Bob Crump was a good friend...
Bob Crump was a good friend.
Great guy.
Passionate about his audio fer sure.
Bought one of my Symphony amps, my cables and interconnects, his reference for many years, exhibited at CES with my amp... he loaded it in the back of his Turbo VW (performance chipped) and drove it up to 'Vegas at least one year... just talkin' with Bob was quite a thing!
I remember when he told me he was going to go into the biz, and about the Blowtorch... ahhh well.
Been too busy to be online here very much over the last weeks...
I spotted the thread on the Blowtorch... :-(
Only to find that Bob has passed away.
I was having a pretty good day up until then.
Bummed my space, big time.
That's a big vacuum to fill up.
crap.
_-_-bear
Bob Crump was a good friend.
Great guy.
Passionate about his audio fer sure.
Bought one of my Symphony amps, my cables and interconnects, his reference for many years, exhibited at CES with my amp... he loaded it in the back of his Turbo VW (performance chipped) and drove it up to 'Vegas at least one year... just talkin' with Bob was quite a thing!
I remember when he told me he was going to go into the biz, and about the Blowtorch... ahhh well.
Been too busy to be online here very much over the last weeks...
I spotted the thread on the Blowtorch... :-(
Only to find that Bob has passed away.
I was having a pretty good day up until then.
Bummed my space, big time.
That's a big vacuum to fill up.
crap.
_-_-bear
Just wonder if John Curl will comment on where audio is going to be in the future. Futurists like Ray Kurzweil talk about how rapidly technologies are advancing. Are there going to be large breakthroughs in audio technology? Major advances in transducers around the corner? Thanks for any thoughts.
John,
I have been embroiled in this Lovoltech power JFET thing and haven't made time to read this thread. Currently, I'm up to page 10 and want to drop in a comment or two and a question before I forget, then I'll go back to reading. I'll ask your pardon if this has been covered between p10 and here.
--I have long been a proponent of using regulators for each channel and for each stage. This has earned me more than one strange look and quite a few unkind comments.
--I have, over time, simplified my regulators. I used to go with one really, really over-the-top regulator coming from the rail directly to the circuit. I now tend towards two, sometimes three regulators, each reducing the voltage a bit, rather than all at once. I'm not sure why it sounds better this way, but I like it.
--I have experimented with electrolytics, films, and sometimes no caps at all after the regulators. Since regulators tend to function as low-pass filters, I find that electrolytics back towards the transformer end of things are not too big a problem, although I don't care for them after final regulation and before the circuit. Film caps are best here.
Now for my question:
Have you found that not having a cap after the final regulation tends to cramp dynamics?
I've observed that it's possible to get really, really beautiful sound at low signal levels, but that a peak falls utterly flat, dynamically.
To the extent that it's possible to design a circuit that has essentially DC current draw (e.g. differential) that can lessen the problem, but it's not always feasible to do so.
Comments?
Grey
I have been embroiled in this Lovoltech power JFET thing and haven't made time to read this thread. Currently, I'm up to page 10 and want to drop in a comment or two and a question before I forget, then I'll go back to reading. I'll ask your pardon if this has been covered between p10 and here.
--I have long been a proponent of using regulators for each channel and for each stage. This has earned me more than one strange look and quite a few unkind comments.
--I have, over time, simplified my regulators. I used to go with one really, really over-the-top regulator coming from the rail directly to the circuit. I now tend towards two, sometimes three regulators, each reducing the voltage a bit, rather than all at once. I'm not sure why it sounds better this way, but I like it.
--I have experimented with electrolytics, films, and sometimes no caps at all after the regulators. Since regulators tend to function as low-pass filters, I find that electrolytics back towards the transformer end of things are not too big a problem, although I don't care for them after final regulation and before the circuit. Film caps are best here.
Now for my question:
Have you found that not having a cap after the final regulation tends to cramp dynamics?
I've observed that it's possible to get really, really beautiful sound at low signal levels, but that a peak falls utterly flat, dynamically.
To the extent that it's possible to design a circuit that has essentially DC current draw (e.g. differential) that can lessen the problem, but it's not always feasible to do so.
