[f] AMB on Facebook

SuSy dynahi v2

Kevin Gilmore's higher-powered version of the dynalo

SuSy dynahi v2

Postby DEFL » January 20th, 2015, 3:57 am

amb wrote:Dynahi v2 is essentially a SuSy (super symmetry a la Nelson Pass patent) implementation of the original circuit. Just like how a β24 is a SuSy version of β22. And a similar problem exists -- low input impedance. For the β24 it's rarely an issue because it's a speaker power amp, and most preamps have low enough output impedance to drive it. But as a headphone amp, dynahi v2 has a problem with putting a volume control pot or attenuator ahead of it. Usually you want the pot resistance to be no greater than 1/10 the input impedance of the amp, or the volume control taper characteristics will be wrong. A pot 1/10 an already-low input impedance will not be an appropriate load for most sources. In the case of the dynahi v2, the input impedance of each of the + and - inputs is 10KΩ, if you use a 1KΩ volume pot, then many sources will have trouble driving it. This is the reason I never made a SuSy β22 headphone amp or preamp.

Also, the dynahi v2 uses the same unobtainium parts as the original dynahi.

Hi Ti,
Maybe I'm addressing wrong question, but do you know what is the latest revision of Dynahi v2 (SuSy) amp?
There are few links KG posted during development, and all of them are still working:
http://gilmore.chem.northwestern.edu/dynahibal1.zip
http://gilmore.chem.northwestern.edu/dynahibal2.zip
http://gilmore.chem.northwestern.edu/dynahibal3.zip
http://gilmore.chem.northwestern.edu/dynahibal4.zip
http://gilmore.chem.northwestern.edu/dynahibal6.zip
http://gilmore.chem.northwestern.edu/dynahibal7.zip
http://gilmore.chem.northwestern.edu/dynahibal8.zip
http://gilmore.chem.northwestern.edu/dynahibal9.zip

And the same concerns PS:
http://gilmore.chem.northwestern.edu/dynahipower2.zip
http://gilmore.chem.northwestern.edu/dynahipower3.zip
http://gilmore.chem.northwestern.edu/dynahipower5.zip

I just changed last digit in numerical order (to a reasonable extent) and discovered that the above links are valid.
DEFL
 
Posts: 175
Likes: 8 posts
Liked in: 4 posts
Joined: August 2nd, 2013, 2:26 am
Location: Saint-Petersburg
Country: Russia (ru)

Re: Dynahi build

Postby DEFL » January 20th, 2015, 2:50 pm

Ok, seems that starting from #6 Motorola devices are used instead of Toshiba.
Power board revision isn't very important since S22 is equivalent, if not better, power supply.

Does anyone have a download link for dynahi rev 9 board but with onboard heatsinks? Like this one:
http://www.andaudio.com/phpbb3/viewtopi ... 8&t=118257
Couldn't find that.
DEFL
 
Posts: 175
Likes: 8 posts
Liked in: 4 posts
Joined: August 2nd, 2013, 2:26 am
Location: Saint-Petersburg
Country: Russia (ru)

Re: Dynahi build

Postby MisterX » January 20th, 2015, 6:01 pm

Does anyone have a download link for dynahi rev 9 board but with onboard heatsinks?


Let me Google that for you.

http://lmgtfy.com/?q=stax+mafia+circuit+boards

http://gilmore.chem.northwestern.edu/boards/
1352.5 and counting.
User avatar
MisterX
 
Posts: 1143
Likes: 52 posts
Liked in: 36 posts
Joined: March 31st, 2010, 2:00 am
Location: Kalamazoo, MI
Country: United States (us)

Re: Dynahi build

Postby DEFL » January 20th, 2015, 10:45 pm

Thanks.
DEFL
 
Posts: 175
Likes: 8 posts
Liked in: 4 posts
Joined: August 2nd, 2013, 2:26 am
Location: Saint-Petersburg
Country: Russia (ru)

Re: Dynahi build

Postby DEFL » January 22nd, 2015, 5:10 am

Sorry if I ask in this thread, but maybe this is the best place.

Ordered small batch of Dynahi v2 boards. This is the latest SuSy revision with Motorola/On Semi parts and onboard heatsinks. Too impatient to hold the boards of this amp in my hands.
Studied a lot of discussion, but still couple of questions remain:
1. Gain. Is gain determined solely by R52, R56 and C1, C3? No other parts are changed (like input resistors R53, R57, which default value is 10k)??? I will use 15K or 10K or even 5K attenuator in front of the amp, if that matters.
Default values of R52, R56 and C1, C3 are 200k, 200k, 2pF, 2pF respectively for Gain 21.
Other options mentioned: 50k, 50k, 5pF, 5pF for Gain 6 and 100k, 100k, 3pF, 3pF for Gain 11. Is it correct?

