If you haven't tried it, get a cheap adum4160 USB isolator and linear dc supply. The one i posted earlier is really nice as it has its own shunt reg for the dac supply, and was pretty cheap. The isolator and linear supply removed a nasty layer of harshness from my dac. I wonder if sensitivity to USB voltage rails noise, and different noise spectra with different pcs, could be causing the forward mids that some hear and others don't.....
USB Isolation
I too have found that USB Isolation works GREAT, with TE7022 anyway.
I have some leftovers from that effort BTW, I have around 30x Opticis
M2-100-03 3M 12Mbps fiber optic based USB isolators, they come with {sigh] a switching PSU for the device-end, at least it has ferrite on BOTH ends of the DC cable. PM me if interested in any at ~1/2 what you'll find on the net. These would NOT be for 480Mbps USB traffic. Perfect for the TE7022 and PCM27## USB receivers.
I also took 2x of these to my Brother-in-law who tried them on his TE7022/Wm8741 and TE7022/CS4398 DACs [we use only USB and FLAC]. I also use a TE7022//CS4398 [with silly USB-powered op amp bypassed and plenty of filtering and bypass caps added, that DAC and the one at my Sister and Bro-in-law's being the same -- and both wired DIRECT OUT to RCAs, not even caps]. Sounds better! Tests were into Magnepan MG3.3R, some studio monitors, and some Genesis Genre I. [The Monitors are modded with some Monsoon planar-magnetic drivers for the mids]. All very revealing. I am not a wine-taster, I know there should be less noise and confusion and, IMHO, isolation is worth doing.
I DID mention to HiFIMEDIY that adding a ADUM4160 to the existing DAC artwork would, in my opinion, take it to the next level. The isolator and a DC-IN jack for power. Just a thought.
I too have found that USB Isolation works GREAT, with TE7022 anyway.
I have some leftovers from that effort BTW, I have around 30x Opticis
M2-100-03 3M 12Mbps fiber optic based USB isolators, they come with {sigh] a switching PSU for the device-end, at least it has ferrite on BOTH ends of the DC cable. PM me if interested in any at ~1/2 what you'll find on the net. These would NOT be for 480Mbps USB traffic. Perfect for the TE7022 and PCM27## USB receivers.
I also took 2x of these to my Brother-in-law who tried them on his TE7022/Wm8741 and TE7022/CS4398 DACs [we use only USB and FLAC]. I also use a TE7022//CS4398 [with silly USB-powered op amp bypassed and plenty of filtering and bypass caps added, that DAC and the one at my Sister and Bro-in-law's being the same -- and both wired DIRECT OUT to RCAs, not even caps]. Sounds better! Tests were into Magnepan MG3.3R, some studio monitors, and some Genesis Genre I. [The Monitors are modded with some Monsoon planar-magnetic drivers for the mids]. All very revealing. I am not a wine-taster, I know there should be less noise and confusion and, IMHO, isolation is worth doing.
I DID mention to HiFIMEDIY that adding a ADUM4160 to the existing DAC artwork would, in my opinion, take it to the next level. The isolator and a DC-IN jack for power. Just a thought.
I have to disagree. While most reviewers preferring uDAC2 over HRT+ (thats explainable considering DACs used) I personally compared mine uDAC2 with this DAC and uDAC is worse mostly in midrange, bass is slightly better defined here too. Thats easily explainable since uDAC has TPA4411 headphone driver on its output which is far from being audiophile with THD+N about 0.1% @ 1 KHz (but it allows to control volume and drive wider range of headphones and has no output caps, voila!). Easily, to me this DAC is a clear winner in all areas, being used with UCD180 amp or AD700 cans.
Not to mention one of mine uDACS amp has failed at once (it is common for TPA4411 which is used on many notebooks btw) and I had to connect ES9023 output directly to RCA. This one is undistinguishable from this DAC and thats perfectly explainable too. There is no magic left these days, at least in DIY forums
mfaughn, since you are only one who noticed this amount of differences between this DAC and buffered version (uDAC2) only explanation could be serious input impedance problems on your amp.
