DIY dac ver.3 and cd-rom controller kit

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
This thread is for Thomas' dac ver.3 and cd-rom controller kit. I purchased the fully assembled dac ver.3 with transformer and the cd-rom controller. Once I figured out the wiring for the transformer for the USA, the dac powered up and worked. I purchased the upgraded caps (which are not installed) and a TDA1541AS2. I also have BB 627 opamps I would like to try also.

My first attemp with the cd-rom controller was not so good. I used an old 4x Creative Labs cd-rom drive. I purchased and briefly tried an ASUS CD-S520/A5 52X cd-rom drive. This drive seems to work well so far. It has not been hooked up to the dac ver.3 kit yet, but the cd-rom controller seems to control it.

I still need to supply a chassis, fix the U4 heat issue, connect the drive to the dac ver.3 and install all upgrades and test.
 
yup mine works well on a LG CDROM. The play push button doesn't seem to work, but i can use the remote. Pray hard the remote doesn't spoil.

other than that, if i press the other buttons too much, my unit will hang and will need a 'reboot'.

however, i have a concern. I was told the digital out of the CDROM has a high current/voltage (5v) than normal digital output from CDP (<2v). Is the non profit DAC design to take that input at that voltage ?

Well, i did use it b4 with no problem. was thinking of using this on a long term, any problem ?
 
Hi Thomas, I might be interested, but I would also like to run the cd-rom too so I assume a +12 would be needed too.

InfinityPWS, I think the TTL input is for the higher voltage digital input. I am not positive, so it would be at your own risk. I chanced mine and it works fine. I don't know the output voltage of the Asus rom, but assume it is the same as other rom drives.
 
Originally by stefanobilliani
Thank you Jean-Charles for putting at the surface the noise problem ... I confirmed myself your findings immediately . Anyway the cure for it is also very easy to implement .

1) look at the Clock little board : Disconnect the GND from J1 to J85(main board )
This immediately eliminate the noise pollution.

2) now look at the SAA7220 dig. filter board :
disconnect GND form J1 to J75 ( main board).
Disconnect GND form J5 to J24 (main board) .

Probably was a ground loop . It would be nice if you measure again the noise at the chips after this trategy.

I tried this today and find it helps. I still get some hum when I turn the volume all the way up with no music playing. For normal listening levels the noise is now low. I need to separate my ground at the output which could help more.
 
I feel like I am writing to my self, but anyway.

I sat down for a listen to the dac and found that the Asus cd rom drive is causing some sort of a distortion issue with the dac. I put in old Genesis cd and heard tizziness in the voice and on some instruments. I then tried another cd and heard the same thing. It was like a high rattle in the sound. I then used my Eastsound E5 as a transport and the problem went away.

I am using the digital out on the Asus drive connected via TTL on the dac. Any comments?
 
Kip said:


I tried this today and find it helps. I still get some hum when I turn the volume all the way up with no music playing. For normal listening levels the noise is now low. I need to separate my ground at the output which could help more.

I received the dac just yesterday so I need to look at it carefully before finding more solutions .
Yes there is still some noise ...but :
a) I do not think it is coming form a ground loop anymore
b) I think the PS need more filtering capacitance just before the diodes
c) consider that the output level of the dac is exceptionally high and no muting circuit involved .

By the way Thomas ( tubelover) ... what type of diodes there are in this ver.3 ? thanks .
 
Need to specfy one thing : Tubelover dac V3 has and inverted phase analog output . I noticed in schematics that the 1541A output is loaded by 50 ohms before entering ( via 2k ohm) the Op Amp .
Elsewhere is explained that loading the output that way produces more distorsion ( and I believe noise ) .
Tonight I will try my JFET Pass -Cascode- Buffered output stage and will let you know how it goes, since it is a very quiet output stage .

:cool:
 
Form the -ecdesigns- linear interpolation thread there examples of I/v stages probably pretty much like the 1541A datasheet . They are using the inverting imput of the OP Amp ...so I think a very quick mod in the Ver.3 board is not possible in this matter . If you just want to decrease the gain you can try different combinations in R63-R64 and R66-R65 . R63 and R66 are the feedback loop to each channel , R64 R65 the load to ground .
 
stefanobilliani said:

Tonight I will try my JFET Pass -Cascode- Buffered output stage and will let you know how it goes, since it is a very quiet output stage


I tested the Jfet Pass output stage ... and the dac is *extremely quiet* with it .

