• Disclaimer: This Vendor's Forum is a paid-for commercial area. Unlike the rest of diyAudio, the Vendor has complete control of what may or may not be posted in this forum. If you wish to discuss technical matters outside the bounds of what is permitted by the Vendor, please use the non-commercial areas of diyAudio to do so.

Introducing Mercury - Achieving escape velocity.

Hi Russ, I just got a ship notification so Mercury is on its way to me. Any instructions/schematic available yet? I would like to see the schematic and know options for gain setting.
As I want to build to 4 VRMS (I am running DSD oversampling in computer with a -3 dB cut to avoid overs). I would also like to know what R is used to figure the corner frequency for the analog filter circuit.
 
Hi Russ, I just got a ship notification so Mercury is on its way to me. Any instructions/schematic available yet? I would like to see the schematic and know options for gain setting.
As I want to build to 4 VRMS (I am running DSD oversampling in computer with a -3 dB cut to avoid overs). I would also like to know what R is used to figure the corner frequency for the analog filter circuit.

For ES9038 use 100-120R (I would go high because of the 14% variance in DACs) to get ~4VRMs - For ES9028 use quadruple that value.

1/4W .1% or better Rs are fine

You should be able to build just knowing those values. I will also put up documentation shortly.

Enjoy! :)

Cheers!
Russ
 
Last edited:
9038 3 SE PRO / Mercury impressions

The end result produced by the TPA 3 SE 9038 PRO / Mercury combo on a revealing ATC system with NC400 amplification, is another league up from the Bufallo II with tridents /IVY III that got replaced.

Salas and Placid HD shunt regulators used in the power section, as before.

The filter used so far is the apodizing one.

Going back to the sound, everything is more natural. Soundstage is more defined, and cymbals are perfect with this combo. Another great surprise was the voices were singers like Sigrid and Dillon are a pure delight to listen too.

Overall very special and thoroughly recommended.

Many Thx Russ for the design and execution.
 
Regs...

May I ask which regulators you are using where?

My plan is to run Salas BiBs for the Mercury (I have them already set up as I used them for the Legato previously), but my experience also suggests that the 330uF caps on the Mercury will have to be eliminated (or changed to low value films) to keep the Salas BiB regs stable. On the Legato I changed the onboard 100 uF electrolytic to .1 uF films to avoid power supply oscillation with the Salas regs, and this approach worked very well (sound great, no lack of bass impact, and no oscillation in the supplies).
 
Infact I use Salas BiB v1.1 for the Mercury, and I kept the 330uF for now. The sense wires terminate on the mercury V+V- inputs. Did not notice any oscillation, however I have not measured with oscilloscope (as I dont have one handy).

For the Bufallo I use a Placid HD and LT3045s 1.3, 3.3, and 3.6 scheme
A Salas BIB 5.2 v powers XMOS WaveIO/ IAN McFiFO.
 
Oscillations

Infact I use Salas BiB v1.1 for the Mercury, and I kept the 330uF for now. The sense wires terminate on the mercury V+V- inputs. Did not notice any oscillation, however I have not measured with oscilloscope (as I dont have one handy).

For the Bufallo I use a Placid HD and LT3045s 1.3, 3.3, and 3.6 scheme
A Salas BIB 5.2 v powers XMOS WaveIO/ IAN McFiFO.

Even a good DMM should show oscillations if you have bad ones. Just measure AC at the outputs of the BiBs. I found that on the Legato with its 300 uF per rail my BiB regs were not stable at all, and had many mVs of AC on their outputs. I would be surprised that these regs could remain stable with the 660 uF per rail on the Mercury? Are you running dual mono regs (L/R) or just two?
 
Member
Joined 2004
Paid Member
Well virtually none - there will still be a very small amount (generally a few mv) because the opamps themselves have some tiny input offset - and even at .1% resistors are not perfectly matched :) The servo reduces the offset to very close to zero - but if you want absolutely no DC your only choice is to AC couple :)

I am reading ~1mv offset at the SE output and a bit less at the balanced outputs.

Cheers!
Russ


Hi,
so if I'm understanding correctly, the single-end outputs are direct coupled not cap coupled?

Does the signal path have any caps?

What path are the big buffers in, only differential outs?
 
Barrows, you were right, on the BIB -V there was 100 mv of AC. Removing the 330 uF Caps reduced this to 10 mv, without ill effects I can observe for Mercury. Also everything seems faster and tighter now. Btw , I forgot to mention I do not run dual mono regs.
Will try that later. Many Thx.
 
Barrows, you were right, on the BIB -V there was 100 mv of AC. Removing the 330 uF Caps reduced this to 10 mv, without ill effects I can observe for Mercury. Also everything seems faster and tighter now. Btw , I forgot to mention I do not run dual mono regs.
Will try that later. Many Thx.

That pretty awful for a regulator... I would suggest replacing it.