The all DHT SET Headphone Amp - Page 58 - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 5th March 2013, 04:29 PM   #571
12B4A is offline 12B4A  United States
diyAudio Member
 
12B4A's Avatar
 
Join Date: Aug 2012
Location: Right here, on this screen
FWIW, I got the Edcor 10k-7k/300-32 in and managed to hook it up to a magnetics analyzer for a very brief time. At 10Vrms @ 30hz connected 10k:32 I measured a Primary inductance of 130H, Primary DCR 123, and a Secondary DCR of 6.5. The point at which it turns from inductive to capacitive is around 2.45khz.
  Reply With Quote
Old 5th March 2013, 04:44 PM   #572
iko is offline iko  Canada
diyAudio Moderator
 
iko's Avatar
 
Join Date: May 2008
Location: Toronto
Quote:
Originally Posted by jpak View Post
iko can you elaborate on why 307a beats the 300b? 300b properly driven is very linear, IMO it rightfully deserves a bad rep when driven from something like 6sn7.

307a tubes are still out there for not much money, you just have to ask dealers. Five pairs pictured, another 6th not shown I think I will be using these on the output of the headamp. Only downside most of these are black glass hiding the beautiful grids
I can only speak from simulating output stages with a "perfect" input signal. The 307A had lower distortions when compared with various 300B models that I found around the net. However, search the forum and you'll also find anecdotal reports that it's better in reality. Perhaps not conclusive, but it's an easy choice for me, as I don't want to fork the $$ for good 300B samples, and I already have a good number of 307A. It's good to know that they can be found for reasonable prices still. Two years ago or so I couldn't find them anywhere below $50 a piece.
  Reply With Quote
Old 6th March 2013, 12:38 PM   #573
regal is offline regal  United States
diyAudio Member
 
Join Date: Jan 2004
Location: MD
Quote:
Originally Posted by 12B4A View Post
FWIW, I got the Edcor 10k-7k/300-32 in and managed to hook it up to a magnetics analyzer for a very brief time. At 10Vrms @ 30hz connected 10k:32 I measured a Primary inductance of 130H, Primary DCR 123, and a Secondary DCR of 6.5. The point at which it turns from inductive to capacitive is around 2.45khz.
What DC current levels? I guess I would consider that transition from inductive to capacitive very high, as one transfo winder would say it will sound like a cap coupled OTL (not a bad thing imo), this must be a small parafeed transformer?



Quote:
Originally Posted by ikoflexer View Post
Perhaps not conclusive, but it's an easy choice for me, as I don't want to fork the $$ for good 300B samples, and I already have a good number of 307A. It's good to know that they can be found for reasonable prices still. Two years ago or so I couldn't find them anywhere below $50 a piece.
There must have been a score of military issue 307A's recently. I guess I am going with this tube on the output ver the 300B mainly because they were all made by Western Electric under contract, and I haven't heard a bad tube from that old company yet. I understand the 306A may have better linearity but less B+ capability? This would put the 306A at the same disadvatage as the 4P1L, 2A3, etc.

Linearity for driving the higher impedances seen with headphones is directly proportional to the B+ and Grid voltage operatig points (why the gm-70 model shows practically zero distortion driving any headphone.)

What makes the 307A "a bit" better than the 300B IMHO is the high operating voltage and a bonus is the hgher mu (less strain on the front end) along with lower current (lower current = better opt.)

The perfect triode for headphones would be the GM 70 or 4-65A with about 5700-800V B+ series or shunt regulated supply. Since we hear distortion more with headphones we want to avoid A1/A2 getting near a 0V grid. But i'm too chicken to strap a pair of headphones on where the only thing separating me from 500V+ is a transformer's insulation.

I'm finding several references to success with a 1:2 stepup transformer, both my DAC's have <30 ohms output impedance and plenty of discrete current so I am really leaning toward a 4P1L-307A and I "think" I would rather steupup at the input than interstage. There are well regarded stepup interstage transfo's from Llundahl and Transcendar and the 4p1L is the perfect tube for this since it has a low Rp but it is much less expensive to buy an input stepup transformer, may turn out I need both, so buying a chassis will be the last thing I do.

I haven't ruled out stacked supplies but I think it will require a custom power TX. When one does a dc coupling with stacked SSHV's what B+ is stacked on what, I'm still confused on the best configuration.
  Reply With Quote
Old 6th March 2013, 01:51 PM   #574
jpak is offline jpak  United States
diyAudio Member
 
jpak's Avatar
 
Join Date: Jun 2007
Location: NH
Looking forward to seeing what you come up with. I want to build another IDHT amp (parafeed) before tackling this, I am thinking about building the circuit iko posted in #138.

