Tube DIY Asylum

Do It Yourself (DIY) paradise for tube and SET project builders.

Return to Tube DIY Asylum

Message Sort: Post Order or Asylum Reverse Threaded

SE 6V6 E-Linear : found my schematic ...

24.140.110.227
Posted on December 29, 2020 at 15:56:24
deathtube 667



Yes, I tried to do a SE DC E-Linear circuit.

Based on the schematic I found dated 4/2019, I didn't have the final screen connected to the UL tap on the OPT.

That might have been the reason I didn't get it to perform properly.

The circuit by Pete Millet shows cap coupled stages.

Perhaps that is need to get this thing to work correctly, not sure.

DT 667




 

Hide full thread outline!
    ...
RE: SE 6V6 E-Linear : found my schematic ..., posted on December 29, 2020 at 17:55:56
PakProtector
Audiophile

Posts: 12360
Joined: May 14, 2002
 Oh yes, has to be the cap coupling that makes or breaks this...are you nuts?

Nice picture.
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: SE 6V6 E-Linear : found my schematic ..., posted on December 29, 2020 at 19:23:19
PakProtector
Audiophile

Posts: 12360
Joined: May 14, 2002
 Or, I could just let you in on the Big Secret: there is a specific coupling cap required. All but this particular one don't work. Some not at all, some even invert the phase all by themselves, in addition to adding just an excruciating amount of eleventeenth harmonic distortion. Others liken one's mother to a gardening implement, and get put right into the fire. There was a code released that would have directed you to The Right Cap had you thought to notice it, and then decode it. Guess that escaped you...
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: SE 6V6 E-Linear : found my schematic ..., posted on December 29, 2020 at 22:00:38
Paul Joppa
Industry Professional

Posts: 7295
Location: Seattle, WA
Joined: April 23, 2001
 Yeah, it's not e-linear without the screen going to the UL tap.

The driver rp and its plate and cathode resistances are critical to getting the feedback right while maintaining drive capability, so either good analytical skills or a ton of experimentation would be needed to make a new design.

Direct coupling can work; I've done it a couple times (without the UL tap). You need a bias servo to maintain the operating point.

 

RE: thanks ..., posted on December 30, 2020 at 02:53:15
deathtube 667
 I knew I had it wired up incorrectly.

Thanks for the reasonable reply.

DT 667

 

RE: SE 6V6 E-Linear : found my schematic ..., posted on December 30, 2020 at 06:49:06
PakProtector
Audiophile

Posts: 12360
Joined: May 14, 2002
 That is curious; I don't recall making that definition when I named this circuit. It could have escaped my memory of course, but since it is mine to define, I will lay it out now: the power tube can be run pentode, triode, or U-L. The local FB bends its plate curves into a linear triode anyway... :)
cheers,
Douglas
Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: who takes credit for e-linear ?, posted on December 30, 2020 at 07:49:26
deathtube 667
 "The term "E-linear" was coined recently by Emotive Audio (www.emotiveaudio.com) as a description for the combination of ultralinear feedback to the screen grid of the output tube and to the driver stage, or to the control grid of the output tube. I'd be surprised if nobody had tried this earlier, but I haven't found any references to this type of output stage connection. It seems that several people started playing with this configuration at about the same time, including myself." - Pete Millet

Pete Millet said Emotive Audio named the "e-linear" circuit.

The article does not mention anyone named "Douglas" as I have read it.

DT667





 

RE: who takes credit for e-linear ?, posted on December 30, 2020 at 09:53:19
PakProtector
Audiophile

Posts: 12360
Joined: May 14, 2002
 Well, since you were not in Fred's Highlander, taking part in the discussion I doubt you'd have any reason to suspect anything regarding the name or the topology we put together. You most certainly were not part of the discussion between Fred and Pete... :)

In any case, Pete got it better than I in his analysis to run a pentode as the voltage amplifier source. I am glad he pointed that bit out to us all.
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: SE 6V6 E-Linear : found my schematic ..., posted on December 30, 2020 at 09:54:44
Paul Joppa
Industry Professional

Posts: 7295
Location: Seattle, WA
Joined: April 23, 2001
 By that definition, it seems little different from an anode follower, or Schade feedback stage from 1938 - unless I'm missing something?

 

RE: SE 6V6 E-Linear : found my schematic ..., posted on December 30, 2020 at 10:08:49
PakProtector
Audiophile

Posts: 12360
Joined: May 14, 2002
 probably not missing anything.

