a class A circuit, are there benefits of using a real well designed active supply over a simple 3 pin reg with a bunch of C-L-C filtering on the outputs?
I only ask this because the sledgehammer technique has always been my answer to power supply issues. Big iron, and 4 pole caps after a simple actice reg. High impedance, with almost zero noise.
Planning on building some real discrete regs soon. Hate to put a couple weeks into copying the Borbely regs to find them no improvement over my current setup.
George
John,
I am a power engineer by degree, also zero work experiance in that area. Started with a slide rule clipped to my belt, saw the first HP calculators. Tells you how complicated the punch card programs modeled in college were.
For a class A circuit, are there benefits of using a real well designed active supply over a simple 3 pin reg with a bunch of C-L-C filtering on the outputs?
I only ask this because the sledgehammer technique has always been my answer to power supply issues. Big iron, and 4 pole caps after a simple actice reg. High impedance, with almost zero noise.
Planning on building some real discrete regs soon. Hate to put a couple weeks into copying the Borbely regs to find them no improvement over my current setup.
George
I am willing to entertain the idea of a better current source, but for all the scads of threads here about current sources, they are nothing but regurgitated ideas at least a couple of decades old. Okay, so you use an LED as the reference...that's not new, not by at least twenty or twenty-five years (minimum) in spite of claims to the contrary. Use of JFETs (for those who think in terms of voltage noise) or bipolars (for those who think in terms of current noise)...I am not exactly falling off my chair. Use of a current source to drive the reference...snore.
Given current technology and available parts, it's not likely that we're going to get any closer to an ideal current source.
With that in mind, John accepts the lower CMRR that goes with the use of low resistance resistors and makes up for it in power supply design. I don't have a problem with that.
To the extent that I am guilty of opening this particular can of worms, I apologize. I thought I knew where John was coming from, but wanted to verify my understanding of the circuit. My thanks to John for verifying my hypothesis.
Grey
Given current technology and available parts, it's not likely that we're going to get any closer to an ideal current source.
With that in mind, John accepts the lower CMRR that goes with the use of low resistance resistors and makes up for it in power supply design. I don't have a problem with that.
To the extent that I am guilty of opening this particular can of worms, I apologize. I thought I knew where John was coming from, but wanted to verify my understanding of the circuit. My thanks to John for verifying my hypothesis.
Grey
Hi Grey,
Thanks for asking the question. Current sources are not new, but the knowledge of their short comings or strengths are more widely understood lately (by more people in general). What John and yourself have actually done is to illustrate the trade offs that engineers make. I think that is valuable to point out.
These discussions help me and I'm sure many others understand the concepts more fully. It may even break the "one tool for all jobs" thing that I see many design with.
-Chris
Thanks for asking the question. Current sources are not new, but the knowledge of their short comings or strengths are more widely understood lately (by more people in general). What John and yourself have actually done is to illustrate the trade offs that engineers make. I think that is valuable to point out.
These discussions help me and I'm sure many others understand the concepts more fully. It may even break the "one tool for all jobs" thing that I see many design with.
-Chris
Each time I have used current sources in various circuits, from preamps to buffers or power amps, to get better PSRR, I come back to resistors. I use batteries (car batteries) to test my circuits, of course when the designs include CCS, switching from batteries to a mains regulated PSU, the audible difference is no or less perceptible but the sound is always worse than when I use resistors. That means, as John pointed out, that with today’s technology and knowledge (at least mine) it’s easier to get better sound using resistors (as CCS) and batteries (or a very very good power supply) than with a more sophisticated design using CCS to get rid of power supply deficiency.
John, bear with me 'cause I'm a slow learner 😉
If the DC servo outputs (+/-) are connected in the second stage as you have mentioned earlier in this thread, are they not somehow in the signal path? Would connections at the JFets sources not be better??
Secondly, you mention 4 separate connections, which, if I'm right, are the 4 cascode points in the line stage. Are you referring ONLY to correction points or is it a mix of correction and sense points??
I know whe have discussed this earlier in the thread, but just ironing out some misinterpretations/ misunderstandings on my side.
Thanks again.
Grey,
..............all this talk about current sources is making feel faint.
In response to your question about me and or Uli trolling for members you could not be further from the truth. My last design (with the help of Mr.Pass) the Mosfet A40 was posted on both sites and is part of a shootout on this site.
................but in the spirit of diy when version 2 is ready I will send you a set of boards if you like.
Regards,
Jam
..............all this talk about current sources is making feel faint.
In response to your question about me and or Uli trolling for members you could not be further from the truth. My last design (with the help of Mr.Pass) the Mosfet A40 was posted on both sites and is part of a shootout on this site.
