DAC-Burr Brown PCM1792

Options
Systems
Systems Posts: 14,873
edited April 2014 in DIY, Mods & Tweaks
After having this board collecting dust for the last year or so I thought it was time to put it to good use.I already have a very capable DAC so I was in no hurry to get it working.It is evaluation board that is essentially a full functional DAC but without the chassis,power supply and voltage regulators.

This board was designed and built by Burr Brown as a platform to evaluate the performance of the PCM1792 and offered to companies interested in using this chip in their CD/DVD players.The PCM1792 is highly regarded and used in a number of hi-end CD/DVD/SACD players.Its a 24 bit /192k Dual Differential device meaning both channels are on one chip and the outputs are balanced.

The eval board is nicely done with lots of ground plane and some high quality parts such as ELNA Cerifine electrolytic decoupling caps and some nice film caps in the low pass filter section.The input reciever chip is the Crystal CS8414 but I would have preferred they used the CS8420 for its upsampling and jitter reduction capabilities.Because the DAC chip has current outputs an op amp is needed to convert the current to voltage.Strangely for this I/V conversion BB used an op amp from a competitor the very old NE 5534. They must think its good if they used it instead of one of their own.

Thankfully all the the op amps are socketed so it will be a simple matter to try different ones.After I get acclimatized to the sound with the 5534's I will try some others.Tops on the list will be BB's own OPA627.(Their kinda pricy at $18ea and I'll need 6.:eek:)

I will be using completely separate supplies for the analog and digital section each with their own toroidal transformer.The regulators feeding the analog section will be a low noise low output impedance discrete design insted of the cheap and noisy 3 IC pin regulators.It is shown in pic #5

Pic #1-3 shows the DAC board, and #4 is the transformer and the 5 volt IC regulator
for the digital section.
Testing
Testing
Testing
Post edited by Unknown User on
«1345

Comments

  • Systems
    Systems Posts: 14,873
    edited September 2008
    Options
    And in the off chance that anyone is interested in such things here is a link to the Users guide/schematics for the board. http://focus.tij.co.jp/jp/lit/ug/sleu037/sleu037.pdf
    Testing
    Testing
    Testing
  • ben62670
    ben62670 Posts: 15,969
    edited September 2008
    Options
    I want one:) How do you hook up SACD to it? The dynamic range is off the charts!
    Please. Please contact me a ben62670 @ yahoo.com. Make sure to include who you are, and you are from Polk so I don't delete your email. Also I am now physically unable to work on any projects. If you need help let these guys know. There are many people who will help if you let them know where you are.
    Thanks
    Ben
  • Systems
    Systems Posts: 14,873
    edited September 2008
    Options
    ben62670 wrote: »
    I want one:)
    No promises but maybe I can help you find one.;)
    How do you hook up SACD to it?
    Unfortunately thats not a simple task because DSD can not be transmitted via the standard coax SPDIF.While the 1792 can convert an SACD/DSD data stream ,getting that data would require surgery on the transport.You would have to run a cable from the DSD processor chip in the player to the input of the 1792 ,so it's likely not worth the effort.:(


    Yes the wide dynamic range/low noise floor makes the use of low noise regulators and op amps a must.
    Testing
    Testing
    Testing
  • ben62670
    ben62670 Posts: 15,969
    edited September 2008
    Options
    I would have no problem sacrificing a cheap SACD transport:) I wouldn't have a problem selling my GDA-700 to finance the project either:) Is the analog servo, or cap coupled?
    Thanks
    Ben

