or Connect
Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › Chameleon DAC listening and modifications
New Posts  All Forums:Forum Nav:

Chameleon DAC listening and modifications - Page 52

post #766 of 1144
Quote:
Originally Posted by Wood View Post
Newk Yuler, I must admit I am a bit puzzled by Michaels reply, I guess he is thinking of the whole product, and a spidf version only, but if removing the usb part of the input board works, and avoids the clock, I would be happy to have this as a second board, and use the original board if I have to use usb only. I am assuming the WM8805 works without the clock, and works as well as the dir90001.
I wish I was better capable of understanding the actual workings of these ICs. Taking into account what Rhodes said about the Wolfson receiver in Chameleon, the way to run it without the TCXO is to use the recovered SPDIF clock from whatever someone is feeding it. I expect feeding it with SPDIF from an Off Ramp will be fine but like Rhodes (and Steve Nugent) stated about most SPDIF, they mostly don't get it right. No telling how it would work with something like the Hiface. A simplified version of the current input board with the Wolfson sans TCXO would be strictly for excellent SPDIF sources which is really what what we're asking for, but there's no wiggle room like the DIR9001 would provide. On the other hand, the Wolfson using the recovered clock might be superior to the DIR9001 when supplied with an excellent source clock. ...Because the DIR9001 has its own clock.

I'd probably be okay with a simplified version of the current input board (sans TCXO), but what about the rest of you guys? I expect Michael will have to have a different board fabricated to suit the simplified design, so it might not be more trouble to ask for the Wolfson or DIR9001.

We probably shouldn't rule out the possibility of finding a way to mod the current input board to do the same thing but I expect most of you would rather have an alternative board to drop into your Chameleons and not have to cut up the proven boards you've been using.

The DIR9001 does a very good job by itself but I believe that because of personal experience when feeding it by an Off Ramp. Bill Allen can tell you about connecting his Pace Car to his Valab and how well it's worked for him. I'm not completely sure it was simple coax SPDIF but I'm reasonably sure it was. He works with serious high end hardware and he's said the modded Valabs can sound magnificent in those systems.

Quote:
Originally Posted by Wood View Post
Steve, excellent 'proof of concept' achievement. So a future iteration of the Teradak series could/should incorporate a I2S input of the bat and so it would be far easier to use with the Pace car and off-ramp, and even Teradaks own reclocker which uses I2S output (I think). But its good to know that this very valuable upgrade path is possible!
I'm very curious to see how this pans out in the coming months but it's painfully clear that properly installing Steve's I2S mod is for an advanced modder. Steve has given kudos to properly implemented SPDIF and he obviously believes it can work well in his own products. I'd be very happy if we got a proper simple SPDIF input board (sans TCXO) from Michael.
post #767 of 1144
Quote:
Originally Posted by driguy View Post
Hi Rhodes,
First off, thanks for doing this mod. I have been thinking about doing something along this line but I wondered about the efficacy as the sound I am getting is pretty impressive but after seeing your scope shots the choice is made pretty clear. Again, it means spending somewhere close to the price of the unit to get better sound but I am willing to do it at this point. Just like laying out the money for the Dueland caps, sometimes you just have to bite the bullet.
That being said, I could not see in your photos where the coax shield is connected. Could you help me out there?
Thanks,
Tony
Hi Tony,

It is very efficient to do this mod. I'm still amazed that so much information gets lost with the stock TCXO. And those highs are finally not hard/harsch anymore. They are transparent and smooth at the same time.
For example: The latest CD of Yello "Touch" is a very good recording but the highs are quite on the bright side. This makes it very detailed but also quite harsch. With a few songs I thought the "s"-sound of the singers where distorting but it tunred out to be the bad clock. Now the "s" sound like it should sound. :-)

The coax shield needs to be connected to a GND spot on the upsampling board. I used the GND spot of the old TCXO. It is located on the top left.

Next thing to test is to replace the 12 Mhz XO for the WM8805 S/PDIF and USB receiver. I'm not sure if it will help a lot when the upsampling baord is used (because of the fifo buffer of the AD1895) but I'm very curious what happens if the WM8805 has a new clock and the upsampling board is bypassed.

