Lynx Hilo - internals

Because I've messed up an attempt to update firmware on the device in question, I had to open it up to flip a dip switch for the secondary PROM.

Here's a list of interesting stuff found inside.

Back IO section (some under LT-USB board):
Burr-Brown SRC43921
Burr-Brown PGA4311U
Texas Instruments OPA1611A (opamp) - x20
Cirrus Logic CS4398-0ZZ x4
Axicom IM41 (relays) - x20
VITEC 46ZR249 x2 (dual audio transformers - these do the dBu/dBv setting for line in/out)

On headphone amp section (separate):
another CS4398
another PGA4311U
Texas Instruments LME49720 x2 (opamp)
LME49600 x2 (buffer)
LM79L regulator

(WHYYYY did you skip OPA1611A you bastards!)

On what is apparently touchscreen control section:
Burr-Brown 7611U
48.1564MHz crystal
48.152MHz crystal

+/-16V Rails for opamps provided by:
2x LM2941CT
2x LM2991T
(Headphone amp has separate power transistors on its part of the board.)

Control section:
Xilinx Spartan-6 XC6SLX45
Xilinx Spartan XC3S50A
JTAG connector
UART pins
DIP switch with PROM selector and JTAG ON/OFF (probably Xilinx ISP)
XpressO crystal

LT-USB board:
Xilinx Spartan XC2C32A
XMOS US1024L2
XTAG connector

Power board:
Sorry can't see. I see MOV and a HUGE cap.
on top side:
Coiltronics VP3-0084 48CH10I choke

 

Some trouble attaching the image, so here's the link:
https://drive.google.com/open?id=1HM6Kg4sPPi18fPgzj2lPtIQCS0QzcXrLtA

 

I think that one is also a hybrid multibit-D/S dac similar to the PCM1794 isn't it?

 

It's similar to the lower 18 bits of the PCM1794 that's a multibit D/S, but the PCM1794 has more strange magicks with the ICOB upper 6 bits contribution. Meant to be a low power product though, unlike the 1794. At least I haven't seen anyone try to squeeze the 1794 into phones unlik the CS4398.

 

Come to think of it, I think that's the chip used in my Bryston integrated (it's a plug-in module; their standalone dacs use AKM). I think it's actually quite decent, though nowhere near the $1k asking price for the module. Maybe $200ish in today's market.

 

As to OPA1611A too new? No way. It has a huge number of them already! They used the LME pair for headphone output instead of 611 with a better buffer...
Incidentally, this makes the back Monitor Out better sounding with headphones that do not require lots of current. Does not have the glassy quality.
It is 12 dB quieter.

Generally comparably the amp sections are (not ABX, AB only, difference is slight and warrants ABX):
- Monitor Out - authoritative, "darker", thumpy classic bipolar amp OPA1611 sound. Best SE.
- Phones Out - glassy, something of a V-shape. Bleh.
- Jotunheim (SE) - teensy bit brighter, lighter sound than Monitor Out, same quality (disappointing)
- Jotunheim (Balanced) - same tonality as Monitor Out but more detailed. Best all around.

In other words, for HD 650 Lynx delivers about the same quality as Jot SE out of the Monitor Out...

The differences are apparent when you drive something with current requirements. Say, HE-6 or HE-500. Jot apparently has a strong buffer on the output. Hilo just struggles on both - back is clearly overdriven and compressed, front gets glassier still.

Personally I rate OPA611 higher than OPA627. The latter is slightly oily FET nature and noisier, higher gain and higher max current. Now 611 is an annoying beast to properly implement But it's mostly a question of taste and not absolute quality.

 

After longer and ABX switcher I'm sure the differences are right.

Now the fun part is that the Monitor Out as well as being better than Jot SE output *also* has a noticeably better stereo separation. Like "10 degrees" better. I guess that comes from the high quality stereo to dual mono jack adapter and having dual mono signal paths.
The front phones jack is not that good but still. Perhaps Jot's SE output jack is just high crosstalk.