Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 26, 2012 14:39:15 GMT
Chris I think CLK from the amanero goes into the PK's BCLK.
You have the choice of using async or sync modes, if you want the first (using local clock) leave MCLK/SCK unconnected, if you want to try sync, then remove power from the local PK's clock and connect MLCK to SCK. It is very easy to forth and back and compare which one you prefer.
|
|
XTRProf
Fully Modded
Pssst ! Got any spare capacitors ?
Posts: 5,689
|
Post by XTRProf on Sept 26, 2012 14:47:22 GMT
Eh, without Async, how do the dac chip play those frequencies that are not a multiple of the source? So must be Async out of no choice, right?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 26, 2012 14:55:12 GMT
Chong, there ia chocie of BOTH sync and async. The Amanero board provides the correct clock frequency for each sampling family on the MLCK, if you play 44.1, 88.2, 176.4, 352.8 KHz or DSD it will provide a 22.5792MHz clock and if you play 48, 96, 192 or 384 KHz it will send a 24.5760MHz one. This is from the ES9023 Datasheet:
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 26, 2012 15:17:01 GMT
Chris I think CLK from the amanero goes into the PK's BCLK. You have the choice of using async or sync modes, if you want the first (using local clock) leave MCLK/SCK unconnected, if you want to try sync, then remove power from the local PK's clock and connect MLCK to SCK. It is very easy to forth and back and compare which one you prefer. A typo? ,
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 26, 2012 16:25:51 GMT
Very Important!!! I believe cannot have simultaneously SPDIF and I2S. I'd suggest you remove power from all SPDIF circuit or even better, remove resistors by the I2S inputs.
Amanero -> PK
White paint/tipex -> MRCLK (only in sync mode, remove power from local clock) FSCLK -> LRCLK BCLK -> BCLK DATA -> DIN
In the future: (I2C bus) pin1-> SDA (not available on Es9023) pin 2->SDL (not available on Es9023)
3.3V output from Amanero board can be used as trigger or PS up to 30mA. DSD output can be used as DSD mode trigger for some DACs (not available on Es9023)
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 26, 2012 21:35:33 GMT
I saw this on the Amanero thread ay DIYAudio and think it looks interesting, instead of connecting a single wire from Amanero's board to the PK's ground why not twist one G wire around every signal one and tie them at the PK's end to connect the lot to ground? it could provide some protection against noise. The 2nd row of pins is mostly grounds so very easy to do.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 26, 2012 21:46:19 GMT
.... or use braided wires the same way?
|
|
pagan
Been here a while!
Posts: 512
|
Post by pagan on Sept 27, 2012 7:50:14 GMT
....or CAT5e lan cable, inside pairs pre-twisted.
or thin coax, something like RG174.
Whatever you use, keep it short, and don't cross them.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 28, 2012 20:58:17 GMT
The red arrow points to the LDO that will need to be removed if one wants to use better quality external power, as can be seen on the pic the onbgoard reg (ADP-151) is a tiny SMD part so not for soldering noobs like me. @will, D For those who don't have it, here is the datasheet: amanero.com/drivers/combo384-D.pdfI was just running through the data sheets and CCTs and found the above needs a little correction/alteration. As I see it the correct positions for the ADP151 (5v > 3.3v) and ADP150 (3.3v > 1.8v) are as follows. (if I'm wrong please say!) Therefore, "U2" is the first one to attack to remove the 5v USB bus power from the board and feed in something superior. Then "U3" could be attacked to feed in a better 1.8v if one feels it necessary. PS, well done Javier on getting first blood for music
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 28, 2012 21:46:21 GMT
That is correct, removing U2 (#2 in your pic) will cut power from USB Vbus. In fact removing L1 will do the same and seems to me a little less risky.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 28, 2012 22:42:59 GMT
Yes, both ways are good. Removing L1 means you can feed in a 5v supply of your own. Removing the ADP151 means you can feed in a 3.3v supply of your own and take a power regulator out of circuit in the process. Of course this depends on how brave one is feeling
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 29, 2012 6:06:56 GMT
Correct me if I'm wrong but even if you leave U2 on board, once L1 is removed and you connect the external PS to the 3.3V and GND pins on the back of the board (by the I2S pins) the ADP-150 won't do anything as it hasn't any power coming in, only the outputs remain connected so it is efectively asleep, right?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 29, 2012 6:19:37 GMT
Correct me if I'm wrong but even if you leave U2 on board, once L1 is removed and you connect the external PS to the 3.3V and GND pins on the back of the board (by the I2S pins) the ADP-150 won't do anything as it hasn't any power coming in, only the outputs remain connected so it is efectively asleep, right? Hi Javier You may be correct, but the data sheet doesn't show much about the internals other than a simple block diagram. Personally, I wouldn't take chances there. Kind Regards Alex
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 29, 2012 6:48:32 GMT
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 29, 2012 6:56:25 GMT
Hi Javier I was referring to the ADP-150 itself Regards Alex
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 29, 2012 7:01:55 GMT
I don't understand what you mean. The regulator is only connected to Vin comming from L1, GND and Vout to the rest of the circuit, unless it takes signal from Vout and routes it to GND I don't quite get what other thing it can do once L1 is removed.
