DACs

Discussion in 'Audio Hardware' started by three_paws, Jul 11, 2021.

  1. Ham Sandwich

    Ham Sandwich Senior Member

    Location:
    Sherwood, OR, USA
    I'm having fun. My Bifrost 2 now has bass after being left to warm-up overnight.
     
    h1pst3r88, three_paws and luckyno13 like this.
  2. three_paws

    three_paws deleted account Thread Starter

    Location:
    USA
    Good stuff. I'll check in later. Need to let my laptop warm up first so that I can connect to the internet.
     
    timind and NettleBed like this.
  3. luckyno13

    luckyno13 Forum Resident

    Location:
    London UK
    If the last few pages have shown anything, it's pointless to discuss whether the thing's better being on for -10 minutes or 5 years.

    The question is does it sound good to you and why it sounds good to you.
     
    h1pst3r88 and rockin_since_58 like this.
  4. three_paws

    three_paws deleted account Thread Starter

    Location:
    USA
    Totally agree that's the most important question. But wanting to understand the physics behind it, in whichever direction, is also reasonable. Don't understand the rabid mob mentality on the "warm up is real" side of the fence.
     
  5. chervokas

    chervokas Senior Member

    With respect to this particular chip, yes (tho there was a test of a different unit of the same item that showed opposite performance change with temp, so it may not be linear that there's improvement or degredation but change with temp). But the point is that there is an optimum operating temperature and changes in performance relating to temperature and those are things for a designer to consider when building a circuit. But temperature does impact performabce
     
  6. saturdayboy

    saturdayboy Forum Resident

    Location:
    Chicago
    A couple of posts ago you said that dacs clocks are designed to be ready to go right out of the box at room temperature, and now it’s there is a warm up period but it’s not that long. Stop pretending like you’re not backtracking.
     
  7. MaxBuck

    MaxBuck Forum Resident

    Location:
    La Quinta, CA
    If people want to warm up their equipment, burn in their cables and speakers, roll their tubes and op amps, that's no skin off my nose. Doesn't affect me in the least, and it apparently gives these people pleasure. Doesn't mean I have to believe it has any objective effect, but my belief shouldn't matter to anyone else.

    I think it's best to reserve strong opinions for stuff that matters. Like healthcare, politics and college football. :cool: Not that I've always followed that very well.
     
    daglesj, bever70 and h1pst3r88 like this.
  8. three_paws

    three_paws deleted account Thread Starter

    Location:
    USA
    Absolutely it does. I agree with that 100%. But too much heat leads to degradation, so it's the opposite of warm up. Projectors and tube amps--sure, I get that. But I haven't been able to find any basis for the assertion that sound improves in SS components.
     
  9. three_paws

    three_paws deleted account Thread Starter

    Location:
    USA
    Have a look here:

    Oscillator Technology

    "TCXO features excellent temperature characteristics, fast warm-up time (typically 50 to 1000 ms)"
     
    Last edited: Aug 31, 2021
  10. chervokas

    chervokas Senior Member

    If you look at the measurements there's an initial period where it goes from cold to warmer and performance improves before going down hill. And another poster measures a different one of the same units and has opposite results with performance improving as the DAC warma. The point is to get to the proper, best operating temp, quickly if possible as a practical matter, and stay there.
     
  11. three_paws

    three_paws deleted account Thread Starter

    Location:
    USA
    Yea, that one is def food for thought. That DAC appears to be unusually sensitive to temp changes--from a distance this just sounds like bad design, doesn't it? I wish there were a way to understand what specifically was causing these results.
     
  12. toddrhodes

    toddrhodes Forum Resident

    Location:
    South Bend, IN
    Without being able to back it up via expertise, but rather just tripping across this stuff every so often, the way it works in my head is that clocks, especially T and OXCOs are temp sensitive, and though they operate at a frequency, they only do so with high precision once up to temp and once temp has stabilized (e.g. that oven analogy earlier - it's not just the air inside a pizza oven that matters, the temp of all the surfaces do as well). It's not a great analogy but it's what I've got.

    Anyway, if we can be onboard with that idea, then from there, my understanding is that any derivation from that target temp produces a slightly different clock frequency. We're talking .001 Hz, I believe. My AccuSilicon clocks are rated out to 4 decimal places of clock accuracy, I think. Anyway, even a slight variance in that frequency leads to jitter and I believe phase noise increases? This is where it gets fuzzy for me on the terminology and how to apply it.

    But that seemed to be the gist, to me. High precision clocks have this narrow window where they are at optimum performance, and temperature is a direct factor in that, same as vibration (for what seems like obvious reasons to me for an oscillating crystal). And when those variables are disturbed, their performance changes.

    But again, this was all relative to Schiit Multibits which are their own sort of animal, I believe. And I'm way out of my depth to understand even the basics of why they operate the way they do.
     
  13. three_paws

    three_paws deleted account Thread Starter

    Location:
    USA
    Yah, that all sounds right to me based on what I've read. I think the warm up period is extremely fast and, absent bad design and outlier conditions, the PPM is very small, like you say. So if all that is true, DACs should sound their true selves at all times.
     
