stacking DACs
Hi,
Could anyone comment on the benefits of stacking three AD1955 in the 963sa? I contacted a local engineering firm to see if they could do the soldering for me (those chips are much smaller than I imagined) and swap out my 1895 for the new 1896 I received, and the electrical engineer who heads the company laughed and told me that these changes could not possibly make any improvement to the sound fo the player, and asked if i had read foolish ideas like this in Absolute Sound. Obviously he is a hopeless skeptic, since he likened these mod's to people who can hear differences in cables!, but, nonetheless, I would appreciate a discussion on what should be the benefit of stacking the DACs and why. I understand the signal should amplify and the noise decrease due to cancellation of the random noise and addition of the constant signal, but a little more in depth explanation would be appreciated.
Thanks
Chris
Hi,
Could anyone comment on the benefits of stacking three AD1955 in the 963sa? I contacted a local engineering firm to see if they could do the soldering for me (those chips are much smaller than I imagined) and swap out my 1895 for the new 1896 I received, and the electrical engineer who heads the company laughed and told me that these changes could not possibly make any improvement to the sound fo the player, and asked if i had read foolish ideas like this in Absolute Sound. Obviously he is a hopeless skeptic, since he likened these mod's to people who can hear differences in cables!, but, nonetheless, I would appreciate a discussion on what should be the benefit of stacking the DACs and why. I understand the signal should amplify and the noise decrease due to cancellation of the random noise and addition of the constant signal, but a little more in depth explanation would be appreciated.
Thanks
Chris
the earth is flat !
There is another engineer who doesn't believe in things that were not in the few books he has read. Science should teach people to have open minds.
There is another engineer who doesn't believe in things that were not in the few books he has read. Science should teach people to have open minds.
output with 3 DACs
Thanks Jaap for your humorous response. I especially like the Santa.
Something else occurred to me, thinking about stacking the three 1955's. If each additional DAC adds 3db of output (I think this is what I read on Audio Asylum or somewhere), is it necessary to make any modification to the output stage when you have the 3 DACs piggybacked, or can you just stack the DACs and leave the rest of the player as is, and the signal will be "louder", i.e., I'll have to turn down (a little) my line amp volume control? I'm sure changing the output would further improve things, but I am just concerned with damaging the player if I stack the DACs and don't do anything else (for now, anyways).
Thanks for any input
Chris
Thanks Jaap for your humorous response. I especially like the Santa.
Something else occurred to me, thinking about stacking the three 1955's. If each additional DAC adds 3db of output (I think this is what I read on Audio Asylum or somewhere), is it necessary to make any modification to the output stage when you have the 3 DACs piggybacked, or can you just stack the DACs and leave the rest of the player as is, and the signal will be "louder", i.e., I'll have to turn down (a little) my line amp volume control? I'm sure changing the output would further improve things, but I am just concerned with damaging the player if I stack the DACs and don't do anything else (for now, anyways).
Thanks for any input
Chris
stacking of DAC's
I've also an additional 1955 dac, but didn't have the time to solder it. Tricky business!!
James (www.effectiveaudiomod.com) has experience with the stacking of the DAC's and can clearify the technical issues on this subject.
I asked him the same question what would be the effect of the stacking, answer: more drive and more inner detail of the sound.
I also ordered the next batch of black gates to experiment with super e-cap configuration on the SPU!!
Saffier
I've also an additional 1955 dac, but didn't have the time to solder it. Tricky business!!
James (www.effectiveaudiomod.com) has experience with the stacking of the DAC's and can clearify the technical issues on this subject.
I asked him the same question what would be the effect of the stacking, answer: more drive and more inner detail of the sound.
I also ordered the next batch of black gates to experiment with super e-cap configuration on the SPU!!
Saffier
After reading the previous stuff and after doing all kind of mods in my 963SA I decided to order 2 free AD1955 samples from Analog Devices and piggyback the original DAC. Soldering these things is indeed tricky. Too tricky for me as a fact
The result after hours was: only SACD playback and a very loud noise in the right front channel. No CD/DVD sound any more. After that i peeled off the 3rd again and after that the second chip too. Now the original DAC is alone again. SACD sound is back again. CD/DVD sound is lost (forever?).
