The all DHT SET Headphone Amp - Page 5 - 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 28th January 2013, 10:30 AM   #41
diyAudio Member
 
Join Date: Sep 2003
Location: London
I've used ECC40 before and it's a really good indirectly heated tube - the origin of the E80CC. But in PP I'd stay with DHTs and look at 3a5 and 1J6. I've used both of those before and they're both excellent - better than ECC40 to my ears. 3a5 to 1J6 would probably do very nicely. In fact that's rather a good idea. But if I could use the PP LL1689 in SE at 3 or 4ma I'd still prefer to go with a 26.
  Reply With Quote
Old 28th January 2013, 10:54 AM   #42
diyAudio Member
 
merlin el mago's Avatar
 
Join Date: Sep 2009
Location: Catalonia - Europe
Quote:
Originally Posted by Rod Coleman View Post
Never!

A compound triode - bipolar stage, with a 2nd-order dc bias controller, to give zero dc offset, no output capacitor at all.

I'll finish drawing it, if there's interest.
+1
  Reply With Quote
Old 28th January 2013, 11:00 AM   #43
45 is offline 45  Italy
diyAudio Member
 
45's Avatar
 
Join Date: Dec 2008
Location: UK
Quote:
Originally Posted by andyjevans View Post
I've used ECC40 before and it's a really good indirectly heated tube - the origin of the E80CC. But in PP I'd stay with DHTs and look at 3a5 and 1J6. I've used both of those before and they're both excellent - better than ECC40 to my ears. 3a5 to 1J6 would probably do very nicely. In fact that's rather a good idea. But if I could use the PP LL1689 in SE at 3 or 4ma I'd still prefer to go with a 26.
As you like of course. For me ECC40 is the by far best of the those and I do prefer Class A PP. Only the 26 can be marginally better in terms of linearity but requires a lot more care in filament supply and is a poor choice in terms of efficiency. The E80CC, regardless of its presentation as a SQ valve, is nowhere near the ECC40 from any point of view.

Cheers
  Reply With Quote
Old 28th January 2013, 11:36 AM   #44
diyAudio Member
 
Join Date: Sep 2003
Location: London
Yes, indeed the E80CC isn't better than the ECC40 despite its high prices. The ECC40 was very popular in Europe for good reason in its day and was used in good equipment. But I think on this thread we'll be looking at DHTs!
  Reply With Quote
Old 28th January 2013, 01:14 PM   #45
работник
diyAudio Member
 
Rod Coleman's Avatar
 
Join Date: Jul 2004
Location: Warwickshire UK
Quote:
Originally Posted by Rod Coleman View Post
Never!

A compound triode - bipolar stage, with a 2nd-order dc bias controller, to give zero dc offset, no output capacitor at all.

I'll finish drawing it, if there's interest.
Here's my serving suggestion, for any DIYers who don't like the cost of a real transformer. The 4P1L + Sowter is very appealing, but any compromise on the transformer will bring disappointment.

Don't be put off by the transistors in this drawing.

Q1 is a little bias source.

Q3 and Q4 are an error servo, with a 2nd-order sub-audio filter. These keep the output within a few mV of ground, without interacting with the music.

M1 is a class-A current sink, set at 250-500mA, depending on the desired peak current into the headphones. Please trade off the peak current capability with the idle power burn. (15 - 30W each in M1 and M2).

The actual output stage is a new JJ E88CC running at 60V and 10mA. It's a cathode follower - but assisted by the audio FET BUZ905, made by Semelab.

The BUZ905 is simply carrying any current demand, above the 10mA standing current in the E88CC. The linearity is much better than a normal cathode follower, because the load variation is so small.

The BUZ905 is chosen because of the high power handling of the TO3 package, and the zero-TC (temperature coefficient) which occurs at < 0,5A (rather than the whole amperes needed by the usual switching FETs). This ensures that the idle current is stable, even at intervals faster than the servo can control it.

The E88CC is chosen, because it has high gm (for good follower performance), and because it likes low voltage. I don't think the 4P1L would work well here.

The overall cost of parts is relatively low, and the performance is very high. No capacitors, chokes, or transformers are needed in the output, because the BUZ905 is performing the impedance transformation from the E88CC to 38 ohms.
Attached Images
File Type: jpg HeadFiCompound.jpg (143.4 KB, 557 views)
  Reply With Quote
Old 28th January 2013, 01:19 PM   #46
работник
diyAudio Member
 
Rod Coleman's Avatar
 
Join Date: Jul 2004
Location: Warwickshire UK
FFT for 1V peak out.