    Edit:
    There are three analog signal output functions onboard as 2-V rms single-ended (RCA),
    4.5-V rms single-ended (RCA), and 9-V rms balanced (XLR).
    This is screaming for the pasive section of my three way pre:)
    I checked OnSemi, and no eval boards:(, but the PCM1792 chip are only $16. Without the board it would be kinda useless to me:( HDCD?
    Please. Please contact me a ben62670 @ yahoo.com. Make sure to include who you are, and you are from Polk so I don't delete your email. Also I am now physically unable to work on any projects. If you need help let these guys know. There are many people who will help if you let them know where you are.
    Thanks
    Ben
  • Systems
    Systems Posts: 14,873
    edited September 2008
    Options
    ben62670 wrote: »
    I would have no problem sacrificing a cheap SACD transport:)
    Well now that might be worth doing if you snag a cheap Sony or Pioneer transport.I will dig out the service manual for my Pioneer and see whats involved.;)
    I wouldn't have a problem selling my GDA-700 to finance the project either:)
    This is much like your 700 but a few generations newer on the digital front.
    Is the analog servo, or cap coupled?
    Neither its direct coupled.:)this is made possible because the dc offset from the 1792 is extremely low due to its having differential outputs.

    I checked OnSemi, and no eval boards:
    It's a TI/BB product not On Semi.The board sells for several hundred $,but I paid much less than that.;)
    but the PCM1792 chip are only $16. Without the board it would be kinda useless to me:
    You don't want to play that game with digital chips,you need a proper board.If you are serious I will try and get you a complete board.:)
    HDCD?
    No the 1792 is not HDCD compatible.
    Testing
    Testing
    Testing
  • fredv
    fredv Posts: 923
    edited September 2008
    Options
    GV#27 wrote: »
    I will try and get you a complete board.:)No the 1792 is not HDCD compatible.
    Any lead for me to get a board? I should be getting a few opa62, but I'll have to see them before I can say for sure. If so, I have enough to cover 1+ boards. I only need 6, right ;)

    -fredv-
  • ben62670
    ben62670 Posts: 15,969
    edited September 2008
    Options
    Ya I'm serious. Let me know, and I'll put my DAC up ASAP. Like I said it would love to share the same case as my pre:) As for the SACD transport would you pull the serial data directly from the laser? Any interesting reads on transports?
    Please. Please contact me a ben62670 @ yahoo.com. Make sure to include who you are, and you are from Polk so I don't delete your email. Also I am now physically unable to work on any projects. If you need help let these guys know. There are many people who will help if you let them know where you are.
    Thanks
    Ben
  • Systems
    Systems Posts: 14,873
    edited September 2008
    Options
    fredv wrote: »
    Any lead for me to get a board?
    I only have a lead on 1,but I will see if I can locate another but honestly I doudt I will be able to.
    If so, I have enough to cover 1+ boards. I only need 6, right ;)
    Yes 6.The pics show 8 but you will only use one of the output sections per channel so the extra 2 are not needed.See schematic on page 21
    http://focus.tij.co.jp/jp/lit/ug/sleu037/sleu037.pdf
    Testing
    Testing
    Testing
  • ben62670
    ben62670 Posts: 15,969
    edited September 2008
    Options
    I am giggly about the analog signal path. There is nothing full in the signal path but 3 R per channel at 4.5v, and 5r at 2v :) No need to spend big money on the analog:)

    Edit: I might spend a couple $ on the 22pf (subsonic filter?) caps.

