AMB Laboratories DIY Audio

[OliverD]

I have recently stumbled upon 2 weird problems with my trusty y2 (pic). As you can see in the picture, I am also using an e24 power switch circuit. This is the root of my first problem:
Ever since permanently hooking it up to my pc and the b22/symasym chain of preamp/poweramp, it was hard to switch it on after turning the power an from the outlet it was plugged into. It switched off immediately once I let go of the switch for 2-3 times sometimes more, before eventually deciding to stay on, while producing a loud screeching noise in the preamp/poweramp chain regardless of whether the input was active or not. Some sort of DC offset i suppose?

This problem led me to alter the y2 to never use power from USB through bridging pin 1 and 2 on J2D, which in turn led to the second weird problem:
Now the screeching high pitched noise appears anytime the power from the y2 DAC is not turned on, it immediately vanishes once I turn it on. This probably means the screeching noise described above was produced while the y2 was switching from USB to external power. This modification didn't help with the turn on problems though. It still sometimes take 2-3 tries to turn it on.

Any ideas what I can do or what can cause this?

[amb]

I have recently stumbled upon 2 weird problems with my trusty y2 (pic). As you can see in the picture, I am also using an e24 power switch circuit. This is the root of my first problem:
Ever since permanently hooking it up to my pc and the b22/symasym chain of preamp/poweramp, it was hard to switch it on after turning the power an from the outlet it was plugged into. It switched off immediately once I let go of the switch for 2-3 times sometimes more, before eventually deciding to stay on, while producing a loud screeching noise in the preamp/poweramp chain regardless of whether the input was active or not. Some sort of DC offset i suppose?

So the ε24 works fine when the preamp/poweramp isn't connected? Please double check to make sure all the resistor and capacitor values used on the ε24 are correct? And you used the correct MOSFET?

The screeching noise occurs even when the preamp is switched to a different input? Can you characterize the screech? Is it just a burst of noise that stops, or is it a continuous noise?

I think your screeching noise may be due to a ground loop. Does it screech when the USB cable is disconnected from the computer (but the DAC is on)?

This problem led me to alter the y2 to never use power from USB through bridging pin 1 and 2 on J2D, which in turn led to the second weird problem:
Now the screeching high pitched noise appears anytime the power from the y2 DAC is not turned on, it immediately vanishes once I turn it on. This probably means the screeching noise described above was produced while the y2 was switching from USB to external power. This modification didn't help with the turn on problems though. It still sometimes take 2-3 tries to turn it on.

In order to truly disable USB power (except for host detect), you not only should jumper J2D 1-2, but you should also remove U1D. Otherwise, U1D may "fight" itself when its output voltage isn't what it's supposed to be.

[OliverD]

So the ε24 works fine when the preamp/poweramp isn't connected? Please double check to make sure all the resistor and capacitor values used on the ε24 are correct? And you used the correct MOSFET?

The screeching noise occurs even when the preamp is switched to a different input? Can you characterize the screech? Is it just a burst of noise that stops, or is it a continuous noise?

I think your screeching noise may be due to a ground loop. Does it screech when the USB cable is disconnected from the computer (but the DAC is on)?

The e24 worked fine for a long time, until I recently decided to permanently hook it up to my PC via USB. This seems to be the problem I hadnt thought of before, and I am feeling kind of stupid now
Once the USB is unplugged, I have no problem turning the e24 and therefore the whole DAC on and there is no screeching noise.

The noise itself is a constant high pitched sound thats at a constant frequency. If its a ground loop, I guess it means the USB power ground isnt perfectly isolated from the chassis. I will look into that once I get some time to take the y2 apart.

In order to truly disable USB power (except for host detect), you not only should jumper J2D 1-2, but you should also remove U1D. Otherwise, U1D may "fight" itself when its output voltage isn't what it's supposed to be.

Ok, thanks for pointing that out, I will remove U1D when I look into the USB ground isolation problem.
Thanks for the very helpful insights as usual!

[OliverD]

I have recently found the time to analyze the noise problem and it seems to be a problem with my special configuration of connecting the ground of the y2 to its Box enclosure and through the mounting screws connecting it to the power supply ground of the surrounding enclosure. I don't think there is a suitable solution for this problem while maintaining USB connectivity so I will opt to use optical input instead (which is better anyway, but I opted out of it due to previously using my iPad as a transport). I might still try to remove U1D (didnt have steady enough hands last time I opened the box to look into the issue) but I guess optical is better anyway due to the possibility of using higher sampling rates.

