Hello all,
I have found a couple of Cambridge Audio Azur CD players (not working, tray issues) locally 540C/640C that have the WM8740 DACs in them. Are they worth buying for the DACs? $50 each.
Just wondered if they are worth picking up for a DIY someday project?
Thanks!
I have found a couple of Cambridge Audio Azur CD players (not working, tray issues) locally 540C/640C that have the WM8740 DACs in them. Are they worth buying for the DACs? $50 each.
Just wondered if they are worth picking up for a DIY someday project?
Thanks!
Maybe…
You’re getting functional power supplies in there also, a cabinet, might be worth it if you can use more than the dac.
As far as the dac itself, I like the WM8740/1, sound decent with everything done right around them.
You’re getting functional power supplies in there also, a cabinet, might be worth it if you can use more than the dac.
As far as the dac itself, I like the WM8740/1, sound decent with everything done right around them.
I have ca azur 640c, with dual wolfsons. Delightful sound. I used that cd player so much mechanics started to wobble. I bought replacement cd drive on aliexpress for $50 and installed it myself. You have to unsolder one spot to activate the laser. CD player is like new again.
Just a quick google:
https://www.ebay.com/itm/143372978439
https://www.ebay.com/itm/154532794783
Moderator Edit: Corrections as requested by poster
Just a quick google:
https://www.ebay.com/itm/143372978439
https://www.ebay.com/itm/154532794783
Moderator Edit: Corrections as requested by poster
Last edited by a moderator:
Thanks guys,
Appreciate the input. I thought the 640C had dual DACs but I think it is only the rev II, the first one, from what I have read had only 1. I'm not sure which rev this one is, maybe you can tell from the picture.
Forgot to mention someone has hacked in a tube output, and the optical and coax output jacks are missing on the 640. The picture is of the 640C. Don't have a picture of the inside of the 540C.
So now what do you think? LOL
Cheers
Appreciate the input. I thought the 640C had dual DACs but I think it is only the rev II, the first one, from what I have read had only 1. I'm not sure which rev this one is, maybe you can tell from the picture.
Forgot to mention someone has hacked in a tube output, and the optical and coax output jacks are missing on the 640. The picture is of the 640C. Don't have a picture of the inside of the 540C.
So now what do you think? LOL
Cheers
What’s the goal?
Myself, Would be thinking a network player using a small computer, running linux (Moode, Volumio), suitable power supply, all tucked into the same case.
Upgrade any older capacitors, op amps to fine tune the sound if desired…
It would probably be tough to buy the parts that will be in those for that price, so would be reasonable, but only if you will use them.
I would want the original output, and likely modify it.😎
Myself, Would be thinking a network player using a small computer, running linux (Moode, Volumio), suitable power supply, all tucked into the same case.
Upgrade any older capacitors, op amps to fine tune the sound if desired…
It would probably be tough to buy the parts that will be in those for that price, so would be reasonable, but only if you will use them.
I would want the original output, and likely modify it.😎
Lampizator is the one who hacked its output with tubes. Who else?
Mine has both, coaxial and optical. That's why this cd player got used so much, it allowed the same cd played on three outputs, two external dacs, with three headphone amps and three different headphones. Its like a gazillion options. Mindboggling.
Cheers!
I made several dual mono 2 x WM8740, and it is one of the best dac sound. Very very good bass and natural mids. But only in dual mono configuration and with good power supply. Also, without OPamps, JFET buffer and analog filter output. Balanced to SE with transformer...
I other stereo configutation where single chip used for booth chennels, sound is clearly worse...
I dont know why, probably about paralleling and something with SE outputs...
...
I am planning to parallel 4 chip per channel in mono-mode, to hear what is happening...
I other stereo configutation where single chip used for booth chennels, sound is clearly worse...
I dont know why, probably about paralleling and something with SE outputs...
...
I am planning to parallel 4 chip per channel in mono-mode, to hear what is happening...
WM8740 have 1K (recommended by WM) at the each output and offset of 1/2 Vpower, about +2.5V.
This is the voltage output DAC and there is no need for amplification of voltage output.
So only use of tubes could be as buffer...
It could be connected (in mono mode only) direct to output line transformer because all balanced lines are in the same +2.5V offset and it will be not current flow trough the windings in the primary layer.
BUT with that 1K off added internal resistance ( it is more if we count only internal of DAC chip) the inductance of the transformer should be huge 🙂
With JFET buffer from 2SK170 output resistence is less than 50ohms an can drive much much lower [Hy].
With ECC88 parallel sections cathode follower buffer, output resistance is about 1/S and for 2 sections is somewhere of 100ohms
That is also not bad for driving, in this case doubled [Hy]
...
This is wery important because the key point is not to loose this excellent bass from WM8740. 🙂
This is the voltage output DAC and there is no need for amplification of voltage output.
