UAE23+ and UAE23
I modified a UAE23 over the holidays with
- separate regulation for the ES9023 + caps
[I used a 5-lead TPS series 3.3v since I do not use headphones, that regulator has a pin for a bypass cap, so that got it's own cap too, on pin 5]. I only wish I had some tiny coils around for the 5v V+ into regulator #2
- bigger pre-filter 5V cap, low ESR
- Fox 924b 3.3v TCXO 4-pad SMD clock
FOX924B-12.000 Fox Electronics | 631-1068-1-ND | DigiKey
it would be nice of the PCB artwork in UAE23+ would support an etch-cut and spot for addition of secondary 3.3 or 3.6V regulator, a 3.3V TCXO... 12V in with 7805 or so pre-regulator [leave the diodes so can feed in AC or DC or battery], and spot for USB vs 12V power switch [for coaxial power jack vs USB]
I may rebuild the second UAE23 I bought into a large enclosure and add a 1/2 level switched attenuator, or switches for a few steps. I feed it now with Opticis m2-100-03 3m fiber-optic USB isolator [12Mbps rated] with far-end 5V injection port. I have 30 of those isolators left at less than 1/2 $ you will find anywhere, I prefer to ship to USA [easy]. PM me if you want one.
I have no idea if it sounds better, it's "nice". I'll have my second UAE23 soon so I CAN then compare, before I mod it and give to my Sister. I wrapped the UAE23 in foil HVAC tape [mostly because cover doesn't fit now hah]
We call it UAE23+, which is usb isolated and need a external power supply. It will be bigger than UAE23. I think UAE23 is very convenient to use but UAE23+ will be little more high-end.
We still use ultra low noise regulater to make es9023 work the best.
I modified a UAE23 over the holidays with
- separate regulation for the ES9023 + caps
[I used a 5-lead TPS series 3.3v since I do not use headphones, that regulator has a pin for a bypass cap, so that got it's own cap too, on pin 5]. I only wish I had some tiny coils around for the 5v V+ into regulator #2
- bigger pre-filter 5V cap, low ESR
- Fox 924b 3.3v TCXO 4-pad SMD clock
FOX924B-12.000 Fox Electronics | 631-1068-1-ND | DigiKey
it would be nice of the PCB artwork in UAE23+ would support an etch-cut and spot for addition of secondary 3.3 or 3.6V regulator, a 3.3V TCXO... 12V in with 7805 or so pre-regulator [leave the diodes so can feed in AC or DC or battery], and spot for USB vs 12V power switch [for coaxial power jack vs USB]
I may rebuild the second UAE23 I bought into a large enclosure and add a 1/2 level switched attenuator, or switches for a few steps. I feed it now with Opticis m2-100-03 3m fiber-optic USB isolator [12Mbps rated] with far-end 5V injection port. I have 30 of those isolators left at less than 1/2 $ you will find anywhere, I prefer to ship to USA [easy]. PM me if you want one.
I have no idea if it sounds better, it's "nice". I'll have my second UAE23 soon so I CAN then compare, before I mod it and give to my Sister. I wrapped the UAE23 in foil HVAC tape [mostly because cover doesn't fit now hah]
hi alkasar,
nope just here
i got the ferrite today and ran several tests. i'm seeing about a 1.5db improvement in noise, dynamic range, and crosstalk with this tdk clip-on ferrite: ZCAT1325-0530A
it's a tight fit on the cable but it goes on. thd was unchanged with the ferrite.
the adum4160 i ordered is the circuits@home kit without the on board regulator.
-matt
I think this is a question for hifimediy. I am just one of his customers
quantran: This new one looks perfect to build into my headphone amplfier, do you have any ETA when it will be avalible to order?
thank you. adum4160 seems to be the only available chip for this purpose.hi alkasar,
nope just here
i got the ferrite today and ran several tests. i'm seeing about a 1.5db improvement in noise, dynamic range, and crosstalk with this tdk clip-on ferrite: ZCAT1325-0530A
it's a tight fit on the cable but it goes on. thd was unchanged with the ferrite.
the adum4160 i ordered is the circuits@home kit without the on board regulator.
-matt
Another kit available here with many top quality through hole components, and excellent regulator.
Alimentations - Audiophonics KIT Isolateur USB + Alimentation
Let's see how hifimediy made the isolation
about availibility, Steinar said : soon for a first batch.
Probably a matter of days to find it on hifimediy.com
Not related to RF problems but ground hum. It,s very difficult to eliminate AC hum (50/60Hz) even from a USB driven system. This hum is easily -30 to -50dB at nominal Freq and some 20-30dB down at multiples.
I can demonstrate this by sowing some Audacity recordings and graps. Even the laptop driven by it's battery is not free of it. At least you will get it in your power amp. The attached graphs shows mainly the hum from a phono cartridge. The Terratec ADC itself gives some -50dB AC-hum
The recordings always have low fq noise and AC noise. Here is a flac 24/96 from HDtracks, you can see the 60/120/180Hz AC-hum and low fq noise propably from microphone amps. Usually this is highpass filtered off (subsonic filter @25Hz)
If you have a laptop, please download the free Audacity and make your own (uncalibrated) measurements. The spectrum of noise is more interesting than the dB of certain specified mV/fq to us non-engineers.
When I connected my Audio-GD DAC via usb and recorded RCA-output to the same PC, there was practically no hum at all. The ground loop did that. But there was more noise from the soundard of the pc.
I can demonstrate this by sowing some Audacity recordings and graps. Even the laptop driven by it's battery is not free of it. At least you will get it in your power amp. The attached graphs shows mainly the hum from a phono cartridge. The Terratec ADC itself gives some -50dB AC-hum
The recordings always have low fq noise and AC noise. Here is a flac 24/96 from HDtracks, you can see the 60/120/180Hz AC-hum and low fq noise propably from microphone amps. Usually this is highpass filtered off (subsonic filter @25Hz)
If you have a laptop, please download the free Audacity and make your own (uncalibrated) measurements. The spectrum of noise is more interesting than the dB of certain specified mV/fq to us non-engineers.
When I connected my Audio-GD DAC via usb and recorded RCA-output to the same PC, there was practically no hum at all. The ground loop did that. But there was more noise from the soundard of the pc.
Attachments
Last edited:
If i get it correctly, the setup for the last picture is :
- output device = usb dac
- input device = motherboard soundcard microphone in
- foobar paused (playing nothing), you record with audacity
In this setup how do you that the noise does not come from the input device ?
Have you tried the same with line-in, instead of mic in. Usually less noisy input.
- output device = usb dac
- input device = motherboard soundcard microphone in
- foobar paused (playing nothing), you record with audacity
In this setup how do you that the noise does not come from the input device ?
Have you tried the same with line-in, instead of mic in. Usually less noisy input.
based on the photos in post #123, the implementation is quite straight forward. The adum4160 was inserted between TE7022 and USB port. It seems to share the power supply with other chips from lt1763. There is AC supply input.
I actually prefer the layout of the previous version without isolation chip.
I actually prefer the layout of the previous version without isolation chip.
alkasar said:
To Alkasar,
The last is PC/usb to dac/Tenor 7022, dac/RCA-out to the same PC/line in (back panel, motherboard/Realtek HD) My mic-in is mono, noise is about the same. The sample is a 24/96 flac played with foobar, silent section.
Yes, there is a ground loop. But I get almost the same when I measure the PC-driven dac with a laptop. And much more AC hum from any other configuration or AC-powered appliance. (I haven't tested the output of my Logitech X-Fi player yet.)
DAC is Audio-GD NFB-12 set to 4x and minimum phase filter.
My Vincent SV-129 stereo amp gives much more noise, like any amp (excluding D-class).
Here is my Terratec ADC box connected by usb to the same AcerPC. No signal (input disconnected) Level set for recording a LP.
The last is PC/usb to dac/Tenor 7022, dac/RCA-out to the same PC/line in (back panel, motherboard/Realtek HD) My mic-in is mono, noise is about the same. The sample is a 24/96 flac played with foobar, silent section.
