|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
82.156.128.160
In Reply to: RE: I have used both the LYNX and JULI@ cards intensively as digital interface posted by theob on October 20, 2011 at 16:15:03
Hi Theo,
Thank you for your response.
Just to check if I understand your post correctly.
You first powered the juli@ digital part with 3.3 Volt through battery’s, but the 5 Volt was still coming from the pci socket (power situation 1).
Later on you cut the 5 Volt PCI fingers and also powered the 5 Volt from battery’s. So both 3.3 V and 5 V are provided through battery’s (power situation 2)
Now about the sound quality.
Are you saying that changing from power situation 1 too power situation 2 again gave a (big?) sound quality improvement?
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Follow Ups:
Yes a big sq pop again.
Hi Theo.
Looks like I have too build myself a second DIY voltage regulator.
This time a 5 Volt one.
Building a 3,3 Volt Voltage regulator was my first electronic DIY project ever.
And… it turned out too be surprisingly easy.
I deliberatly build a prototype with only the LT regulator and 2 caps on the PCB.
Like this:
Van on filtering, de-coupling, demping, ect
This construction with P4 connectors allows for easy swithing in and out all kinds of filter types, before and after the DIY regulator.
Van on filtering, de-coupling, demping, ect
And guess what?
In my setup there is no sound quality difference between feeding the 3.3 V DIY regulator with 5 Volt DC coming from a Pico or 5 Volt DC coming from a linear PSU.
Provided (!) ….. that you filter the input too the regulator board.
Ripple rejection of the LT regulator is arround 75 dB at 1000 Hz but only 15 dB at 300.000 Hz.
(the Pico swithing frequency = 300.000 Hz)
But if you filter this 300.000 Hz swithing frequency before the LT regulator,
than there is no sound quality difference between 5 Volt used from the Pico or 5 Volt from a linear PSU.
I (mis-)use a simple Schurter 5500.200 netfilter for this.
Specs see data-sheet: http://datasheet.octopart.com/5500.2014-Schurter-datasheet-2547.pdf
The Schurter netfilter has 2 coupled ferrite based 40 mH inductors inside and filters from 100.000 Hz too 10.000.000 Hz with 50 dB.
When you ad extra (Oscon) caps on the outside, the 2nd order effect (filterslope) can be bettered/altered/tailord
I haven’t used any batterys and thus I don’t know this for sure, but from this experiment (no diffrence between a linear PSU and a switching PSU when the input is proparly filtered) I think there is no need too use a battery as a source for digital components. The noise level only needs to be on a low level. It doesn’t necessarily need to be approaching almost zero.
Since you already constructed DIY regulators before, give it a try and feed your DIY regulator with filtered input from a pico PSU or a linear PSU and compair that too a battery source.
Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hi Mark!
Here is the standard schematic of the stabilized power supply unit.
The explanatory notes to schematics:
1. TV1 – the power transformer
2. VD – the Schottky rectifiers
3. C1 and C2 – the filtering capacitors, C3 and C4 – output capacitors
4. С2 and С4 - the qualitative film capacitors with small capacity (for example Wima) shunting electrolytic capacitors for filtering HF noise (the film capacitor capacity is around 1/100 part of the electrolytic capacitor capacity). С2 and С4 are possible not to install in case of use of electrolytic capacitors Black Gate, Elna Cerafine, Elna Silmic, Panasonic FC and similar.
5. R1 – the load resistor providing the minimum load current (5-10 mA) in the psu’s output.
6. U1 – the regulator.
In your case:
the filtering capacitor capacity (C1) should be not less than 4700 µF and C3 should be not less than 470 µF approximately. Oscon does not approach here.
R1=500-1000 Ohm
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
Is R1 mandatory in the design?? I have made my Linear for the Juli@ long back following this schematic, but without R1. Also the Filter caps I have used are 20000uf Elna Silmic II & output caps are 1000uf Elna Silmic II. The regulator I have used is a Dexa UWB series 3.3v in the linear psu itself & de-soldered the onboard reg on the juli@, & powered it through it's pin 10 & pin 8 straight with this 3.3v linear psu I made...
Junaid
= Is R1 mandatory in the design?? =
No. R1 is necessary for maintenance of stabilisation state of ps when loading is disconnected. For example if you want to measure correctly an output voltage of ps without loading connection.
= Also the Filter caps I have used are 20000uf Elna Silmic II =
You are cool :).
Serge.
http://cmp2-mihaylov.narod.ru/
My cMP2: Windows XP SP2, Gigabyte GA-H55M-UD2H, Intel Core i3-530, Corsair CM3X160C9DHX 1GB, system drive - Transcend IDE FLASH MODULE TS2GDOM40V-S, ESI Juli@, full linear PSU, NAS - WD My Book Live, iPad
:)
Junaid
Hi Serge,Thank you for the schematics.
But it is not at all my intention too build or construct a separate high quality DIY linear PSU for several reasons.1.
I would need too buy a whole lot of extra other materials like: an extra transformer, rectifier diodes, housing/casing, fuse/holder, on/of swith, cables/wires, sockets/plugs, ect etc. I also would have too spend a considerable amount of time too for assembling and constructing a DIY linear PSU.2.
I don’t think it is always necesarry too strive for lowest possible power supply noise.
I can fully understand the need for the lowest possible power supply noise in analog audio.
But this is NOT ALWAYS necessary for digital circuits.
For example see this article.
http://www.analog.com/static/imported-files/rarely_asked_questions/RAQ_tamingADC.pdf
The article clearly demonstrates that lowest possible power supply noise is a ‘very nice thing too have’. But striving for lowest possible noise levels, is not always necessary.
It can also be a waist of time, materials, money, ect.
The article outlines a situation where striving for lowest possible noise levels is not necessary at all. The article explains that in given situation, power supply noise levels only have too stay below a certain level. Only TOO MUCH noise wil degrade performance. Striving for lowest possible power supply noise levels is not necesarry at all (given the situation in the article)Now back too the Juli@ digital part.
The logic on the Juli@ card will produce noise itself. And the Juli@ card will also receive noise through the PCI-bus from the MoBo.
Considering these noise levels: is it really necessary too construct an external DIY linear PSU with lowest possible power supply noise? Or is it enough/sufficient too construct an external power supply for the Juli@ digital part, that keeps power supply noise levels below a certain level?However not exactly knowing these noise levels, it is a safe approach too construct an external DIY linear PSU which has lowest possible noise levels. One never can go wrong with this approach.
But, ….. if one doesn’t want too build a DIY PSU or one is not capable of building one with lowest possible noise level, than may be also good results can be achieved when using a LT regulator as a post regulator after a switching PSU or after an off the shelve linear bench PSU.
The filtering effect of the LT regulator, together with pre- and post filtering before and after the regulator, might reduce power supply noise levels sufficiently.I guess for many inmates building there own high quality, low noise linear PSU, will be a ‘bridge too far’. However they may be capable of construction a simple (post) voltage regulator with pre and post filtering. Constructing such a (post) regulator, with pre and post filtering is rather easy.
(Also there is no dangerous 230 AC high voltage is involved.)Although when using these simple DIY (post) regulators, power supply noise levels are not lowest possible, the effect on sound quality is nevertheless quite remarkable (!!).
Hoping you are not the only one who is reading my posts, I post my flimsy constructed filters and voltages regulators too inspire other inmates, who cannot build a high quality linear DIY PSU, but who might be able too construct such a simple (post) voltage regulator.
It is defenately worth the effort and costs for materials.Mark
fully optimized cMP2 PC -> ESI Juli@ -> Van den Hul Optocoupler MkII-> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 10/21/11
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: