KG Balanced Dynahi build discussion thread

[kevin gilmore]

No plans to do that

[Victor Chew]

Picture of my almost completed Dynahi. Top plates still have to be painted and lacquered. It sounds lovely. Thank you Kevin for this lovely design.

 

 

[Pars]

Looks good. Did you get the offset down to more manageable levels? I try to get the offset to within 1mV on dynalo/dynahi before I put the servo opamp in. Not sure if anything has changed on these boards with respect to that versus the old single boards or not? I'm aware of the input section being true balanced now vs. the previous mono boards. I wouldn't expect that to change anything.

[Victor Chew]

Thanks Pars. Manually, I was able to get it to about +-3mv for the dc offset. Patience was the key. With insertion of the opamps the dc offset achieved 0mv. Was also able to get balance to 0mv.

[Pars]

Good to hear. Yes, patience is required

[RiStaR]

Congrats Victor

[Victor Chew]

Thanks Cal. Come over and have a listen to it.

[johnwmclean]

Lovely build Victor.

How are dynamics holding up to the stat rig?

[Victor Chew]

Thanks John.

 

The presentation to me for both these amps (KGSShv with 007 vs Dynahi with HD580) are different and it would be difficult to say which is better. I guess it boils down to a matter of taste. Personally, I still prefer the music out of the electrostatic rig. The dynahi is no slouch and it is probably the best I have heard for any dynamic headphone amplifier by a substantial margin. 

 

The Dynahi puts the listener closer to the musicians than the Stax. It is fast, exciting and engaging. I find myself tapping to the music most instantly. I am not really a bass head but this thing gives out bass that is tight, authoritative and voluminous. The sense of space around the instruments and vocalist makes them clearly identifiable. As such, clarity must be the biggest strength of this design. Treble is extended but lacks the "airiness" of the Stax. As for the midrange I can feel the realism and emotions of the singer.

 

It has everything the Stax rig has but I just prefer the way the stax presents it. Others may feel otherwise  . My impression is based on the HD580 and may be different with different headphones.

 

Oh yes, one other thing, I feel it is really powerful. Its like when you step on the car accelerator and the power never seems to puncture.

 

I know I am bad a describing music, so please do forgive me.

Edited October 30, 2013 by Victor Chew

[Kerry]

Congrats on a nice build Victor.

 

I thought your impressions were well stated.

 

If you have a chance to invite some friends with Audeze's LCD-3 or the HD800's I'd like to get your feedback with those two as well.

[Victor Chew]

Thanks Kerry. If I do get my hands of either of the headphones will certainly post my impression again.

[congo5]

Hi   I have the boards populated (from the latest Lil' Knight GB)

 

do not understand how the adj pots work,

both seem to change bias and offset.

 

Also does .375V across the emitter resistors sound right?

 

Is there a different thread to look in?

 

Am using it SE, - input   grounded......- output   floating.

 

Thanks

[Pars]

Amb's procedure in this thread is the clearest I know of on how to adjust these. Since you are doing a fully balanced board, I'm not sure if the fully differential front end will have any effect on doing this or not.

[congo5]

Thanks Pars,  had not seen that.

If I hook up enough meters, it should become apparent

what the extra pots do.

[Zashoomin]

Hi everyone long time luker on head case and was looking into building an on board heat sink balanced dynahi and just wanted to clarify some things.  So it looks like it will work very nicely with a Sigma 22 power supply powered by a 100VA transfromer.  The questions I do have are if I wanted to power speakers out of this how many watts would I get into 8ohms,  Do I need 2 sigma 22's or just one, and are the gerber's available for the board.  

[spritzer]

While the amp is stable into 8ohms I'd recommend proper power amp for this role.  The PSU is always the bottleneck so dual S22's with big heatsinks are the bare minimum. 

 

Personally I'd just build a F5.... 

[Zashoomin]

While the amp is stable into 8ohms I'd recommend proper power amp for this role.  The PSU is always the bottleneck so dual S22's with big heatsinks are the bare minimum. 

 

Personally I'd just build a F5.... 

The F5 is in the works as well.  So many projects so little time.  I was hoping to make another beta 22, dynahi, KSA5 and compare as those seem to be the top of the line when it comes to headphone amps.  

[jose]

Hello everyone.

 

It´s my first post here and I need a bit help, if it´s possible of course.

 

I have got a Balanced Dynahi (Tran´s boards, using the THAT340). I just finish the amp and set-up but I have a problem with the output:

 

I´m using a SE input with Input- to ground and output-  floating. The sound is very poor and the volumen is little. With balanced input and the output- floating, the result it´s the same, poor sound and little volume. In this time, I haven´t volume pot.

 

Have the same problem?

 

Thanks

[kevin gilmore]

something very wrong. Suggest you measure output and power supply voltages.

[jose]

Thanks Dr. Gilmore. Sorry, I forgot give you the settup of my amp.

 

My Dynahi used two Sigma 22 with two toroids (160VA each one). The output volt. is 29,5 Volt DC.

 

I followed the Ti Kan´s guide for settup. The Bias is 0,75 Volt, the working temp is arround 70º centigrades. I had problems with the offset setting but now it´s controled.

 

I´ve used my DAC with balanced output and I checked some Input configuration:

 

Balanced input with SE output (I used the O+ and PSU ground)
SE input with I- to ground with SE output

 

With those options the sound is very poor and the volume is "minimal" but with balanced input I believe that the test is a bit best. I use for dummy test an old AKG.
 

If you need more info, photos or other check/test, I can post in this thread
 

 

Thanks  

 

Edit for error, sorry

[kevin gilmore]

input impedance of amplifier is 10k, so this might be an issue.

 

if you have a scope and signal generator, might test with this first.

[jose]

Hello

 

I made a test with help of Torpedo and Picaudio (Thank you guys).

 

I'm sending a calibrate signal from my PC and DAC (1Khz and 1 volt) to I+.

 

On the O+  I just check 0,0791 volt.

 

If I send the I- to ground, I check 0,0500 volt.

 

I would like post photos of my work.

 

 

 

 

Thanks 

[justin]

with a THAT340 input i wouldnt be surprised to see dc offset

 

next time bent the resistors with the value facing up!

[jose]

with a THAT340 input i wouldnt be surprised to see dc offset

 

next time bent the resistors with the value facing up!

 

    ok, Thanks Justin, the next time.

 

 I checked the boards but the resistors values are correct wiht this:

 

 

I only changed the cap. for 1000 uF/35 volt

 

 

Yes, I had problem with the offset but I could solve with patience... and a matl whisky.

 

Thanks

[kevin gilmore]

are you sure those are 2pf caps, not 2uf caps. because that would definitely kill all the gain.

 

also mospec semiconductor...  I use real Motorola parts...

 

 

edit: yep 2uf would take the gain down to about -50db at 1khz.

Edited December 31, 2013 by kevin gilmore