Just a means of delivering power stage plate signal to the driver stage( aka the power stage's grid ). Only a single B+ supply is needed for a 2-stage amp.

It is very much a plate-to-grid arrangement...and cares not at all what sort of power stage is used. Just needs a tap along the OPT's primary in the right spot.
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: did you get a patent for the "e-linear" circuit?, posted on December 30, 2020 at 10:48:57
deathtube 667
 Where I work, a few of the engineers hold US Patents for various products or technical processes.

Did your engineering team get a patent for the "e-linear" circuit?

DT 667




 

RE: e-linear vs. RH circuit , posted on December 30, 2020 at 11:15:48
deathtube 667
 What are the advantages of using the e-linear connection vs. RH's plate to grid feedback in SE circuits?

All you need is a resistor or resistor + cap for direct coupling with the RH feedback network. It is pretty simple to implement.

 

RE: did you get a patent for the "e-linear" circuit?, posted on December 30, 2020 at 11:50:49
PakProtector
Audiophile

Posts: 12360
Joined: May 14, 2002
 I hold several...I confess I quit counting at 5, and no it was decided not to patent. At the time, there did not appear to be enough profit. Given the results of another contribution I made, that choice may have been in error.

Also, I think some misunderstanding is contained in your use of the term, 'engineering team'. There were three of us, if Senior is counted... :)

cheers,
Douglas
Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: e-linear vs. RH circuit , posted on December 30, 2020 at 11:53:10
PakProtector
Audiophile

Posts: 12360
Joined: May 14, 2002
 RH is tied somehow to triodes, albeit high plate resistance ones...

Go dig up some of his responses to the suggestion to use pentodes...LOL
cheers,
Douglas
Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: SE 6V6 E-Linear : found my schematic ..., posted on December 30, 2020 at 12:31:07
Stuben
Audiophile

Posts: 669
Location: Guber Ohio
Joined: December 30, 2005



Our Guy at Glassware shared something like this with us a while back. As stated by our gurus, takes some artistic patients...

Stuben

 

RE: SE 6V6 E-Linear : found my schematic ..., posted on December 30, 2020 at 13:43:19
Paul Joppa
Industry Professional

Posts: 7295
Location: Seattle, WA
Joined: April 23, 2001
 Thanks for clarifying. FWIW, I've been calling it a transimpedance stage - current in, voltage out. I don't expect it to catch on though ... :^)

 

RE: SE 6V6 E-Linear : found my schematic ..., posted on December 30, 2020 at 18:23:06
PakProtector
Audiophile

Posts: 12360
Joined: May 14, 2002
 Part of the way things catch on is with the Name I find. Couldn't think of another one at the time. Voltage FB using the U-L features seemed to have done well enough.

I confess I may do away with the front end pentodes all together as I build the 26HU5 amp. There is currently 5W dissipation on each plate, and I do have some worry keeping a TO-220 cool at that rate. Perhaps on a lower voltage variant that runs the 'HX5 as its power stage. Three cascode cards; one for the LTP current reg, and two more for the amplifier elements. After that, might as well do away with the power tubes...and that brings me right into Gary Pimm territory...a fair distance from where I first started( with a Foreplay manual, and a pile of parts to deal with 5687's instead of 12AU7's )...LOL
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

I got a question, posted on December 31, 2020 at 14:49:44
Ralph
Manufacturer

Posts: 4769
Location: Minnesota
Joined: April 24, 2002
 This is from the article linked:


- less distortion is better than more distortion
- even-order distortion (2nd, 4th, and so on) is less audible than odd-order distortion (3rd, 5th, and so on)
- low-order distortion (2nd, 3rd, and so on) is less audible than high-order distortion (5th, 6th, and so on)
- low-order distortion tends to mask higher-order distortion

I think most people would agree that the ideal amplifier would have no distortion. You can' [sic] achieve that, so the next best thing is to minimize distortion, but keep the distortion profile such that low, even-order harmonics dominate, and don't allow large amounts of high-order odd harmonics.

I agree with a bit of it, but one part I do not is the special case of the 3rd harmonic. The ear treats this much the same way as the 2nd; to this end the comment about the 3rd being more audible than the 2nd appears to be false (and to be clear, the author does state that some will disagree with him on these points).