It is pretty ridiculous of you to state that you disagree about a statement I made about my own perceptions. I did say, "to my ears". Anyway, I have pretty good ears and a pretty good pre, a Nak CA-7A. It may not be the pinnacle of preamps but its no slouch and works just fine -- no input impedance issues. I'm not using headphones, I'm using the RCA outs -- which should mean I am bypassing said headphone driver. The differences between the uDAC2 and HifiMeDIY DAC are quite small. I couldn't have heard them without playing well recorded, demanding tracks through a fairly resolving system. I find both the uDAC and the HifiMeDIY DACs to be somewhat colored / forward in the midrange. I wouldn't have noticed this unless I had a reference against which to compare this -- in this case, the clearly superior MS+. I just think the HifiMeDIY DAC is a little bit more forward in the midrange and that the peak is a bit higher in frequency than that of the uDAC2. I am only saying that this is how it sounds in my system, one that I think is pretty good, but still just what I've got. Heck, I know for sure that different ostensibly stellar systems can still sound different. In general, I think a lot of people, myself included, don't know what they are hearing/missing until they hear something that demonstrates how much better things can be. Not saying that anybody here (other than myself) personally might be subject to this phenomenon but I don't think it is uncommon. Anyway, it is entirely conceiveable that others might have rigs that have a great synergy with the HifiMeDIY DAC and sounds "just right" to them.
The HifiMeDIY DAC is a great piece of kit, especially since it is so affordable. I'm actually giving one as a gift to a audio-enthusiast friend of mine. I am not attempting to disparage it, just to give my analysis of it as honestly as possible.
FWIW, My friend gave me a grubDAC this past week and I compared it to this DAC. This DAC is clearly better than the grubDAC that I have in hand.
Last edited:
For this dac, does the es9023 run on 3.3v supply? I see only one LT1763 chip so i this it is the case. I read some where about using 3.6v for es9023. I have just ordered this DAC and I am thinking about making a separate power supply for it rather than using USB power. I have also order the teralink isolator.
Can't help, but you are presuming something which is merely not true, and that even affects your impressions. RCA outs on uDAC2 are simply paralleled with headphone output (they are switched mechanically in the socket to be true). No doubt differences are small, but HifiMeDIY DAC is better than uDAC2, it's both easy to feel and explain, just could not imagine how it could be related to someone ears or gear. If you put something of less quality into signal chain it will limit overall performance, I guess it is obvious.
From the other hand, HRT sound signature might and should be different from them both and that's another story, I did not hear it on my gear and can only rely on reviewer's impressions. Did Sabre DAC based units have forward mids or HRT+ have recessed ones - that could be subjective and ear/gear related. But anyway PCM1794 is good old but not as stellar DAC as Sabre is.
I have read this document, not much information inside. I am looking for the datasheet ...
I understand that it works better with 3.6v but the design of hifimediy uses only one regulator for 3 ICs including the es9023 so i think it runs on 3.3v due to sharing ps with other chips.
This dac looks to have a great potential with power supply upgrade. Some member already mentioned the improvement when used with usb isolator. That way the dac is actually powered by the usb isolator...
I understand that it works better with 3.6v but the design of hifimediy uses only one regulator for 3 ICs including the es9023 so i think it runs on 3.3v due to sharing ps with other chips.
This dac looks to have a great potential with power supply upgrade. Some member already mentioned the improvement when used with usb isolator. That way the dac is actually powered by the usb isolator...
http://www.esstech.com/PDF/ES9023 PB Rev 0.2aPB 110117.pdf[/URL
Quan:
http://www.yoyodyneconsulting.ca/downloads/General/ODAC/ODAC-release.pdf
I see 1.8V 3.3V 3.6V and LOTS of caps! A choke...
I do not know the rules about USB power. Even though I have USB isolation now to my various TE7022 I wonder about the poor man's solution? Can you just cut the 5V from PC and insert say a 7805 and external power for down-stream? Does one have to place a fake load on the PC +5V?
http://www.yoyodyneconsulting.ca/downloads/General/ODAC/ODAC-release.pdf
I see 1.8V 3.3V 3.6V and LOTS of caps! A choke...
I do not know the rules about USB power. Even though I have USB isolation now to my various TE7022 I wonder about the poor man's solution? Can you just cut the 5V from PC and insert say a 7805 and external power for down-stream? Does one have to place a fake load on the PC +5V?
