Who knows about USB?
-
- Shed dweller
- Posts: 2300
- Joined: Fri Aug 03, 2007 10:57 am
- Location: From the land of the Bodgers
#1 Who knows about USB?
Trying PC computer audio again. USB power lines and grounds are incredibly noisy so on the little PCM2704 based USB to SPDIF converter Im messing about with, I've isolated Vbus power from the USB interface and injected clean 5v into it. Noise is still high but disk access now no longer causes a voltage drop / noise pulse....
....But I cant isolate ground. If I lift ground the device is not recognized by the PC....the D+ line is pulled high to indicate a full speed device, I don't really want to convert the PCM2704 to self powered mode IE: I'm looking for a quick lazy way around the problem! I suspect the HOST pin needs to be routed to D+ as well but that definitely would mean converting to self powered mode.....maybe I've just answered my own question!
....But I cant isolate ground. If I lift ground the device is not recognized by the PC....the D+ line is pulled high to indicate a full speed device, I don't really want to convert the PCM2704 to self powered mode IE: I'm looking for a quick lazy way around the problem! I suspect the HOST pin needs to be routed to D+ as well but that definitely would mean converting to self powered mode.....maybe I've just answered my own question!
- Mike H
- Amstrad Tower of Power
- Posts: 20189
- Joined: Sat Oct 04, 2008 5:38 pm
- Location: The Fens
- Contact:
#2
I assume you mean USB audio out?
AFAIK USB devices need to use the computer's 5V from its USB port.
I get noise as well, it's caused by the computer's SMPS. Albeit this is a notebook with no mains earth. You'd think that would be an advantage wouldn't you?
Either I need to have earth lift resistors on ALL the sound system stuff that the USB audio o/p connects into, OR use stereo 1:1 isolation transformers for the L/R audio channels. These have to be good quality though.
AFAIK USB devices need to use the computer's 5V from its USB port.
I get noise as well, it's caused by the computer's SMPS. Albeit this is a notebook with no mains earth. You'd think that would be an advantage wouldn't you?
Either I need to have earth lift resistors on ALL the sound system stuff that the USB audio o/p connects into, OR use stereo 1:1 isolation transformers for the L/R audio channels. These have to be good quality though.
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
#3
Neil wrote:
Mike wrote:
I reckon so too....maybe I've just answered my own question!
Mike wrote:
not so sure about this......USB2 is specced at 5 units drawing 100ma which means 1 unit can use all 5 allowances giving a max of 500ma.....older external HDDs were self powered to get around this limitation......any unit can be self powered, which may lend more noise isolation....AFAIK USB devices need to use the computer's 5V from its USB port.
There's nowhere you can be that isn't where you're meant to be
-
- Shed dweller
- Posts: 2300
- Joined: Fri Aug 03, 2007 10:57 am
- Location: From the land of the Bodgers
#4
Mike USB audio devices can be self powered which is probably what I'll have to do with this PCM2704...this way I hope to be able to separate the ground plane of the PC from the converter. Injecting 5v with the grounds connected helps a touch but theres still a lot of noise present...or I could just use an isolator....hmmmm
#6
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
-
- Shed dweller
- Posts: 2300
- Joined: Fri Aug 03, 2007 10:57 am
- Location: From the land of the Bodgers
#7
Thanks Nick that may be a solution. Clive, iFI is too expensive, does not fit what I'm trying to do
I made up a 3v3 supply today and lifted the earth again. I repositioned the D+ lift resistor and used it to lift D+ to my new 3v3 supply, the board is now recognized by the PC but it's 'unknown'. iE the pc is not enumerating the device correctly...but it is detecting insertion and removal.....
I made up a 3v3 supply today and lifted the earth again. I repositioned the D+ lift resistor and used it to lift D+ to my new 3v3 supply, the board is now recognized by the PC but it's 'unknown'. iE the pc is not enumerating the device correctly...but it is detecting insertion and removal.....
- Mike H
- Amstrad Tower of Power
- Posts: 20189
- Joined: Sat Oct 04, 2008 5:38 pm
- Location: The Fens
- Contact:
#8
I think this is an earthing problem rather than supply from where. Presumably the USB will still have to be 'earthed' to the computer, that's where the problem is
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
#9
Yep, what I do is let the USB interface be powered from the USB host, then take the i2s out via those isolators, so the DAC ground is isolated from the computer ground.Mike H wrote:I think this is an earthing problem rather than supply from where. Presumably the USB will still have to be 'earthed' to the computer, that's where the problem is
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
-
- Shed dweller
- Posts: 2300
- Joined: Fri Aug 03, 2007 10:57 am
- Location: From the land of the Bodgers
#10
Yes, there appears to be a dependency on the ground being connected before the device can be enumerated. I converted the muse board to self powered mode but it still has the same issue if the ground is lifted. So it appears it may not be possible with this little board to separate the ground from the DAC...certainly not with any ease... Plan B time...
-
- Shed dweller
- Posts: 2300
- Joined: Fri Aug 03, 2007 10:57 am
- Location: From the land of the Bodgers
#11
...I let the magic smoke out of plan B... Valves are so forgiving but 3v3 silicon really does not like 12v. I really should know by now after taking out an SB touch with the wrong power supply...sigh!
-
- Shed dweller
- Posts: 2300
- Joined: Fri Aug 03, 2007 10:57 am
- Location: From the land of the Bodgers
#12
Plan B is back up and running, a little isolator based on the ADuM4160. whats interesting is the little USB to SPDIF board I've been messing about with and is now self powered still shows 100mV or more of noise on the 3v3 supply even with the isolator in place, these's something very odd with it. Substituting a different board with very different layout but still based on the PCM2704 is a lot cleaner and this board is bus powered...
Oh well, I've gone from this sort of noise on the USB VBUS (the drop out is when the hard disk is accessed)
To this: a much more stable and cleaner supply although there are some small spikes remaining but I' not sure if this is the scope or not.
Oh well, I've gone from this sort of noise on the USB VBUS (the drop out is when the hard disk is accessed)
To this: a much more stable and cleaner supply although there are some small spikes remaining but I' not sure if this is the scope or not.