C135 and C142 are 1500 pf NP0 ceramics, together with the 640 ohms impedance from the R-2R network they set the -3 dB 165 Khz point of the RC filter.
Hi, so you replace (or by pass?) the on-board regulator with the Sparko? What power supply do you use to feed the sparko? Thanks.
I would suggest if possible you may create a standalone AES67 interface card with I2S output, preferbly with the same form factor as Amanero (though card can be wider) which make them interchangable. It will be the coolest thing since the invention of UAC2 XMOS and Amanero.
Last edited:
I'm sorry, but you severely degrade the dac1541 by:
Replace switchers with other switchers probably feed from a switcher ?
Bypass the ultra low impedance and ultra low noise discrete regulators, instead the output amplifiers are connected directly to the switchers ?
Add high impedance (relative) capacitors to low impedance discrete regulators ? Directly to the transistor collectors, which might affect stability ?
Disconnect chassis from ground ? Ok, it's probably connected at the bottom of the PCB.... But why then the plastic screws ?
Add extra high impedance (relative) capacitors to the low impedance vref buffer ? Ok, at least that don't hurt....
Remove the output mosfet relays, instead getting clicks and thumps ?
Hi JJansen,
I agree with you, the 1541 should be a great DAC and it needs even more attention and modification to the board before issuing to the market.
We did some listening (subjectively, btw) by comparing to the 1541 and my friend's DIY on dam1021, and the dam1021 perform better all the way. I wish Soren could provide some tips on improving the dac1541 with some modification by DIYers!
Besides, what have you done to the 1541? I am about to do some modification to the 1541 as well, could you provide some direction for me? Thanks alot!
Last edited:
If there is a weak point on the dac1541, it's probably the filters. I do have plans to increase the number of filter taps and create new filters, that will eventually be implemented on all of my dams and dacs as it will fit inside the existing fpga....
You can load other filters to suit your taste, the dac1xxx series use same filters as the dam1021, that will probably have the largest effect.
Some people for some reason also seems to want old fashioned linear power supplies, there is actually a connector on the PCB you can use for that, the J7, connecting to an external linear supply.... You don't even need to remove the onboard switchers, just don't power them at the same time....
The dac1541 is already VERY well designed, trying things that worked on the dam1021 will most likely have no effect, or even degrade the dac1541, although others might have other beliefs....
When you say "Soeren should take a little more time before bringing its products on the market." I kinda get a little insulted....
But I, and probably many others, are interested to hear about the details....
Thanks Soren for your reply! And hey, it is Saturday, time for relax man!
I do believe the dac1541 is a good product, and what degrades the whole design was the choice of installing the onboard switchers with 200mV ripple and noise (max)!
I honestly believe if the board would perform way better with LDO Voltage Regulator such as Tps7a47 series which I am about to start the DIY process.
Besides, just being curious, would you be kind enough to provide more details on the dac1541 board such as those junctions and the case openings size and locations? Thanks!
You also need to remember that there are different requirements for diy products and commercial ones, to ex. commercial dacs needs the output buffers and mosfet relays, and there are all kind of usability, emi/emc and safety requirements.... And even when doing high end equipment there still are cost concerns, it's easy to do a "no cost concern" dac, but you will not sell very many....
Just in case you don't know, over at SBAF they have tested and measured the dac1541, one of the least noisy ever....
The main reason for using switchers is the support for direct world wide powering, and to have a small and light unit, shipping is expensive....
There was a time I transported my main computer forth and back between US and Denmark, once I forget to change it between 115V and 230V, and puff, a dead power supply.... Then I decided that everything I designed would not have a switch to choose input voltage, or even worse, have two different versions just for different input voltage....
I haven't decide what info to release for diy, but simple question will most likely get answers....
There is one issue that might need more attention: there is quite a lot of noise at the RCA output.
red: PicoScope reading from DAC1541 RCA out left
blue: PicoScope reading from R&S UPD monitor of input from DAC1541 XLR out left
note: noise amplitude changes from time to time
Had the impression to have a fault in my measurement...
May be, JJansen has addressed this (and possibily other stuff).
Attachments
Last edited:
Agree on your stand point where business perspective for world wide powering arrangement!
So some simple linear regulators connecting to J7 for 12v-0, 12v-0 and 5v-0 would perform the magic.
I am more curious on the details for the following:-
- Junction J2;
- Junction J3;
- Junction S2;
- As I am trying to build a new case together with the linear regulators stuff and the dac board, I may wish to obtain, if possible, the dimensions for front and rear panels so that I could apply them to the new case.
Thanks Soren.
Here is what I get on the dac1541 RCA line outs with my Tek scope, input active with nothing playing. The level changes with what 230V outlet I'm using, the level goes up when output is set to "headphones", in that case the line outs are disconnected from the output amplifier and instead connected to gnd with 1K5 resistors.
Note that the scope is set for 1mV/div, the repeat rate is 100 Hz. I believe it's in the air or ground and is what you can expect with switchers around on a cluttered desk, note how ungly the power input is, I'm located at the end of a small road....
There is absolute nothing to hear in my speakers when volume on the stereo amplifier is turned all the way up.
Yellow is power input, Pink is right output, Green is left output,
Note that the scope is set for 1mV/div, the repeat rate is 100 Hz. I believe it's in the air or ground and is what you can expect with switchers around on a cluttered desk, note how ungly the power input is, I'm located at the end of a small road....
There is absolute nothing to hear in my speakers when volume on the stereo amplifier is turned all the way up.
Yellow is power input, Pink is right output, Green is left output,
Attachments
My question is: a "naked" transformer followed by a diode bridge would have, of course, much worse ripple, but then you filter and regulate. So with a MeanWell you actually start with a better output. The only problem would be the noise spewed before the internal PSU back in the power chain.
So I am curious (and this is for @soekris ) about filtering at power input. I know that the Meanwells have an ingress EMI filter - is this sufficient to not pollute other audio devices?
Roberto
I am currently looking for the same Push Button (Tactile?) Switch as per the attached photos below in RS but I am out of luck.
Would some expert be kind enough to advise the same push button (Tactile?) switch? Thanks!
Would some expert be kind enough to advise the same push button (Tactile?) switch? Thanks!
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Last edited: