Opening the new DacMagic????

Jump to Latest
#404
hi all i am new to dacs and higher quality audio equipment.
so i am interested in the dacmagic or the new dacmagic plus but i read the wolfson dacs are not the best so can they be upgraded/replaced with better ones?
i am new to this so please correct me if i am mistaken on any of these terms or things that can actually be upgraded.
thanks in advance
 
#405
extrafuzzyllama said:
hi all i am new to dacs and higher quality audio equipment.
so i am interested in the dacmagic or the new dacmagic plus but i read the wolfson dacs are not the best so can they be upgraded/replaced with better ones?
i am new to this so please correct me if i am mistaken on any of these terms or things that can actually be upgraded.
thanks in advance
Click to expand...

The wolfson is not that bad. Propper internal power supply for the dac chips and op-amps, is both easier and more important.
The wolfson dac-chip is smd and has several very small pins, that is difficult to unsolder and resolder. By the way, you will not gain much, compared to other mods allready mentioned - even if you find a correct replacement.
I made the following mods:
Nichicon Muse for outputcaps, dexa 5v for dacs and external power for digital chips(bypassing MC34063)
Then tried the dac in high end setup. And even though it lack in detail, clarity, and most other aspecs. then it still peforms ok, it's price considered.
Ofcourse my new dac blows it away. But it also cost's 16 times more 😉

If you aint got the experience and want different sound, dont open it up. Try something else.
remember to read about speakerplacement and acoustics, before judging to much. these things can cheat alot 😉
 
#406
Tordenguden said:
Not sure what you mean

As i wrote on page 31.

First I replaced U404 with a Dexa 5V.
Then I replaced output caps with Nichicon Muse ES.
And yesterday I replaced the original clock, with the D-clock from dexa/newclassd. The clock is build with a dedicated powersupply, as seen on the picture:
http://newclassd.com/images/dclk1.jpg
And the rest of the specs:
NewClassD Audio Clocks
But mine is whitout the HALO transformer.

I have powered the clock from the output of U402. whitch I think should be ok, since this is where the U404 gets its power to. But I am thinking of replacing the MC34063 (U12) with a dedicated linear regulator or powersupply, to totally elliminate switching noise from the circuit.

Another thing. Last night before i went to bed. I realised that I had been listening for an straight hour, with no listening fatiuqe at all. And this was with music, from a 20 year old cd, that I used to think was badly recorded. But last night, was the first time I really enjoyed this cd again.

I dont know for sure what happend with the sound, and maybe my system simply is the right combination with this mod and my ears. but something did happen... and I like the change
Click to expand...

Hi,

Just bought a audio gd clock to improve sound on DM. Can you tell me if I connect it to power out pin of dexa 7815 regulator for cleanest power?

cheers
 
#409
mutil-Opamps trial

Have just finished trialing a number of opamps on the XLR section of the output stage. Bare in mind I have removed the RCA section (OpAmps/Caps) from my setup.

So far and in addition to LM49720 I tried:

OPA2211 - running cool, no oscillations, after 2wks the sound was muddy, with no clarity and distorted highs.

OPA1612 - running cool, no oscillations, sound after 2wks was details but stage feels backwards and sound lucks clarity on busy passages with highs getting distorted.

ISL28118 -no oscillations, running cool, interesting low/mid end giving some nice detail however stage compressed and distorted highs.

MC33078 - this OpAmp used in the application example inside the Wolfson DAC datasheet. Didn't listen as I measured oscillations on its output; negative signal phase was oscillating but not the positive one.

AD797 -burning hot :flame:, same stage and instrument definition as LM49720, forward sound, details are easier to listen and sound is less cluttered on busy passages than any of the above opamps.


So AD797 stays for now, and will let it burn.... literally !! is it normal to be that hot, I mean you cannot touch for more than 5''. Would this do anything ?
 

Attachments

  • IMG_20120203_234243.jpg
    IMG_20120203_234243.jpg
    693.4 KB · Views: 637
Last edited:
#411
The AD797 runs hot on U55/U56, U58/59 position however the highs are very detailed, and music sounds much better than other opamps tried, especially complex passages in classical & rock recordings.

Also running a freq sweep from 10Hz-20Khz didn't point any oscillations when measured with my oscilloscope. I recall LM49720 was also hot but cannot recall a "burned finger".

