MDAC First Listen (Part 00101001)

[clivem2]

Given the high margin that UpTone are making, I've no doubt that yours could be cheaper as well as better: as they appear to be under-capitalised, serious competition might persuade them to sell the design to a high-volume manufacturer. And who would be surprised if it was already in the process of being ripped off in China? Either outcome would put a crimp on sales. What's the initial minimum production run?
A low cost trial run to test qc for the FDAC is obviously desirable, depending on how much additional risk capital is required.
If salvaged MDACs were not part of the equation, would you still be keen to go ahead?

Just a few random thoughts.

I expect the design will be in dacs very soon....whether Uptone are involved or not.

I don't think their markup is high. It the world of consumer electronics it's incredibly low volume production. Mine arrived with a stated value of $45 which may be close to their cost. If so the markup is 389%; with the costs of running a business this seems reasonable. 1000% is not untypical but that would be dealer retail price.

 

[JohnW]

A low cost trial run to test qc for the FDAC is obviously desirable, depending on how much additional risk capital is required.
If salvaged MDACs were not part of the equation, would you still be keen to go ahead?

Just a few random thoughts.

The salvage MDAC situation (or more importantly repaying those who decide to use the FWC as a reason to drop out of the FDAC project) is a really concern for me. I feel 100% that the FWC is the correct route - since my return from HK I've been basically "removing" the limitations and compromises I made to the analogue PCB to squeeze the design into the original MDAC package.

The more I rework the PCB the more I realise how I had "Crippled" the design - the FWC version is really looking beautiful!

Ironically, I'm having to modify the design back to earlier 'prototype' layouts - where I had since to remove some ideas and features due to the MDAC size limitations - such as the option for Kelvin sense feedback. I believe Kelvin sense feedback is very important for the ultimate performance and just another example of how the FDAC design pushes the limits and "out of the box thinking"

Even if there was not an issue of the FWC / Salvage MDAC's to consider - its good to have a small very low risk project to get the ball rolling, this also gives be a revenue stream so that I don't feel pressured into making bad decisions that could ultimately effect the quality of the FDAC.

 

[JohnW]

I expect the design will be in dacs very soon....whether Uptone are involved or not.

The prime advantage I see is that RF filtering and "Data Cleaning" is performed external to the DAC enclosure - building this internal to the DAC would remove this prime feature.

I don't think their markup is high. It the world of consumer electronics it's incredibly low volume production. Mine arrived with a stated value of $45 which may be close to their cost. If so the markup is 389%; with the costs of running a business this seems reasonable. 1000% is not untypical but that would be dealer retail price.

I also don't see there cost as too high when all things are considered - I'd target our version with clock-locking and isolation at around GBP130.... its a much more advanced design with linear PSU etc...

 

[davidjt]

Only playing Devil's advocate - the Jitterbug and Regen can go on the junk pile in due course.

 

[Rune]

How much interest would there be in a ReGen type product combined with a RF filter?

I'm thinking of putting a simple design through the FDAC Manufacturing channel to get the ball rolling.... we could even clock-lock the design to the MDAC's internal clock via the MDAC's Clock output it could be used standalone or clock-locked....

The idea is to build up a working relationship with the SMD vendor and supply chain with a simple project - a Clock-Locked ReGen / USB RF Filter seems a good place to start

I will support the project and take one.

It will also be interesting to compare it with the Uptone Regen I have on order.

 

[JohnW]

Do you think this is a better solution (assuming there is a problem) than using
an optical usb cable?

Optical USB would be a perfect solution WRT Electrical isolation - although I'm not aware of such a product that is affordable.

The USB Hub also attenuates the USB host devices clock Jitter - I'm not sure how the Clock is handled with an optical USB link.

 

[adamdea]

Optical USB would be a perfect solution WRT Electrical isolation - although I'm not aware of such a product that is affordable.