The reason I bring it up is that if you're going to build an amplifier design that is based on a 2nd harmonic as the dominant distortion product, it has what is called a 'quadratic non-linearity'. An amplifier that has a cubic non-linearity (3rd harmonic as dominant) instead will have overall lower distortion across the audio band. And, as stated above, the 3rd will mask the presence of the pesky higher ordered odd harmonics.

The way to do this is to avoid a single-ended input into a push-pull output (make the entire circuit differential). In this way the main HD product will be the third (assuming proper operating points). This strikes me as a more practical approach as you get more usable (IOW, lower distortion) power.

What am I missing?




 

RE: I got a question, posted on December 31, 2020 at 15:56:05
Paul Joppa
Industry Professional

Posts: 7295
Location: Seattle, WA
Joined: April 23, 2001
 "I agree with a bit of it, but one part I do not is the special case of the 3rd harmonic. The ear treats this much the same way as the 2nd; to this end the comment about the 3rd being more audible than the 2nd appears to be false ..."

I'm not following your logic here - specifically, what exactly do you mean by "The ear treats this [3rd harmonic] much the same way as the 2nd ..." ?

My own disagreement with Millett is his statement that "even-order distortion (2nd, 4th, and so on) is less audible than odd-order distortion (3rd, 5th, and so on)" - I think it's more likely that the audibility is monotonic with the order, i.e. 3rd is more audible than 2nd, 4th is more audible than 3rd, 5th more audible than 4th, 6th more audible than 5th, etc.

 

Most astute observation & fwiw, agree.nT, posted on December 31, 2020 at 18:04:59
Cleantimestream
Audiophile

Posts: 7550
Location: Kentucky
Joined: June 30, 2005
 !
The Mind has No Firewall~ U.S. Army War College.

 

Isn't this based on the harmonic series in music?, posted on December 31, 2020 at 19:36:13
Rick W
 If not I admit my ignorance. But if it is, I don't see why anything below the 7th harmonic would be much of a problem, at least in terms of harshness. Below that is basically just a triad. Some might feel too much lower harmonics fattens the sound too much but I don't see how it'd contribute to harshness. Too much odd number harmonics from the 7th upwards could contribute to a perception of harshness though.

Do you agree?

 

RE: I got a question, posted on December 31, 2020 at 21:06:58
PakProtector
Audiophile

Posts: 12360
Joined: May 14, 2002
 Part of the 'idea' of this basic topology I like is the FB is differential too. Not sure why this seems to leave me a lintenable amp...but here we are.

That it also works by effectively lowering the output Z of the finals directly, and shortens the FB loop length, leaves me listening to these for hours when nothing else gets in the way.

Happy New Year to y'all!
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: I got a question ... me too, posted on January 1, 2021 at 09:13:35
deathtube 667
 "The way to do this is to avoid a single-ended input into a push-pull output (make the entire circuit differential). "

My thought was to modify an existing SE 6V6 UL amp to SE E-Linear.

The driver stage is 6SL7 with a CCS plate load @ 1-2mADC.

The output runs UL mode and also uses plate to grid feedback.

The amp measures and sounds well enough that I might not want to modify it.

I did find other builders of SE E-Linear who said that they had some hum issues with implementation of this type of feedback.

Not sure if it was an issue on inadequate filtering at the PS node or something else.

Still not too clear on what the major advantages of E-Linear are over other feedback modes when applied to SE amplifiers.

More study will be required before moving ahead.

DT 667

Edit : the 6SL7 with CCS apparently is not the way to go for E-Linear driver stage. I did find an SE EL84 E-Linear schematic that I can adapt for 6V6 finals, so it should be easy enough to try out.


 

RE: I got a question ... me too, posted on January 1, 2021 at 14:12:28
rage
Audiophile

Posts: 792
Joined: December 17, 2010
 deathube:

my perceived advantages are:

- no global feedback loop to adjust

- potential for less parts in the amplifier with one B+ node, no RC stages etc

- short local feedback loop and my understanding again being that you could theoretically have an output transformer with multiple UL taps - and then experiment with what you prefer. seems the biggest change would be gain depending on the tap used.


- potential for balanced input is a plus for me as well (with push pull)


 

RE: SE 6V6 E-Linear : found my schematic ..., posted on January 1, 2021 at 14:14:14
rage
Audiophile

Posts: 792
Joined: December 17, 2010
 dt - you really need a pentode front end.

