AC vs. DC umbilical power cord
May 28, 2015 at 1:30 PM Thread Starter Post #1 of 25

ginetto61

1000+ Head-Fier
Joined
Sep 28, 2009
Posts
1,185
Likes
53
Location
Milan _ ITALY
Hi !
i have a question about units with external power supply.
Some have a DC umbilical (most of all) power cord (i.e. they have a separate DC power supply) but some have a AC umbilical power cord like the unit here below. 
 

 
which would be the best solution from a technical point of view for an umbilical power cord ? DC or AC ?
Thanks a lot.  Regards,  gino 
 
May 28, 2015 at 1:55 PM Post #2 of 25
The power from the wall is AC, and the power required to drive the amplifier is DC, so the difference is whether the rectifier is in the external supply or in the amp. I don't think it makes much difference.
 
The reason mains power is AC is because it is more efficient to transmit AC power across long power lines. For a short, low power cord that is probably insignificant.
 
May 28, 2015 at 2:16 PM Post #3 of 25
  The power from the wall is AC, and the power required to drive the amplifier is DC, so the difference is whether the rectifier is in the external supply or in the amp. I don't think it makes much difference.
The reason mains power is AC is because it is more efficient to transmit AC power across long power lines. For a short, low power cord that is probably insignificant.

 
Hi and thanks a lot for the valuable reply
I asked because i have a little DIY project in mind in which i will extract the PS from the unit and i will built a separate PS.
But i wonder if to take out only the transformer would be smarter.  I do not like the stock transformer at all. I have one in mind much better.
It seems to me that DC power umbilical are more common .. i surely do not know if there is a technical reason at all. 
Me too i was thinking that AC transmission is more efficient. 
Yes the power cord will be very short ... less than one meter for sure.  The power supply will be very close to the main unit. 
Thanks again,  gino   
 
May 28, 2015 at 2:54 PM Post #4 of 25
External DC supplies are probably more common because they reduce the cost, complexity, size, temperature, etc. of the device. It is easier to throw an off the shelf DC supply in the box with your product than design one yourself. If your project is to redesign the power supply then it could make more sense to do it the other way.
 
May 28, 2015 at 4:31 PM Post #5 of 25
External DC supplies are probably more common because they reduce the cost, complexity, size, temperature, etc. of the device. It is easier to throw an off the shelf DC supply in the box with your product than design one yourself.   If your project is to redesign the power supply then it could make more sense to do it the other way.

 
Hi and thanks again.  My main goal is to use another and better transformer instead of the stock one. 
I would like to use one of this type used here   ... but i cannot find them 
 

 
 
actually i think i have found them !!!
 
http://www.hammondmfg.com/pdf/229Insert.pdf
 

 
 
if they are good for Berkeley Audio they will be fantastic for me 
Thanks again,  gino 
 
May 28, 2015 at 7:05 PM Post #6 of 25
you shouldn't use that xfmr - its not speced for high enough Hipot V for double/reinforced insulation required for a Class II floating product - it should only be used with a 3 wire protective earth gnd
 
if look at the manuals for the Berkley DACs you should see that they all have 3 wire IEC mains cabling
 
and the semi-toroidial "hum bucking" construction doesn't do anything useful for a physically separated application and "costs" more in pri-sec parasitic coupling C
 
 
I would likely not bother changing the Schiit supply
 
but if I did I'd be looking for a 4000 V Hipot double/reinforced insulated EI split bobbin with the required insulation, creepage and clearance distances
 
May 28, 2015 at 7:28 PM Post #7 of 25
Have a search for Tamura transformers on eBay, those split bobbin come up regularly for pocket change.  Only issue is they do need PCB mounted.
 
