Y A A B - Yet Another Aikido Build - Page 4 - 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 14th May 2008, 04:22 PM   #31
Cappy is offline Cappy  United States
diyAudio Member
 
Join Date: Feb 2004
Location: Southern Willamette Valley
In my last post I mentioned possible impedance issues with the proposed LCLC supply.

The problem as I understand it is the supply output impedance will vary with frequency, where a better situation is more of a "brick wall" low impedance. But that is just theoretical for me until I play around with different power supplies.

My next step is to attempt to simulate the LCLC supply in a Spice program and see what the output impedance looks like. I'm going to build the proposed LCLC supply in any case and see how it sounds.

Speaking of John Swenson again, his road led him to try "low DCR" supplies, and then various regulated supplies, finally settling on a regulated supply of his own design. I don't plan to try a low DCR supply, i.e. beefy low impedance power transformer, low DCR chokes, low value caps. If anyone is interested in the subject, there is endless discussion on the AA tube diy forum. I do plan to try a few choke loaded LC->reg supplies as part of this project. I'll need a higher voltage power transformer to do this.

I'm still ordering parts, but in the meantime I've been able to work on the DAC side of the project a bit. I modified my pcm63-based Lite Dac 50 with jumpers so that I can get the raw signal off the DACs, to be input into the Aikido. With the jumpers I can also input a DAC signal into the SRPP stage of the Lite DAC, for comparison purposes.

I also built up the Twisted Pear COD DAC and tried it with the SRPP stage. It sounds good. I was able to get SPDIF working, but I2S from my Empirical Audio Pace-Car (with Squeezebox as source) causes a ground loop -- probably because there are so many different modular components and no overall grounding strategy yet: Squeezebox, Pace-Car, COD, COD Power supply, Pace-Car Power Supply, Lite DAC 50 output stage. Apart from the hum, I could hear a definite improvement from I2S.

The Buffalo DAC should arrive shortly. With it will also come the Twisted Pear IVY solid state output stage with active I/V which will be interesting to compare to the Aikido with passive I/V.

Speaking of SRPP stages, I emailed a bit with Lukasz Fikus since he has a great deal of experience modding CD players to work with tube output stages. Among other things, I asked his opinion on balanced vs. single ended output stages for differential DACs. His suggestion was to stick with single ended unless my amp is balanced; that the slight gain in sound quality is trumped by the added complexity. My amp isn't balanced, it's a SET. Broskie now sells a balanced Aikido with 8 (!) tubes that would work with the differential outputs of the COD's pcm1794 and Buffalo's ESS Sabre chip, but I'm not going to go there... Perhaps a simpler tube differential circuit, not Aikido, would be interesting to do at some point, however.

One advantage of differential, other than eliminating common mode noise, is twice the current out, which is useful when using resistor-based passive I/V like I plan to do, since it halves the load the DAC sees, and lower is better, as I verified modding my pcm63 DAC.

I also have some good Lundahl LL1931 step-up transformers -- I could raid my phono stage -- to do differential to single ended conversion. My TVC preamp also can be wired for balanced input and single ended output, so there are various possibilities to try here.
  Reply With Quote
Old 26th May 2008, 06:42 AM   #32
diyAudio Member
 
tubetone's Avatar
 
Join Date: May 2008
Default I just started an Aikido build

Hi, I'm new here, this is my first post. I just started an Aikido 5687 headphone amp using 6CG7's and 5687's so this is a very interesting & timely thread. I haven't decided on my power supply yet or my attenuator, so thanks for the good ideas. I'm planning another one with 6H30pi's and 6CG7's. Here's what I have so far. tt
Attached Images
File Type: jpg dsc04210.jpg (58.4 KB, 1193 views)
  Reply With Quote
Old 26th May 2008, 07:37 AM   #33
Cappy is offline Cappy  United States
diyAudio Member
 
Join Date: Feb 2004
Location: Southern Willamette Valley
Tubetone,

Welcome to the thread. Your Aikido is looking good so far. What type of resistors are those? And what are those exotic green caps?

If you haven't purchased a power transformer yet, and plan to try a choke input power supply, this transformer might be a good fit:

Angela Instruments Universal Power Transformer

Broskie has mentioned 6CG7 tubes in the Aikido like a 250-300 volt B+ and the Angela transformer has a higher output voltage than the Lundahl I have. It is less expensive too. Another possibility is upping the "transitional" first small cap slightly to get the voltage up. I have a few charts explaining this that I'll post tomorrow.

************************************************

Most of my parts have now arrived so I am going to start building this week. I have a design for an LCLCLC power supply that adds a choke, has the same capacitors, and costs just $25.00 more, but lowers the ripple by an order of magnitude over the LCLC version I posted previously. The impulse response is almost as good. I'll upload the images tomorrow. As usual, any design feedback would be appreciated.

The Buffalo DAC has been running in my system for a week now. Currently I have it going through the SRPP 6922 valve output stage of my pcm63 DAC. It is sounding rather good with impressive clarity, resolution, and soundstaging. Sound is balanced with zero glare. Tone and bass realism is not as good as the pcm63, but there are plenty of things to do to try to improve that.