5V
One TE7022//WM8761 DAC I've modded numerous times comes with a full-wave bridge and 7805 after a coaxial power plug and "USB vs external" toggle switch. Batteries would be an option, many of us have dead laptop batteries with a few good 18650 left -- charging those... one guy:
LED Bike Light projects
[3/4s the way down the page] he just mods chargers... I like the chargers with separate monitoring LEDs-per-battery-bay. 2 banks placed in series after charging could hit around 8.2Vdc and eventually sag to maybe 7.6Vdc . I have never had a Lithium exPLODE but I hear they DO so I would not suggest too crude a DIY.
One TE7022//WM8761 DAC I've modded numerous times comes with a full-wave bridge and 7805 after a coaxial power plug and "USB vs external" toggle switch. Batteries would be an option, many of us have dead laptop batteries with a few good 18650 left -- charging those... one guy:
LED Bike Light projects
[3/4s the way down the page] he just mods chargers... I like the chargers with separate monitoring LEDs-per-battery-bay. 2 banks placed in series after charging could hit around 8.2Vdc and eventually sag to maybe 7.6Vdc . I have never had a Lithium exPLODE but I hear they DO so I would not suggest too crude a DIY.
I kinda suspected that was the case but found some other threads that suggested it might not be, that they were set up to work with different impedance. Anyway, it doesn't affect my impressions. How could it? I'm using my ears, not technical specs, to judge.
You know, I've seen this discussions before. I've just never been a participant. Being a participant drives the point home just a little more. Sound is subjective. I am talking about what I hear. What I hear is that the example of the HifiMeDIY DAC that I have sounds, to me, slightly less pleasing than the example of the uDAC2 that I have. This is utterly inarguable. I don't know what your basing your belief that this can't be true on? specs? I don't care what is better on paper. In fact, I think it is more likely that knowing what "should" be better is more likely to be a bias when listening than ignorance of the specs is. I'm using my ears. I know that I have pretty good ears when it comes to analytical listening but what we like or dislike is subjective. If it weren't, then we would all agree on the best speaker and whether we prefer tubes or solid state and exactly which circuit with which components was best. FWIW, I'd love to try out a UCD-180 and am actually considering building that or something very similar. As for cans...I've been through *many* of them in the past several years. I don't use them much anymore but still retain a set of AT AD2000 that I'm quite happy with.
Hi enjoybiking
Thanks for the schematic. That is for the ODAC which use the same chips set but more complex power supply. The Micrel regulators used there is low noise type. The specs of ODAC is quite amazing considering it using usb bus power.
A full wave bridge and a 7805 should work fine but not optimal. I would go for LT1963 which I can get for very reasonable price and much better than 7805. No need to have fake load for USB bus supply I believe.
Battery is a good idea. My friends use battery with their moded airport express with good result but the still use several low noise regs after the battery.
However, an Isolator is still good for decoupling noise on data signals. The isolator must have good power supply too.
Thanks for the schematic. That is for the ODAC which use the same chips set but more complex power supply. The Micrel regulators used there is low noise type. The specs of ODAC is quite amazing considering it using usb bus power.
A full wave bridge and a 7805 should work fine but not optimal. I would go for LT1963 which I can get for very reasonable price and much better than 7805. No need to have fake load for USB bus supply I believe.
Battery is a good idea. My friends use battery with their moded airport express with good result but the still use several low noise regs after the battery.
However, an Isolator is still good for decoupling noise on data signals. The isolator must have good power supply too.
Last edited:
I see your point. Of coarse good sounding definition is very subjective (as the good ears, too). Maybe I was excessively offensive and would like to apologize for that. But one thing remains constant - I do HEAR the opposite and I can logically explain that, both in terms of specs and simple logic. Anyway someone who will read this discussion should be aware of both points of view, I guess.
I was not talking about the inside build quality of uDAC2 yet, that's probably subject for different thread. I was unpleasantly surprised when opened it to fix dead output. Manually added cheap electrolythic caps, matte solder, unwashed flux, wire jumpers... Pretty unexpected for Made in U.S.A. stuff. But I dunno maybe exactly these things make it sounding that special pleasant (or unpleasant?) way.
ES9023 are perfect demistyfing devices - they built to be used in extremely simple way and there are no real method to "improve" their sound against simple direct connection (at least when used with amp not cans). Adding something should be of same or better specs so the total quality will be not compromised. I know so far only one such approach - Patrick's EUVL.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.