Last edited by jpak; 6th March 2013 at 02:13 PM.
  Reply With Quote
Old 6th March 2013, 02:15 PM   #575
12B4A is offline 12B4A  United States
diyAudio Member
 
12B4A's Avatar
 
Join Date: Aug 2012
Location: Right here, on this screen
Quote:
Originally Posted by regal View Post
What DC current levels? I guess I would consider that transition from inductive to capacitive very high, as one transfo winder would say it will sound like a cap coupled OTL (not a bad thing imo), this must be a small parafeed transformer?
No DC since this is intended for parafeed. I'm not sure how someone can say a transformer will give a circuit the sound characteristics of a circuit topology based solely on the inductive/capacitive transition. I have seen it mentioned years ago that audio winders intend to keep this transition out of mid-range band (above 2khz?) but I had not seen a technical explanation as to why. I have tested some toroid power transformers as parafeed outputs and their transition point has varied from 120-200hz and the resulting sound quality seemed quite alright to me. The hammond 119DA's transition point (a 600:8 that's been mentioned several times for parafeed headphone circuits) was just short of 7khz.
  Reply With Quote
Old 6th March 2013, 02:37 PM   #576
работник
diyAudio Member
 
Rod Coleman's Avatar
 
Join Date: Jul 2004
Location: Warwickshire UK
Quote:
Originally Posted by regal View Post
......... I guess I am going with this tube on the output ver the 300B mainly because they were all made by Western Electric under contract, and I haven't heard a bad tube from that old company yet. I understand the 306A may have better linearity but less B+ capability? This would put the 306A at the same disadvatage as the 4P1L, 2A3, etc.

Linearity for driving the higher impedances seen with headphones is directly proportional to the B+ and Grid voltage operatig points (why the gm-70 model shows practically zero distortion driving any headphone.)

What makes the 307A "a bit" better than the 300B IMHO is the high operating voltage and a bonus is the hgher mu (less strain on the front end) along with lower current (lower current = better opt.)

.
Although the pentode-mode specs allow 500V, I don't think this will translate into 500V for triode.

The WE data sheet gives maximum operating conditions as an envelope in the attached diagram, for class-A pentode.

See that the 60mA rating is already derated to 50mA at 300V.

For triode, the voltage can probably increase a bit, maybe 340-380V, but at 300B levels, I would expect real trouble.

Not only that, but the linearity will get much worse at high voltage. We need some real measured curves to see where the degradation sets in (this happens for almost any power tube, but the onset voltage varies a lot).

Personally, I would find some way of obtaining real measurements before spending money on transformers.

The fact that the data-sheet says explicitly that the class-A performance of the 307A was compromised in order to get good class-C operation is a warning. It gives us exactly zero confidence that the WE 307A can compete with the same company's 300B design, where class-A linearity was the actual design goal. But still, measured curves will answer this question, too.
Attached Images
File Type: jpg 307A_curv.jpg (164.6 KB, 238 views)
  Reply With Quote
Old 6th March 2013, 10:24 PM   #577
diyAudio Member
 
Join Date: Mar 2003
Location: Zürich
Quote:
Linearity for driving the higher impedances seen with headphones is directly proportional to the B+ and Grid voltage operatig points (why the gm-70 model shows practically zero distortion driving any headphone.)
I still don't get this.

A 300B, for example, can produce 8W of power in a (reflected) load of 3k2. If the secondary of the OPT is designed for an output of 8R, the turn ratio is 20:1, and there will be 8V RMS at the secondary output. If the secondary is designed for 40R, the turn ratio will be about 9:1 and there will be 18V RMS on it. 18V RMS in 40R produces, of course, about 8W (18^2 / 40). So the 300B produces the required swing on the secondary to drive the HE6. If the headphones were 3k2 one could make a 'parafeed' arrangement, with a choke in the 300B place and couple to the headphone with a cap, and still have enough swing to pump the 8W into the headphones.

I am quite sure that indeed the GM70 can produce 6W with less distortion than a 300B does, and maybe that difference can be heard better on headphones and that is what you are after (lower distortion at 6W output). But it is not because a GM70 'matches' better with the headphone due to higher operating voltage, I think ?!

Can somebody explain what I am missing?

many thanks, Erik
__________________
my surname is indeed 'de Best': neither misspelling nor snobbism! Ask SY!

Last edited by ErikdeBest; 6th March 2013 at 10:26 PM.
  Reply With Quote
Old 7th March 2013, 01:48 PM   #578
regal is offline regal  United States
diyAudio Member
 
Join Date: Jan 2004
Location: MD
Quote:
Originally Posted by Rod Coleman View Post
The fact that the data-sheet says explicitly that the class-A performance of the 307A was compromised in order to get good class-C operation is a warning. It gives us exactly zero confidence that the WE 307A can compete with the same company's 300B design, where class-A linearity was the actual design goal. But still, measured curves will answer this question, too.
Yes and a WE engineer making a slight compromise in linearity in the 40's is better than anything we can expect with todays materials. Making god tools is largely materials engineering/science and they just don't make 'em like they used to ?



Quote:
Originally Posted by ErikdeBest View Post
I still don't get this.