The USB Hub also attenuates the USB host devices clock Jitter - I'm not sure how the Clock is handled with an optical USB link.

http://www.corning.com/opcomm/OpticalCablesbyCorning/products/USB-3.Optical.aspx#.Vd76FdyrS71

I seem to remember that the corning ones cost $109 on amazon.com not sure what they cost in uk.
They only seem to make 10m minimum, but perhaps shorter lengths may be possible.
You may find this interesting

http://archimago.blogspot.co.uk/2015/05/measurements-corning-usb-3-optical.html

I can get the possible need for electrical isolation. The need for usb signal regeneration....well, how do we know what is and isn't good enough? An optical cable with electrical usb signal transmitter to acceptable spec would be interesting. Whatever acceptable spec is.

 

[LarsS]

Optical USB would be a perfect solution WRT Electrical isolation - although I'm not aware of such a product that is affordable.

Wouldn't a photo/optocoupler connection between the digital and analog board solve this?

 

[JohnW]

http://www.corning.com/opcomm/OpticalCablesbyCorning/products/USB-3.Optical.aspx#.Vd76FdyrS71

I seem to remember that the corning ones cost $109 on amazon.com not sure what they cost in uk.
They only seem to make 10m minimum, but perhaps shorter lengths may be possible.

This is a PC to a Host device - USB A to USB A Receptacle - I'm not sure it offers electrical isolation as power seems to be transmitted from the PC's USB "A" to the circuits in the USB A Receptacle...

 

[Bob L]

Do you think this is a better solution (assuming there is a problem) than using
an optical usb cable?

There is a lot about the Corning cable on the (several!) Regen threads over at CA. One of the points made was that the impedance loss in the long (10m) very thin power leads to the DAC end was part of the reason for the improvement with this cable, and this led to one of the tricks in the Regen. So far as I know, the Corning is not available anywhere with less than 10m length. Several people have reported that the much cheaper Supra is better as a feed to the Regen; also there are several reports of the Corning being very fragile.

 

[JohnW]

Wouldn't a photo/optocoupler connection between the digital and analog board solve this?

The FDAC has an isolation barrier between the Digital and Analogue Sections, however this helps little with higher frequency isolation which is why its better to prevent RF entering the FDAC enclosure in the first instance.

Its easy to understand how RF energy radiates and how its default to screen it once its allowed to "escape".

 

[JohnW]

BobL,

How can this cable offer electrical isolation when it has A to A USB connections - am I missing something here?

Edit:- it seems to small to include an isolated DC to DC converter, has anyone confirmed it offers electrical isolation - its not claimed as such in the datasheet (unless I missed it)?

 

[clivem2]

There is no simple way to isolate USB 2.0 at the full High Speed data rate of 480Mbps which is required for Audio sample rates above 96KHz.

There are simple isolation solutions for USB 1.1 (12Mbps) which is able to support the MDAC's 96KHz USB port.

The FDAC has an isolated USB that works on the Audio Data path so while it would be great to have a second external isolation barrier - its not as beneficial as it would be for the non isolated MDAC's USB port.

I see the design looking like this:-

1. RF filtering of the USB HOST input.

2. Regenerated USB Data from a "Clean" Clock - with support to Clock-lock the USB BUS to the MDAC / FDAC Audio Clock.

3. USB2.0 ReGen output port with a output filtering.
.
4. USB1.1 Galvanically Isolated ReGen output port with output RF Filtering.

John, I wanted to check, will you include 5V for powering USB cards or even some DACs? From the linear PS.

 

[MPerkins]

hi Guys

It's been awhile since I've been on the forum.

I noticed the posts regarding the bulging cap issue with the MDAC. I would like to take the cover off to check if this is a problem with mine.

Can anybody give me any tips on how to remove the cover, including the torx driver size.


Thanks
Mike

 

[JohnW]

John, I wanted to check, will you include 5V for powering USB cards or even some DACs? From the linear PS.

Yes it can do, but we would need a larger linear AC adaptor for USB "Powered" DACs that require greater then 300mA or so.

The MDAC (requires no power from USB) and FDAC will only requirearound 300mA or so, so would not need a high Current AC adaptor - I'm trying to keep costs down.

 

[clivem2]

Yes it can do, but we would need a larger linear AC adaptor for USB "Powered" DACs that require greater then 300mA or so.

