Amplifier design: Truths or Myths?

0 Members and 1 Guest are viewing this topic. Read 29048 times.

HT cOz

Re: Amplifier design: Truths or Myths?
« Reply #20 on: 29 Jul 2008, 01:37 pm »
Items 1-12 are all true and Items 1-12 are all false.   :thumb:

pubul57

  • Jr. Member
  • Posts: 50
Re: Amplifier design: Truths or Myths?
« Reply #21 on: 30 Jul 2008, 12:20 am »
One reason it would be good to hear from Roger is that I think he has a bit Paul Klipsch in him, but stepping on toes in public isn't always fun. Another myth[?]: Good active preamps are better than passive preamps (assuming some proper matching between source and amp). If this myth[?] prooves false there sure is a lot of money at stake for owners and manufacturers. Secondary, related myth: Transformer-based passive are better than resistor-based passives.

6BQ5

  • Jr. Member
  • Posts: 75
Re: Amplifier design: Truths or Myths?
« Reply #22 on: 30 Jul 2008, 04:20 am »
Well, how's about a few more myths and truths (to fall on deaf ears)?  :oops:
What are we up to ... 25 now?


1. Transformer step-ups for MC stages are better than an active gain stage in a preamp
2. Massive power supplies with huge filtering caps are needed to get the most of an amp
3. A stepped resistor pot is best, preferably using Vishay type resistors
4. Some tubes sound "better" in a preamp than others, such as Telefunkens with little diamonds
5. Amplifier feet have a big influence on the quality of sound
6. Hybrid amps using MOSFETs sound warm and fuzzy
7. Tube amps have no bass, especially SET designs
8. SE amps have a purer midrange than PP types
9. Amps with metal fabrication (cases) have a hard sound because they are prone to RFI and EMI
10. Handwound transformers sound better than machine made transformers
11. High-current amps sound best
12. High voltage amps sound best

I can go on. No wonder the typical audio consumer is cynical with this kind of stuff floating around in the ether.

 

konut

  • Full Member
  • Posts: 1574
  • Came for the value, stayed for the drama
Re: Amplifier design: Truths or Myths?
« Reply #23 on: 30 Jul 2008, 04:33 am »
I am shocked that you havn't mentioned that......opamps are the devils' children.

6BQ5

  • Jr. Member
  • Posts: 75
Re: Amplifier design: Truths or Myths?
« Reply #24 on: 30 Jul 2008, 04:56 am »
I am shocked that you havn't mentioned that......opamps are the devils' children.

13. Op-amps are full of negative feedback and are the devils' children (even OPA627s biased into Class A).
13A. Op-amps have no place in high-end audio gear, a good designer should be able to design an optimized discrete circuit, which will always sound better.




NealH

  • Full Member
  • Posts: 373
Re: Amplifier design: Truths or Myths?
« Reply #25 on: 30 Jul 2008, 10:52 am »
The biasing, internal or external, of an op-amp has nothing to do with the class of operation.  And, if an op-amp did not have NFB implemented in or around it, it would be totally unstable.  It would operate more like a comparator. 

A well designed discrete circuit topology will usually outperform op-amp based topologies.  Usually, not always.

One of the biggest myths I ever heard was; "I have a well designed amplifier but, I need a boutique power cord for it to sound its best".


GMuffley

  • Jr. Member
  • Posts: 37
Re: Amplifier design: Truths or Myths?
« Reply #26 on: 30 Jul 2008, 12:25 pm »
One reason it would be good to hear from Roger is that I think he has a bit Paul Klipsch in him, but stepping on toes in public isn't always fun. Another myth[?]: Good active preamps are better than passive preamps (assuming some proper matching between source and amp). If this myth[?] prooves false there sure is a lot of money at stake for owners and manufacturers. Secondary, related myth: Transformer-based passive are better than resistor-based passives.