Probably the chip became too hot.
Decided to order two more at Digikey (12.50 dollar each +..).
Conclusion: After all kinds of mods (BG's, mute relays, AD8066, analog psu etc.) this was a bridge too far for me.
You have been warned.
Rob
The result after hours was: only SACD playback and a very loud noise in the right front channel. No CD/DVD sound any more. After that i peeled off the 3rd again and after that the second chip too. Now the original DAC is alone again. SACD sound is back again. CD/DVD sound is lost (forever?).
Probably the chip became too hot.
Decided to order two more at Digikey (12.50 dollar each +..).
Conclusion: After all kinds of mods (BG's, mute relays, AD8066, analog psu etc.) this was a bridge too far for me.
You have been warned.
Rob
SMD soldering technique
Nice link for SMD soldering technique....
http://www.infidigm.net/articles/solder/
Btw, Rob, I'm sorry to hear about your AD1955 stacking, again proof for me that this is very hard to do and I'm still reluctant to get the job done.....
Saffier
Nice link for SMD soldering technique....
http://www.infidigm.net/articles/solder/
Btw, Rob, I'm sorry to hear about your AD1955 stacking, again proof for me that this is very hard to do and I'm still reluctant to get the job done.....
Saffier
soldering SMT parts
Hi everyone,
I came across the following link for building a hot air pencil for soldering surface mount ICs. Maybe this will be of use to some of you who are contemplating changing some of the small chips on the 963sa. I'm still waiting for a quote from a local electronics company to swap my upsampling chip and piggyback my 1955's. I'm not brave enough to try to do this myself.
Anyway, here's the link:
http://www.usbmicro.com/odn/documents/46.html
Cheers
Chris
Hi everyone,
I came across the following link for building a hot air pencil for soldering surface mount ICs. Maybe this will be of use to some of you who are contemplating changing some of the small chips on the 963sa. I'm still waiting for a quote from a local electronics company to swap my upsampling chip and piggyback my 1955's. I'm not brave enough to try to do this myself.
Anyway, here's the link:
http://www.usbmicro.com/odn/documents/46.html
Cheers
Chris
Saffier and Chris, thanks for your advice. I think I will try making the desolder iron in the next weeks.
By the way: suddenly (while playing) the CD sound came back again! Something to do with Pentacoste?
By the way: suddenly (while playing) the CD sound came back again! Something to do with Pentacoste?
stacking DACS
A (lengthy) question to all those DAC stackers out there:
I'm very astonished that nobody mentions any adaptations to the pull up resistors of the U/I section when stacking DACs. Following the datasheet of the 1955 there is a bias current of -3.24 mA at the DAC's output. To bring the DC level of the U/I op-amp output to zero in order to gain maximum headroom this bias current plus the 2.4 V bias voltage of the U/I stage should be compensated by the pull-ups.
Suprisingly the original pull-up resistors of 1.8 kOhm (e.g. resistor 3438) compensate for the 2.4 V only (is there an argument for that or did the Philips people just not recognize the bias current?!?). In deed if you measure the DC output level of opamp 7403 or 7404 in the original state you'll find 2.1 V which corresponds to the bias current of 3.24 times the feedback resistor of 650 Ohm (e.g. resitors 3403+3405).
Now we start stacking DACs. In my interpretation then all output currents add up and thus are doubled (or trippled when you even put 2 additional DACs). The 1.8 kOhm still compensate for the 2.4V but now I'd expect double (tripple) bias current leading to 4.2 V (6.3 V) DC at the output of U/I. The whole section is fed by +/- 8V only leaving a (in my view) very poor headroom to the positive supply rail. In tripple DAC config I'd even expect some clipping whithout further modifications since the output voltage swing is also trippled.
Can anybody confirm my interpretation or am I completely wrong? Do I miss anything?
If that is correct, the pull-ups should be 550 Ohm in double respectively 417 Ohm in tripple config to bring the ouput to zero.