    Edit Edit: It's kinda weird that they spec OpAmps 1-4 at 5534, 5-6 5534 or 134, and no spec on 7-8.
    Please. Please contact me a ben62670 @ yahoo.com. Make sure to include who you are, and you are from Polk so I don't delete your email. Also I am now physically unable to work on any projects. If you need help let these guys know. There are many people who will help if you let them know where you are.
    Thanks
    Ben
  • Systems
    Systems Posts: 14,873
    edited September 2008
    Options
    ben62670 wrote: »
    Let me know.
    I will.
    As for the SACD transport would you pull the serial data directly from the laser?
    No,the data must be taken from the SACD processor IC.For instance my Pioneer uses the Phillips SSA7893HL chip.From it you need to get the clock and the left and right channel data from the apropriate pins.
    Any interesting reads on transports?
    You mean for info about doing mods?
    Testing
    Testing
    Testing
  • Systems
    Systems Posts: 14,873
    edited September 2008
    Options
    ben62670 wrote: »
    It's kinda funny how they call it a balanced out. It is one XLR with the normal unbalanced left, and right signals?
    Yes thats just for testing its not a real balanced signal.However you can derive a proper balanced out if you use the extra op-amps for the inverted half of the signal.It would require a slight mod.
    I am giggly about the analog signal path. There is nothing full in the signal path but 3 R per channel at 4.5v, and 5r at 2v :) No need to spend big money on the analog:)
    Yep fairly simple signal path which is a good thing.
    Testing
    Testing
    Testing
  • ben62670
    ben62670 Posts: 15,969
    edited September 2008
    Options
    You mean for info about doing mods?
    Info on where to pull data from(which you just basically outlined), and The process of processing.
    However you can derive a proper balanced out if you use the extra op-amps for the inverted half of the signal.It would require a slight mod.
    I don't really care about balanced outs. My signal run will be less than a foot. If I can get this board it will reside in my pre:)
    Please. Please contact me a ben62670 @ yahoo.com. Make sure to include who you are, and you are from Polk so I don't delete your email. Also I am now physically unable to work on any projects. If you need help let these guys know. There are many people who will help if you let them know where you are.
    Thanks
    Ben
  • Systems
    Systems Posts: 14,873
    edited September 2008
    Options
    ben62670 wrote: »
    I might spend a couple $ on the 22pf (subsonic filter?) caps.
    Those are compensation caps.They are required when using the 5534 as it is not unity gain stable.The OPA134 would not need them.
    It's kinda weird that they spec OpAmps 1-4 at 5534, 5-6 5534 or 134, and no spec on 7-8.
    7&8 have the same requirements as 5&6.
    ben62670 wrote: »
    Info on where to pull data from(which you just basically outlined), and The process of processing.
    Don't know of any good reads on this' but would be interested even though I don't plan to run DSD to mine.If you happen to find something let me know.
    I don't really care about balanced outs.
    Nor do I, but I will include them even though I will likely never use them.
    If I can get this board it will reside in my pre:)
    Ahhh too bad, its much too BIG to fit.:p


    Actual size is 8"x8.75"
    Testing
    Testing
    Testing
  • ben62670
    ben62670 Posts: 15,969
    edited September 2008
    Options
    GV#27 wrote: »
    Ahhh too bad, its much too BIG to fit.:p


    Actual size is 8"x8.75"
    I kid you not. That is exactly what I have! I'll post a pic later. If it doesn't fit I will order a new case. I send you a link to a supplier I found;)
    Please. Please contact me a ben62670 @ yahoo.com. Make sure to include who you are, and you are from Polk so I don't delete your email. Also I am now physically unable to work on any projects. If you need help let these guys know. There are many people who will help if you let them know where you are.
    Thanks
    Ben
  • Systems
    Systems Posts: 14,873
    edited September 2008
    Options
    ben62670 wrote: »
    I kid you not. That is exactly what I have!
    Cool:)

    That measuement is the board itself.The connectors stick out beyond that a bit.I also removed a testing port and will remove the rest of the connectors as I will use chassis mounted ones instead.

    Im stoked about getting this working as it is a nice piece of digital engineering. I just need to order a little 2x18 volt toroid from digi key and I'll have all the parts.
    Testing
    Testing
    Testing
  • Systems
    Systems Posts: 14,873
    edited October 2008
    Options
    I have made some progress and have the chassis almost complete.I'm undecided whether to put the power switch on the front panel or rear panel.I plan to paint the front panel black ,somehow the grey just doesn't do it for me.
    The pics only shows the transformer for the digital section there will be another for analog power supply.Also not installed yet is the true 75 ohm BNC connector for the digital in. Using a 75 ohm connectors and cables is the best way to transmit via the SPDIF interface as there is less reflections in the cable and therefore a reduction in jitter.
    Testing
    Testing
    Testing
  • Systems
    Systems Posts: 14,873
    edited October 2008
    Options
    Some pics.
    Testing
    Testing
    Testing
  • ShinAce
    ShinAce Posts: 1,194
    edited October 2008
    Options
    GV#27 wrote: »
    I have made some progress and have the chassis almost complete.I'm undecided whether to put the power switch on the front panel or rear panel.I plan to paint the front panel black ,somehow the grey just doesn't do it for me.
    The pics only shows the transformer for the digital section there will be another for analog power supply.Also not installed yet is the true 75 ohm BNC connector for the digital in. Using a 75 ohm connectors and cables is the best way to transmit via the SPDIF interface as there is less reflections in the cable and therefore a reduction in jitter.

