• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Review and Measurements of WesionTEK Khadas Tone Board DAC

Veri

Master Contributor
Joined
Feb 6, 2018
Messages
9,596
Likes
12,036
In the ad it mentioned 5V-32V and I thought it was adjustable range. Then when I foolishly ;D emailed the seller AFTER I bought the PSU about adjustability he said the tunable range was only 5V-12V.

I'm going to be using the Meiyan to power multiple devices now and in the future I will most likely use it for any other devices, perhaps a DAC that might require 5V-12V. The benefits of these LPS' have been nothing short of amazing so the cost is low in comparison to the satisfaction it provides to my listening experience.

When I say multiple devices I mean :

*PPA V2 which Paul claims to want 3A but I'm currently running the V2 using an alarm clock SMPS which I fashioned a 2.1mm plug to, with a LT3045 5V 500ma in post. I would guarantee a significant gain in SQ once I use either the Teradak or Meiyan.
*Khadas Tone Board via USB.
*Startech/ICRON LAN extender in time when I can afford one :)
*Behringer UMC404HD. It takes 5V 1A and still want to see if changing the SMPS to an LPS will give my microphone better SQ on the input. Maybe with an LPS it will provide cleaner sound? I don't know, we'll see!
*Any future devices needing some good clean power :)
This is how OCD begins! Take care man :p

Then... I found the Ultra USB Chain lead by Rob and the wonderful community involved.
Wait, you mean this?
9487449.jpg


Oh. Oh no...
 

graz_lag

Major Contributor
Joined
Oct 13, 2018
Messages
1,296
Likes
1,583
Location
Le Mans, France
In the ad it mentioned 5V-32V and I thought it was adjustable range. Then when I foolishly ;D emailed the seller AFTER I bought the PSU about adjustability he said the tunable range was only 5V-12V.

I'm going to be using the Meiyan to power multiple devices now and in the future I will most likely use it for any other devices, perhaps a DAC that might require 5V-12V. The benefits of these LPS' have been nothing short of amazing so the cost is low in comparison to the satisfaction it provides to my listening experience.

When I say multiple devices I mean :

*PPA V2 which Paul claims to want 3A but I'm currently running the V2 using an alarm clock SMPS which I fashioned a 2.1mm plug to, with a LT3045 5V 500ma in post. I would guarantee a significant gain in SQ once I use either the Teradak or Meiyan.
*Khadas Tone Board via USB.
*Startech/ICRON LAN extender in time when I can afford one :)
*Behringer UMC404HD. It takes 5V 1A and still want to see if changing the SMPS to an LPS will give my microphone better SQ on the input. Maybe with an LPS it will provide cleaner sound? I don't know, we'll see!
*Any future devices needing some good clean power :)

My journey began when I threw out the SMPS on my Geshelli Labs ENOG PRO2 and powered it with a Circuit Specialists LPS with unbelievable results. Then... I found the Ultra USB Chain lead by Rob and the wonderful community involved.

Trust your ears
Take care!

1552080631318.png
 

bequietjk

Senior Member
Joined
Jan 4, 2019
Messages
462
Likes
474
@Veri Yes... Oh yes hahahaha. But that's only the beginning to the madness. That goes back some years. The chain has been MUCH improved and I'm mostly speaking from the experience of only a fraction of the additions. There is so much sound to be had! In the time of that picture it is before the LT3045's were implemented into the chain. The most recent chain looks like medusas headdress compared to that in all it's devices and connectivity.

@graz_lag <3 your display pic!

Now, not to steer away from this thread.

I absolutely love my KTB!

Powered and data streamed through USB it sounded magnificent compared to my ol dac. Powered through a USB port on my PC and using SPDIF's input fed by the Topping D10 it sounded even greater! Truly exposing what the KTB is capable of, at least in my experience. YMMV.

Super excited to power with an LPS.
0308191746.jpg
 
Last edited:

Dominique

New Member
Joined
Dec 13, 2018
Messages
2
Likes
14
Hello,,
You knew my DAC Khadas in his "prototype" version.


Here it is in a more complete version.
This small DAC is now incorporated in a good big Goldmund Nemesis style box that also integrates a passive volume/input selector with remote control.


The 5V power supply of the DAC is provided by a module with a comparator NE5534 followed by a HexFet transistor.
The volume control is operated by a ramp of Dale RN55 resistors selected by relays. Constant input impedance of 10Kohms, attenuation from -64 dB to 0 dB in 128 steps of 0.5 dB. The input selector manages 4 inputs, each of which can be either analog (unbalanced) or digital (SPDIF).