Best regards and happy listening
Rainer
😕
A (lengthy) question to all those DAC stackers out there:
I'm very astonished that nobody mentions any adaptations to the pull up resistors of the U/I section when stacking DACs. Following the datasheet of the 1955 there is a bias current of -3.24 mA at the DAC's output. To bring the DC level of the U/I op-amp output to zero in order to gain maximum headroom this bias current plus the 2.4 V bias voltage of the U/I stage should be compensated by the pull-ups.
Suprisingly the original pull-up resistors of 1.8 kOhm (e.g. resistor 3438) compensate for the 2.4 V only (is there an argument for that or did the Philips people just not recognize the bias current?!?). In deed if you measure the DC output level of opamp 7403 or 7404 in the original state you'll find 2.1 V which corresponds to the bias current of 3.24 times the feedback resistor of 650 Ohm (e.g. resitors 3403+3405).
Now we start stacking DACs. In my interpretation then all output currents add up and thus are doubled (or trippled when you even put 2 additional DACs). The 1.8 kOhm still compensate for the 2.4V but now I'd expect double (tripple) bias current leading to 4.2 V (6.3 V) DC at the output of U/I. The whole section is fed by +/- 8V only leaving a (in my view) very poor headroom to the positive supply rail. In tripple DAC config I'd even expect some clipping whithout further modifications since the output voltage swing is also trippled.
Can anybody confirm my interpretation or am I completely wrong? Do I miss anything?
If that is correct, the pull-ups should be 550 Ohm in double respectively 417 Ohm in tripple config to bring the ouput to zero.
Best regards and happy listening
Rainer
😕
stacking DACS
Is Guunther correct? Do I need to worry about this if I plan to stack my DACs? Has anyone who has stacked the 1955's made such modifications, or did you just stack the DACs and leave the pull up resistors alone? I'd really appreciate hearing from someone who has stacked to know whether it is necessary to do further mod's for the stacking to work correctly. Sorry; my background is science, but not electrical engineering (I just love music and would like the best from my 963sa), so I have to rely on all of you to point me in the right direction with this.
Thanks
Chris
Is Guunther correct? Do I need to worry about this if I plan to stack my DACs? Has anyone who has stacked the 1955's made such modifications, or did you just stack the DACs and leave the pull up resistors alone? I'd really appreciate hearing from someone who has stacked to know whether it is necessary to do further mod's for the stacking to work correctly. Sorry; my background is science, but not electrical engineering (I just love music and would like the best from my 963sa), so I have to rely on all of you to point me in the right direction with this.
Thanks
Chris
Hi,
when you refer to the datasheet page 13 you will get more information. The pullup resistor has to bring the output of the I/V converter to 0V. So what happens if you have more current because more than one dac are used? But anyway you will not get more voltage - the supply voltage is constant, the reference voltage must be constant too - the only thing is that you have to deliver more bias current and more signal current.
Michael
when you refer to the datasheet page 13 you will get more information. The pullup resistor has to bring the output of the I/V converter to 0V. So what happens if you have more current because more than one dac are used? But anyway you will not get more voltage - the supply voltage is constant, the reference voltage must be constant too - the only thing is that you have to deliver more bias current and more signal current.
Michael
Attachments
Hi Mchael,
yes right, as you said: you have to deliver more bias current. However the bias current has to be delivered by the pull-ups. And the pull-ups don't if you do not change them.
Thus if it's correct that with stacking DACs the bias current is doubled the formula on page 13 for the pull-ups changes too since - as you stated correctly- all other values in this formula stay the same.
Is there somebody who stacked the DAC who could measure the DC level at the output of U/I at either of testpoints I411, I417, I418 or I424?
Regards
Rainer
yes right, as you said: you have to deliver more bias current. However the bias current has to be delivered by the pull-ups. And the pull-ups don't if you do not change them.
Thus if it's correct that with stacking DACs the bias current is doubled the formula on page 13 for the pull-ups changes too since - as you stated correctly- all other values in this formula stay the same.
Is there somebody who stacked the DAC who could measure the DC level at the output of U/I at either of testpoints I411, I417, I418 or I424?
Regards
Rainer
Hi,
Has anybody tried the power supply or clock upgrade from TentLabs? http://www.tentlabs.com/
I am interested in the clock upgrade in particular and I was wondering if anybody has succesfully installed it. In general it seems to get rave reviews, but I have not seen where someone has tried this in the 963.