|
|
|
Post by gommer on Sept 29, 2012 11:14:59 GMT
Javier,
Absolute maximum ratings in ADP-151 datasheet: VOUT to GND −0.3 V to VIN
That's why there is a problem. VIN is undefined (zero? or higher because of reverse leakage?) when no power is fed at VIN, but 3V3 is at output.
Marc
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 29, 2012 11:27:44 GMT
Marc,
You beat me to it!
The best course of action looks to be the removal of U2 (ADP151) and then take advantage of pins 19/20 (back set of 3.3v) and pin 18 GND. Thus saving messing around with the tiny pads that the removed 151 reveals.
The 150 may be tad more fiddly...
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 29, 2012 12:07:28 GMT
Point taken. Thanks for the info. Anyway, my skills (and sight) are not up to that task, if someone voluteers to remove mine I wll consider getting one of those Salas BIB kits that come with two positive and one negative board.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 29, 2012 12:30:32 GMT
Hi Javier Provided that the VREGs are low noise types, you will often find that a good shunt regulator preceding them can give excellent results too. I get excellent results in my X-DAC V3 with a JLH supplying them, despite Allan's reservations originally. I would have preferred to have used non current limited JLHs after the 5V and 3.3V regulators to clean up residual noise.That would have been awkward to implement, but the results were far better than expected. Regards Alex
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 29, 2012 23:58:48 GMT
Hi Alex,
You lost me a bit there, especially in relation to the Amanero?
Chris
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 30, 2012 0:13:26 GMT
Hi Chris What I was trying to suggest is that the ADP151 IS a low noise device, and there may be only minor gains in bypassing it.Perhaps if the supply rails were derived from a higher voltage Salas or other low noise source, then the onboard low noise regulators would have virtually no work to do other than supply the appropiate lower voltage rails.I haven't closely followed the thread, so my comments may not be applicable, or helpful, in which case please ignore them. My comment was mainly addressed to Javier, who doesn't feel comfortable about modifying areas like this. I wouldn't either. Kind Regards Alex
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 30, 2012 0:20:26 GMT
Ah, OK.
I was thinking my first stop would be supplying a JLH 5v in, most likely at the USB socket pins under the board while using a 5vbus-disabled USB lead....
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 30, 2012 0:46:58 GMT
For those contemplating bypassing the ADP151 etc. , I would point out that the ADP15x is a new breed of Ultralow Noise,200mA, CMOS Linear Regulator. The data sheet states "Ultralow noise : 9uV RMS." The original JLH PSU Addon design quoted 4uV typical. My first step would be to replace the existing incoming USB +5V SMPS with a +5V JLH or Salas supply , and in addition, disconnect the Vbus lead at the PC end of the USB cable if doing so does not affect it's operation.Sometimes these circuits need to see the ïncoming +5V from vbus, in which case methods must be used to simulate that, such as used in the S.C. USB power PCB where the presence of incoming USB power switches the power on and off in the replacement power feed.
Alex
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 30, 2012 10:46:19 GMT
The guy I mentioned on page 2 specifically ordered his board without the U2 ADP-150 regulator. He powered it using the pins at the end of the card and had no problem whatsoever with the card not being recognised by Windows nor had any other malfunction. I'm no expert but according to the schematics it would seem power from the Vbus goes through U2 only (can't see any other trace for incoming power) so I would asume cutting the Vbus wire of the USB cable won't stop the card from working properly.
As the ADPs seem to have rather nice specs so I won't be trying any external PS for a while. Later on, if I decide to take that step I also think the correct way to do it is removing U2 and use the PS pins at the back, though I don't expect that even a Salas would make any huge improvment, sure it'll perform better but not a night and day difference.
|
|