Comments?
Grey
John,
Ah...in your post on the next page (#108), you make much the same point that I do about cascading (I have a mental image of water cascading down a mountainside, each step lower--you, quite correctly, say series) regulators.
But why does this sound better? I can hear it, but the reason has eluded me.
Grey
Ah...in your post on the next page (#108), you make much the same point that I do about cascading (I have a mental image of water cascading down a mountainside, each step lower--you, quite correctly, say series) regulators.
But why does this sound better? I can hear it, but the reason has eluded me.
Grey
Possible, but I lean more towards the idea that a single big regulator, meaning one hefty enough to drop a reasonable amount of voltage and current at the same time, is going to require a larger device, be it FET, bipolar, or MOSFET.
All other things being equal, a large device is generally noisier. (Yes, I'm going to try the Lovoltechs in regulators, but bear in mind that there are no P devices for the negative rails.) That device-generated noise will require filtering, which will mean caps, inductors, or another stage of regulators. So you end up with this infinite regression of regulators to filter the residual junk from the upstream regulators.
In other words, the regulators themselves become the problem.
Smaller regulators, needing to pass the current (in the case of series regulators), but not necessarily a lot of voltage can use lower noise, higher quality pass devices.
However, this is only a hypothesis. I was hoping that John might comment on this and my earlier question about dynamics.
Grey
All other things being equal, a large device is generally noisier. (Yes, I'm going to try the Lovoltechs in regulators, but bear in mind that there are no P devices for the negative rails.) That device-generated noise will require filtering, which will mean caps, inductors, or another stage of regulators. So you end up with this infinite regression of regulators to filter the residual junk from the upstream regulators.
In other words, the regulators themselves become the problem.
Smaller regulators, needing to pass the current (in the case of series regulators), but not necessarily a lot of voltage can use lower noise, higher quality pass devices.
However, this is only a hypothesis. I was hoping that John might comment on this and my earlier question about dynamics.
Grey
Here's a clue guys,
When you follow series regulators with shunt regulators... you are talkng about???... constant power.. The series regulator deals with LINE induced deviations... the shunt regulator deals with LOAD induced variations... constant currents are induced here and there... where, you should physically LOCATE the shunt reg, can make a difference here... where should it be???... and why???
Think about it... HARD, BEFORE you blast me a load of rebutal.
When you follow series regulators with shunt regulators... you are talkng about???... constant power.. The series regulator deals with LINE induced deviations... the shunt regulator deals with LOAD induced variations... constant currents are induced here and there... where, you should physically LOCATE the shunt reg, can make a difference here... where should it be???... and why???
Think about it... HARD, BEFORE you blast me a load of rebutal.
It's not necessary to shout, poobah. Doing so doesn't make your point any more or less true.
In my case, I'm speaking of using voltage regulators only. I have no quarrel with current/shunt regulators. However, my observations were based on one, two, and three voltage regulators in series.
Grey
In my case, I'm speaking of using voltage regulators only. I have no quarrel with current/shunt regulators. However, my observations were based on one, two, and three voltage regulators in series.
Grey
Really? I'll have to listen more carefully the next time I'm near some heavy machinery.
(That was a joke. You don't have to laugh, but at least make a feeble attempt to smile.)
I have seen it said that current/shunt regulators make the most sense supplying circuits that have the least variation in current draw. The obvious endpoint of this line of reasoning is a circuit that draws DC, or as close to it as you can get.
I don't remember why the writer was saying so, but it's pretty much of a truism--at least in my estimation--that voltage regulators perform best when they don't have to regulate. This perhaps echos John's comment (I think it was John) earlier in the thread about designing the output to need as little DC correction as possible before adding the servo.
Of course, taken reductio ad absurdum, that would imply that amplifiers amplify best when they don't have to amplify at all!
Actually, come to think of it, that's true.
Grey
(That was a joke. You don't have to laugh, but at least make a feeble attempt to smile.)
I have seen it said that current/shunt regulators make the most sense supplying circuits that have the least variation in current draw. The obvious endpoint of this line of reasoning is a circuit that draws DC, or as close to it as you can get.