Has anybody tried the power supply or clock upgrade from TentLabs? http://www.tentlabs.com/
I am interested in the clock upgrade in particular and I was wondering if anybody has succesfully installed it. In general it seems to get rave reviews, but I have not seen where someone has tried this in the 963.
@bearbearloo
nice little picture - but what shows it? A double XO with internal power supply? Interesting kind of mounting the pcb with the XO- is it a "lord of the rings" mounting?😀
nice little picture - but what shows it? A double XO with internal power supply? Interesting kind of mounting the pcb with the XO- is it a "lord of the rings" mounting?😀
Re: Re: Clock Upgrade
Woops... I see. I missed the picture. I guess I wasn't logged in...Hans L said:
Tentlabs XO
Hi Bearbearloo,
Can you report on the sound of the Tentlabs XO in your 963SA. What's the sonic benefit? Did you just implement seperate XO's or the dedicated 963SA solution where Guido is still working on, as to quote his own website.
I tried to contact him before through email, but never got any reply. I do not think he's serious about the 963SA upgrades anymore. It's probably to complicated with all PLL's and seperate clocks.
Regards,
Saffier
Hi Bearbearloo,
Can you report on the sound of the Tentlabs XO in your 963SA. What's the sonic benefit? Did you just implement seperate XO's or the dedicated 963SA solution where Guido is still working on, as to quote his own website.
I tried to contact him before through email, but never got any reply. I do not think he's serious about the 963SA upgrades anymore. It's probably to complicated with all PLL's and seperate clocks.
Regards,
Saffier
Message to Jon,
Please check email.
Message to mvollb
Another clock module (Neoclock) for your reference
http://www.nasotec.com/acecart/bin/shop.cgi?action=view&itemID=ns0201&cate=110100000
Message to saffier
- My XOs not from Tentlabs, these clocks are bought from Taiwan which the design is similar to Neoclock & with 2ppm TCXO.
- About the sonic outcome .... details, pin point, sound stage seems have been improved (not yet run in because I just mod it at Sunday)
- What I did at the past are
1 SPSU caps upgrade,
2 Output/Opamp bypass/DAC bypass caps upgrade,
3 24.x & 33.x XO
4 addtional RCA output Jack
Message to all,
Did anyone try to seperate the +/- 8v power supply? Any sharing?
Please check email.
Message to mvollb
Another clock module (Neoclock) for your reference
http://www.nasotec.com/acecart/bin/shop.cgi?action=view&itemID=ns0201&cate=110100000
Message to saffier
- My XOs not from Tentlabs, these clocks are bought from Taiwan which the design is similar to Neoclock & with 2ppm TCXO.
- About the sonic outcome .... details, pin point, sound stage seems have been improved (not yet run in because I just mod it at Sunday)
- What I did at the past are
1 SPSU caps upgrade,
2 Output/Opamp bypass/DAC bypass caps upgrade,
3 24.x & 33.x XO
4 addtional RCA output Jack
Message to all,
Did anyone try to seperate the +/- 8v power supply? Any sharing?
different types of blackgates
I'm a bit confused as to which types of blackgates to use where. Some people appear to have replaced most of the electrolytics with Blackgate STD's while other have used only Blackgate N's. Are these essentially interchangeable (aside from the lower noise and higher quality [and much higher price] of the N's)? In particular, I am interested in output caps # 2421 and 2429, since I accidentally ordered Blackgate STD for these and want to make sure I can use them at these locations without problems, i.e., does the output cap have to be nonpolar?
Thanks
Chris Porada
I'm a bit confused as to which types of blackgates to use where. Some people appear to have replaced most of the electrolytics with Blackgate STD's while other have used only Blackgate N's. Are these essentially interchangeable (aside from the lower noise and higher quality [and much higher price] of the N's)? In particular, I am interested in output caps # 2421 and 2429, since I accidentally ordered Blackgate STD for these and want to make sure I can use them at these locations without problems, i.e., does the output cap have to be nonpolar?
Thanks
Chris Porada