Also testing at max volume - no music it is hardly possible to hear any noise close to the ( 96db ) Fostex Fe206 MLTL.

So I think lot of noise is picked up by the OP amps in the Ver3 and by the fact that 1541A outputs are loaded with 50 ohm and working at 100mV .

I reccomand to use a different output stage for max dynamics and much less noise . :D
 
hi kip,

the linear power supply will had three section of output.

Large current 5V,12V & one varible volatge for USB or others.

hi stefanobilliani,

Low jitter VCXO I was contact NSK china to oem for me.
it was low jitter & 5ppm. If i choose lowered ppm the price looks Diyers will not easy to afford. They charge me USD26 per one low jitter VCXO in 5 ppm. I will use this in my comming ver.3 ultimate version.

regarding the little noise problem. I improve some point for it.

1. reference the project D-1 laout, change the power section & DAC layout, prevent remain ground loop.

2. increase the 3300uf of the power supply to 4700uf but large caps was parallel buy one large & one small cap(3300uf +1000uf +0.1uf).

3. I understand the Jfet or mosfet, NPN output stage will better. But close match for the cascode circuit will cost much money. I will prepare this in the top model of my dac, full balance only. since I guess that they will use two AS2 chips. They will not care paid little money more for match the fet. Output stage of my balance dac I choose BC109 & 2N930. NPN only.

4. for the ultimate ver.3 I will update above improvement & use transformers output stage for balance or SE output to lower the sound stage more.

5. I thought some of the noise was bad contact of the adapter to the pcb. So in this Ultimate version onlythe CS8414 will remain adapter, 7220,Op, reclock will all fit into the PCB.



BTW, diode that i use in powe supply was FR107 . parallel two FR107 with small caps can lower noise, more dynamic. most 1541a dac was little soft sound after caps was 78XX series. Ultimate version will use 317 & 337.

hope this can help, any more comments for improve.

thx

thomas
 
I am giving up on the cd-rom as a transport. I have posted it for sale. I am going to use an older cd player for the transport. I would preferr an older Philips or such that can be modified to use an I2S bus out to this dac.

Thank you stefanobilliani, Maybe I should have waited for the ultimate dac. I can just reverse the speaker cables on my amp to fix the inverting or phase issue. I don't have enough knowlage to design my own output stage.
 
I don't have much technical experiance and I am choosing the easier way. The cd-rom connected to the TTL input adds some sort of high tizzyness or buzzing distortion that the player used as a transport does not. I don't know where to go with this and would rather sell the set to someone who has more knowlage. The Cd-controller seems to control the Asus rom fine with one small issue being, the controller doesn't show how many tracks on the cd inserted until stop is pushed. It only shows one track. Other then that, no lock-ups or any other issues experianced. The Asus rom outputs IEC-958 spec. I contacted the chip manufaturer (MediaTek) to see if their 1189 chip outputs I2S, but received no responce.
 
hi kip,

pls let me know which asus model U are Using. let me check that I can help or not. Cd-rom controller will had some limitation. will not the control interface full identical to cd-machine. but good cd-rom was better than cheap cd-machine. many expensive cd-machine use cd-rom or dvd rom for the reading machine.

Looks I must only choose one model of the cd-rom & detail to tell diyers how to modify. so diyers not need to waste money.

Pls let me know the model of your cd-rom.

thx


thomas
 
Hi! stefanobilliani

QUOTE]I tested the Jfet Pass output stage ... and the dac is *extremely quiet* with it .[/QUOTE]

Your test interested me a lot. I was planning to try the pass Jfet i/v as soon I find time, but your report boosted my curiosity and I plan to try it sooner now.

Just a few questions. How did you connected the i/v to the dac? Dirrectly to the output, removing the two 100 ohms resistors and shunting capacitor?

Jean-Charles
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.