Should I infer the AD797 is stable on this circuit topology besides running hot ?

Re: bypassing, although datasheet recommends bypassing with 0.1uf + 4.7uf, would there be an issue with only the 47nf ECHU bypass cap already in place ?
 
Last edited:
#412
bicefalo said:
Hi Evan,

try UNREG 2 should be better option. Tell us the improvement, if any...

ciao
Click to expand...

Hi, i am still waiting on the audio gd clock to be delivered from china! But what do you mean by Unreg 2 is that 7815 reg? I have in the mean time put nichihon es muse series in all bipolar positions, only cost 11 euro. I think this was positive for sound but subtle. Better bass and vocals and slightly better clarity. Hope the clock has some effect?

regards
 
#413
gvelim said:
The AD797 runs hot on U55/U56, U58/59 position however the highs are very detailed, and music sounds much better than other opamps tried, especially complex passages in classical & rock recordings.

Also running a freq sweep from 10Hz-20Khz didn't point any oscillations when measured with my oscilloscope. I recall LM49720 was also hot but cannot recall a "burned finger".

Should I infer the AD797 is stable on this circuit topology besides running hot ?

Re: bypassing, although datasheet recommends bypassing with 0.1uf + 4.7uf, would there be an issue with only the 47nf ECHU bypass cap already in place ?
Click to expand...

Have you tried 9v power supply instead of 12 v? I bought one of ebay 9v 2.6A and runs great with dacmagic , alot cooler than before.

regards
 
#414
9V AC adaptor doesn't work for me as it causes voltage drop which in turn makes my SRegs not function properly. I have ordered some clip-on TO220 heat sinks to try out. Also I'll get the bypaasing fixed as per datasheet + the feedback cap at pin 7 + 6
 
#416
gvelim said:
UNREG2 is the output of R107.

Feeding from R5's output is a much cleaner option with less noise as long as the voltage is good for your purpose. I am feeding from R5 the DACs' SRegulator

Nice idea, I will try it.

About AD 797 : Check their power absorption across R400, if any oscillation is suspected.

Like a guide line, sum the quiescent current ( from data sheet ) and see what you have with no signal ( 50% more or less should be ok ), then try with signal to see if you get an increased value across R400.

ciao.
Click to expand...
 
#417
gvelim said:
UNREG2 is the output of R107.

Feeding from R5's output is a much cleaner option with less noise as long as the voltage is good for your purpose. I am feeding from R5 the DACs' SRegulator so that way I avoid overloading the analog +SReg (7815) and improving a bit on heat dissipation.
Click to expand...

thanks , unfortunately too late reading your message as my tech installed the audio g clock, used the dexa 7815 output for power and silver cable i bought cheap of ebay. First impressions are good, not night and day but a bit more detail especially on 16bit/44.1 cd rips. sound is clean. Since i changed to 9v adaptor not as much heat as before.

cheers
 
#418
bicefalo said:
About AD 797 : Check their power absorption across R400, if any oscillation is suspected.
Click to expand...

With my R400 at 68ohm I am getting 5.86Vdc which gives me around 86mA for the 4 AD797 + 2 OPA2120 + SReg current. So assuming 16mA go to the 2 OPA + SReg then it will be 17mA for each AD797.

17mA translates to 0.5W which calculates as 86C on the packaging, which resembles reality. The datasheet reads -40 to 80C as normal operating temperature, so what should I infer ?

Since the browndog adaptors are being toasted from both sides, I have put a clip-on heat sink per channel for regulating the temperature.

Otherwise music is sweet, full in midrange and better than before, even when the LM49720s were on.
 
Last edited:
#419
AD797 is a bad beast, looks like yours are oscillating.....
IC adaptors are very bad for stability with hight speed op-amps, no heat transfer to large cooper lines too...
The safe option is to put LM49720s back. In my opinion a couple of silmic II ( 100uf/25v ), at the output of Yours regulators, will give You sweeter HF, creamy mids and strong sub bass, like I have in my dacmagic.

ciao
 
#420
I agree browndogs could be the reason for the oscillation so I am planning to fix 2 x 1206 film caps (0.1uf + 0.47uf) per rail on the adaptor to see whether this fixes the problem.

re: Silmic II, I already have super regs installed so would they still make any difference ? also where do you buy them from ?
 
You must log in or register to reply here.
Top Bottom