One more imporant thing: Be sure that the powersupply of the new clock you will install is very clean! I have chosen for the Tentlabs XO because of the basic but very efficient design: Galavanic Isolation and filtering.
post #768 of 1144
is there any chance of using I2S from a Philips 960? Should have ample room if I scrap the digital board and insert the valab board, only a short run of wire required (I recall I2S is only any good below 20cm)
post #769 of 1144

More isolation

Talking about isolation psu, this is good news for those of us using any kind of usb input.

ADUM4160 from analog devices completly isolates the dac from the polluted computer power supply.

ADUM4160 | Full/Low Speed USB Digital Isolator | Digital Isolators | Interface | Analog Devices

I have ordered a completed product to use on my async device and will try it on the usb input to the Chameleon as well. I got it from here.

Ultravox - SGD60.00 : Welcome to the diyparadise e-store!

Will report back when its up and running.

With an ultra clean usb input to my async device it may even be possible to remove the pulse transformers.
post #770 of 1144
IMHo will an ultraclean input not replace the pulse transformers unless you can also improve the signal transmission with the input that drives your interlink...
post #771 of 1144
Quote:
Originally Posted by Wood View Post
Talking about isolation psu, this is good news for those of us using any kind of usb input.

ADUM4160 from analog devices completly isolates the dac from the polluted computer power supply.

ADUM4160 | Full/Low Speed USB Digital Isolator | Digital Isolators | Interface | Analog Devices

I have ordered a completed product to use on my async device and will try it on the usb input to the Chameleon as well. I got it from here.

Ultravox - SGD60.00 : Welcome to the diyparadise e-store!

Will report back when its up and running.

With an ultra clean usb input to my async device it may even be possible to remove the pulse transformers.
That's a nifty little device Wood and well worth the cost IMO ! I'd add a small film bypass cap to the large *electrolytic on that board

* Maybe swap out the two electrolytic caps for Black Gates or Rubycon ZAL/ZL's if you have them on hand.

It might be well worth hard wiring the USB output of this supply to your USB input board discarding the extra connection jacks. Might also be worth making a custom USB cable that eliminates the V+/V- runs of wire from your transport cable to your DAC.

I predict this little gem will improve your DAC's USB performance by a pretty fair margin.

Peete.
post #772 of 1144
Seem a very interesting solutions prior to amplify signal, use this Optical Attenuator (Lightspeed) but I have read that there is an impedance matching issue:

"For a good match in a system the Lightspeed Attenuator should see:
1 A CD player or Dac that has an output impedance of less than <1kohm (1000ohms)
2 The poweramp solid state or tube should have an input impedance of more than >47kohms (47K)

When these parameters are met you will have the deepest cleanest bass, most dynamic transients, most transparent sound stage and cleanest sweetest highs, you have heard."

How about Chameleon impedence output wired with Lightspeed?

Many thanks,
Andrea
post #773 of 1144
brianmay, I use a 50k pot on my amp with no issues, so I should think the Chameleon is pretty standard, in this regard, when it comes to system matching.

There is a lot of buzz around Optical Attenuators at the moment, I would think the only other contender, that would give this a run for its money would be a TX2575 resistor based pot that somes guys on the Valab forum have been threatening to build (we are still waiting!). For now a TKD pot does it for me.

I guess its going to get harder to keep this thread from drifting a little, I am having to work now on my other system components to reap the full benifits from the Chameleon as it progresses. So if a suggestion of a good attenuator, etc along the way, helps that process, its welcome in my book Andrea (as long as we don't get to astronomical about it ).
post #774 of 1144
Quote:
Originally Posted by Wood View Post
brianmay, I use a 50k pot on my amp with no issues, so I should think the Chameleon is pretty standard, in this regard, when it comes to system matching.

There is a lot of buzz around Optical Attenuators at the moment, I would think the only other contender, that would give this a run for its money would be a TX2575 resistor based pot that somes guys on the Valab forum have been threatening to build (we are still waiting!). For now a TKD pot does it for me.

I guess its going to get harder to keep this thread from drifting a little, I am having to work now on my other system components to reap the full benifits from the Chameleon as it progresses. So if a suggestion of a good attenuator, etc along the way, helps that process, its welcome in my book Andrea (as long as we don't get to astronomical about it ).
I bought the Tentlabs Volume Control a few months ago and I'm very satisfied with it. I had a Goldpoint before but the Tentlabs beats it easily, even with the stock Vishay Dale resistors. I recently replaced them with TX2575 resistors and that took it to a level I never heard before.