So only use of tubes could be as buffer...
It could be connected (in mono mode only) direct to output line transformer because all balanced lines are in the same +2.5V offset and it will be not current flow trough the windings in the primary layer.
BUT with that 1K off added internal resistance ( it is more if we count only internal of DAC chip) the inductance of the transformer should be huge 🙂
With JFET buffer from 2SK170 output resistence is less than 50ohms an can drive much much lower [Hy].
With ECC88 parallel sections cathode follower buffer, output resistance is about 1/S and for 2 sections is somewhere of 100ohms
That is also not bad for driving, in this case doubled [Hy]
...
This is wery important because the key point is not to loose this excellent bass from WM8740. 🙂
Hello Zoran, I have built a dual DAC with CS4398 controoled via software with arduino. I have set one chip with the A Output Left channel and the B Output MUTE.
For the other chip the same , of course for the right channel. I am not using the B output. It is worth to use it ? And how I should use it ?
Paralleling in one chip aL + with bL+ and aL- with bL+? I can understand for DAC with current output the advantage to have more current, but how should I connect the 2 Chips?
I have also thought to use 4 chips just to experiment, but I am not sure how to connect them.
Thanks
For the other chip the same , of course for the right channel. I am not using the B output. It is worth to use it ? And how I should use it ?
Paralleling in one chip aL + with bL+ and aL- with bL+? I can understand for DAC with current output the advantage to have more current, but how should I connect the 2 Chips?
I have also thought to use 4 chips just to experiment, but I am not sure how to connect them.
Thanks
Attachments
Hi I didn work with CS4398. Sorry.
I took a very bref look at the CS4398 datasheet and i didnt find option for mono mode?
And I think that is voltage output DAC, not current output?
I took a very bref look at the CS4398 datasheet and i didnt find option for mono mode?
And I think that is voltage output DAC, not current output?
Yes it is a voltage output DAC like WM8740, very similar to this chip. With I2C bus I change the register 0x03 values like in the attached files, disabling the analog output B (PIN 19 and 20). So every chip is working only with one channel. It sounds very good and bass are amazing.
I tested it with opamp discrete output and with Bisek output transformer. Next step is also a jfet plus cathode follower configuration.
Well my question is: should I use also Analog output B (PIN 19 an 20) and connecting together PIN 24 with PIN 19 and PIN 23 with PIN 20?
In this case I will set in the register aL bL for Chip 1 and aR bR for chip Two.
Or schould I try something else? What if I use 4 DAC? How should work with them?
Thanks for your time.
Michele Isaia
I tested it with opamp discrete output and with Bisek output transformer. Next step is also a jfet plus cathode follower configuration.
Well my question is: should I use also Analog output B (PIN 19 an 20) and connecting together PIN 24 with PIN 19 and PIN 23 with PIN 20?
In this case I will set in the register aL bL for Chip 1 and aR bR for chip Two.
Or schould I try something else? What if I use 4 DAC? How should work with them?
Thanks for your time.
Michele Isaia
Attachments
If You can set with registers to supply booth internal dacs with same data L or R then You can try?
But You have to set 2 different register words
.
For one L cnannel chip use:
aL bL with 01010 (word left to right from ATAPI4 to ATAPI0)
For one R cnannel chip use one of 2 options:
aR bR with 00101
aR bR with 10101 (that is inverted word from aL bL)
al these from the table.
Please check
.
But first not connect the outputs togeher. First test each ouput for dedicated channel
If it is working, same channel at 2 ouputs, then connect 100R resistors at each output and parallel tham.
.
Useful information if exist will be output reistance of analog output.
But You have to set 2 different register words
.
For one L cnannel chip use:
aL bL with 01010 (word left to right from ATAPI4 to ATAPI0)
For one R cnannel chip use one of 2 options:
aR bR with 00101
aR bR with 10101 (that is inverted word from aL bL)
al these from the table.
Please check
.
But first not connect the outputs togeher. First test each ouput for dedicated channel
If it is working, same channel at 2 ouputs, then connect 100R resistors at each output and parallel tham.
.
Useful information if exist will be output reistance of analog output.
Thank you Zoran, I Have already done what you said last night and also verified the phase between the output with the oscilloscope. I connected them together without resistor and I measured again the output: nothing changed in the Vpp voltage.
Could you please make a sketch of your following sentence?
But first not connect the outputs together. First test each output for dedicated channel
If it is working, same channel at 2 outputs, then connect 100R resistors at each output and parallel them.
According to datasheet at page 9 the output Impedance of analog output is 118 Ohm.
Could you please make a sketch of your following sentence?
But first not connect the outputs together. First test each output for dedicated channel
If it is working, same channel at 2 outputs, then connect 100R resistors at each output and parallel them.
According to datasheet at page 9 the output Impedance of analog output is 118 Ohm.
What I think is pretty much You have been done.