Paul,

I would enjoy hearing your experience with the RM-9 SE, now that you have been listeneing to it for several months--particularly since you were a long term user of the CAT JL-2.

pubul57

  • Jr. Member
  • Posts: 50
Re: Amplifier design: Truths or Myths?
« Reply #27 on: 30 Jul 2008, 01:50 pm »
GMuffley, I came to the RM9SE because I was looking for something that ran cooler than the CATJL2, which was a room heater in the summer. I started with the RM10MKII and loved it, but felt I needed a little bit more power so I tried an RM9MKII and the added drive was an improvement for my speakers, though I think the RM10 may be all you need with speakers 92db and up. I went to listen to an RM9SE in Virginia and fell in love with. I certainly had no complaints with the sound of the CAT, very powerful, dynamic, and transparent, but I find the RM9SE gives me every bit of that (perhaps a touch smoother), especially with a passive preamp, though it certainly works very well with the Joule LA150MKII I owned when I first got it. To my ears, the amp improved when I moved from the NOS Siemens EL34s to the KT88 RAM tubes, more dynamic, better bass, and larger sense of the soundstage. I take it is ultra reliable like all of Roger's designs (the CAT had no fuses to protect the resistors from tube failure, ugh.), and with the additional power and low impedance drive of the SE versus standard, can drive just about any speaker. My only hesitation was making a move from Class A to A/B and triode to ultralinear, the myths again. I strongly recommend the amp, but I think Roger only made 16 or so and maybe he has a handful left. I also had him add a ponytail extension for use with my own power cords, another myth I find hard to let go of.

6BQ5

  • Jr. Member
  • Posts: 75
Re: Amplifier design: Truths or Myths?
« Reply #28 on: 30 Jul 2008, 02:21 pm »
The biasing, internal or external, of an op-amp has nothing to do with the class of operation. 

Interesting. I often see DIY people saying they bias an op-amp into Class A mode.

Roger A. Modjeski

Re: Amplifier design: Truths or Myths?
« Reply #29 on: 1 Aug 2008, 04:49 am »
Thanks to 6BQ5 for stating this thread and those who have participated. Going by the number of replies in a just a few days I see this is a hot topic.

Before I answer his 12 points I would like to share a recent experience in modding two RM-4 head amps that had been extensively modified previously. In one of them all 8 rectifier diodes were replaced with high speed diodes. If high speed diodes make a difference that is where to put them. Yet, on the other RM-4 the 8 original rectifiers were still in place and 3 other diodes were replaced instead. This mod replaced 3 diodes that are never "on" in normal use. They are there to protect the high voltage power supply from damage no matter what a technician does when he is poking around measuring voltages. I think we can safely say that replacing rectifiers that are never on in normal use cannot change the sound. Does anyone care to argue this point?

I have found the following:

1. Smaller output transformers have less iron to magnetize and less copper to go around a smaller core: thus lower magnetizing loss and lower copper losses. Music Reference output transformers generally have half the loss compared to others. It is interesting to note that the amount of iron determines the low end of the frequency response and the amount of copper and the way it is wound determines the high frequency response.

2. Tube rectification has certainly attracted a lot of attention along with the additional belief that a premium tube rectifier sounds better than a common one. A high power amplifier like an RM-9 or RM-200 would need 4 rectifier tubes as the power drawn from the supply is quite high at 500 mA per channel x 2 is 1 amp. Tube rectifiers cannot work into first filter caps larger than 40 uF. The first filter in most of my amps is 500 uF or more.  Tubes work best with choke input filters but these supplies have very poor regulation and are not suitable for high power class AB amps.

3. Such a general statement is not likely to be true and how old does a NOS tube have to be to be NOS? More than anything, tubes vary from batch to batch just like cookies do. I unintentionally provide QC for several tube producers by letting them know when I get a batch that is not like the usual from that factory. The last time this happened I had bought some tubes and not tested them for a year. When I got around to testing them I found they had low gain and more noise than the usual from this factory. They sent me another batch made 3 months later. These were no better. They complained to the factory and a new sample was made for approval. This time the tubes were right again. These same tubes are used by other tube amp makers who apparently hadn't noticed the change in performance.

stay tuned for the next 3

6BQ5

  • Jr. Member
  • Posts: 75
Re: Amplifier design: Truths or Myths?
« Reply #30 on: 1 Aug 2008, 11:33 pm »
I would like to share a recent experience ...




  :D

 

Roger A. Modjeski

Re: Amplifier design: Truths or Myths?
« Reply #31 on: 5 Aug 2008, 10:53 pm »
I'll take that as agreement.

 :D

I was liking the discussion you were having amongst yourselves. Please continue. I'll chime in now and then.
« Last Edit: 6 Aug 2008, 04:20 am by Roger A. Modjeski »

Ericus Rex

Re: Amplifier design: Truths or Myths?
« Reply #32 on: 6 Aug 2008, 11:39 am »
I'll take that as agreement.