Current is strong out of the DAC, what with the four ganged channels, even running single ended with half the output. Iím not sure Iím reading the ESS Sabre data sheet correctly, but it sure seems it is more than 2.1 times the pcm63 (4 ma pp vs 4.224 ma * 2). This is good for passive I/V as the resistor value can be quite low. Iím running with a 15 ohm Mills resistor at the moment, but could go quite a bit lower, maybe even around 4-6 ohms for the gain structure of my system. Those who need 2 volts out could probably use 50 ohms or less as a single ended i/v resistor with the Aikido or other tube output stage with typical 18-26 db gain.
  Reply With Quote
Old 26th May 2008, 10:51 PM   #34
Cappy is offline Cappy  United States
diyAudio Member
 
Join Date: Feb 2004
Location: Southern Willamette Valley
Here is the proposed LCLCLC supply. Ripple is down to 1.25 mV from above 10 mV:
Attached Images
File Type: png aikido lclclc promising a static.png (26.1 KB, 969 views)
  Reply With Quote
Old 26th May 2008, 11:02 PM   #35
Cappy is offline Cappy  United States
diyAudio Member
 
Join Date: Feb 2004
Location: Southern Willamette Valley
This simulation shows the impulse from a 10% change in the output current. The horizonal units are 1 second so from one current level to the other is about 220 msecs:
Attached Images
File Type: png aikido lclclc impulse.png (35.4 KB, 929 views)
  Reply With Quote
Old 26th May 2008, 11:06 PM   #36
diyAudio Member
 
tubetone's Avatar
 
Join Date: May 2008
Default Thanks Cappy, glad to be here

I don't know what the parts are exactly, I got a complete kit from Broskie. The caps are paper/oil I beleive (K42V-2)? and the resistors are 5% carbon film. Thanks for for the tranny reference I'm still looking, someone else suggested a toroidal. I'll check it out and let you know. tt
  Reply With Quote
Old 26th May 2008, 11:15 PM   #37
Cappy is offline Cappy  United States
diyAudio Member
 
Join Date: Feb 2004
Location: Southern Willamette Valley
This simulation shows the same LCLCLC supply with the higher voltage Angela transformer. This might be more copacetic with the 6cg7.

I'm guessing at the Angela transformer secondary resistance, so a different value might move the B+ plus or minus 5 or 10 volts.

Both the Lundahl and the Angela have the desirable center taps. Notice I threw in a tube rectifier which drops more voltage than the Crees, but I'm still at about 265 volts B+. I intend to test power supplies with both Schottky and tube rectification.
Attached Images
File Type: png aikido lclclc static angela.png (27.8 KB, 922 views)
  Reply With Quote
Old 26th May 2008, 11:25 PM   #38
Cappy is offline Cappy  United States
diyAudio Member
 
Join Date: Feb 2004
Location: Southern Willamette Valley
Tubetone,

Ah, Russian caps, makes sense. Thanks for the info, I haven't tried those.
  Reply With Quote
Old 26th May 2008, 11:42 PM   #39
Cappy is offline Cappy  United States
diyAudio Member
 
Join Date: Feb 2004
Location: Southern Willamette Valley
Here is another way to increase the B+ to 250 volts or thereabouts -- increase the first small "transitional" cap. By changing it slightly from 0.47 uf to just 1.1 uF I increase the B+ 35 volts, to 250 volts. Increasing it to 2 uF ups the B+ all the way to 280 volts, which is kind of interesting.

Well, that seems like a good way to tune the output B+ voltage for a given power transformer. But is there a catch? Well, maybe, since the impulse response goes to hell. Ripple gets even better though. Does it matter in practice if the change in current is slow? No idea at this point, so I'll try it. If anyone has insights I'd like to hear them. Here is the slow impulse simulation with the 1.1 uF first cap:
Attached Images
File Type: png aikido lclclc impulse larger first cap.png (29.8 KB, 919 views)
  Reply With Quote
Old 27th May 2008, 12:06 AM   #40
Cappy is offline Cappy  United States
diyAudio Member
 
Join Date: Feb 2004
Location: Southern Willamette Valley
The BOM for the power supply currently is:

Lundahl LL1683 power transformer or
Angela Instruments Universal power transformer

Lundahl LL1673 20H choke
Hammond 157R 2H choke
Hammond 153N 0.6H choke

Diy HiFi Supply Obbligato 47 uF Film (500v) and 70 uF (630v) film/Oil caps
470 nF Auricap (since I have some -- don't know how they sound -- I've heard this small cap is important so I may try others)
20 uF cap composite of smaller Solen and Obbligatos I have lying around.

No electrolytics will be used in the power supply.

I think the Obbligatos are excellent in power supplies, based on a bunch of them being in my Lux 300b amp. I also have a 70uF Obbligato in the B+ of my SRPP stage. I'd have more in there, but I could only cram one in the chassis -- they are big. The picture shows the 47uF film power supply caps and the 70uF film/oils.

The potentiometer in the picture is a PEC 100k pot, which will replace the R15 resistor that adjusts the voltage divider that controls power supply noise to the top and bottom triode's grids. It allows on-the-fly adjustment for lowest noise with different tubes.
Attached Images
File Type: jpg obbligato caps a1.jpg (95.7 KB, 944 views)
  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
aikido build thread- nate Tubes / Valves 1 27th February 2009 05:01 AM
WTB: pwer transformer for my aikido build wendell Swap Meet 0 12th February 2009 03:58 PM
24V aikido build! woodturner-fran Tubes / Valves 10 3rd December 2007 06:11 PM
build my first full ranger,let's build another one:a small fullrange high end pc-spea then_dude Full Range 2 30th November 2005 07:54 AM
Do i need to build main supply (240v) filter?Or build power distribution? thomgun_lc Chip Amps 9 16th September 2005 09:52 AM


New To Site? Need Help?

All times are GMT. The time now is 05:33 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