A nice case is currently being made. :-)
The Tentlabs volume control has some nice extra's like 4 inputs and remote control.
post #775 of 1144

jeepers!

rhodes54, you are truly the original non-compromise guy. This pot could well be the best pot on the planet as of now! But wot u gonna do with just one arm and leg?
post #776 of 1144
Quote:
Originally Posted by Wood View Post
rhodes54, you are truly the original non-compromise guy. This pot could well be the best pot on the planet as of now! But wot u gonna do with just one arm and leg?
LOL. :-) I just try to get the best possible quality within' a certain budget. Although this will cost me like 600-700 Euro in total it will beat every pre-amp in the same price range (and probably far above).
Luckily my tube monoclocks don't mind to get driven by a passive controller like this.

It could be one of the "best" volume controls in the world though it weakest point is the relays. But even then, a relay is still far better than a potentiometer or switch.

Anyway: One of my friends claims he found out how to control the volume in the analogue domain without one single part in the signalpath! He didn't want to tell me (yet) how he did it but if he really manages to get it working this will be a very valuable invention. Can you imagine a volume control which is totally uncoloured and not affecting the signal in any way? :-)
post #777 of 1144
Quote:
Originally Posted by Bill Allen View Post

The next logical step is to move on to building your very own power amp. Keep it simple and build it yourself ... or mod an amp that has potential with quality parts.

That my friends is not a job ... it's an adventure!
yeah Bill!! Thanks to hanging around on these threads, after moding my valab dac, I am now moding my Conrad Johnson PV3 with stepped attenuator, and russion teflon caps. I also am looking at a new power supply, and chokes to replace grid and anode resistors.

In the meantime i am studying the preamp signal paths and reading what i can about tube amplifiers. It is all very interesting stuff. It would be a more difficult process without the comments everyone leaves on these threads.
post #778 of 1144
Errnst, that is building a new amp, but you are using parts of the old one.....

Not sure why you would go that route over building a proven design, as a good tube amp is designed and tweaked for best sound. In f.e the well known Onganku (Kudos to Bill as I'd give my right arm for a pair after having auditioned them a lonf time ago) every little detail was auditioned and tweaked and redesigned etc.....
Modding a PV3 this drastically may get you something completely different for sure but whether it is better?
Not saying you should'nt but I;d like to keep anyone from tinkering with a perfectly fine system only to discover that it was fine to begin with the hard way.
post #779 of 1144
Quote:
Originally Posted by marcelnl View Post
Errnst, that is building a new amp, but you are using parts of the old one.....

Not sure why you would go that route over building a proven design,
It is pretty much building a new amp, still this route is easier and can be taken in steps, maybe next time i will start closer to scratch. The pv3 is not all that perfect, it was the bottom of the conrad johnson line in the first place, i already ran it for 15 years after buying it used. The volume control was old cheap and noisy. So far a new volume control and replacing half of the coupling caps has improved it a lot.

As far as the comment about keeping anyone from tinkering with a perfectly good system.... what???.... isn't that why people come here in the first place? because they already tinker. I did make some comments about how reading these threads can be very encouraging to me, i guess it can go the other way as well.
post #780 of 1144
Now here is a fun thing. I have only used the usb once, just for a few minutes, after I got the Chameleon, because the spidf input with the musiland async device was clearly superior.

But for the last 3 days I have been running on direct usb from the pc only, this is to burn in anything that might need burning in and also to get used to the sound from the usb.

The reason being; I can compare the isolator thats on its way, working with the musiland vs working with usb only.

http://www.head-fi.org/forums/f7/cha...ml#post6537353

But the usb sounds much better then I remember, it doesnt have the soundstage, and detail of the Musiland, but it is more musical and natural sounding. The Musiland sounds more synthetic/strident by comparison, and we dont like that kind of thing here do we?

If the isolater reduces jitter with the straight usb input, I might well find it works better, in all ways, than using the fancy pants async Musiland device. Then again the usb powered Musiland might improve with the isolator.

So its usb isolator vs cheap async device.

Watch this space!
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Dedicated Source Components
Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › Chameleon DAC listening and modifications