PP Amplifier Advice Sought: EL34/KT88 UL/Triode - 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 25th August 2014, 04:11 PM   #1
danhei is offline danhei  Hong Kong
diyAudio Member
 
Join Date: Dec 2004
Location: Hong Kong
Default PP Amplifier Advice Sought: EL34/KT88 UL/Triode

Hello All,

Iíve recently picked up a pair of Tamura F-925 PP output transformers (5K, 30W, 200mA, UL taps) to build a stereo amplifier. Iíve built a number of guitar amps before but only one stereo amp (6V6GT PP UL, simple but sounds good). Since the Tamuras should be nice transformers, Iíd like to have a correspondingly nice circuit to let them shine.

All the usage information Iíve seen on these transformers are in Japanese, which I will link to for reference as the schematics are understandable:

Given those designs, and with the parts I have on hand, Iím leaning towards an amplifier that can use EL34s or KT88, in either UL or triode mode (possibly switchable), although not necessarily following one of the Japanese designs. After much searching, I see three amplifier types frequently recommended at DIY Audio: Eli Duttmanís El Cheap, Gingertubeís Baby Huey, and Poindexterís Musical Machine. I understand all of those designs could require modification for the bigger output tubes.

I would like approx 30W per channel output, and will build as a stereo amplifier with one power transformer. I have a set of Shuguang EL34 output tubes to be used, as well as one pair of Svetlana KT88s (which I can try to decide if I prefer the sound to EL34). Would like to use noval socket preamp tubes, hoping to keep it to a total of three preamp tubes for both channels. I have access to all of the 12A_7 family, ECC88, 12BH7, 6AN8, among other common and not-so-common noval preamp tubes.

Iím willing to use solid state components where preferred (e.g. regulated voltage, bias servo, MOSFET CCS, etc). Would like to keep the signal path itself all-tube, if only to be able to say itís an ďall-tube amplifierĒ. But that requirement is not firm, if it turns out that, say, MOSFET drivers make a better design fitting with the preamp tube limitations.

Thinking B+ should be around 430-440V with fixed bias.

I would appreciate any suggestions on the amplifier topology, including UL vs triode (is it worth having this function switchable or better to design around one more of operation only?), output tube choice, preamp tube choice, phase-inverter design, gain stage(s), driver stage(s), global NFB, and which, if any, of the frequently recommended DIY Audio designs would be a good starting point. And I do realize that this amplifier will require a lot of work on my part, both in the design and construction.


Thank you for any help.

Last edited by danhei; 26th August 2014 at 12:44 PM.
  Reply With Quote
Old 26th August 2014, 04:33 AM   #2
kward is offline kward  United States
diyAudio Member
 
Join Date: Sep 2013
Location: Western USA
Well this is just my opinion but I would pick one single output stage topology using one given tube type and optimize the entire amp around it. It gets much more difficult to optimize for two different output tube types or two different output topologies simultaneously.

For example, I built an amp once that was switchable between kt88 and el34 tube types. Both ran UL connections. I made it switchable between tube types for fixed bias voltage and for feedback and high frequency stabilization parameters. It worked but was quite a bit more complicated. In the end I always preferred one of the output tube types over the other anyway in that particular amp, probably because unknowingly at the time I had picked the output transformers with primary impedance that better suited one of the tube types over the other at the operating parameters I had selected. So I guess in that regard it was a useful exercise, but I've never wanted to do it again.
  Reply With Quote
Old 26th August 2014, 11:37 AM   #3
diyAudio Member
 
artosalo's Avatar
 
Join Date: Feb 2010
There are a lot of stuff in Tamura's circuit, too much I think. I also think that NFET driver stages are not needed.

I have used simple Dynaco-style topology a lot with very good results.
Many of my friends have built the following design:

https://www.dropbox.com/s/o6i7ofcfex...2P_PP.GIF?dl=0

The test results of this amplifier are here: (first are for UL):

https://www.dropbox.com/s/gahw2em2u0...okset.gif?dl=0

The output transformer I have used is 4k to 8 (Indel TGL40/001), but I assume 5k Tamura will perform almost identically
  Reply With Quote
Old 26th August 2014, 04:05 PM   #4
diyAudio Member
 
Join Date: Apr 2004
Location: Monroe Township, NJ
Some sort of loop feedback is in order, when UL mode "finals" are employed. It's a damping factor thing. The 30 W. rating of the Tamura O/P "iron" rules a design containing a global NFB loop out. No magnetic headroom is available. Therefore, you come down to a "Baby Huey" using EL34s.

Contact Ian (GingerTube) and get his thinking. A line stage with gain may be necessary. EL34s require more drive than "12" W. types and 7591s.
__________________
Eli D.

Last edited by Eli Duttman; 26th August 2014 at 04:06 PM. Reason: grammar
  Reply With Quote
Old 27th August 2014, 03:36 AM   #5
diyAudio Member
 
Join Date: May 2005
Location: Pretoria, South Africa
I cannot find any reason not to use UL. Some prefer the "sound" of triodes; without giving offense, I am not much into the hearing evidence thing - too personal. In any case, in the circuits I have used distortion for UL at say 90% of maximum triode output was always equal to or lower than that for triodes, plus the advantage of UL giving more headroom.