 

RE: single node PS ..., posted on January 1, 2021 at 14:28:52
deathtube 667
 "- potential for less parts in the amplifier with one B+ node, no RC stages etc"

A simple LCLC "flywheel" filter would fit here. Not sure about LSES.

Go for some real "low DCR" , without the R dropper, LOL.

Actually, feedback is pretty easy to adjust with the ARTA program, IME.

DT 667

 

RE: pentode driver, posted on January 1, 2021 at 14:33:39
deathtube 667
 Yeah.

Been reading about "E-Linear" driver requirements on another audio site.

I got a boatload of 6AU6, 6SJ7 and 5879's.

Waiting for EF86 to get delivered.





 

RE: I got a question ... me too, posted on January 1, 2021 at 14:50:20
PakProtector
Audiophile

Posts: 12360
Joined: May 14, 2002
 You can indeed build an OPT with more than one set of taps. The S265Q offers end-of-layer increments of 10%. Makes for a fine Class A amp due to its 10k a-a load. With a stocker, you're stuck with 50%. Same for a stock S271S. IIRC the S271S goes in 12.5% increments( 8 layers for each half of the primary ).
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

Charlie Parker turned the harmonic series {arithmetic progression} on its head ~nT, posted on January 3, 2021 at 11:28:28
Cleantimestream
Audiophile

Posts: 7550
Location: Kentucky
Joined: June 30, 2005
 ~!
The Mind has No Firewall~ U.S. Army War College.

 

RE: Isn't this based on the harmonic series in music?, posted on January 3, 2021 at 14:42:26
elblanco
Audiophile

Posts: 3486
Joined: August 20, 2004
 not the same as triad.

for instance e major triad on guitar would be 82, 104, and 123 hz respectively (E, G#, B)

the first, major third and fifth.

harmonics of E would be octaves at ~82,164,328, etc.....

 

The 3rd harmonic, posted on January 4, 2021 at 10:59:36
Ralph
Manufacturer

Posts: 4769
Location: Minnesota
Joined: April 24, 2002
 I'm not following your logic here - specifically, what exactly do you mean by "The ear treats this [3rd harmonic] much the same way as the 2nd ..." ?

What I mean by that is exactly as you quoted me (so the logic should be easy to follow): The ear treats this [3rd harmonic] much the same way as the 2nd. The 3rd is as innocuous and audible/inaudible as the 2nd (however you want to put it) in that it adds 'warmth' and 'bloom'; both audiophile terms for the presence of either of these harmonics. To this end, you cannot hear the difference between the two if that is all that is present. IOW the 3rd is not any more obvious than the second. The audibility of the higher orders begins, it seems, with the 5th.

But in any system both harmonics are always present to some degree. Perfect cancellation does not occur in push-pull circuits and single-ended circuits will always produce a 3rd harmonic.

But since the presence of the 2nd and 3rd can mask higher ordered harmonics, what happens is that a circuit that expresses a 3rd harmonic as its main distortion product will be perceived as being more neutral than that which expresses a 2nd as the main product. This is because is a circuit that expresses the 3rd, the actual product is often a lot less than it is with one that produces a 2nd, and the succeeding harmonics are lower in amplitude as well.

I know that people really like SETs for what they do, but to be clear a lot of that is simply how they make distortion. What people almost never do is to compare apples to apples: when comparing an SET to a push-pull amp, the latter is usually some multiple of power of the former. To add to that, usually the push-pull amp has a pentode output circuit. What happens if the output circuit is the some DHT that the SET in comparison uses? What if the P-P amp has the same full power as the SET? These issues seem unexplored but most SET aficionados.

But I have explored this a bit; I've compared a type 45 SET to a push-pull SET, using the same input tube (6SN7) and also compared several SETs to a P-P amp using EL95s that only makes 5 watts. In all cases the sonic merits of the P-P amp over the SET were obvious. Now my results are anecdotal; I encourage others to make similar comparisons, doing as much as possible to eliminate the variables. When you think about it, you really don't see that happening; nearly all comments made by people that prefer SETs are anecdotal as well.

This is why I brought up the issue of the 3rd harmonic. If people understand how the ear perceives the various harmonics, then they can build circuits that suit without any loss of smoothness or detail. To this end, this is part of why I think when people compare SETs to P-P, its apples and oranges, since if you have a P-P output but a single-ended input, you'll get a bit more of the 5th, more than you would get with an SET. But if you use a differential input this doesn't happen so you get a smoother sound, not unlike an SET, even if the amp employs pentode outputs.

