"The Wire" Ultra-High Performance Headphone Amplifier - PCB's

opc

Member
Joined 2004
Paid Member
Hi Guys,

I've updated the list with everyone up to this post.

Soooooo...

The group buy is closing this evening. but there will be exceptions made.

Because of the skewed interest toward the SE-SE and BAL-BAL PCB's, I'm going to be stuck with 20 extra BAL-SE boards. So on that note, if you want BAL-SE boards or kits, then you're more than welcome to place orders.

There will be one more BAL-BAL and one more SE-SE PCB or kit available to whoever is the next person to ask for one, and then those buys are officially closed. That will make 38 of each of those boards, and I want to keep two of each for myself.

I will also not be ordering parts for the kits for another week or so, which means you can change your orders to full kits, but you cannot change the number of PCB's.

I've attached the "Final" list, but I'll update it once more when the last two BAL-BAL and SE-SE PCB's are sold. I will also continue to update any orders for BAL-SE kits.

Cheers,
Owen
 

Attachments

  • THE WIRE - SECOND RUN - UPDATE 9-25-2011.pdf
    39.1 KB · Views: 159

opc

Member
Joined 2004
Paid Member
Hi Guys,

I've also got measurements for the SE-SE version of The Wire that I took using the proto-board shown a few posts back.

This is powered by the PSU setup exactly the way the kit has it.

The first 5 images are FFT's under various conditions. There's one with open inputs, grounded inputs, and three at varying drive levels. All of these are taken into a 50 ohm load.

The ground floor is astonishingly low at -155dBV. You can see some 60Hz noise poking above the floor, along with all of the harmonics. The worst harmonic is at 300Hz, but it's still -131dBV, and as such it certainly won't be audible even with the most sensitive headphones. To put it into perspective, the highest harmonic here would produce about 3.12 femto-watts into a 32 ohm headphone.

The noise floor also stays incredibly low regardless of the output level. The amplifier will start to clip at around 8.5VRMS if you use a +/- 15V supply which is enough to drive even absurd headphones. The noise floor remains at least 150dB below the output, and the second and third harmonics are each down 117 and 118dB respectively. Harmonics above the fifth are more than 135dB down even at full output.

SNR as measured by the AP is 132.5dB at 8.5V output. Channel crosstalk is -91dB which would improve somewhat with dual mono supplies, but it's probably limited by the fact that all SE output systems share a common ground at the output which contributes to crosstalk. I would expect a BAL-BAL system with dual mono supplies to measure significantly better, but this is still extremely good.

There are also two frequency response graphs taken at 1V output and 8.5V output. Both are better than +/- 0.0005dB from 5Hz to 60kHz. the measurements BW on the AP limits the measurable range.

All in all, you give up nothing at all by using the SE-SE version of The Wire, and like its BAL-SE sibling, it really does live up to its name! The only real difference between the two is the CMRR offered by the BAL-SE version which is really the better choice if you have good balanced outputs on your source.

Cheers,
Owen
 

Attachments

  • FFT Spectrum open input - 50R.png
    FFT Spectrum open input - 50R.png
    94.8 KB · Views: 655
  • FFT Spectrum grounded input - 50R.png
    FFT Spectrum grounded input - 50R.png
    92.1 KB · Views: 642
  • FFT Spectrum Monitor 500mV - 50R.png
    FFT Spectrum Monitor 500mV - 50R.png
    94.8 KB · Views: 632
  • FFT Spectrum Monitor 1V 50R.png
    FFT Spectrum Monitor 1V 50R.png
    95.9 KB · Views: 621
  • FFT Spectrum Monitor 8.5V 50R.png
    FFT Spectrum Monitor 8.5V 50R.png
    97.2 KB · Views: 616
  • Signal to Noise Ratio 8.5V.png
    Signal to Noise Ratio 8.5V.png
    44 KB · Views: 134
  • Relative Level (1.00000 kHz) 8.5V.png
    Relative Level (1.00000 kHz) 8.5V.png
    70.6 KB · Views: 114
  • Relative Level (1.00000 kHz) 1V 50R.png
    Relative Level (1.00000 kHz) 1V 50R.png
    70.6 KB · Views: 116
  • Crosstalk.png
    Crosstalk.png
    39.4 KB · Views: 114

opc

Member
Joined 2004
Paid Member
A few more measurements
 

Attachments

  • Distortion Product Ratio 500mV 50R.png
    Distortion Product Ratio 500mV 50R.png
    47.1 KB · Views: 120
  • Distortion Product Ratio 1V 50R.png
    Distortion Product Ratio 1V 50R.png
    47.2 KB · Views: 125
  • Distortion Product Ratio 8.5V 50R.png
    Distortion Product Ratio 8.5V 50R.png
    47.2 KB · Views: 99
  • THD Ratio vs. FR 1V percent.png
    THD Ratio vs. FR 1V percent.png
    63.6 KB · Views: 105
Hi Owen,

>>I'm going to be stuck with 20 extra BAL-SE boards.

No way!

I recall a wish for "a bridged version using a pair of LME49600 per channel, each driven by a single LME49990. . [into] . . headphones setup for bridged output". It would seem that a pair of BAL-SE boards, each following one TP Opus running in dual differential mode, would deliver on that wish. Over kill? Possibly :)

Please add two BAL-SE (pcb only) to my order.

'gardz, Dick