I see some people have changed the on-board NE5532 for a LME49720, how have you found this upgrade now you've have time to listen to it - worthwhile?
Has anyone replaced the NE5532 with a discrete solution? If so, how are you finding it?
What about the LME49860, has anyone tried using one of those as I read that it is better than the LME49720.
Has anyone replaced the NE5532 with a discrete solution? If so, how are you finding it?
What about the LME49860, has anyone tried using one of those as I read that it is better than the LME49720.
Last edited:
Interesting. The specs suggest it is schmitt triggered based on input switching thresholds and output transition times, but they don't say that it is in plain language. Probably an excellent choice to speed up slow clock transitions.
Would the way to get a very clean rail for such a buffer be to give it it's own, isolated power supply?
I ask because I'm about to start putting my power supply for the main dac board together and while I'm aware that multiple, isolated power supplies to different parts of the system is a superior design path to take, I'm still not quite clear on which parts of the system should be given their own power supplies.
Am I right in thinking that the most basic approach to multiple PSUs is to power the dac board and analogue stage separately, each from it's own isolated PSU?
I ask because I'm about to start putting my power supply for the main dac board together and while I'm aware that multiple, isolated power supplies to different parts of the system is a superior design path to take, I'm still not quite clear on which parts of the system should be given their own power supplies.
Am I right in thinking that the most basic approach to multiple PSUs is to power the dac board and analogue stage separately, each from it's own isolated PSU?
About the su5 first impressions, on a not so revealing system, it is an upgrade from a lower spec and price dac. At first .. not excited but not disappointed either, so not an exceptional for the price dac, but not bad either. Previously Markw4 compared it with Benchmark reference DAC , and found this to have some kind of "distortion" (joke?)...
I would say you need to warm the unit a bit, more than 10 hours, so do not blame it for the very beginning. Some good parts inside, some maybe good quality capacitors, unit as previously described with 2x op275 opamps at RCA/headphone output and 2x ti TL072c at the so called I/V.
I cannot express a clear opinion neither this is an objective review, mostly because I use it with an external xmos USB to SPDIF with better TXCO crystals: http://essabre-90xx-rpi.sfb2.com/2018/01/skysong-xu208-first-expressionsexperiences-win10/
+
A tube buffer with 2x Voskhod 70s USSR valves
+ extensive use of VSTs specially selected for my needs/room corrections (console emulations, EQ, Center/LR stereo imaging etc) as a result of music production experience, not electronics.
tested a bit up-sampling to see if there is real benefit at my case, cannot say anything special right now or hear an obvious difference (only 44.1 to 88.2 as 176 is a bit unstable at SPDIF)
So,
1. I was wondering if someone can have a view or idea if changing the RCA out OPAMP with an adapter like this: 10pcs DIP8 to SOP8 Adapter SOIC8 socket pcb 1.27mm PCB converter 300mil | eBay
could get any better/different sound out of this...
2. Is there a real benefit by creating an i2s link between the 2 devices instead spdif? What kind of plugs/cable will I need? As only USB 44.1/16 audio will be used, so I guess spdif is ok. But then I could avoid spdif chip and get some higher and stable upsampling for experiments. I do not know if it is a worthy move. Foobar2000->real time DSD translation has drop outs at spdif (DSD64 only) so no real test can be done for DSD .
Finally I would like to note that the internal xmos USB board works only with the class2 3.20 driver and not with the 4.36 that happens to work with the skysong board.
Singxer_USBAudio_v4.36.0_2017-11-14_DSD512
I would say you need to warm the unit a bit, more than 10 hours, so do not blame it for the very beginning. Some good parts inside, some maybe good quality capacitors, unit as previously described with 2x op275 opamps at RCA/headphone output and 2x ti TL072c at the so called I/V.