 

RE: The 3rd harmonic, posted on January 4, 2021 at 11:38:15
Paul Joppa
Industry Professional

Posts: 7295
Location: Seattle, WA
Joined: April 23, 2001
 Thanks, Ralph - that's a very clear expansion, and a new perspective for me. Much appreciated!

I've only heard a couple push-pull DHT amps with no feedback, and I think both of them had substantial 2nd harmonic from the driver stages. I'll do a little more exploring.

 

RE: The 3rd harmonic, posted on January 4, 2021 at 13:07:27
Ralph
Manufacturer

Posts: 4769
Location: Minnesota
Joined: April 24, 2002
 I put up a photo of a little amplifier I made for my bedroom system elsewhere on this forum. The amp is based on EL95 power tubes and makes 5 watts.

Since the output section is cathode biased using a common cathode resistor, it is differential. What attracted me to the tube was its small size and the fact that you could run it with a fairly high value grid leak resistor. It also has a fairly high mu so its easy to drive.

So I built a 12AT7-based differential voltage amplifier/driver for it, using a solid state CCS tied to a B- of 300V. The CCS is a two stage device, allowing it to reject power supply noise as well as regulate current very effectively (single-stage CCS circuits always leave some performance on the table). In this way the 12AT7 has measurably the same gain from either input, and its performance does not measurably change despite AC line changes from 95 to 125VAC. One side is used for signal and the other for feedback, the latter mostly to reduce gain as the entire circuit has good linearity owing to the ultra-linear output. I find it amusing that the negative feedback has to be in phase with the input signal for it to work. The idea here was to avoid IMD generated at the feedback node; this technique does that effectively. Owing to the output transformer being fairly small, it has open loop bandwidth to 100KHz. It has enough gain that I can drive it to full output with an old tuner; its got a volume control built in along with two inputs.

The primary distortion product is the 3rd harmonic. It does have some 2nd as the circuit isn't perfect of course. Overall its very smooth and despite running pentodes and feedback (although its class A) it easily challenges SETs of similar power in terms of soundstage, detail, smoothness and bandwidth. It lacks the 'dynamic' character for which most SETs are known, but that character is really caused by higher ordered harmonics showing up on transients. Since the ear uses those harmonics to sense sound pressure, you get distortion masquerading as 'dynamics'. Since this amp lacks the higher ordered harmonics at those levels, it also sounds less 'dynamic' but its not boring either.

Once I realized that the 3rd harmonic is as benign as the 2nd, it simply became a matter of designing fully differential circuits to avoid the 2nd and the often attendant 5th generation (present when single-ended and balanced circuits are combined) as much as possible.

 

and yet..., posted on January 4, 2021 at 15:17:24
PakProtector
Audiophile

Posts: 12360
Joined: May 14, 2002
 your NFB signal is SE.

I slightly disagree with your compare based on power.

PP is going to need two finals, so double the SE amp(pse). Not quite an apples-apples comparison. Part of the benefit to PP is more power...if you want SE for some reason, live with that limitation...:)

I think you'll find that the magnetic behaviour of a gapped signal Iron noticeable. For the amps I have built of late, this gap is delivered to its CT'd grid choke. The L vs signal magnitude plot is different. SE nearly has to do it with its OPT( even if the OPT is 'distributed', parafeed style ).

I have found it easier to build an amp that is Class A. AB makes a radically different load on the PS. In Class A the non-linearities are what presents an AC load, so a performance restriction from the PS could be considered a benefit. Now I don't tend to build a lightweight PS, even for thoroughly Class A amps, so take that as supposition.

While we are at this, who is making Class A PP for sale?
cheers,
Douglas


Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: and yet..., posted on January 4, 2021 at 16:02:27
rivervalley817
 Border Patrol?