A question to top it off:
Can I wire the TORX147PL from my main enclosure to the y2 board instead of mounting it on the board if I keep the wires short (e.g. 1-2 inches max)?

And a little off topic as well:
@LinuxWorks: I have seen you using 1-pin sockets for exchangable resistors in the d1/d2 threads iirc. Do you have a mouser article # at hand? I couldn't find the one you used and any hint would be greatly appreciated. I am currently building a phono stage and would like to keep the gain easily exchangeable to accomodate MC and MM cartridges.

[amb]

I have recently found the time to analyze the noise problem and it seems to be a problem with my special configuration of connecting the ground of the y2 to its Box enclosure and through the mounting screws connecting it to the power supply ground of the surrounding enclosure.

You can try to isolate the two enclosures from each other...

Can I wire the TORX147PL from my main enclosure to the y2 board instead of mounting it on the board if I keep the wires short (e.g. 1-2 inches max)?

If you're going to panel-mount then use a PLR135/T or PLR135/T9 (made by Everlight, available from Digi-Key). These have a mounting hole that makes panel-mounting easy. You should put the power supply filter inductor and capacitor (L1D and C12D) next to the receiver module, maybe put all three on a small home-made PCB.

And a little off topic as well:
@LinuxWorks: I have seen you using 1-pin sockets for exchangable resistors in the d1/d2 threads iirc. Do you have a mouser article # at hand? I couldn't find the one you used and any hint would be greatly appreciated. I am currently building a phono stage and would like to keep the gain easily exchangeable to accomodate MC and MM cartridges.

There are strips of machined pin sockets that are "break-apart" so you can use them for this purpose. For example the Mill-Max 310, 311, 315 or 316 series, available from Digi-Key and Mouser. Other manufacturers also offer similar products. I wouldn't recommend this on 2-pin devices if the unit will be subjected to vibration or transportation, to prevent parts from working loose.

[OliverD]

Thanks alot for the pointers as to which components to use for the sockets and the panel-mounted Toslink option! I was actually trying to find both in the mouser webshop but find it extremely difficult to do if you don't know the right search-terms. I usually hate it when people ask for stuff which can be found via a search engine, so I didn actually spend some time trying to find those parts at mouser.

You can try to isolate the two enclosures from each other...

I will do that eventually, but I figured adding the optical input had additional benefits and is actually easier to implement right now.
For a perfect isolation I need to get the correct (isolating) mounting screws (the y2 enclosures hangs on the backpanel inside the PSU enclosure) and I guess I would need to rework the holes I made for the connectors.
I am planning to move the y2 (or build a new desktop DAC *hint* ) to a nicer case (e.g. hifi2000.it) anyways and will use panel-mounted inputs and outputs. But thats still a little distant since I first need to get my b22 preamp/headamp and my monoblocks inside a case

[mikeg88]

For a perfect isolation I need to get the correct (isolating) mounting screws (the y2 enclosures hangs on the backpanel inside the PSU enclosure)

I had to do this on my build as well. I actually copied your method for mounting the DAC...thanks for the clever idea.

To make sure that the y2 enclosure was isolated from the backpanel of the chassis, i used a very thin sheet of polypropylene inserted in between, with small holes on the corners to let the screws pass through.

To ensure the mounting screws were isolated from touching the backpanel, i used something called 'shoulder washers'. They form a sort of 'sleeve' between the threads and the hole in the panel, but also go outside the hole so that heads of the screws dont touch the exterior of the back panel.

The ones i used were black nylon and can be seen in the 'post pics of your builds' thread.

[OliverD]

Thanks for the compliment, it may have been a clever idea, but the implementation wasn't really all that clever
I didnt think about possible isolation problems between the 2 enclosures like you obviously did.

Thank you for pointing out your solution, I should take a look inside my parts bin to see if I have some shoulder washers from TO-220 components left and apply those.

[OliverD]

OK, just to report back and state once again that I have next to no clue what I am doing:
I opened the case with the intention to isolate, but while I was at it removed U1D anyway.
The result beeing:
The high pitched noise is completely gone, so I guess you were right amb, it had something to do with U1D fighting itself...
There is still humming going on when USB is connected, so I will try the isolation but will ultimately go to optical input anyway.