Hi,
I can testify to the problems that the PCM2707 has with some computers or versions of Windows (German windows xp) . As soon as my friend and I upgrated to SP2 both his and my dac that had been working flawlessly up to then did not work in self-powered mode anymore.
I have had so many failures to get the PCM2707 working in self powered mode, that I just gave up on it.
Brent was trying his best to help me, but I just could not succeed even with the Bord that Brent send me (thanks again Brent).
I had to change the chip so often that some traces came off.
At the moment I am designing a board that will be a using ideas of Brent's and Doede's designs. Everything is going a bit slow as I have very limited time.
I found a board manufacturer in the States that does 40 square inches for $14, but I have to order 5 of these boards, so I try to cram as many projects as possible on to the available area. I finished the Tripath TA2020 board, an electronic crossover board and are currently working on the USB dac.
I am also half way done with the Tripath RB TK2051 Evaluation Bord. I just have to try it because of the 6Moons review of Vinnie Rossi's Signature 30 which is obviously just using the Eval.Bords with the minor change of getting rid of the LED, using Jensen coupling caps and using a better and bigger supply cap. The author puts the sonics of the Signature 30 in the same league as $40.000 tube amps.
MMM.6moons.com/audioreviews/redwine4/30.html
I might have to go up to 60 square inch boards to do all this but it would only be three dollars more.
I slowly develop a passion for designing boards.
Anyway I will end up with some spare boards, so I have some stuff to trade if anyone is interested.
I have not forgotten you Brent as soon as the boards are back ( Probably end of Sep.) I will send the TA2020 to you to try.
Klaus
I can testify to the problems that the PCM2707 has with some computers or versions of Windows (German windows xp) . As soon as my friend and I upgrated to SP2 both his and my dac that had been working flawlessly up to then did not work in self-powered mode anymore.
I have had so many failures to get the PCM2707 working in self powered mode, that I just gave up on it.
Brent was trying his best to help me, but I just could not succeed even with the Bord that Brent send me (thanks again Brent).
I had to change the chip so often that some traces came off.
At the moment I am designing a board that will be a using ideas of Brent's and Doede's designs. Everything is going a bit slow as I have very limited time.
I found a board manufacturer in the States that does 40 square inches for $14, but I have to order 5 of these boards, so I try to cram as many projects as possible on to the available area. I finished the Tripath TA2020 board, an electronic crossover board and are currently working on the USB dac.
I am also half way done with the Tripath RB TK2051 Evaluation Bord. I just have to try it because of the 6Moons review of Vinnie Rossi's Signature 30 which is obviously just using the Eval.Bords with the minor change of getting rid of the LED, using Jensen coupling caps and using a better and bigger supply cap. The author puts the sonics of the Signature 30 in the same league as $40.000 tube amps.
MMM.6moons.com/audioreviews/redwine4/30.html
I might have to go up to 60 square inch boards to do all this but it would only be three dollars more.
I slowly develop a passion for designing boards.
Anyway I will end up with some spare boards, so I have some stuff to trade if anyone is interested.
I have not forgotten you Brent as soon as the boards are back ( Probably end of Sep.) I will send the TA2020 to you to try.
Klaus
Hi Klaus,
Sorry to hear you've had so many problems with this chip - in my experience so far its more sensitive than most of my DIY chips - but I blame my failures so far with this chip to be my stupidity.
I haven't had any computer related issues yet - on XP UK version (Ireland) but then again I haven't yet gone the self powered route - I intend to but still waiting to hear the TDA1543 output as the base configuration & evaluate modifications.
However, if XP turns out to be a problem I will go the Linux route on my final music dedicated minimal PC client - music server is going to be Linux based anyway, networked through LAN.
You've got to put together a lot of projects to fill 4 40 inch boards - keeps you busy I bet - still good prices for manufacture
Good luck with all your work - lets know how they sound
John
Sorry to hear you've had so many problems with this chip - in my experience so far its more sensitive than most of my DIY chips - but I blame my failures so far with this chip to be my stupidity.
I haven't had any computer related issues yet - on XP UK version (Ireland) but then again I haven't yet gone the self powered route - I intend to but still waiting to hear the TDA1543 output as the base configuration & evaluate modifications.
However, if XP turns out to be a problem I will go the Linux route on my final music dedicated minimal PC client - music server is going to be Linux based anyway, networked through LAN.
You've got to put together a lot of projects to fill 4 40 inch boards - keeps you busy I bet - still good prices for manufacture
Good luck with all your work - lets know how they sound
John
There is what seems to be an error in the datasheet. When powered by the usb bus (bus powered) the datasheet says to us 5V, but in reality you should regulate the 5V usb supply down to 3.3V. I am in the midst of a conversation about this, but no real answers yet., but it does not work with 5V.
When it is self powered (i.e., not by the usb bus) it should also get 3.3V.
When it is self powered (i.e., not by the usb bus) it should also get 3.3V.
I will only fill one board with the 60 square inches, but the manufacturer will produce five of them.
The PCM2707 will internally regulate the 5-6V of the USB-bus to 3,3V. You can not pre-regulate, it will not work when it does not have more Voltage on the unregulated side as on the regulated one.
Klaus
Attachments
Radian said:
jkeny said:
Yes, I have read the data sheet over and over and I cannot explain this at all. My last post was missing a word, which is that I am in the midst of a conversation with TI about this.
On my 1543 project (http://www.ecp.cc/HPDAC.html) you have to use a 3.3V regulator. I tried with the raw USB and with a 4.5V reg, and neither worked. In fact, one person toasted a chip using 5V.
Please have a look at the schematic, but it works with a 3.3V reg and not with 5V.
-d
According to datasheet, the only supply pin that tolerate 5V is Vbus. The other supply pins (Vccp, vdd) have "absolute maximum rating" of 4V, and in your schematic they are all linked together. Even if it don't fry, it may simply not work because of the too high voltage on the digital input pins (D+, Psel, etc.) They have absolute maximum of 4V too.dsavitsk said:
If you supply regulated 4.5V or raw 5V from USB port only to Vbus pin (no supply to the other V pins) I guess it will work fine.
It's my guess... let us know what TI says.
Now I see 
. Because you are feeding Vbus pin with 3.3V the internal regulator probably is not active (it requires minimum 4.35V). This means Vccp and Vdd are being powered exclusively by the reg101 
.
Seems to be a clever way to have the "configuration descriptor" set to "bus powered - 500mA" and at the same time have the pcm2707 working like in "self powered" mode, fed by external regulator.
My circuit (schematic in the first post) has the "descriptor" set to "self powered - 20mA", but it draws more, approximately 40mA
. I think it's not a problem, because the exceeding current is small. But in a more "hungry" circuit, like the dsavitsky's one, it becomes important to set the "descriptor" properly.
. Because you are feeding Vbus pin with 3.3V the internal regulator probably is not active (it requires minimum 4.35V). This means Vccp and Vdd are being powered exclusively by the reg101 .Seems to be a clever way to have the "configuration descriptor" set to "bus powered - 500mA" and at the same time have the pcm2707 working like in "self powered" mode, fed by external regulator.
My circuit (schematic in the first post) has the "descriptor" set to "self powered - 20mA", but it draws more, approximately 40mA
. I think it's not a problem, because the exceeding current is small. But in a more "hungry" circuit, like the dsavitsky's one, it becomes important to set the "descriptor" properly.- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.