................but in the spirit of diy when version 2 is ready I will send you a set of boards if you like.
Regards,
Jam
Given my views on current sources and their limitations, I try to use them only in places like the tail of a differential where the AC signal is minimal and the noise component has a chance to cancel via CMRR. Even then, a resistor is often better. People don't seem to realize that it only takes a few thousand Ohms to coax a differential into doing its job--not multiple Meg. Once you get above 10 or 20k, there's no real performance difference. At that point the CCS sometimes wins as a practical matter if you're drawing significant current because the rail has to be pretty high to accomodate the resistor.
One feature of resistors that rarely gets mentioned is that they're a great deal more reliable.
Chris,
Sometimes a current source is a useful thing to have...it's the "Well, of course we're going to use a current source there!" attitude that I'm trying to combat. There's no 'of course' about it. It's not the only option. Folks tend to use it out of mental laziness or a preconceived notion that it's superior. T'ain't necessarily so. If you choose to use one...great...but be aware that it's not a no-brainer selection. There are trade-offs inherent in any of the choices that we make.
I've been almost pathologically suspicious of current sources for about ten or fifteen years now. I needed one for a phase splitter in an amplifer and found that it did dreadful things to the sound. That lead to a six or eight month detour wherein I tried numerous CCS topologies and did a great deal of pondering as to exactly what the proper role of a CCS should be. It delayed the completion of my amp project considerably, but I learned a thing or two in the process.
I was already trying to simplify my circuits, having learned the hard way that the "technically correct" way of doing things didn't always give good sound. However, I had not gotten so far as to consider the CCS as a potential culprit. Given John's comments here, I had begun to suspect that he felt somewhat the same way, but wanted to check. I just chose a roundabout way of asking.
Jam,
If the same schematic is posted here, then by all means post a link to the appropriate thread.
There's a thread on the other site that is quite enlightening. Again--I have serious reservations about some of the things that happen here, but there's a right way and a wrong way to go about things.
As for feeling faint...I generally get that way when contemplating topics like Jessica Alba or Angelina Jolie. Current sources don't exert the same attraction.
Grey
One feature of resistors that rarely gets mentioned is that they're a great deal more reliable.
Chris,
Sometimes a current source is a useful thing to have...it's the "Well, of course we're going to use a current source there!" attitude that I'm trying to combat. There's no 'of course' about it. It's not the only option. Folks tend to use it out of mental laziness or a preconceived notion that it's superior. T'ain't necessarily so. If you choose to use one...great...but be aware that it's not a no-brainer selection. There are trade-offs inherent in any of the choices that we make.
I've been almost pathologically suspicious of current sources for about ten or fifteen years now. I needed one for a phase splitter in an amplifer and found that it did dreadful things to the sound. That lead to a six or eight month detour wherein I tried numerous CCS topologies and did a great deal of pondering as to exactly what the proper role of a CCS should be. It delayed the completion of my amp project considerably, but I learned a thing or two in the process.
I was already trying to simplify my circuits, having learned the hard way that the "technically correct" way of doing things didn't always give good sound. However, I had not gotten so far as to consider the CCS as a potential culprit. Given John's comments here, I had begun to suspect that he felt somewhat the same way, but wanted to check. I just chose a roundabout way of asking.
Jam,
If the same schematic is posted here, then by all means post a link to the appropriate thread.
There's a thread on the other site that is quite enlightening. Again--I have serious reservations about some of the things that happen here, but there's a right way and a wrong way to go about things.
As for feeling faint...I generally get that way when contemplating topics like Jessica Alba or Angelina Jolie. Current sources don't exert the same attraction.
Grey
Hi Grey,
Yes, I agree with you with regard to current sources being used as a magic bullet. Making the options and ramifications more visible to the rest of us is called teaching. Many of us are here to learn (I am) as well as help. The average person is very lazy mentally. The average manufacturer is just plain cheap.
Regarding the perceived problems with the site, we all try to do our best. Frankly, I don't expect everyone to agree all the time. It's the personal attacks and wars that we need to stop.
Thanks,
-Chris
Yes, I agree with you with regard to current sources being used as a magic bullet. Making the options and ramifications more visible to the rest of us is called teaching. Many of us are here to learn (I am) as well as help. The average person is very lazy mentally. The average manufacturer is just plain cheap.
Regarding the perceived problems with the site, we all try to do our best. Frankly, I don't expect everyone to agree all the time. It's the personal attacks and wars that we need to stop.
Thanks,
-Chris
- Status
- Not open for further replies.