The rear panel is equipped with a dozen of CINCH for analog and digital inputs and 2 additional pairs for outputs. There is also a pair of XLRs, connected to a balun module very useful to attack a pro SUB.


There is still a very important space available in the box so enough to integrate a preamplifier (tube?) or an amp (K209?).
- I'm thinking about it.
 

redjr

Member
Joined
Oct 20, 2018
Messages
72
Likes
63
Some of the songs on MP3 at 128kbps sounded really weird but no problem with 320kbps especially the bass that I ripped from the same CD. No problem with FLAC, SACD, or CD player. While the Benchmark is fine
Any music at 128kbps will sound weird. :)
 

dwalme

Member
Joined
Feb 19, 2018
Messages
41
Likes
25
Anyone else have the problem that the board disappears from Windows when the computer sleeps and wakes up?

I have to unplug / plug the USB to the tone board for Windows to see it again. This happens with both the UAC compliant drivers and the Xmos drivers from Khadas.
 

soundwave76

Addicted to Fun and Learning
Forum Donor
Joined
Dec 28, 2018
Messages
730
Likes
1,364
Location
Finland
Damn. I still can't wrap my head around the fact that this $99 DAC is performing this good and yet the designer provides the schematics on the net, FOR FREE?! When was the last time something like this happened in audio industry? The ol' ODAC days?

NwAvGuy. Never forget.
 

rajapruk

Senior Member
Forum Donor
Joined
Sep 22, 2018
Messages
472
Likes
489
Hi, I would like to contribute with my two cents on the topic of the mysterious "Sabre hump" in the THD+N vs. output level (and IMD vs. output level) measurements. I came across this issue in 2014 when I was developing a DAC module for my master's thesis and had an opportunity to spend a lot of time in my university's lab with an audio analyzer. There was no Sabre PRO series at the time and the flagship of ESS DAC lineup was the ES9018S, so that is the chip I'm referring to, but I believe the same applies to other Sabre DAC chips exhibiting this phenomenon.

The hump starts occurring when DAC output surpasses a level of around -36dB full scale. It isn't clear from the THD+N vs. output level measurement what exactly happens, for that the spectrum of a tone with peak level slightly above that needs to be measured. So I did and found very interesting modulated noise patterns changing both frequency and amplitude with the output tone level.

I discussed that on DIYaudio.com back then and some people with a lot of experience in sigma-delta technology pointed out that the culprit is probably mismatch in the physical elements of the DAC output, while -36dBFS is a region, when the modulator transitions from single-bit operation to a multi-bit (-36dB corresponds to 6 bits). The difference between the channels is then caused by a different amount of mismatch. Note that I only measured channels 1 and 2 (best and worst in my case) while channels 3 to 8 stayed unused.

So it is actually a noise component of THD+N, not a distortion, what creates a hump. I believe that IMD vs. frequency measurements were affected just because the audio analyzer isn't able to recognize the noise from the intermodulation products.

The good news is that with ES9028PRO, I didn't find any evidence of the hump happening when I tried to recreate this. I suppose the same applies to ES9038PRO and I'm really glad ESS Technology addressed this issue in the PRO series chips.

Kind regards,
Pavel Krasensky, www.oktoresearch.com

View attachment 18412View attachment 18413View attachment 18414View attachment 18415View attachment 18416View attachment 18417

Is it unit-to-unit variation also maybe in the chips from manufacturing in this specific matter?
 
D

Deleted member 6802

Guest
Hallo, is the Khadas Tone Board good enough to be used in High-end floorstanding speaker setup?
 
D

Deleted member 6802

Guest
About sound you bet :) about how it looks can be the culprit if you know what i mean when a user talks High-end ;)...

Hahaha I know :) ;) For me looks matter according to the price, the more you pay the more you are allowed to expect.

My plan is to use it with raspberry pi and hide both in a case at the back of the shelf :)

There is a very poor choice of emojings here
 

ngs428

Active Member
Joined
Nov 21, 2018
Messages
268
Likes
230
And I say HEY! what a wonderful kind of day...

View attachment 23456
https://dl.khadas.com/Firmware/ToneBoard/

  • Khadas_Tone_Board_dfu_20190309_Upgrade_Notes
  • 1. Upgrade the XMOS firmware version to v1.02.
  • 2. Fixed issue with boards becoming bricked after firmware upgrading.