.
Attached Images
File Type: jpg HeadFiCompound1_FFT.jpg (32.4 KB, 536 views)
  Reply With Quote
Old 28th January 2013, 01:24 PM   #47
работник
diyAudio Member
 
Rod Coleman's Avatar
 
Join Date: Jul 2004
Location: Warwickshire UK
all the files for LTSpice fun, with this circuit.
Attached Files
File Type: zip compoundE88CC_headFi.zip (154.8 KB, 44 views)
  Reply With Quote
Old 28th January 2013, 01:35 PM   #48
diyAudio Member
 
Join Date: Sep 2003
Location: London
Wow - good work. How would this work with a 4P1L? Would it need 2 stages?
  Reply With Quote
Old 28th January 2013, 01:44 PM   #49
работник
diyAudio Member
 
Rod Coleman's Avatar
 
Join Date: Jul 2004
Location: Warwickshire UK
4P1L would not like the low +60V supply. You could increase the supply to 100V, and although the 4P1L might work better, the power burn in the BUZ905 would then be 25-50W.

Going further into the class-A room heater world, you could upgrade to the more expensive BUZ906 and up the supply to 150V, if the 4P1L likes that better. the BUZ906 heatsink would then need to handle 40-75W in this case!

But it would still have a gain of almost 1, like the E88CC version. So you can drive it with whatever stage gives the peak voltage you need, for the headphones it used with.

If 3V peak is high enough, you can connect directly to a DAC-level output. The input capacitance is very low, since Mr Miller has been denied access to this party.
  Reply With Quote
Old 29th January 2013, 11:52 AM   #50
regal is offline regal  United States
diyAudio Member
 
Join Date: Jan 2004
Location: MD
Quote:
Originally Posted by 45 View Post
.

The original Class A PP amp in the datasheet uses 30K plate-to-plate and about 5 mA for 0.5W Pout while the single ended version uses 15K at 250V/6mAfor .28W Pout.
A class A PP 4P1L all DHT design would be interesting it could double as a nice speaker amp and also power HE-6's.

But I don't know about the filaments do the Coleman regs work with PP, if so how?




Quote:
Originally Posted by andyjevans View Post
Wow - good work. How would this work with a 4P1L? Would it need 2 stages?
Good question.

Rod, I appreciate the input but there are a million hybrid headphone amp designs out there and none of them are DHT, neither is yours. In fact I bought a used "Lyr" for $250 that performs just as well as what you posted, so there is zero incentive to DIY another indirect heated hybrid.

Change that tube to a #26 or something DHT and at least make an attempt to contribute to the thread subject

Btw is a Buz905 a lateral ? I am just busting your balls a little, we wouldn't even be TALKING about an all DHT headphone amp were not for you and iko sharing your filament power designs


On another subject in my quest for an output transformer this company got back to me. http://goldenmiddle.com/images/is6.pdf Looks interesting sort of old school preampish transformers, I think I'm reading that the primary is 3K which would work for PSE 4p1l or 2A3. Price is $230 eur not including postage.


One thing I am afraid of is investing in quality OPT's and then in a few months wanting to hook the amp up to the pair of Fostex127E's I have hidden in the garage, or one day I open our local classifieds and I see a pair of Klipsch La Scala's for $100. I don't think it would be too much scope-creep to want an 8 ohm tap on the secondary ?


Btw I also think it is possible to DC couple #26 to 4P1l and still use filament bias, here the cap under the 4p1l is below the signal and just for noise reduction, I think it can be left out.
Attached Images
File Type: png 26 to 4p1l DC fil bias.png (15.2 KB, 493 views)

Last edited by regal; 29th January 2013 at 12:03 PM.
  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 11:46 AM
300b all-DHT SET andyjevans Tubes / Valves 16 24th August 2012 08:06 PM
Help on designing full DHT SET power amplifier milen007 Tubes / Valves 56 11th August 2011 09:38 PM
Class A, DHT driven, headphone amp iko Headphone Systems 100 26th March 2011 02:54 PM


New To Site? Need Help?

All times are GMT. The time now is 09:33 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