Hi Eldam
In this supply I used 70R 1/2W PRP resistor. Looking back at it, I realize it is only CRC. I may go back and try it with CRCRC or even CLCRC with a filament choke. I did this on the +15v supply to the 1541a with good results. Thanks for causing me to look closer.
I pretty much used what I had on hand. A good engineer would have done all the math and simulations so I will not be offended if others chime in to suggest an alternate approach. I kept in mind the advice from ECDESIGNS to keep transformer small. I used a small 6VA Hammond 229 here wired for 12v output. I had to balance keep resistor small to still get enough current to drive the Salas shunt.
In this supply I used 70R 1/2W PRP resistor. Looking back at it, I realize it is only CRC. I may go back and try it with CRCRC or even CLCRC with a filament choke. I did this on the +15v supply to the 1541a with good results. Thanks for causing me to look closer.
I pretty much used what I had on hand. A good engineer would have done all the math and simulations so I will not be offended if others chime in to suggest an alternate approach. I kept in mind the advice from ECDESIGNS to keep transformer small. I used a small 6VA Hammond 229 here wired for 12v output. I had to balance keep resistor small to still get enough current to drive the Salas shunt.
Can't speak for Walter but I usually use 0R5 or 1R0. WWs are ideal.
That said, I use John Swenson's C-L-C design to power my WaveIO c/w a pair of Jon Risch-style isolation trannies. The difference is audible (which I suppose it should be as the PS cost more than the board . . .
).
Ryelands
Thanks for this add. I think I will keep playing around with this. On the WaveIO 5V PS, I too went CLC repurposing a Peter Daniel supply. It did have a very noticible improvement over a simpler linear supply that used a standard bridge and CRC.
This one just supplies the 5V to the WaveIO isolator. I think it is time to try a choke. Had really good success with choke on the +15V supply inspired by the DDDAC thread.
Thanks for this add. I think I will keep playing around with this. On the WaveIO 5V PS, I too went CLC repurposing a Peter Daniel supply. It did have a very noticible improvement over a simpler linear supply that used a standard bridge and CRC.
This one just supplies the 5V to the WaveIO isolator. I think it is time to try a choke. Had really good success with choke on the +15V supply inspired by the DDDAC thread.
Attachments
Lucien informed me the connector is of the type:
10P IDC 2.54mm; and I saw this generally is used for flat cable.
see for instance this FleeBay cable
Separate solderable female connectors are also available. Then you can connect your own coax to it.
See 10Pin 2.54mm Double Row Female Straight Header
I expect a flat cable to work: computers and CD's use them too
Anyway it might be convenient.
I checked the connector to work with one from a Philips CD player.
And these U.FL cables are very handsome too, and affordable.
Grab them while they last :roll eyes:
[ The vendor posted this as to my request ]
Grab them while they last :roll eyes:
[ The vendor posted this as to my request ]
@triode_al - I've tried a few u.fl leads and as far as I can make out the thicker (and shorter) the better the sound. I'm currently using 178 diameter like the ones that you can see here:
IPX Female to IPX U FL Female Crimp RG178 RF Cable Pigtail 50cm 20" Gold Plated | eBay
Please don't take this link as any sort of recommendation for the vendor, I'm just did a quick hunt to illustrate my point.
IPX Female to IPX U FL Female Crimp RG178 RF Cable Pigtail 50cm 20" Gold Plated | eBay
Please don't take this link as any sort of recommendation for the vendor, I'm just did a quick hunt to illustrate my point.
The interesting part is they are all 50 ohms; but the damping factor differs; on 5cm the effect should be negligible.
More important is to have a correct damping at the receiving side (such as with a 5-10 ohm resistance in series).
Some TDA1541 schema's I have seen have the damping approach (series), and I remember ECDesigns did that too.
But probably even better is a true termination of the 50 ohm HF line (50 ohm SMD-1nF??). There are no boards I know that have that.
Then of course I myself did it once, but I can't measure anything like that and it always sounds good (the Newness Effect).
More important is to have a correct damping at the receiving side (such as with a 5-10 ohm resistance in series).
Some TDA1541 schema's I have seen have the damping approach (series), and I remember ECDesigns did that too.
But probably even better is a true termination of the 50 ohm HF line (50 ohm SMD-1nF??). There are no boards I know that have that.
Then of course I myself did it once, but I can't measure anything like that and it always sounds good (the Newness Effect).
After reading again the impressions saying that non-isolated output sounds better i can't help but want to try it. Although i already tried once and burned I2S multiplexers on my Buffalo dac. Maybe i should ensure that capacitors are fully discharged and measure voltage on signal wires before connecting to be sure..
Anyway - if i connect non-isolated outputs to DAC, do i have to connect ground of those mini-coaxials to dac signal ground? Or it mostly serve as shielding there?
Anyway - if i connect non-isolated outputs to DAC, do i have to connect ground of those mini-coaxials to dac signal ground? Or it mostly serve as shielding there?
Last edited:
Hi Wlowes, I would look at a bit more capacitance. This was designed for 50mA or so draw. your load peaks at nearly 10 times that.
Thanks Luke, right you are.
I did rebuild it with different caps & resistors to accommodate the 10x current. I am still tinkering with all of the supplies to the WaveIO and 1541a and having good results with clclcl.. filters employing 12mh high current inductors and large # of Panasonic FC 1800uf caps. I'll post findings when done.
I did rebuild it with different caps & resistors to accommodate the 10x current. I am still tinkering with all of the supplies to the WaveIO and 1541a and having good results with clclcl.. filters employing 12mh high current inductors and large # of Panasonic FC 1800uf caps. I'll post findings when done.
thanks for the idea, I'm finally getting a bom together for my dac, its been a while now. I will order some common mode chokes for mine as well.
Did you not use regulators on your tda1541? I remember someone on the forums saying they preferred clc and no reg.
Will do!
I put some text describing the latest driver version available to use, thank you for the hint!
Speaking of drivers, I got 3.00 pack and I'll customize it soon enough. I'll get back to you when I have a stable/tested versions so don't flood me with requests
. From what I know, with 3.00 drivers, all support for older Windows up to Vista (included) was deprecated.From 3.00, only Windows 7 and up will get "attention".
Regards,
L
I use salas shunts on 1541a and on waveio isolator. Did not want to risk the 1541a chip.
actually using fixed radial lead inductors inspired by ecdesigns PS. Found them at surplus but similar to digi part 445-3823-ND. well worth the extra space and effort.
To update, my PSU problem seems to be in the -15v circuit only; the +/-5 lines are stable. New trimpots have arrived that will (hopefully) set the correct OP values. One thing at a time.
Meanwile, just to note that upgrading resistors and adding by-pass caps in the Lampizator I/V stage as recommended in your blog is a top notch suggestion. My eyesight is no longer up to making Managnin resistors but Charcrofts (aka Vishay VAR-Series "naked" Z201 Z-foils) proved easy to fit and very well worth the effort. Thanks!