I cannot express a clear opinion neither this is an objective review, mostly because I use it with an external xmos USB to SPDIF with better TXCO crystals: http://essabre-90xx-rpi.sfb2.com/2018/01/skysong-xu208-first-expressionsexperiences-win10/
+
A tube buffer with 2x Voskhod 70s USSR valves
+ extensive use of VSTs specially selected for my needs/room corrections (console emulations, EQ, Center/LR stereo imaging etc) as a result of music production experience, not electronics.
tested a bit up-sampling to see if there is real benefit at my case, cannot say anything special right now or hear an obvious difference (only 44.1 to 88.2 as 176 is a bit unstable at SPDIF)
So,
1. I was wondering if someone can have a view or idea if changing the RCA out OPAMP with an adapter like this: 10pcs DIP8 to SOP8 Adapter SOIC8 socket pcb 1.27mm PCB converter 300mil | eBay
could get any better/different sound out of this...
2. Is there a real benefit by creating an i2s link between the 2 devices instead spdif? What kind of plugs/cable will I need? As only USB 44.1/16 audio will be used, so I guess spdif is ok. But then I could avoid spdif chip and get some higher and stable upsampling for experiments. I do not know if it is a worthy move. Foobar2000->real time DSD translation has drop outs at spdif (DSD64 only) so no real test can be done for DSD .
Finally I would like to note that the internal xmos USB board works only with the class2 3.20 driver and not with the 4.36 that happens to work with the skysong board.
Singxer_USBAudio_v4.36.0_2017-11-14_DSD512
Last edited:
Basically I do not have Benchmark DAC, but at some stages of the modifications I also heard this distorsion. (Possibly it still exists, but at the end on a minimum level)
you can check my post (1577 & 1306) where I summarize the mods.
I think the I/V Stage + applying everywhere LME47920 OPAs decreased the distorsion.
Sz.
For me as well the NE5532 replacement was a big step.
First I changed to OPA2134, and after this to LME49720.
(both changes were audiable)
As I know Janos changed to burson audio. (but not 100% sure.)
Cheers babolcs, it was your original post about changing op amps that made me ask the question, so it's good to hear months later that you found it was an audible improvement. I have a MUSES8920D I can try, but it sounds to me from your experience and some other comments that the LME49720 is a better choice.
Is this LME49720NA the right one to get? I ask because of the NA suffix, I have no idea what it indicates....
NATIONAL SEMICONDUCTOR - LME49720NA/NOPB - OP AMP, AUDIO, 17VIN, 55MHZ, 8DIP | eBay
Is this LME49720NA the right one to get? I ask because of the NA suffix, I have no idea what it indicates....
NATIONAL SEMICONDUCTOR - LME49720NA/NOPB - OP AMP, AUDIO, 17VIN, 55MHZ, 8DIP | eBay
Last edited:
No joke. It is distorted even if you don't recognize what makes it sound the way that it does.
Mostly what happens is the owner gets used to the new sound, not necessarily so much that the unit warms up (at least for not this type of circuitry).
The opamps are not so good compared to what should be there.
All that stuff has to be changing the sound in certain ways that may obscure some details of what makes the DAC sound like it does. It is distorted though, although maybe you don't find it all that bad. Wouldn't work for me, but then again it doesn't have to.
Tested by what means?
Don't know how you would solder such a thing in there, not that those adapters are good to use. You might have to remove some other components to remove the existing opamps and solder in adapters. Problem with adapters like that is they have a lot of inductance with the long pins and traces, and a lot of capacitance between the leads. Could easily adversely affect performance, maybe cause opamps to oscillate, etc. Film caps in the board are not good for bypassing faster opamps anyway, and there is no ground plane to return the bypass to, only a thin strip of ground trace shared by power and signal returns. What a mess.
Did you say you already tried your own USB to I2S with this DAC? I was going to ask you how you did it. There would be no place to hook it up except to remove the internal XMOS USB to I2S board and plug it there.
Has anyone else had difficulty soldering components on the blue version of the ES9038q2m board?
I ask because I just tried to replace the caps on mine with Rubicons and on the first cap, I desoldered and removed the cap and when I went to clean up the solder with a wick so the leads of the new cap could be pushed through the hole, the tiny solder pad for the positive pin lifted right off the board. This meant I was able to solder the negative lead of the new cap but there is nothing to solder the positive lead to and this board is now dead.