A 300B, for example, can produce 8W of power in a (reflected) load of 3k2. If the secondary of the OPT is designed for an output of 8R, the turn ratio is 20:1, and there will be 8V RMS at the secondary output. If the secondary is designed for 40R, the turn ratio will be about 9:1 and there will be 18V RMS on it. 18V RMS in 40R produces, of course, about 8W (18^2 / 40). So the 300B produces the required swing on the secondary to drive the HE6. If the headphones were 3k2 one could make a 'parafeed' arrangement, with a choke in the 300B place and couple to the headphone with a cap, and still have enough swing to pump the 8W into the headphones.

I am quite sure that indeed the GM70 can produce 6W with less distortion than a 300B does, and maybe that difference can be heard better on headphones and that is what you are after (lower distortion at 6W output). But it is not because a GM70 'matches' better with the headphone due to higher operating voltage, I think ?!

Can somebody explain what I am missing?

many thanks, Erik

Because there is no such thing as a 50 ohm FIVE Watt secondary, they only exist on web forums and possibly AudioNotes $2500 HE6 KIT.

Designing for 50 ohm phones that require 5 watts is the equivalent of asking for a 35 Watt SET into 8 ohms same requirements as far as primary voltage and trying to shave this down with a high DCR Secondary (lower ratio) does not work without going to extreme lengths such as double C-core amorphous whatever, stuff nobody can afford. Its easy to end up with a transformer thats about 60% efficient unless you have a deep wallet. Start looking at the 300B curves where 60% of the power is wasted. Only chance is A2 with 300B and under tubes.

Do you see why the GM70 and similar HV group of tubes is ideal for this task?
  Reply With Quote
Old 7th March 2013, 02:17 PM   #579
diyAudio Member
 
Join Date: Mar 2003
Location: Zürich
Hi Regal,

thanks for the explanation, so it is indeed mostly about DCR on secondary.

I agree that one needs a large double C core for 8W of Single Ended power when one wants wide response and low DCR. I had a pair of audionote OPTs that confirm that. IIRC Pieter (from tribute audio) spoke that his SE transformers for 300B have about 0,2R on the 8 ohm secondary. If that secondary is made of 4 0,8R windings in parallel, rearranging for series parallel would result in 0,8R DCR output on a 32 ohm tap. But indeed, those are expensive transformerts (even though a pair is still probably much cheaper than the HE-6).

But a 35W OPT for a GM70 will also be expensive. Maybe a 35W (SE) EI GOSS core costs less than a smaller (8W SE) double C-core, but it will be a monster, and those dimensions will make that each secondary winding will be long and therefore contribute to high DCR as well, I think (sorry, can't do the transformer math).

And a high voltage amp brings all other sort of costs - and risks - with it.

Now this all sounds very rational (and maybe it is just wrong), and at least in the audio hobby I am not rational myself (just trying to not kill myself and others with high voltages, but for the rest all sort of extravagancies are allowed), therefore I repeat that I only asked because I didn't understand.

many thanks! Erik
__________________
my surname is indeed 'de Best': neither misspelling nor snobbism! Ask SY!
  Reply With Quote
Old 7th March 2013, 02:20 PM   #580
работник
diyAudio Member
 
Rod Coleman's Avatar
 
Join Date: Jul 2004
Location: Warwickshire UK
Quote:
Originally Posted by regal View Post
Yes and a WE engineer making a slight compromise in linearity in the 40's is better than anything we can expect with todays materials. Making god tools is largely materials engineering/science and they just don't make 'em like they used to ?
It's possible, but this could go either way!

My point was that we need some kind of reliable characterisation of the 307A - in triode mode - before dropping cash on the iron. Nothing about the data sheet suggests more than middling performance, and 50-70 years lying around soaking up gas does not improve anything - gassy grid current is noisy, for starters.

(I speak for myself, but) I don't see the worth in designing with an unknown quantity. Real curve tracing, like the Audiomatica samples linked earlier, will allow:

- meaningful (rather than anecdotal) comparison with 300B;
- checking grid current in ancient samples;
- checking the operating point (especially voltage) that gives reasonable linearity - before the high-voltage degradation sets in.

OTOH, the latest 300B EH Gold is very well made (a major improvement on previous issues), despite moderate cost, and sounds excellent, provided only that the B+ and Filament supplies are worthy of it.

And this 300B is a 40W anode, rather than 15W+ some fraction of 6W for G2. If we are looking for headroom in our head-fi, the 300B gives much more.

I suppose one could build with a 360V B+ and a 5000-ohm OT, and try them both.

But relying on a driver circuit that can only support 307A and not reach the extra for a 300B, seems like asking for trouble, IMV.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
High end Ortho Amp on a basic Russian Tube DHT SET DIY design regal Headphone Systems 2 7th December 2012 12:46 PM
300b all-DHT SET andyjevans Tubes / Valves 16 24th August 2012 09:06 PM
Help on designing full DHT SET power amplifier milen007 Tubes / Valves 56 11th August 2011 10:38 PM
Class A, DHT driven, headphone amp iko Headphone Systems 100 26th March 2011 03:54 PM


New To Site? Need Help?

All times are GMT. The time now is 01:33 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2