Please add me to the list for both also, thanks!
Estimated price for 2 adapters less than $40 shipped.
How do the adapters work with the factory or Edcor 4400ās?
No adapters needed for them.
Do they replace the existing transformers? Sorry if I missed something somewhere.
Yes they do. T
Edcor has 5 and 6 pins, lundahl has 4 and 5.
can be handled without an adapter if what it stands for is known
Actually Edcorās has 12 pins.
1 Like
Measure it? No. Hear it? Yes. Hear it in a transformer application? I do not know, never tried it. T
I actually have several of those from a previous project. Theyāre way too big for most applications and hugely expensive. And I think you can usually achieve what you want without that level of cost and size. They do have nice clarity and transparency though. My opinion, the stock DS 75R are fine there audibly. SMD metal films tend to be pretty good IME.
Resistors absolutely have āsoundā. Some are bad. Iāve spent years hearing them and finding the best sounding ones in my apps. Wirewound, tantalum, thick/thin metal film, carbon film, etc. It all started with a passive i/v converter (ie resistor) in my first NOS DAC projectā¦ What better place to hear the differences than right at the very heart of the sound, right?
Turns out there are even better places to hear the differencesā¦ In fact you can amplify the resistor noise (acck) and hear it even more lolā¦ For example, one well known critical location is the feedback resistor of a power amplifier. Well it turns out there is an even more critical location in a typical power amp topology, so critical in fact it almost derailed my own efforts to try and create a better sounding power amp. I call it the resistor noise amplifier location 
Thankfully thereās at least one R that was quiet and transparent enough that it allowed us to get over that hurdle and move forward, but itās not cheap (nude Z foil). I think the best tantalums would probably also work there, but generally not available in the value it needed.
BTW here ānudeā means un-encapsulated, not flaunting a nice set of hootersā¦
Anyway, I found that even the DC leakage bleeder resistors routinely put at the input and output of a cap coupled device matter in the sound. Which makes sense because the full signal voltage is developed there, so of course it matters.
Hereās a free tip Vishay- if you would ditch the ceramic substrate and use something like teflon instead, your nude Z foil would probably be unbeatable 
But hey, what does a crazy old guy like me know Haha
T
DAC#2 update. Itās playing in stereo now with 1 OA/channel and still old fw on it. And wow is what I can say for starters. Even my sister agreed via just cell phone clips. No trafos does exactly what I recalled hearing. Thereās another blanket layer removed leaving an organic much more natural sounding connection, brain to the sound. Less like hi-fi, more like real music. Everything in the music sounds more real to another level.
And thereās SO much more to each song than I ever realized. I mean detail, but other sounds and textures that I never realized were present. And those sounds/textures/shading add meaning to the composition now, which seem much deeper and more complex and enjoyable as a result. More of the artistry is coming through and it all adds up to WAY more musical enjoyment. Again, wow.
And it seems we got the grounding scheme right as well, thankfully. Iām using common mode/differential mode filtering. It was a choice that took a lot of pondering and sweating a bit, in case it wasnāt right (then what?)ā¦ A choice like that could derail the whole thing. With the trafo itās a no brainer really, you just use the output center tap. But now its different.
Weāre in uncharted waters here now sonically IME. Yes, yes, yes!
I do feel the need to check another inductor (much larger physcially, the same as in DAC#1) vs the one I have put in. Yes itās more work but need to get this right as much as we can now. Itās a labor of passion anyway. Weāll start the trial process today.
BTW what weāre doing is succeeding thankfully, but really all weāre doing here is getting the analog the h-ck out of the way as much as possible (easy to say, butā¦), which then allows us to hear how good the DSD approach can be and Tedās gifted digital handiwork.
I also find it interesting that I can hear the new stuff breaking in in real time basically, after the first few songs it becomes audibly better.
And thanks for bearing with my ramblings. T
2 Likes
Oh it definitely adds to the sound. But it doesnāt improve it. As usual YMMV T
2 Likes
Turbo, better than DAC #1?
Tony yes, Dac#1 and #2 are the same basically except in DAC#2 there is nothing in the way of the D/A converter ā output. Ie, no trafo. DAC#1 is no slouch, not at all. But DAC#2 takes it up another notch. I believe there will be no going back.
Checked a few things before I take the board out again for the larger inductors install. Itās looking great IMO. Square wave is nice. Noise pickup is way down, even unshielded so far; I moved some things apart now in the design (digital further away from the analog). Freq response is flat 20Hz-20KHz (~ -0.2dB).
I think the AD8132 maybe cannot provide the bits output amplitude Iām asking for it. We donāt want to saturate it, though as is the bits look ~ ok to me and it is sounding great. Time to put in the 8139 I think.
We may need to raise the Vsupply to the OA; Iām thinking 10v would be nice. But by the time the 12v main makes it through the RLC filter stages thereās only about 8.6v input supplied to the Vsupply regulator (which makes 7.0v currently).
Iām wondering if maybe we cant change that main 12v regulator to say, 15v instead(?) Ted? It would be nice if we could also drop the heat load on that main 12v regulator as well IMO. Itās the only one in the system that is really stressing thermally as far as I can tell.
T
(All images from single ended outs)
352K FS 441 Hz square wave- looking good IMO:
Zoom of the edge- ~ perfect step response makes Turbo happy:
Bit buffer output; looks pretty good. Iām happy as long as the bits look uniform. But not achieving the little bit of amplitude increase we wanted:
20KHz sine; the horizontal cursors mark the output level for reference vs the next image:
20Hz sine; the horizontal cursors mark the output level for reference vs the 20KHz (response is flat):
2 Likes
The 12V was a restriction caused by having to have an upgrade path from the PWD. Iād have loved to get 15, 16 or even 17V to the analog card. I donāt think the current power supply board in the DS can support 15V well, but I donāt have the schematics for it. The design wasnāt meant for a lot of current at 12V, the PWD had a separate supply for its analog output and this 12V supply was unused.
Also, just in case it isnāt obvious you can only get the output within about 1.4V of the rails with the AD8132. The AD8139 can get much closer to the power rails.
If you arenāt familiar with it ADI-DiffAmpCalc can be useful here.
2 Likes
Thanks Ted. Do you happen to know how much DC level is fed from the rectifiers into the 12v main reg?
T
Great work, Turbo!
Itās not that simple 
The 12V supply is weird. Personally, Iād invest in a good quality linear supply and bypass the stock power board. Thereās plenty of room in the case for another supply.
1 Like
Oh noooo Itās almost never ~ straightforward, is it? Iāll take your word for it of course, but can you also tell why I cant just put in a 15v linear there?
T