2. Two small 0.1uF bypass caps near OPA445. Didn't get clear understanding from available BOMs what type they should be. Mostly missed for unknown reason.
But found this answer posted by Ti:
http://www.head-fi.org/t/95079/polar-ta ... st_1107131
Does this message concern those caps?
Is so, then Kemet Golden Max (or similar to used in beta22/sigma22) will be OK I guess.
DEFL
 
Posts: 175
Likes: 8 posts
Liked in: 4 posts
Joined: August 2nd, 2013, 2:26 am
Location: Saint-Petersburg
Country: Russia (ru)

Re: Dynahi build

Postby amb » January 22nd, 2015, 5:17 am

This really isn't the best place for SuSy dynahi questions, and I haven't followed the development very closely. I don't know how many revisions there are, and how they're different from each other. Nevertheless I'll try anyway. You need to point me to a schematic that matches the circuit you describe.

I also think this stuff should be in its own thread. We've strayed far off topic -- this has nothing to do with Stixx's build.
User avatar
amb
Site Admin
 
Posts: 10355
Likes: 175 posts
Liked in: 478 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: Dynahi build

Postby DEFL » January 22nd, 2015, 5:39 am

This is the board:
http://gilmore.chem.northwestern.edu/bo ... ibal8f.zip

Same is here along with BOM (incomplete and may contain mistakes, just FYI. I'm not very much relying on it):
http://tee8tee4388.blogspot.ru/2013/07/ ... ifier.html
It's Lil Knight's site.

Funny enough, I couldn't find the schematic of the revision with THAT340 chip and Motorola/ON Semi parts. I really tried hard.
This repository doesn't contain required schematic:
http://gilmore.chem.northwestern.edu/boards/
I searched a lot of forums and didn't see it published by KG.

So, this is the schematic I'm looking at currently. It's for Toshiba devices but most part of the circuit should be the same:
http://gilmore.chem.northwestern.edu/dynahiver2a.pdf

Update:
Schematic is here:

http://gilmore.chem.northwestern.edu/bo ... balbjt.PDF
This is a version with Motorola/ON Semi parts and THAT340 IC which substituted all Toshiba devices.
Last edited by DEFL on February 4th, 2015, 5:50 am, edited 1 time in total.
DEFL
 
Posts: 175
Likes: 8 posts
Liked in: 4 posts
Joined: August 2nd, 2013, 2:26 am
Location: Saint-Petersburg
Country: Russia (ru)

Re: Dynahi build

Postby DEFL » January 22nd, 2015, 5:43 am

amb wrote:I also think this stuff should be in its own thread. We've strayed far off topic -- this has nothing to do with Stixx's build.

I don't mind of course. Should I post further to new thread or will you move the messages to a new thread please?
DEFL
 
Posts: 175
Likes: 8 posts
Liked in: 4 posts
Joined: August 2nd, 2013, 2:26 am
Location: Saint-Petersburg
Country: Russia (ru)

Re: SuSy dynahi v2

Postby amb » January 22nd, 2015, 6:09 am

This thread has been split.

The gain of the amp is (R52 / R53), and R52 = R56, R53 = R57.
Thus, with R52 = 10K and R53 = 200K, the gain is 20.

R53 and R57 also sets the input impedance of the amp. In this case it's 10K ohms at each of the + and - inputs.
The pot or attenuator value before this input should have a substantially lower resistance than the input impedance (usually 1/10 or less), but in this case, it would mean using a 1K ohm pot or lower, which is going to be a difficult load for many sources to drive. You could increase the value of R53 and R57, but that means R52 and R56 also needs to increase in value to maintain the gain. These resistor values are already pretty high, enough so that resistor noise may become an issue. This is the problem that I mentioned (and you quoted in the first post).

C1 and C3 are compensation capacitors to help stabilize the amplifier at ultrasonic frequencies. The RC constant of R52 * C1 and R56 * C2 sets the corner frequency. If you change the value of R52 and R56, these capacitor values probably also need to change. By how much? Maybe in proportion to the resistance change, but whether that's correct would require examining the amp's square wave response on an oscilloscope.

0.1uF X7R multilayer ceramic capacitors are good for bypassing the opamp supply rails. Be sure to choose one with sufficient voltage rating.
User avatar
amb
Site Admin
 
Posts: 10355
Likes: 175 posts
Liked in: 478 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: SuSy dynahi v2

Postby DEFL » January 22nd, 2015, 6:20 am

Thanks!
amb wrote:Thus, with R52 = 10K and R53 = 200K, the gain is 20.

And balanced 40, I believe.

I think I'll try both 100K/3pf and 50k/5pf combinations, will not change input resistors from default 10K, and will use Goldpoint 5K attenuator or build Delta1 with Zatt something around 5K (will have to check available resistors values @ Mouser/Digikey).

Again, thanks.
DEFL
 
Posts: 175
Likes: 8 posts
Liked in: 4 posts
Joined: August 2nd, 2013, 2:26 am
Location: Saint-Petersburg
Country: Russia (ru)

Next

Return to dynahi headphone amplifier

Who is online

Users browsing this forum: No registered users and 7 guests

Quick Links

AMB audio DIY resources
AMB audio shop