[no affiliation & never heard one]

regards,

 

RE: and yet..., posted on January 4, 2021 at 16:07:04
PakProtector
Audiophile

Posts: 12360
Joined: May 14, 2002
 soo...if you can get 8W from a SE 300B, that leaves you 16 if you go PP. Same thing as bridging a pair of SE amps( not paralleling ). So, no I don't think that one is Class A.
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: and yet..., posted on January 4, 2021 at 17:58:59
dave slagle
Manufacturer

Posts: 5430
Location: NYC
Joined: April 27, 2001
 Not sure about who is currently making class A PP ams but I do recall Allen Wright offering some PP 300B amps with his "black box bias" for the finals which essentially was a CCS. This would by definition force the output stage signal loop into a series circuit mandating class A behavior.

dave

 

Yup. nt, posted on January 4, 2021 at 18:40:38
Tre'
Industry Professional

Posts: 17292
Location: So. Cal.
Joined: February 9, 2002
 .
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Yup. nt, posted on January 5, 2021 at 03:35:57
PakProtector
Audiophile

Posts: 12360
Joined: May 14, 2002
 OK, so that makes exactly one so far. It is not common, that is for sure...LOL It does follow the traditional, 'power game' build/marketing philosophy.
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: Yup. nt, posted on January 5, 2021 at 06:44:11
dave slagle
Manufacturer

Posts: 5430
Location: NYC
Joined: April 27, 2001
 I think the issue here comes down to grey area and definitions. I know tre' has a definition of "Class A" which essentially boils down to tubes run in the "most linear region" opposed to the more traditional definition which only requires conduction. The "most linear region" caveat is still a huge grey area open for interpretation and exploitation. It simply comes down to how much specsmanship one wants to play. I agree with 8W/16W as the theoretical max for class A SE/PP 300B designs but sadly the consumer thinks there is a huge difference between 8 and 12W so marketing departments start bragging how their brilliant engineers can get 10W, 12W or even 20W out of a single SE300B. The only two legitimate number games that can be played here involve relaxing the distortion spec or allowing the driver to distort with the outputs to null the evens. The SE case is pretty simple and clear cut since the distortion is easily quantifiable and there is not "class AB" lurking in the shadows to fix the numbers. Moving the PP and the availability of class B to cook the books opens a whole new can o' squiggly things. The transition from deep class A 16W PP 300B's to 20Wpp 300B's like gary at Border Patrol still seems to me to be safe within the confines of class A. Taking that number to 24W or 30W just exploits the grey area of class A even more and at some point you need to cry foul.

The reason I brought up Allen Wright is his method with the CCS is unique in that it removes the possibility of class B from the picture and falls into the realm of SE class A. I also seem to recall a group of people from the Pacific Northwest doing something similar at about the same time (gary pimm?) and some pretty heated discussions about the merits of this approach decade or more ago.

dave

 

"a whole new can o' squiggly things" :-) Nice post dave, thanks. nt, posted on January 5, 2021 at 07:24:29
Tre'
Industry Professional

Posts: 17292
Location: So. Cal.
Joined: February 9, 2002
 .
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Isn't this based on the harmonic series in music?, posted on January 5, 2021 at 07:43:34
Tre'
Industry Professional

Posts: 17292
Location: So. Cal.
Joined: February 9, 2002
 "harmonics of E would be octaves at ~82,164,328, etc....."

If 82Hz is the fundamental, the second HD would be 164Hz and the third HD would be 246Hz, the forth HD would be 328Hz, etc.

The octaves would be 2nd, 4th, 8th, 16th, etc.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: and yet..., posted on January 5, 2021 at 08:51:45
kyle
Audiophile

Posts: 1839
Location: London Ontario
Joined: September 29, 1999
 I have no idea how they run their tubes but according to some old 300B operating point sheets, you can get 10.5 watts at 400V and 60mA into a 3500 ohm transformer. Apparently you can over 17watts out of a 300B at max conditions. I wouldn't bet on good sound or a tube lasting more than a few weeks.
I have made amps using operating points to match existing iron that I had sitting around from these sheets and they sound good and in one case I used the amp for several years without undue wear on the tubes. They still tested as close to average.

 

What I wanted was low power and high quality in a small package, posted on January 5, 2021 at 09:32:11
Ralph
Manufacturer

Posts: 4769
Location: Minnesota
Joined: April 24, 2002
 I looked at SE amplifiers that were about the right physical size and power and they were all junk, mostly made in China. Funny, you want good quality parts and execution in something lower power and there really aren't many out there.

So I designed it myself. Its working well enough that we are going to make more of them, probably as a short run (it cost nearly as much to make 10 chassis as was going to cost for just one).

Its class A1: at full power the current draw is identical to idle.

PP is going to need two finals, so double the SE amp(pse). Not quite an apples-apples comparison. Part of the benefit to PP is more power...if you want SE for some reason, live with that limitation...:)

I'm not making out what you mean in the first sentence.

