At the moment I am listen mostly to BBC Radio 3 high definition sound, 320 kps mp3, on an older Toshiba laptop.
Post 55 interested me with regards to the noise caused by the laptop charger.
I have my earbuds plugged directly into the HiFimeDIY ES9023+TE7022. Not expecting ground loop problems, as my earbuds are not connected to ground, I still decided to try running the computer off the battery by unplugging the charger..
When the charger is unplugged, and the computer runs off the battery, there is a slight, but noticeable reduction of a slight harshness in the music and the voice. The battery is quite old however and I listen for quite long periods. Running the computer without being plugged into the charger does not seem a good idea.
I know that the earbuds have a sharp rise in the high frequencies that probably exaggerate the harshness.
I am not able at the moment to compare the harshness with my desktop, as I am not at home.
I love the portability of the HiFimeDIY DAC. I am not interested in bringing along a larger box and extra power supplies for portable use. The USB isolator is a good suggestion for home use.
Does anyone think that cutting the power cord between the charger and the laptop and running the positive and negative through the windings of a small 1:1 iron core transformer to cancel the noise might help? I do not have the tools or parts to try it at the moment. I will not be home for another 2 weeks.
Post 55 interested me with regards to the noise caused by the laptop charger.
I have my earbuds plugged directly into the HiFimeDIY ES9023+TE7022. Not expecting ground loop problems, as my earbuds are not connected to ground, I still decided to try running the computer off the battery by unplugging the charger..
When the charger is unplugged, and the computer runs off the battery, there is a slight, but noticeable reduction of a slight harshness in the music and the voice. The battery is quite old however and I listen for quite long periods. Running the computer without being plugged into the charger does not seem a good idea.
I know that the earbuds have a sharp rise in the high frequencies that probably exaggerate the harshness.
I am not able at the moment to compare the harshness with my desktop, as I am not at home.
I love the portability of the HiFimeDIY DAC. I am not interested in bringing along a larger box and extra power supplies for portable use. The USB isolator is a good suggestion for home use.
Does anyone think that cutting the power cord between the charger and the laptop and running the positive and negative through the windings of a small 1:1 iron core transformer to cancel the noise might help? I do not have the tools or parts to try it at the moment. I will not be home for another 2 weeks.
Hello from Portugal, everyone.
First of all, I am not audiophile, by any means, so take my feedback with a grain of salt
Sometime ago, I got a cheap PCM2704 DAC/board that did seem to improve on my (high-end/ recent) onboard sound somewhat.
Obviously though, it was not great either...
Having found the hifimediy sabre DAC with good reviews and a great price (maybe it's not a audioquest dragonfly, but its a hell of a lot cheaper), I decided to order it.
Compared to the PCM, some aspects are very obvious: more dynamics, instrument separation, and detail.
soundstage also seems to improve somewhat.
Still, there was a slight "harshness" to it.
One of the causes, is a noise USB line (on the PCM board without any sound output, the noise is very obvious while connected to the Amp)
Also, I have found that some burn-in time did actually improve some of the harshness, and the sound also seems somewhat clearer as well.
All in all, I am *very* pleased with this DAC.
I am also considering getting an usb isolator (olimex?) and connect a proper power supply to it.
First of all, I am not audiophile, by any means, so take my feedback with a grain of salt
Sometime ago, I got a cheap PCM2704 DAC/board that did seem to improve on my (high-end/ recent) onboard sound somewhat.
Obviously though, it was not great either...
Having found the hifimediy sabre DAC with good reviews and a great price (maybe it's not a audioquest dragonfly, but its a hell of a lot cheaper), I decided to order it.
Compared to the PCM, some aspects are very obvious: more dynamics, instrument separation, and detail.
soundstage also seems to improve somewhat.
Still, there was a slight "harshness" to it.
One of the causes, is a noise USB line (on the PCM board without any sound output, the noise is very obvious while connected to the Amp)
Also, I have found that some burn-in time did actually improve some of the harshness, and the sound also seems somewhat clearer as well.
All in all, I am *very* pleased with this DAC.
I am also considering getting an usb isolator (olimex?) and connect a proper power supply to it.
The harshness I noticed with BBC Radio 3 a few days ago is gone.
It did not show up when I tried some other 320 kps radio stations.
I could not detect it while playing a flac file.
I am guessing that there was a defect in the BBC signal that has been corrected.
Unplugging the power supply now makes no noticeable difference.
I continue to be impressed with the sound quality of such an inexpensive DAC and how well it directly drives phones of both high and low impedance without an extra headphone amplifier.
It would be really informative if someone could make distortion measurements while the HiFimedDAC it is connected to 30 and 600 ohm headphones, different real loads in place of the ideal resistor.
It did not show up when I tried some other 320 kps radio stations.
I could not detect it while playing a flac file.
I am guessing that there was a defect in the BBC signal that has been corrected.
Unplugging the power supply now makes no noticeable difference.
I continue to be impressed with the sound quality of such an inexpensive DAC and how well it directly drives phones of both high and low impedance without an extra headphone amplifier.
It would be really informative if someone could make distortion measurements while the HiFimedDAC it is connected to 30 and 600 ohm headphones, different real loads in place of the ideal resistor.