Good thing is that You measure individual outputs, before paralleling, and checked the phase.
.
118ohm is good lower output impedance, and in parallel mono mode it will be cut to half. of About 60ohms.
With that walue of generator You can drive even a interstage transformer?
.
The simpiest way is to put C at the output.
Value depends on next input resistor of the next stage.
for 10K next stage use 10uF
for 100K use 1uF, and so.
.
But even better will be to use 2 chips, (4 DACs) in parallel per channel.
.
Is there any DC offset voltage at the outputs?
Maybe the 1/2 of Vanalog supply?
.
cheers
Good thing is that You measure individual outputs, before paralleling, and checked the phase.
.
118ohm is good lower output impedance, and in parallel mono mode it will be cut to half. of About 60ohms.
With that walue of generator You can drive even a interstage transformer?
.
The simpiest way is to put C at the output.
Value depends on next input resistor of the next stage.
for 10K next stage use 10uF
for 100K use 1uF, and so.
.
But even better will be to use 2 chips, (4 DACs) in parallel per channel.
.
Is there any DC offset voltage at the outputs?
Maybe the 1/2 of Vanalog supply?
.
cheers
Hello Zoran, yes there is a DC offest of 2,5 V at the PIN output, that is way I adopted a "Bisek" Transformer followed by a Borberly Buffer stage. The transformer deleted the problem of the DC offset. This I have done for my "official" version already assembled in a beautiful Box.
The other version of the DAC where I do know all these tests has in the output stage a 2nd order Multifeedback Low pass filter with Gain = 1.
So If I have right understood I should put before the Multifeedback low pass filter an Extra 100 Ohm resistor for every analogue output PIN which I connect in parallel to the Input + and - of the op amp? At the end if I use two chip per channel I should have for one channel four conductor with R100 in series bringing the + signal and other 4 conductor with R100 in series bringing the - signal. Could you confirm this?
The other version of the DAC where I do know all these tests has in the output stage a 2nd order Multifeedback Low pass filter with Gain = 1.
So If I have right understood I should put before the Multifeedback low pass filter an Extra 100 Ohm resistor for every analogue output PIN which I connect in parallel to the Input + and - of the op amp? At the end if I use two chip per channel I should have for one channel four conductor with R100 in series bringing the + signal and other 4 conductor with R100 in series bringing the - signal. Could you confirm this?
Conceptually it is good solution with transformer. It is OK until same DC offset value on the primary, then no current flow trough the primarry winding and no magnetization of sensitive high permeability core... But if there a slight diference in DC ofset at the outputs, when turning on, or off, or during the operation, the core will saturate with DC and stay in that mode. so check eith some analog instrument (without transformer connected, if there some difference in voltage at the outputs.
With a transformer is good passive way to get Single ended output from differential...
.
You can also make a passive filter directly at the outputs?
Advantage is very low output impedance of paralleled dac voltage outputs.
1x dac section 118ohm
2x dac section 59 ohm (1 dac chip per channel, mono mode)
4x dac section 30 ohm (2 dac chip per channel, mono mode)
.
these 30 ohm is about to be smaller than Rdc-on of some JFETs...
So it is a very good driver.
.
I do not see the point of using OP amps in any eay with this dac?
If You need a filter You can use passive RCC one, 5 or 7 order, bessel, or other type?
You have low output imedance alrady and suficient voltage level.
please take a look at the link:
https://rf-tools.com/lc-filter/
shunt configuration resulting with lower L values. For the start You can use ferrite beads (low Rdc values as can be) close values.
In this example this is 47uH and 100uH
there is single ended filter but it is easy way to rearange fir balanced version,
Pr use 2x for each diff. output to ground.
You can use DC block output C of 10uF to each output if You want to eliminate DC, and You can use even without the transformer direct...
With a transformer is good passive way to get Single ended output from differential...
.
You can also make a passive filter directly at the outputs?
Advantage is very low output impedance of paralleled dac voltage outputs.
1x dac section 118ohm
2x dac section 59 ohm (1 dac chip per channel, mono mode)
4x dac section 30 ohm (2 dac chip per channel, mono mode)
.
these 30 ohm is about to be smaller than Rdc-on of some JFETs...
So it is a very good driver.
.
I do not see the point of using OP amps in any eay with this dac?
If You need a filter You can use passive RCC one, 5 or 7 order, bessel, or other type?
You have low output imedance alrady and suficient voltage level.
please take a look at the link:
https://rf-tools.com/lc-filter/
shunt configuration resulting with lower L values. For the start You can use ferrite beads (low Rdc values as can be) close values.
In this example this is 47uH and 100uH
there is single ended filter but it is easy way to rearange fir balanced version,
Pr use 2x for each diff. output to ground.
You can use DC block output C of 10uF to each output if You want to eliminate DC, and You can use even without the transformer direct...