It is true that P-P can make more power, but how does it sound if it makes the same power? One of the problems with P-P is that at some low power point, the distortion starts to go up as power is decreased. This is why if you really want to compare SET to P-P the amps should be very similar power levels. This is all about that first watt, which this amp does very well. We know that OPTs tend to be wider bandwidth when they are lower power, so this could be a benefit (as well as not really having a limitation with core saturation). The circuit runs a total of 3 tubes, a 12AT7 and a pair of EL95s. Its pretty simple. I could have run balanced feedback easily enough but it really wasn't going to make any difference since the CCS circuit is very effective, combined with the mu of the 12AT7. This makes for a high performance differential amplifier as far as tubes are concerned, with a CMRR that's in the high 90dB region.

Most of the compact SE amps I looked at were not as compact as I was able to get with this design. Nor are they as low distortion or as wide bandwidth. The type 45 SET to which I compared it to is nearly 4x the size, owing to it being monoblocks with considerably larger magnetics. It sounds no-where near as transparent as the EL95 amp!

The link is to the AA post of this amp. I can supply a schematic. FWIW the corners of the chassis are welded and ground; since its made of 0.050 steel, it will polish and chrome quite nicely. I used pressed-thru PEM studs for mounting parts so as to minimize hardware on the apron. Since the OPTs were supplied with a blue finish, I had the powdercoater match it and do the same with the end bell on the power transformer. The rear panel has an IEC connection and the amplifier meets EU directives so it meets UL as well.

So that's 'whose selling'... Although I really didn't make this to sell- this is for the system in my bedroom. I have the matching STA-175C amplifier that goes with that tuner in the photo, which I had completely refurbished (it uses an output circuit that makes a 2nd harmonic as its primary distortion product, so sounds pretty smooth as long as you don't push it). That is where this started; I began to wonder if I could build a tube amp that would be lower distortion, the same power and wider bandwidth that could fit in the same space. Turns out that is a 'yes'. Another consideration is I wanted something that really wouldn't get in trouble if the power tubes failed and simply wasn't going to make all that much in the way of heat (I used ceramic sockets 'cause I could...). To that end, the OPTs are rated for double the power- you could run 6AQ5s in this amp because the pin base is the same (I've not tried that, a potential issue I've not looked at is the value of the cathode resistor for the 6AQ5s).

I thought about using a 6SN7 or 6BL7 as a power tube since they would load correctly with this OPT and make enough power for my purposes (its also meant to be suitable as a desktop or headphone amp). I'm a fan of those tubes but they literally wouldn't fit.

 

RE: What I wanted was low power and high quality in a small package, posted on January 5, 2021 at 13:03:58
PakProtector
Audiophile

Posts: 12360
Joined: May 14, 2002
 Parallel SE( with just two ), should make for a power matching amp to a Class A PP amp. Since you think it is power that should be matched, that is one way to do it.

In any case, I've not much interest in SE amps, except as a target held up by their devotee's to eclipse with a PP amp during a listening session... :) Or perhaps a geetar amp.
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

I probably could have gotten the power SE, posted on January 5, 2021 at 13:45:36
Ralph
Manufacturer

Posts: 4769
Location: Minnesota
Joined: April 24, 2002
 But that makes the OPT trickier, and you have to be careful about filaments for low noise, and the power supply becomes more critical. It worked out to be a bigger unit, and very likely not sounding or measuring as well.

I used the Dyna ST-70 as a bit of a standard (although I used better parts); I used the same filter capacitance values it does, along with a choke for smoothing. So the plates of the power tubes are seeing fairly clean power at the CT of the OPT, although they have some rejection of power supply noise anyway due to the differential effect. But instead of making 35 watts/channel, this amp only makes 5 so I expect the power supply has far more capability than needed. I seriously doubt I run it past even 1/2 watt.

 

RE: Yup. nt, posted on January 6, 2021 at 03:39:52
PakProtector
Audiophile

Posts: 12360
Joined: May 14, 2002
 mmmmhhhh...squiggly things.

In order to remove some of the grey area, 'none' is pretty absolute, so 'few' might be a better way to categorize Class A, PP designs. 'Uncommon' is just as good.

In any case, it is far better than the 'Class A to x Watts, and B( or even worse, AB) at XX Watts.
cheers,
Douglas
Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

Page processed in 0.044 seconds.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2024, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]