[stemiki]

Would be nice to get in touch with @Reactcore and @stemiki to have more details on battery supply

Having the entire system completely isolated from home electricity leads to results that, once heard, you can no longer do without!

 

[Reactcore]

Having the entire system completely isolated from home electricity leads to results that, once heard, you can no longer do without!

You are using Opto DX or toslink from Scaler to Dave? To fully isolate

 

[Reactcore]

Hi. Reading this thread about the optical link some weeks ago, I decided to give it a try and started to order all the necessary things to make it . This is my first project in many years , because even if I'm an electronic engineer , I was too busy with my former job so , now that I'm retired I have a lot of time to dedicate to my hobby, my music system.

I made some changes to the original Reatcore idea:
-I decided to use aluminum cases to have a better shielding
-I decided to not take the power from the MScaler and TT2 power supplies , but to make them run on battery

The reasons are relatively simple: I spent a lot of money on 3 LPS from Sean Jacobs , an original DC3 I purchased second hand from the same person who sold me the MScaler , and 2 others I personally made using SJ original boards and purchasing all the other parts to make them aesthetically and technically almost identical to the original DC3.
I took that decision as I was totally disappointed by the sound of the original SMPS provided with the devices from Chord.
I found also that even replacing the umbilical cable between LPS and MScaler was impacting the sound quality.
For these reasons I decided to use 2 Lithium 9V rechargeable batteries .
They are rated 9V 650mAh: in reality they give around 8.7 V when they are fully charged , then the voltage decreases as the charge is depleting .
For this reason I slightly changed the power voltage regulation circuit replacing the trimmer with a 5.6V Zener diode polarized with a 50 Ohm resistor: this keeps the 5V relatively stable for a long time.
The transmitting box absorbs around 120mA giving the battery around 5 hours of running time before charge is gone.
The receiving box absorbs around 80mA (less components to run).
Just a note about this solution: when the battery level reaches the lowest level the link becomes unstable so the TT2 starts making very bad cracking noises. I think I will work on a different solution that gives more hours running, or I will try with a separate LPS (I'm not sure if this can give problems).

The batteries have an internal charging circuit , they can be charged using a standard phone charger , so I bought a cheap one with a double output , one for each box, and 2 long cables with USB A termination on one side and a micro USB termination on the other side, to keep it far from the system.
The power goes to a double diverter switch that in ON position powers only a blue power led and a small relay that closes the contacts between the battery output and the board.
When it's set in OFF position, it sends power only to the battery charging circuit ; it can be kept in this way for days without issues as the charge is automatically interrupted by the circuit inside the batteries.
In this way the battery charger is totally isolated from the digital circuit.

The aluminum box required some manual milling work to keep the BNC connectors as close as possible to the right lateral side.
This is needed to avoid interferences between the box and the cables connected to the TT2 and the MScaler.
I used a small Dremel-like mill to make that.
To fit the micro USB female connection on the back plate I had to manually mill the hole: the result is not nice at all, but it's on the back side so nobody can see it
The front plate required just to drill the holes to fit the BNC connectors : their position is not exactly in the center of the plate but is the same for both the boxes even if the TT2 and the MScaler have different height : I used 2 different size sets of rubber feet to have the proper height for each box, in this way the 2 boxes are staying stably on the back of the devices without forcing on the BNC connectors.

I used the same board also for the receiver to have a mechanical support for the components, obviously I had to cut some tracks and use some jumpers to make it working.

A note about trimmers adjustment : I found that the most critical adjustment to have a stable connection was when I used DSD 64 (I don't have DSD at higher resolution, these files are from ripped SADCs), so I first followed the procedure from Reatcore (I was SO HAPPY when the link appeared on TT2 display) and then I observed , as he said , that increasing the resolution it becomes more and more difficult to have a stable connection. It's happened that the A fiber connection was the most critical, so I worked first on that and then set the B fiber trimmers on the same position. Follow his recommendations to avoid to damage the optical transceivers, they are the most expensive part. If you can reach a stable connection with DSD64, then all the other lower resolutions will work flawlessly.

And now some comments on the most important thing: how they sound?
For sure MUCH MUCH MUCH better than the original cables provided by Chord: much more informations , more clarity and details, they definitely worth the investment in money and time and it was a lot of fun, but also a challenge, making them.
I never had the possibility to test for comparison the Wave cables (the MScaler/DC3 previous owner was using the Storm between it and his Qutest, but I never heard them and I think he decided to sell everything because he wasn't totally happy with his system sound).
I decided for this way essentially for the cost , because I thought it is a very clever and non conventional idea and because I liked to challenge me on this project.
It allowed me also to understand that EMI can be a great problem.
Installing the optical link, I probably removed a big source of EMI and very likely the data transmission is also better as probably is not affected by cables capacitance, but I wasn't totally satisfied as some harshness and "digital artificial sound" was still present .

Reatcore pointed my attention toward EMI effects.

Now I would like to share my experience as I believe the optical link had a very important role on the benefits the sound had.

A couple of weeks ago I bought a 1.5 mt power cable from Puritan Audio, the Classic Plus. They are claiming an EMI absorption effect of their Classic Plus and Ultimate cables, so I decided to give it a try, it doesn't cost a fortune.
I tried it first on my preamp, then to the other devices and realized it was giving the best effect when it was used to power the MScaler LPS: slight less harshness and more calm sound.
All the other power cables I use are very cheap shielded stuff with a good section of apparently good copper.
PLEASE don't laugh , I know many people believe these are bul...its: this is my first experiment with "hi-fi power cables" and I didn't try with an expensive one, but my system allowed me to clearly hear the difference.
The effect may depend from system quality, but also from the problems that may affect it's performances, and this is valid for each component including all the cables (that's my personal opinion).
Reatcore suggested me to move the plug to a different power line: again BIG improvement, this is proof that EMI were coming from the MScaler through the LPS and affecting the other components ).
Observing how the "hi-fi power filters" are made, they normally contains DC filters and common/differential mode noise filters.
I've also seen that in some hi-fi components that uses linear PSUs similar to the ones I use , the power sockets are filtered against common/differential noise. So I searched on the web for informations about these devices that are available in several types and shapes : I decided to give a try to a 6A 250V double stage filter (33 euro+VAT and shipment) that claims a very high reduction of both common and differential mode noise, not only from the power line but also from the load side to prevent it pollutes the power line.
I can say that this filter made the miracle: a HUGE improvement!!
Now I can clearly hear that the sound , which is still slightly "metallic" after powering up the digital chain, opens up in some minutes revealing details I never heard before: it is natural and detailed, with a great soundstage, something I never heard before.
Now it's really nice to crank-up-the-volume knob to enjoy the musicians playing in my room.

This is in my opinion a confirmation that the optical link makes a great job.
Thanks Reatcore for your idea and your hints!!

I will post some pictures soon!

Thanks for posting the pics!
Its proof others can create this optic link too!

Still sounding as good?

 

[mpino61]

Thanks for posting the pics!
Its proof others can create this optic link too!

Still sounding as good?

It’s difficult to say. Every time I make an upgrade, in a few days I hear that there are some other problems.
Still a long way to run to achieve perfection

 

[SteveHulk]

It’s difficult to say. Every time I make an upgrade, in a few days I hear that there are some other problems.
Still a long way to run to achieve perfection

The nice thing about upgrades is that even though a particular upgrade might not be exactly night and day in itself, the cumulative effect of several upgrades could well be.