I have to say, the quality of this PCB is rubbish, I've done a hell of a lot of soldering in my life and never have I seen a solder pad just come off a board like this and almost never have I seen such tiny solder pads for THT components.
So it looks like this board I have is going to have to go in the bin and I'll have to buy one that isn't such crappy quality.
I ask because I just tried to replace the caps on mine with Rubicons and on the first cap, I desoldered and removed the cap and when I went to clean up the solder with a wick so the leads of the new cap could be pushed through the hole, the tiny solder pad for the positive pin lifted right off the board. This meant I was able to solder the negative lead of the new cap but there is nothing to solder the positive lead to and this board is now dead.
I have to say, the quality of this PCB is rubbish, I've done a hell of a lot of soldering in my life and never have I seen a solder pad just come off a board like this and almost never have I seen such tiny solder pads for THT components.
So it looks like this board I have is going to have to go in the bin and I'll have to buy one that isn't such crappy quality.
350 degrees I think, it's an older unit with no digital readout, just a knob.
I'll take a picture in the morning when there's daylight, it's just past midnight here.
I would think it would be possible to use some thin wire to connect to the trace on the board, the question is, can I do it, I'm not entirely convinced this screw up wasn't my fault, to err is human...
I'll take a picture in the morning when there's daylight, it's just past midnight here.
I would think it would be possible to use some thin wire to connect to the trace on the board, the question is, can I do it, I'm not entirely convinced this screw up wasn't my fault, to err is human...
Last edited:
Could you provide a short flac file with a characteristic part that can reveal this? or this is so obvious? This way I could try to compare with other dac also.
What would you propose?
Foobar has some DSP plugins that do up sampling, 2-3 different algorithms, with different options. So using internal xmos you can up-sample up to 384khz 32bit, and hear/test what it can give you. With spdif I should stay 88.1/96 because of possible drop outs (protocol/cable limitations).
An other option is to translate any PCM source signal to DSD (64 - 512) on the fly with other add-on plugins for DSD/SACD and a kind of ASIO to DSD wrapper driver. []Page not found
- SourceForge.net Maybe this way the ess chip avoids some processing, I do not have a clue.
It is impossible through DSD64 and SPDIF here, cable cannot retain sync. You will need i2s I guess.
For off line conversions:
TASCAM Hi-Res Editor
Product: TASCAM Hi-Res Editor | TASCAM
Album player is an other good player with bit perfect capabilities and upsampling (see some posts above), and some others also I have never used
https://vv.uka.ru/aplayer_eng.html
but I always keep in mind posts like this: 24/192 Music Downloads are Very Silly Indeed
Thanks, I will keep that in mind
Not really i2s but I am using spdif only, this external xmos board has spdif/optical output. But as you can see (link) there are 20-pins ready to use, some of them are for i2s. Of course you need to remove existing xmos board and use an external cable maybe out of the USB port, my external board has its own box.
Last edited:
Unfortunately, no. Sorry. You would need to find someone or maybe a store with a really good, more or less state of the art DAC. That's about what my modded DAC sounds like. You could play a CD without distortion insofar as humanly possible. I think you must have no idea what they can sound like, otherwise it would be hard to understand you saying some of the things you do about your present playback system. Things like using room correction plugins and tube preamps. You should be able to clearly hear the distortion each of those adds. If you can't, you can't. No flac file could reveal that. If you were around were I am I would invite you over and show you, but looks like we are too far apart for that.
If you have a good sound card you could record a clean signal and a distorted one to 2 different files using exactly the same recording parameters. Then we could listen and compare them.
you can check this software method, never used by me though:
Audio DiffMaker
Audio DiffMaker example files
http://www.libinst.com/AES Audio Differencing Paper.pdf
Diffmaker - digital audio measurement tool | Headphone Reviews and Discussion - Head-Fi.org
Audio DiffMaker freeware for checking for equipment effects
there should be more tools and methods out there, but this one is free
so, there might be a chance that we can find out if the distortion you refer to, is audible to someone else's system, if it is critical or subtle.
Last edited:
Maybe I am not being clear. The clean recording is a CD. The distorted recording is the CD being played out of your DAC. How are you going to tell the difference between the two since they both get played through your DAC?