I installed this, all went well. Should it say “eval” in the device info in foobar? I assume that is referring to the eval USB installer we have to use? I just heard talk of a eval firmware that beeped once in a while.. that isn’t what this is, correct?
FA3FFCB4-248B-486F-9588-D1323F9EC9D2.png
 

BYRTT

Addicted to Fun and Learning
Forum Donor
Joined
Nov 2, 2018
Messages
956
Likes
2,452
Location
Denmark (Jutland)
I installed this, all went well. Should it say “eval” in the device info in foobar? I assume that is referring to the eval USB installer we have to use? I just heard talk of a eval firmware that beeped once in a while.. that isn’t what this is, correct?...

Had not yet updated firmware myself in my came as v6.F2 aka 1226, but as i understand about procedure is you need that EVAL driver and yes its the one that beeb once in a while used for audio streaming, EVAL install is needed only for one reason that is it installs a firmware icon shortcut in program group that is the key and command to upgrade firmware. Then as soon as firmware is upgraded one should uninstall EVAL installation and revert to either Win10 native UAC2 driver or install the good v224 driver.

Myself had not had any issues at all running v6.F2 on Win7 platform so think will stay there but will be curious hear if there is anything to notice about after latest upgrade.
 

ngs428

Active Member
Joined
Nov 21, 2018
Messages
268
Likes
230
Had not yet updated firmware myself in my came as v6.F2 aka 1226, but as i understand about procedure is you need that EVAL driver and yes its the one that beeb once in a while used for audio streaming, EVAL install is needed only for one reason that is it installs a firmware icon shortcut in program group that is the key and command to upgrade firmware. Then as soon as firmware is upgraded one should uninstall EVAL installation and revert to either Win10 native UAC2 driver or install the good v224 driver.

Myself had not had any issues at all running v6.F2 on Win7 platform so think will stay there but will be curious hear if there is anything to notice about after latest upgrade.

So, I now have the latest firmware on the TB after following their supplies process on windows 10. https://docs.khadas.com/toneboard/H...#Upgrade-On-Ubuntu?tdsourcetag=s_pctim_aiomsg

I should uninstall the USB installer eval program and make sure the v224 driver is still installed? Sounds like that is what you are recommending.

Not sure if someone who has upgraded can chime in. Thanks!
 

milw50717

Active Member
Joined
Jan 25, 2019
Messages
162
Likes
156
It seems that you can power the KTB from the USB port on a Raspberry Pi. I just installed RoPieee (Roon endpoint with USB audio out) on a Pi 3 powered by a 1 2 Amp Samsung phone charger. The KTB is then connected via USB cable to the Pi and all works as expected within Roon.

This saves me buying another hifiberry Digi+ and means that my headphone stack now becomes PI + KTB + Atom.

Edit - The 1A charger was not sufficient and the Pi would crash. The recommended supply for the Pi 3 is 2.5A, bur I have been using a 2A charger for a day now and all seems OK.

20190313_211408.jpg
 
Last edited:

BYRTT

Addicted to Fun and Learning
Forum Donor
Joined
Nov 2, 2018
Messages
956
Likes
2,452
Location
Denmark (Jutland)
So, I now have the latest firmware on the TB after following their supplies process on windows 10. https://docs.khadas.com/toneboard/H...#Upgrade-On-Ubuntu?tdsourcetag=s_pctim_aiomsg

I should uninstall the USB installer eval program and make sure the v224 driver is still installed? Sounds like that is what you are recommending.

Not sure if someone who has upgraded can chime in. Thanks!

Guess if you used that firmware upgrade shortcut in program group fram EVAL installation and process said okay you should be there.
Then there seems two ways to check your current firmware, one is if you have the shortcut at bottom right side for TUSBAudio control Panel then open that and go to its "Info" tab to see firmware version, example is the blue below which is how it looks on my Win7 for v6.F2 aka 1226, If one is on Win10 native UAC2 driver then there is no TUSBAudio control Panel then go to Win10 device manager and open XMOS device that have symbol as a USB terminal (forget the XMOS device that have speaker symbol) then on details tab shift proberties to same as below and version should be listed in lowest window there, one could imagine after upgrade a unplug replug of USB connector will refresh OS to probe for the new firmware version.

1000.png


Yes you should uninstall EVAL version, but whether to use native Win10 UAC2 driver or the rather old v224 driver is up to you. Would imagine a native driver code written by same company that have written code for OS itself should be the preference but was it myself running Win10 i would do a physical analog loopback test to another input port using for example free REW to check everything looks right and if it doesn't then try v224 and make the same loopback test to get peace in mind, say this because first Win10 build that had build in native UAC2 driver worked not very good on a ASUS U7, sound was different than used to and after a loopback test one could see impulse response was scary weird and phase was nothing but grass, but maybe they fixed that since or that U7 hardware is not up to protocol other than use driver from ASUS.
 
Top Bottom