I don't remember why the writer was saying so, but it's pretty much of a truism--at least in my estimation--that voltage regulators perform best when they don't have to regulate. This perhaps echos John's comment (I think it was John) earlier in the thread about designing the output to need as little DC correction as possible before adding the servo.
Of course, taken reductio ad absurdum, that would imply that amplifiers amplify best when they don't have to amplify at all!
Actually, come to think of it, that's true.
Grey
Grey,
The trick employed here is that if you give voltage regulators a constant current load; they behave better.
About the hydraulic systems... always fluid in motion... ready to do work... investment in kinetic energy (inductance - momentum) already made. This parallels the constant current approach...
The trick employed here is that if you give voltage regulators a constant current load; they behave better.
About the hydraulic systems... always fluid in motion... ready to do work... investment in kinetic energy (inductance - momentum) already made. This parallels the constant current approach...
You're saying the same thing I'm saying, just in different words.
Conversely, a current regulator will function more smoothly if it is presented a steady voltage load.
This gets back to the design of the circuit. If it presents a steady load to the regulators, then they will be happier. This is one of several ideas that I've been playing with for the past few years. Wide bandwidth (I generally shoot for at least 200-250kHz), low to no feedback, and trying to remove the power supply from the equation by only asking it to deal with 0Hz voltage/current. So far the results have been worthwhile.
Grey
Conversely, a current regulator will function more smoothly if it is presented a steady voltage load.
This gets back to the design of the circuit. If it presents a steady load to the regulators, then they will be happier. This is one of several ideas that I've been playing with for the past few years. Wide bandwidth (I generally shoot for at least 200-250kHz), low to no feedback, and trying to remove the power supply from the equation by only asking it to deal with 0Hz voltage/current. So far the results have been worthwhile.
Grey
Folks, I don't have answers to all your questions. The future? I don't know any more than anyone else. Actually I may, at most, use 3 regulators in series. One is feedback and zener controlled to remove hum and line breathing, the second (if used) in order to remove high frequency glitches, and the third is always open loop and is actually part of the individual stage. It is usually designed to isolate that stage from other signals. It is simple, usually a cap follower, so it MUST have a stable DC input voltage that is supplied by the conventional high feedback zener referenced voltage regulator at the input.
The main factor that seems to be forgotten is that the regulators must remove both very low and very high frequencies . This is difficult with one regulator, at least, I don't know how to do it easily.
The main factor that seems to be forgotten is that the regulators must remove both very low and very high frequencies . This is difficult with one regulator, at least, I don't know how to do it easily.
At the moment I tend towards two regulators. One in an external power supply and one next to the circuit. It seems to get the job done. I don't have the low frequency power problems that others have (line voltage drift, etc.), so my eye is more on higher frequency noise artifacts. Even more so since I'm working on a phono stage.
I keep meaning to build a big enough regulator to handle a small to moderate power amp, but I'm perpetually short on heatsinks. With that in mind, all my comments should be understood to refer to low level circuitry only.
Grey
I keep meaning to build a big enough regulator to handle a small to moderate power amp, but I'm perpetually short on heatsinks. With that in mind, all my comments should be understood to refer to low level circuitry only.
Grey
Sorry, everyone about the off topic posts... last one promise... It seems to me that perhaps in ear transducers may be the thing of the future... Even now, a Blowtorch into a set of Ultimate Ears must sound pretty good... However, in the future, there needs to be some microprocessors involved to simulate changes in the reflection patterns as the head moves. A speaker system would benefit, if there were interaction--the sound changes as we move around the room. I think that in the future sensors and processors to modify the sound will provide added realism. If nothing else, something that at least changes the performance at least slightly each playback.johnferrier said:
john curl said:
Except for the pre-regs* this is also similar to Allen Wright's SuperReg. Even down to replacing the 317 CCS with a resister once the current draw is known (resister sounds better than the 317).
*(is the purpose here to suupply regulated supply to the control OpAmp? -- Allen's reg being high voltage he uses a seperate series regulated supply for the comtol IC)
I'd be very interested in seeing this reg with example values.
dave
- Status
- Not open for further replies.