May 29, 2015 at 2:03 AM Post #8 of 25
you shouldn't use that xfmr - its not speced for high enough Hipot V for double/reinforced insulation required for a Class II floating product - it should only be used with a 3 wire protective earth gnd
if look at the manuals for the Berkley DACs you should see that they all have 3 wire IEC mains cabling

 
Hi and thanks for the advice.  I understand they are not completely safe.  I think i will pass.  I do not want to take the risk.
 
and the semi-toroidial "hum bucking" construction doesn't do anything useful for a physically separated application and "costs" more in pri-sec parasitic coupling C

 
as i see these x-formers used in a very high end unit i thought they have been carefully selected.   It is true that i saw also used toroidals.
A lot depends also on what is before and after the x-former ... there is not just that.
Another honorable case ...
 
http://schiit.com/public/upload/general/imagecache_thumbnail_product_gallery_large/yggdrasil_pcb_1000.jpg
 
not the very same type but with split bobbins ...
 
I would likely not bother changing the Schiit supply
but if I did I'd be looking for a 4000 V Hipot double/reinforced insulated EI split bobbin with the required insulation, creepage and clearance distances

 
of course it is not my intention to mess with the internals.  I could be wrong but i guess that the external PS is nothing but a transformer, or AC-AC converter
rolleyes.gif
 as someone call them
Given the extremely low price of the Magni the transformer cannot be the end of the world.  It must be a very cheap (i.e. poor quality) device.  It is just normal.
A better speced x-former cannot do any damage.  More powerful, better regulation, lower impedance in the secondaries  ... and so on.
A 100 VA thing ... just to start.  Overkill ? is it not all high end overkill ?
By the way i do not know how many VA is the stock one ...
Thanks again,  gino 
 
May 29, 2015 at 2:10 AM Post #9 of 25
Have a search for Tamura transformers on eBay, those split bobbin come up regularly for pocket change.  Only issue is they do need PCB mounted.

 
Hi and thanks a lot for the very helpful advice.
Also the stock toroidal is for pcb mount.  But yes i will have to make a new PCB to support the x-former.
i have a very trivial approach.   I look at what the best units in class use.
In order to get hints, suggestions on the best parts available for a specific use.
Of course there is much more than what meets the eyes.
The schematic is never avaialble. And this is just understandable.
I can only say that toroidals are not good at filtering mains noise.  This i know for sure.   EI types are much better.
And i also measure a quite high impedance on the secondary of a toroidal transformer.
Unfortunately i do not have the instrument at hand.  My friend owns it.
Low impedance in the secondaries i think it is very good.
Thanks again for the valuable suggestion,  gino
 
May 29, 2015 at 2:25 AM Post #10 of 25
the safety ratings of the transformer absolutely matter for the system of the separate AC mains transformer and the DAC powered by its 14 VAC secondary
 
for Class II floating, reinforced/double insulated equipment there are other construction considerations too - everything, your wires, any pcb, vero/perf board, conductive mounting screws, hardware, case has to meet the pri-sec creepage and clearance requirements of the standards, all connection have to be (doubly) secured so any single point joint failure still won't allow a flexing wire or component to violate the creepage and clearance
 
May 29, 2015 at 2:48 AM Post #11 of 25
  the safety ratings of the transformer absolutely matter for the system of the separate AC mains transformer and the DAC powered by its 14 VAC secondary
for Class II floating, reinforced/double insulated equipment there are other construction considerations too - everything, your wires, any pcb, vero/perf board, conductive mounting screws, hardware, case has to meet the pri-sec creepage and clearance requirements of the standards, all connection have to be (doubly) secured so any single point joint failure still won't allow a flexing wire or component to violate the creepage and clearance

 
Hi do you maybe mean that the stock transformer is the best solution available ? that absolutely no benefits can be obtained by replacing it with a better specified one ?
Unfortunately i do not have the actual part here to check what type of transformer they provide with the unit, but i read enough about hum and noise coming from it to guess that for sure there must be something better.
Just think ... the all unit costs 100 USD ... what kind of transformer quality they can humanly provide for this low price ? a joke of transformer ... and it is perfectly fine because the price is nothing.
They are very good at Schiitt ... but i dont think they can walk on water. Or they ?
rolleyes.gif