    *sigh*

    I used to work at an electronics shop here in Ottawa that has since been violated by a new owner...no names.

    We used to have amphenol and another well known brand of true BNC's. We also had a ton of bnc terminated precision video cable(beefed up RG6 before RG6 was around) lying around. It still wasn't the cheapest, but at 2$ a connector versus 10$ and then 1$/foot for the wire, we were laughing.
  • Systems
    Systems Posts: 14,873
    edited October 2008
    Options
    ben62670 wrote: »
    it would love to share the same case as my pre:)
    Do u still have room in that chassis?;):D
    Testing
    Testing
    Testing
  • ben62670
    ben62670 Posts: 15,969
    edited October 2008
    Options
    Sure do!
    Please. Please contact me a ben62670 @ yahoo.com. Make sure to include who you are, and you are from Polk so I don't delete your email. Also I am now physically unable to work on any projects. If you need help let these guys know. There are many people who will help if you let them know where you are.
    Thanks
    Ben
  • ben62670
    ben62670 Posts: 15,969
    edited October 2008
    Options
    Well I have been doing a little reading, and...
    http://www.diyaudio.com/forums/showthread.php?s=&threadid=120464&perpage=25&highlight=&pagenumber=1
    I think I will go with LM4562
    It is a relatively cheap OpAmp, and should be a step up from the BB OPA-2134. From the reviews it sounds like what is needed here. The OPA627 is about 3 times more expensive with reviews only being ever so slightly favorable.
    I have also been doing some research on ganking the data from an SACD transport unit that I may rip out of an SACD player to incorporate into my Pre. That would make it a Pre(active, tube, passive)/active XO/Phono pre/SACD Player in one unit. There are a few SACD players that convert SACD to PCM within the unit before converting it to DSM before it leaves the unit. This may be easier, but I would rather pull the full data from the SACD. We'll see.
    Ben
    Please. Please contact me a ben62670 @ yahoo.com. Make sure to include who you are, and you are from Polk so I don't delete your email. Also I am now physically unable to work on any projects. If you need help let these guys know. There are many people who will help if you let them know where you are.
    Thanks
    Ben
  • PolkClyde
    PolkClyde Posts: 662
    edited October 2008
    Options
    I have a Audio Alchemy HDCD DAC and a jitter reduction amp DTIpro. The sound is unbelievable.
    PolkAudioClyde
  • Systems
    Systems Posts: 14,873
    edited October 2008
    Options
    ben62670 wrote: »
    I think I will go with LM4562
    The LM4562 is a good chip but it can't be used in this DAC as it's a dual,you need singles.Im not aware of the availability of a single version of the 4562.The closest I've seen is Nationals LME49870.http://www.national.com/pf/LM/LME49870.html Im considering giving it a try.
    I would rather pull the full data from the SACD.
    Yes I think preserving the DSD data is the best route.
    Testing
    Testing
    Testing
  • Early B.
    Early B. Posts: 7,900
    edited October 2008
    Options
    GV#27 wrote: »
    The LM4562 is a good chip but it can't be used in this DAC as it's a dual,you need singles.Im not aware of the availability of a single version of the 4562.The closest I've seen is Nationals LME49870.http://www.national.com/pf/LM/LME49870.html Im considering giving it a try.

    Hey Ben and GV#27 --

    I have 2134's in my CDP. Should I consider an upgrade to the LME49870?