Check this isolator
Teralink ADuM4160 USB Isolator board
I think that use they share the same circuit...but it cost little less and apparently is more "audio grade" (MJE15030 , BC548 NPN low noise linear voltage circuits)
Last edited:
Thanks for the tip, that looks very interesting, and inexpensive as well!
I will certainly have a look.
Without knowing the circuit, though, I'm a bit weary of the vreg... having a known LDO part would be easier to look at.
Kit with SMDs done should be easy enough to solder too.
This seems even more interesting in the power filtering options it provides.
If i read it correctly, it should be supplied with either AC or 9V/12VDC power, and it can feed upto 1A trough the USB Line?
In any case, with postage and tax it ends costing about twice as much as the DAC
Also, I was ondering if using a long (5m+), active (bus powered) USB cable would be a terrible idea. If not, the isolator would go at the end of the cable for best results, I guess?
It looks very nice but I'm agree with you.
At similar price is possible to get to the new board made by hifimediy 9023 + coax and an amanero board...
Except that's apples to oranges.
The amanero board will have the same problem with USB/PC noise, except you can't use a USB isolator like this unless you're willing to live with full speed, which kind of defeats the purpose of it.
I just bought an Olimex usb iso board and have just started to play with it. My initial impression is good, but it is diminishing returns, it costs me almost as much as this dac ($40), but the pair is still under $100.
BTW, I was thinking about trying to feed the Olimex better power, but I think I decided it's a waste of effort. The isolator chip is powered by USB power, and is tied to USB/PC ground, so it is noisy.
The Olimex has a buck converter so you can generate +5 for the isolated USB, but that is obviously less then ideal. But, you can just disconnect USB power at the DAC, and feed the DAC cleaner power from your own PS. If you do this, it doesn't matter what kind of reg the usb isolator has because you aren't going to use it.
I haven't decided if I want to feed the DAC 5V, or defeat the reg on the dac and feed it 3.3V.
BTW, right now I am planning to put a little PS, along with the DAC and USB ISO in a little case so its all self contained.
Randy
You right!
I also thought to split the usb cable with something similar...
MikroTik 5V USB Power Injector - €2.69 - WirelessConnect.eu MikroTik wireless / network routers, proxies, firewalls and VPN Concentrators
I also thought to split the usb cable with something similar...
MikroTik 5V USB Power Injector - €2.69 - WirelessConnect.eu MikroTik wireless / network routers, proxies, firewalls and VPN Concentrators
I would expect with that cable splitter you could supply a better +5 for USB power, but you will still get ground noise.
The ADuM4160 chip that everyone uses in these USB isolators creates seperate grounds, so I would hope the peripheral ground is cleaner than the host ground.
And after reviewing the 4160 datasheet, I retract a previous statement, there might be gains by providing the 4160 +3.3V on the 2 or peripheral side. The 1, or PC side would still run from USB power.
Randy
the amanero app note calls for an isolator on the i2s, not the usb. this adds jitter but it can be cleaned up to a degree by the ESS dac. you can just supply clean power to the amanero and you have only ground left as mentioned.
the ultimate solution is to use the i2s fifo buffer and the clock isolation board, then you would not need the isolator on the i2s or USB and the clock and i2s would be clean as a whistle. not the cheapest solution going, but cheaper than some tweaks ive seen.
the ultimate solution is to use the i2s fifo buffer and the clock isolation board, then you would not need the isolator on the i2s or USB and the clock and i2s would be clean as a whistle. not the cheapest solution going, but cheaper than some tweaks ive seen.
I went ahead and ordered the Teralink ADuM4160 board - for less than 20€ shipped, it's a bargain.
I don't have any measuring equipment, but will provide my subjective feedback (for what it's worth) when it arrives.
On a side note, my USB data lines - not just the power - are noisy, so an improvement on that area is expected.
(I also have a FiiO D3, connected to my TV, and that thing produces zero background noise using the PC onboard spdif)
I've also heard issues of PC not recognizing "peripherals" attached to the USB ISO, and having to reconnect them to force detection.
If the gear is all inside a case, having a separate PSU - and a switch - can be handy in this situation.
I don't have any measuring equipment, but will provide my subjective feedback (for what it's worth) when it arrives.
On a side note, my USB data lines - not just the power - are noisy, so an improvement on that area is expected.
(I also have a FiiO D3, connected to my TV, and that thing produces zero background noise using the PC onboard spdif)
I've also heard issues of PC not recognizing "peripherals" attached to the USB ISO, and having to reconnect them to force detection.
If the gear is all inside a case, having a separate PSU - and a switch
- can be handy in this situation.
Really?!?
As the chip isolates the data lines, I would assume that electrical noise/interference would simply be rejected as well?
The ADUM will reject common mode noise by virtue of the transformers, and the CMOS switching may go some way to cleaning up the signal too.
ahh fair call, I didnt check which one of the isolation methods its using. neither opto or capacitive isolators would do anything much. hmm cmos switching? thats a bit of a stretch isnt it?
ok so common mode noise will be cleaned up somewhat. the adum datasheet does not specify or even mention any CMRR for noise, just massive transients.
ok so common mode noise will be cleaned up somewhat. the adum datasheet does not specify or even mention any CMRR for noise, just massive transients.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.