Yes, there is a ground loop. But I get almost the same when I measure the PC-driven dac with a laptop. And much more AC hum from any other configuration or AC-powered appliance. (I haven't tested the output of my Logitech X-Fi player yet.)
DAC is Audio-GD NFB-12 set to 4x and minimum phase filter.
My Vincent SV-129 stereo amp gives much more noise, like any amp (excluding D-class).
Here is my Terratec ADC box connected by usb to the same AcerPC. No signal (input disconnected) Level set for recording a LP.
Attachments
Last edited:
but think of how much the UAE23+ can $ave money for people wanting isolation, and there won't be some silly uncovered PCB inline ahead of the DAC, because the DAC maker integrated it. It's a big win for the consumer, and who else HAS this at a DIY price?
I use RCA jacks so am happy about THAT in the UAE23+ -- but I wonder about headphone people [which I am NOT]. I modify almost everything I buy so some room to do the mods would be nice -- if the PCB artwork supports modding so much the better. Pads for a clock, etch-cut and pads for a second regulator...pads for a variety of caps...[some designers filter the heck out of the 1.8V, the DAC AVCC, and use a mixture of caps for the negative charge-pump] so people will ask for the SPDIF to USB, USB to SPDIF [maybe that is there], a pre-regulator [say 7806?] off the diodes [allow any V in up to 12 AC or DC], spot for a USB vs external PWR SWITCH. I'll hack it either way though, fun stuff!
I may breadboard a ES9023 very soon; I have a few now... and 3 spare TE8802 USB receiver assemblies, also a CM6631. I also ordered some 4PDT 5V relays, so I can try I2s switching while keeping leads very short [as I have been advised]. Last eve I made TE8802-->AK4396 "Miracle DAC" -- it would be nice to switch fast to ES9023 and compare, or WM8742...AK4430... [I have them]. Maybe it will work.
And here tne same TErratec ADC with phono cartridge connected, player not running. The AC hum now -50dB
I just want to demonstrate that AC hum is a real problem. You will get minimal (no?) hum only if you play a battery-driven player and dac and amp (portable player)
I just want to demonstrate that AC hum is a real problem. You will get minimal (no?) hum only if you play a battery-driven player and dac and amp (portable player)
Attachments
You are right, it was just my preference from a DIYer perspective. UAE23+ is a very cost effective solution for consumer.
I have just bought a chassis. The cost is equal to the UAE23. Adding the isolation board and separate power supply, it's not cheap but fun to DIY.
I have just bought a chassis. The cost is equal to the UAE23. Adding the isolation board and separate power supply, it's not cheap but fun to DIY.
I personally think it is ridiculous, AC hum is no way related to DAC, check ground loops cables power supplies etc (but being already caused by any reason of above it CAN change between DACs). -50 dB AC hum is EASILY noticable, I can hear nothing like this on any of my DACs/sound boards/players including UAE23. That 0-10 Hz bump is even more strange, most likely it caused by some 'click' on input/power switching or alike.
Yubing : I can't read from the picures
Hi, there are rectifier bridge and caps to AC-DC, and LT1763 to regulate. The LT1763 accept upto 20V input. So you will know how to choose the power adapter. We suggest AC6~9V, 500mA min adapter. Low voltage will save power for a linear regulator.
BTW, DC input also accepted.
Oh, sorry - i mixed you both together because both had red flags at the names. Anyways, then.
hifimediy: Any ETA on the new dac?
You can place order now. Just contact Steinar if you can't find the link. First batch already reached him.
LT1763 accepts 20v input, but not for a 5v output at the currents here (which are exactly?). it will thermally limit at Junction Temp 150c well before that.
read the 'Calculating Junction Temperature' example in the datasheet.
I know you are clearly not recommending 20v input, but I wouldnt even mention a 20v input being a possibility.
read the 'Calculating Junction Temperature' example in the datasheet.
I know you are clearly not recommending 20v input, but I wouldnt even mention a 20v input being a possibility.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.