 :D

I was liking the discussion you were having amongst yourselves. Please continue. I'll chime in now and then.



We're waiting for your 4, 5, and 6.

6BQ5

  • Jr. Member
  • Posts: 75
Re: Amplifier design: Truths or Myths?
« Reply #33 on: 9 Aug 2008, 06:33 am »

Point  :wink:     Counterpoint   :icon_twisted:

1. Transformers
The HK Citation II lovers seem to attribute a lot to "iron"
http://www.quadesl.com/refurb/refurb_hkCitation2.html
http://db.audioasylum.com/cgi/m.mpl?forum=vintage&n=161589&highlight=citation+Iron&r=&session=

2. Tube Rectification
Supposedly quieter than diodes, but do have their own drawbacks so it seems
Air tight tube preamp description http://www.katli.com/preamplifiers.htm
Both sides of the story ... http://www.welbornelabs.com/drd.htm

3. NOS
Tons of examples ... but this one is typical.
http://www.tonequest.com/articles/article3.htm
Maybe there is more money to be made selling NOS?



Ericus Rex

Re: Amplifier design: Truths or Myths?
« Reply #34 on: 9 Aug 2008, 12:51 pm »
Don't know enough about 1 and 2 to reply...still learning  :D

However, I do have an opinion about NOS versus current production tubes.  Do NOS tubes sound better?  Many, I'm sure,  do.  But most, I'm sure, do not.  Add to that the counterfeiting, the cross-labeling, the extortionate pricing, questionable testing and the lack of significant guarantees and I find that the risks involved with buying NOS aren't worth the 10X to 30X price difference.  If after six months my $13 tube dies I won't sweat it.  But if my $150 NOS tube fizzles in that time I'm going to be ticked.  Current production still sounds way better than solid state to me.

pbrstreetgang

  • Full Member
  • Posts: 604
Re: Amplifier design: Truths or Myths?
« Reply #35 on: 9 Aug 2008, 02:46 pm »
Don't know enough about 1 and 2 to reply...still learning  :D

However, I do have an opinion about NOS versus current production tubes.  Do NOS tubes sound better?  Many, I'm sure,  do.  But most, I'm sure, do not.  Add to that the counterfeiting, the cross-labeling, the extortionate pricing, questionable testing and the lack of significant guarantees and I find that the risks involved with buying NOS aren't worth the 10X to 30X price difference.  .  Current production still sounds way better than solid state to me.

Well the good one sound much better in most circuits. Reports that the superbly engineered MR and VAC gear fly against that statement but in the vast majority they for sure are better. If you buy NOS tubes you pay the premium from the good vendors or you become familiar with construction, date codes and vintages. If not you will be burned.


"If after six months my $13 tube dies I won't sweat it.  But if my $150 NOS tube fizzles in that time I'm going to be ticked" Thats the beauty I find they last much longer in most cases :thumb:

6BQ5

  • Jr. Member
  • Posts: 75
Re: Amplifier design: Truths or Myths?
« Reply #36 on: 9 Aug 2008, 04:19 pm »
Re: NOS

I'm leery of tubes that have not been matched (if needed) or tested/graded for quality. I notice that some sites selling NOS tubes make no mention of the quality of the tube.
An example of good practice would be Upscale Audio, where Keven Deal categorizes every NOS tube, as to standard, low-noise and super low-noise quality.

Some tubes are just plain expensive because they have become rare as stocks have been depleted. So its basic supply and demand, and paying a huge premium simply based on scarcity rather than quality seems ridiculous to me. I once paid a premium for a pair of cryo'd Sylvania 6SNGT's which were nothing special if not noisy, whereas a cheap Mazda 6SNGT for 1/10 the cost sounded incredibly better. I learned my lesson, you don't always get what you pay for.