    Here's an old pic of the board (before the Sonicaps were installed):
    HT/2-channel Rig: Sony 50” LCD TV; Toshiba HD-A2 DVD player; Emotiva LMC-1 pre/pro; Rogue Audio M-120 monoblocks (modded); Placette RVC; Emotiva LPA-1 amp; Bada HD-22 tube CDP (modded); VMPS Tower II SE (fronts); DIY Clearwave Dynamic 4CC (center); Wharfedale Opus Tri-Surrounds (rear); and VMPS 215 sub

    "God grooves with tubes."
  • Systems
    Systems Posts: 14,873
    edited October 2008
    Options
    Early B. wrote: »

    I have 2134's in my CDP. Should I consider an upgrade to the LME49870?
    If it's a OPA 2134 not an OPA134 then it's a dual so the LM4562 will work but the LME49870 won't.I don't if its worth the trouble unless the 2134 is socketed allowing easy removal.The 2134 is a decent op amp.



    Edit.I see in the pic that the op amps are soldered in place so removal of the circuit board would be required.
    Testing
    Testing
    Testing
  • Early B.
    Early B. Posts: 7,900
    edited October 2008
    Options
    GV#27 wrote: »
    The 2134 is a dual so the LM4562 will work but the LME49870 won't.I don't if its worth the trouble unless the 2134 is socketed allowing easy removal.The 2134 is a decent op amp.

    Oh, it's a dual? OK. Nope, the 2134 is not socketed. If I decided to "upgrade," I'd ask Ben to do the soldering.

    GV -- Have you checked out the Burson Audio discrete opamps?

    http://www.partsconnexion.com/Index/burson.php

    They're expensive. Based on the pic, I'm assuming I'd need two of them? What do ya think?
    HT/2-channel Rig: Sony 50” LCD TV; Toshiba HD-A2 DVD player; Emotiva LMC-1 pre/pro; Rogue Audio M-120 monoblocks (modded); Placette RVC; Emotiva LPA-1 amp; Bada HD-22 tube CDP (modded); VMPS Tower II SE (fronts); DIY Clearwave Dynamic 4CC (center); Wharfedale Opus Tri-Surrounds (rear); and VMPS 215 sub

    "God grooves with tubes."
  • Systems
    Systems Posts: 14,873
    edited October 2008
    Options
    Early B. wrote: »

    GV -- Have you checked out the Burson Audio discrete opamps?
    You bet ya I have and it looks good on paper atleast.I use something very similar in one of my modded Pioneer DVD/SACD players.
    see here http://www.polkaudio.com/forums/showthread.php?t=55733&highlight=pioneer+dv563

    With the discrete OA I used,I found that it sounded significantly better than an identical unit moodded with an LM4562.


    As you say the Burson is pricey and you would need one of the duals to replace a 2134.
    Testing
    Testing
    Testing
  • Systems
    Systems Posts: 14,873
    edited October 2008
    Options
    Early B. wrote: »
    Oh, it's a dual?
    Dual meaning with the OPA 2134 there are 2 complete op amps contained within the one 8 pin chip,where as the OPA134 has a single opamp in the 8 pin chip.
    Testing
    Testing
    Testing
  • ben62670
    ben62670 Posts: 15,969
    edited October 2008
    Options
    Early B. wrote: »
    Oh, it's a dual? OK. Nope, the 2134 is not socketed. If I decided to "upgrade," I'd ask Ben to do the soldering.

    GV -- Have you checked out the Burson Audio discrete opamps?

    http://www.partsconnexion.com/Index/burson.php

    They're expensive. Based on the pic, I'm assuming I'd need two of them? What do ya think?

    I got ya bro. Bring over some pain killers, an ice pack, and I will get right on it;) I just found my gas soldering iron:eek::D
    Please. Please contact me a ben62670 @ yahoo.com. Make sure to include who you are, and you are from Polk so I don't delete your email. Also I am now physically unable to work on any projects. If you need help let these guys know. There are many people who will help if you let them know where you are.
    Thanks
    Ben
  • Systems
    Systems Posts: 14,873
    edited October 2008
    Options
    ben62670 wrote: »
    I just found my gas soldering iron:eek::D
    I have heard that these are known to cause minor cosmetic damage if not used with care.;):D
    Testing
    Testing
    Testing