Any unit can be improved ... any.
Regards,  gino
 
May 29, 2015 at 3:53 AM Post #13 of 25
to evaluate a electrical system like the mains transformer supplied with the DAC you would need to know the PS circuitry inside the DAC
 
as an example
 
while "better transformer" in isolation may seem to be higher VA rating, bigger core, lower winding resistance from more copper - it could also give larger inrush current, lunch the diodes if their DAC PS design, component sizing actually allows for, expects a certain higher winding R in the mains transformer they spec
 
 
and a big mains transformer design option can't really be had without getting them custom wound - lower operating B in the core - all off the shelf mains transformer designs push the core B to the margin to save on core and copper weight, size
 
May 29, 2015 at 4:21 AM Post #14 of 25
to evaluate a electrical system like the mains transformer supplied with the DAC you would need to know the PS circuitry inside the DAC
as an example
while "better transformer" in isolation may seem to be higher VA rating, bigger core, lower winding resistance from more copper - it could also give larger inrush current, lunch the diodes if their DAC PS design, component sizing actually allows for, expects a certain higher winding R in the mains transformer they spec
and a big mains transformer design option can't really be had without getting them custom wound - lower operating B in the core - all off the shelf mains transformer designs push the core B to the margin to save on core and copper weight, size

 
Hi and thanks again.  I completely agree with the scientific approach. But we have in the end to select a brand and a part.
 
I have here in my hand a very very cheap product.  It is a usb to spdif converter.
It sounds ok to me but i am not in the position to judge in a dependable way beacuse my present system is not resolving enough.
And it uses a toroidal transformer in the PS. 
For curiousity i looked at one of the best similar devices available in the market.  It uses a different type of mains transformers.
I tend immediately to give much more credit to the design choice of the better unit.  Immediately without any exitation.
I agree that the transformer is just a part and not the all unit but it is an important element.
Clearly for the more expensive unit any transformer around could have been selected because cost is less of a constraint.
For what i know the Talema in the cheap unit could even be a fake one ... it would be not the first time.
I am completely sure that an Hammond transformer with split bobbins would be a better part also for the cheap unit. Like some of the filters/chokes used in the better unit.
I would never think to mod an almost perfect and usually very very expensive unit (of course)
wink.gif

Like by instinct i think to mod a very very cheap unit already decent because i am sure that some mods on strategic components can have a great outcome.
It is fundamentally sane and with some mods can be very good.
And i think i am not alone in this thinking.  And for sure i cannot afford the better unit.
If this were not possible, to elevate the performance of a cheap unit, i would agree with leaving it as it is.
But i am instead sure that it is possible. So why not do it ?
maybe the problem is to do it properly ... of course.
Just to make an example i will try a external filter and listen for any change in sound. Is this a mod ? i would say yes.
I am quite sure that the same filter on the better unit will have very little effect because the unit has a so much better filtering already.
Usually cheap units have to skip on something for cost reason. A 10 USD mod will increase the total cost of the unit of 100 USD in the end of the process.
For me instead it will cost 10 USD.  Modding when done properly on a already promising unit is a very smart decision.  Just look at how many modders are around.
Sometimes the mod is just a part replacement.
I remember to have read a very interesting comment about a mod on a dvd player. The mod consisted in just bypassing some parts.
The sound was better !!!!   with less parts the sound was better ... unbelievable.
These non needed parts are like some dressing on a very good food ... they just ruin it.
Often the parts are not what makes the final cost. Is the labour that costs.
Something they skip on parts that instead are fundamental for performance.
I would always look at what is fantastic and try to copy the design solution.
I bought for instance some service manuals to try to understand something of the design.
Another problem is that i do not have necessary instruments to see for noise or similar.
This is another top class dac ... and it does not have toroidals ... and i see a mains filter at the input.
 
http://schiit.com/public/upload/general/imagecache_thumbnail_product_gallery_large/yggdrasil_pcb_1000.jpg
 
So my reasoning is that the same parts (with the right ratings of course) put in the cheaper unit could give some benefits
For the best units that provide the best sound possible the parts selected can be the best available. Cost is less of a constraint.
There is a lot to learn looking to the best units around. To the benchmarks. I think.
But if i spotted the same Talema (supposing that mine is not fake) in a similarly top unit i would be willing to think again about the mod.
Thanks again,   gino
 

Users who are viewing this thread

Back
Top