If I were a gear manufacturer, I just can't see being able to rely on NOS tubes, where a steady and reliable supply is important to meet production requirements. RAM cherry-picking new production is good practice, but I have seen postings here and there by people who have done some rolling that they have preferred NOS in quite a few cases over RAM tubes. I also question why RAM needs to be doing this in the first place ... policing the quality of tube manufacturers. Name me one other industry that does this ... I can't imagine a BMW dealership saying we pick out the best Beemers off the production line and only sell those. The QA in places like China and Russia must be then terrible if RAM is having to do this.


pbrstreetgang

  • Full Member
  • Posts: 604
Re: Amplifier design: Truths or Myths?
« Reply #37 on: 10 Aug 2008, 01:47 am »
Re: NOS

I'm leery of tubes that have not been matched (if needed) or tested/graded for quality. I notice that some sites selling NOS tubes make no mention of the quality of the tube.
An example of good practice would be Upscale Audio, where Keven Deal categorizes every NOS tube, as to standard, low-noise and super low-noise quality.

Some tubes are just plain expensive because they have become rare as stocks have been depleted. So its basic supply and demand, and paying a huge premium simply based on scarcity rather than quality seems ridiculous to me. I once paid a premium for a pair of cryo'd Sylvania 6SNGT's which were nothing special if not noisy, whereas a cheap Mazda 6SNGT for 1/10 the cost sounded incredibly better. I learned my lesson, you don't always get what you pay for.

If I were a gear manufacturer, I just can't see being able to rely on NOS tubes, where a steady and reliable supply is important to meet production requirements. RAM cherry-picking new production is good practice, but I have seen postings here and there by people who have done some rolling that they have preferred NOS in quite a few cases over RAM tubes. I also question why RAM needs to be doing this in the first place ... policing the quality of tube manufacturers. Name me one other industry that does this ... I can't imagine a BMW dealership saying we pick out the best Beemers off the production line and only sell those. The QA in places like China and Russia must be then terrible if RAM is having to do this.



Your right a manufacturer cannot rely on NOS tubes. It just isnt possible, Quicksilver did that once and now owners have to convert at great cost and different sound. Mr. Modjeski is a superb engineer and quite possibly has no time or patience for nothing but super tight matches in his tubes. That is not saying however that the same tight matched NOS tubes wouldnt sound better, or just sidestepping different in any case. If there was huge supply and sane prices of all NOS tubes, I wonder which ones he would prefer? I bet it would be a NOS of some vintage.

Also there is huge variance in cars and BMWs, I know Im in the industry of aftermarket performance tuning. There are few "ringers" with higher HP/TQ, and turbos that are just a bit more efficient. There are cars that are quite a bit more knock prone and put out less power than the norm too. You just would never know unless it is tested and compared to many other like vehicles under controlled conditions. (Like Roger does with tubes). Of course BMW doesnt say anything and doesnt consider it a problem. They are more picky and higher QC with the M cars that are mass produced then undergo hands on evaluation by BMW. Also special spec cars and cup racers are tightly matched- though they dont often stay matched because of a huge variety of reasons.
This is evident in the Mazda star series were everything is matched to a "T" though some specd racers put out 5% or more power after just a few laps. Complicated machines made by man are bound to be different even when they are the same.

It would be interesting to test long term validity of the special matching from RAM by testing a quad of KT88s when new, in 6 months then 18 months to see what variability is over the long term

6BQ5

  • Jr. Member
  • Posts: 75
Re: Amplifier design: Truths or Myths?
« Reply #38 on: 10 Aug 2008, 04:08 am »
re: Rectifiers

It seems to me, that the main motivation for going back to tube rectification is to avoid some of the downsides with SS diodes, specifically the noise they can generate. However, tube rectification has its downsides too. Michael Elliott formerly of Counterpoint, when designing his new Aria preamp examined the tube rectification option and created a prototype.



In the end, he said naw, and went with a conventional SS rectified power supply. Looking at that pic, can you blame him?

When I look at my RM-5, I see IN4007 diodes. These are run-of-the-mill diodes that go for as little as 3c bought in bulk.

The newest generation of Cree Schottky rectifiers (silicon carbide) introduced in the past couple of years, supposedly don't have the noise problems of standard diodes like an IN4007.
Would a Schottky in lieu result in audible improvement?

Wish I knew. Certainly, they are way more expensive, going for several dollars each.


pbrstreetgang

  • Full Member
  • Posts: 604
Re: Amplifier design: Truths or Myths?
« Reply #39 on: 10 Aug 2008, 04:57 am »
Any word on using SS regulator with tube rectification? I picked up a pre using this and it is impressive, It has replaced some decent classic pres and even went to personal preference against a stereophile class A pre