The MDAC (requires no power from USB) and FDAC will only requirearound 300mA or so, so would not need a high Current AC adaptor - I'm trying to keep costs down.

Hopefully powering USB cards rather than DACs would be ok with the standard PS. I'll have to find out what an M2Tech card requires or measure it.

 

[jamjar784]

There is no simple way to isolate USB 2.0 at the full High Speed data rate of 480Mbps which is required for Audio sample rates above 96KHz.

There are simple isolation solutions for USB 1.1 (12Mbps) which is able to support the MDAC's 96KHz USB port.

The FDAC has an isolated USB that works on the Audio Data path so while it would be great to have a second external isolation barrier - its not as beneficial as it would be for the non isolated MDAC's USB port.

I see the design looking like this:-

1. RF filtering of the USB HOST input.

2. Regenerated USB Data from a "Clean" Clock - with support to Clock-lock the USB BUS to the MDAC / FDAC Audio Clock.

3. USB2.0 ReGen output port with a output filtering.
.
4. USB1.1 Galvanically Isolated ReGen output port with output RF Filtering.

Hi John

This is an interesting idea. Many posts back on this long thread there was discussion about HQ PSU's - I think there are at least two 3rd party versions for the MDAC - and that these could improve SQ by feeding clean stable PSU voltages even though the MDAC already has built in filtering and regulation.

Would it be possible to kill two birds with one stone and plug the existing MDAC rectified AC adapter into the new USB isolator? So we'd have 2 outputs: clean DC into MDAC PSU input, and isolated/regenerated 96KHz USB. Hopefully this would be cheaper than building another complete PSU as only rectification/smoothing/regulation would be required?

Just a thought

KR

 

[Bob L]

The salvage MDAC situation (or more importantly repaying those who decide to use the FWC as a reason to drop out of the FDAC project) is a really concern for me. I feel 100% that the FWC is the correct route - since my return from HK I've been basically "removing" the limitations and compromises I made to the analogue PCB to squeeze the design into the original MDAC package.

.......
Even if there was not an issue of the FWC / Salvage MDAC's to consider - its good to have a small very low risk project to get the ball rolling, this also gives be a revenue stream so that I don't feel pressured into making bad decisions that could ultimately effect the quality of the FDAC.

John,

Since I'm one of those owning one of your salvaged MDACs, I would be very keen to support this effort and take one of your ReGens, so long as it will be happy with MDAC-1, and Sovereign, Toy and (old) Fusion modded versions. [ Edit- ignore this; I see you have already mentioned the MDAC-1]

Would the clock-locking mean we couldn't use the clock-lock with, say, a modified Oppo, or does the clock-lock o/p on the MDAC-1 have enough current drive for two devices?

It will be well worth it to compare with my Regens (small g: California, capital G: Czech?)

 

[adamdea]

BobL,

How can this cable offer electrical isolation when it has A to A USB connections - am I missing something here?

Edit:- it seems to small to include an isolated DC to DC converter, has anyone confirmed it offers electrical isolation - its not claimed as such in the datasheet (unless I missed it)?

If i understand this correctly the host powers not only the electrical to optical conversion, but also the optical to electrical conversion. The output data is isolated electrically from the input data, but not actually from the host's power.

 

[JohnW]

John,

Since I'm one of those owning one of your salvaged MDACs, I would be very keen to support this effort and take one of your ReGens, so long as it will be happy with MDAC-1, and Sovereign, Toy and (old) Fusion modded versions. [ Edit- ignore this; I see you have already mentioned the MDAC-1]

Would the clock-locking mean we couldn't use the clock-lock with, say, a modified Oppo, or does the clock-lock o/p on the MDAC-1 have enough current drive for two devices?

It will be well worth it to compare with my Regens (small g: California, capital G: Czech?)

Bob,

The Oppo modifications (if they happen as HDMI might be on the cards) will be designed for the MDAC2 (now called FDAC) so the Oppo clock-lock is not applicable to the MDAC1.

The MDAC1 Clock-Lock connection is via optical link.