Depending on how Tamura defines the maximum output rating, you might also consider the use of 6L6GCs with lower heater current. My personal favourite topology would be a pentode input stage (EF86 or such) followed by a l.t.p. (Long-tail pair) phase inverter. From the valves listed, an ECC81 or ECC88.
  Reply With Quote
Old 27th August 2014, 08:37 AM   #6
diyAudio Member
 
jazbo8's Avatar
 
Join Date: Jan 2011
Location: In Transient
Check out the Turner 5050 PP UL, you can scale the design down to 30W using the Tamura OPT.
  Reply With Quote
Old 27th August 2014, 04:55 PM   #7
danhei is offline danhei  Hong Kong
diyAudio Member
 
Join Date: Dec 2004
Location: Hong Kong
Thanks for all of the responses!

Here's a bit more information on the Tamura OT:
Impedance: 5K:0-4-8-16
Freq. response (-2dB): 10-100K
Power: 30W
Primary max current: 100mAx2 (balanced), 8mA (unbalanced)
Primary Inductance (H): 170H (0mA DC), 100H (10mA DC)
Weight: 5kg
After checking the data sheets, the KT88 in triode mode at 422V is using 94mAx2, which seems a bit too close to the 100x2 limit. Looks like I should just stick to EL34s for this amplifier.

kward, I'm definitely getting the message that optimizing for one tube and mode of operation makes more sense.

artsalo, I've been intrigued by the Dynaco designs as they seem relatively simple, although the gain-stage phase inverter combination (in the 7199 and 6AN8 amps) don't seem to be as well regarded. Have you found any limitations on that aspect of the circuit?

Eli, I'm a little confused on the global NFB magnetic headroom issue. It looks like the three Japanese designs all use global NFB, even using higher powered tubes like the KT120 with 45W output. Is this poor design given the 30W rating of the tube? Also, I see a feedback loop on Baby Huey from the output transformer to the phase inverter. Is the Baby Huey feedback not considered global?

Johan, good to know on the UL. I'll have to check how many EF86s I have on hand. It may not be more than one or two. But I do have a bunch of 6AU6s and the previously mentioned 6AN8 (with a pentode section).

jazbo8, that Turner design is quite nice. Looks like it's a well-considered design.
  Reply With Quote
Old 27th August 2014, 09:53 PM   #8
diyAudio Member
 
Tom Bavis's Avatar
 
Join Date: Jun 2002
Location: Macedon NY
The Dynaco design is excellent for UL operation, but changing to a long-tail pair phase inverter will improve performance with triode outputs. There are a number of PCBs available with this circuit, all in the original Dynaco PCB size. With a pentode front-end, this is the traditional Mullard circuit. A triode will provide enough gain, since the phase inverter has gain, unlike the split-load inverter in the Dynaco circuit.
  Reply With Quote
Old 27th August 2014, 10:13 PM   #9
diyAudio Member
 
Join Date: Apr 2004
Location: Monroe Township, NJ
Quote:
Eli, I'm a little confused on the global NFB magnetic headroom issue. It looks like the three Japanese designs all use global NFB, even using higher powered tubes like the KT120 with 45W output. Is this poor design given the 30W rating of the tube? Also, I see a feedback loop on Baby Huey from the output transformer to the phase inverter. Is the Baby Huey feedback not considered global?
"Baby Huey" uses short loop NFB. The O/P trafo is outside the loop.

In a design, like "El Cheapo", that employs a GNFB loop, which encompasses the O/P "iron", a distinct vulnerability to core saturation exists. O/P trafo core saturation can occur, when the LF error correction signal tries to "flatten" deep bass response and any infrasonic noise that is present exacerbates matters. For a 30 W. design that uses GNFB, the O/P trafo needs to be rated for at least 45 W. and 60 W. would be better.
Attached Images
File Type: gif EC big.gif (38.8 KB, 155 views)
__________________
Eli D.
  Reply With Quote
Old 28th August 2014, 02:25 PM   #10
diyAudio Member
 
Tom Bavis's Avatar
 
Join Date: Jun 2002
Location: Macedon NY
Usually the coupling caps are selected to limit the overall gain at the lowest frequencies - if the output transformer is rolling off, the feedback is decreasing and the gain is INCREASED, making transformer saturation more likely if very low frequencies are present. Fortunately, there is little content in music below 40 Hz - low E on a string bass.

The low end of the transformer is set by its inductance, which is not a fixed value - it generally increases with signal level up to the saturation point. Imbalance in the output tubes will reduce inductance and degrade bass performance.
  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
PP-UL amp with KT88 misterno Tubes / Valves 1 27th July 2011 05:47 PM
Advice sought on medium power triode for tube headphone amplifier rjm Tubes / Valves 13 16th May 2010 05:05 AM
Which do you prefer - EL84 UL PP or EL34 Triode PP - and why? ray_moth Tubes / Valves 4 23rd February 2007 01:22 PM
PP-Triode , PP-UL , PP-Pentode-strapped-as-Triode percy Tubes / Valves 23 22nd January 2007 01:36 AM


New To Site? Need Help?

All times are GMT. The time now is 03:41 AM.


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