  14. toddrhodes

    toddrhodes Forum Resident

    Location:
    South Bend, IN
    This is where I defer to guys like Mike Moffat and the guy who runs Ian Canada, or the guys from Aries Cerat. They all disagree, and they design this stuff for a living.

    For me, I just leave it on and not worry about it any longer. But I am curious to see if my Holo DAC sounds any different in a couple of days or a week, like everyone says it will.
     
    Kyhl, saturdayboy and three_paws like this.
  15. three_paws

    three_paws deleted account Thread Starter

    Location:
    USA
    I'm going to keep digging for info. I'm really fascinated by this, both in terms of physics and psychology. I'll share any useful information I find.
     
  16. chervokas

    chervokas Senior Member

    Yeah, maybe not an ideal implementation. Maybe a chip that would have done well with a heat sink. I don't know. Not my area of expertise. Just an example that operating temps, high, low and just right, do make a measurable difference in performance. And, to get back to the original case of R2R DACs, resistor values can drift, both with temp and with time. Just one of a set of things to consider in designing and sometimes even in operating audio gear. None of it is a perfect ideal. All of it involves real-world compromises and limitations. And some of these DACs and HF oscillators in particular seem to be relatively temperature sensitive devices sometimes.
     
    Kyhl and three_paws like this.
  17. three_paws

    three_paws deleted account Thread Starter

    Location:
    USA
    Right, though there is the possibility that these differences are so noticeable in this case precisely of bad design. I would really love to find an electrical engineering take on this, but I haven't come across that anywhere. My background is in computers, and I've not come across any parallel reasoning there, even though a clock is essential in that context, too.
     
  18. Nice Marmot

    Nice Marmot Nothin’ feels right but doin’ wrong anymore

    Location:
    Tryon NC
    um….. a processor doesn’t get warm unless it’s processing. Leaving power on and not listening would be the equivalent of sleep mode on a computer. It would actually be “time playing” to start producing any type of useful heat in a DAC. I have to believe that a DAC is engineered to warm the circuitry off the limited heat of the processor. It can only get so warm. Any warmer and it’s a not good, break down issue.

    Isn’t this the basic difference between tubes and solid state that’s always been?
     
    three_paws likes this.
  19. three_paws

    three_paws deleted account Thread Starter

    Location:
    USA
    From what I've read, a TCXO uses a dedicated compensation circuit that handles temperature control. Warm up is typically achieved between .1 to 3 seconds, so pretty much at startup. So, it's not borrowing heat from elsewhere--it has a dedicated mechanism that controls the process.
     
    Last edited: Sep 1, 2021
    Nice Marmot likes this.
  20. Nice Marmot

    Nice Marmot Nothin’ feels right but doin’ wrong anymore

    Location:
    Tryon NC
    Agreed. I’m reading that a TCXO controls its own temperature against ambient temperature increases. It sounds like it protects itself from the heat of the processor. I think we are discussing circuitry updated from transistors to processors .... the lauded “quartz technology” of yore .... hahaha
     
    three_paws likes this.
  21. three_paws

    three_paws deleted account Thread Starter

    Location:
    USA
    Right, exactly.

    And I think even if warm up took longer, increased clock accuracy wouldn't translate to more bass, etc. It would just translate to better accuracy in output. I also have a feeling that the PPM would have to be pretty sizable for any of the imprecision to be audible anyway.

    As for the rest of the board, I don't see why increased temperatures would improve sound. Electrons travel better in cold temperatures. Warm is usually a bad thing in electronics. And there's also the question of why heat would enhance DAC performance but no other electronics. So, I'm still totally unpersuaded that warm up is a physical rather than psychological phenomenon.
     
    Nice Marmot likes this.
  22. h1pst3r88

    h1pst3r88 Forum Resident

    Location:
    Boston, USA
    Stevie Ray Vaughn did. Cesar Diaz tells a story about recording SRV -- he was using 32 tube amps concurrently that were left on all day and night for weeks.
    Heat'em if you got'em! :edthumbs:
     
    three_paws likes this.
  23. three_paws

    three_paws deleted account Thread Starter

    Location:
    USA
    That still doesn't mean they actually sounded any better, though. Only that he had enough money to cover the electric bill.
     
  24. h1pst3r88

    h1pst3r88 Forum Resident

    Location:
    Boston, USA
    ...and only that they sounded good to him (well, I guess to millions of others who idolize SRV, too).

    Cesar all tells a story of when he met a young SRV who had just finished a set -- obviously with a rig and tone that SRV liked. Cesar said to him, "you're playing is great but you sound like ****." Thus began a great partnership.

    However, the anecdote underscores that what sounds good to one of us can sound terrible to others... and they both had very good ears.

    OK, back to DACs -- sorry for the sidestep ;-)
     
  25. three_paws

    three_paws deleted account Thread Starter

    Location:
    USA
    Ever listen to the Albert King/SRV sessions? I love listening to their banter. Funny and sweet. They clearly liked and respected each other. Great recording, too--sounds really good on vinyl and CD (after you warm 'em up, obviously, har har).
     

Share This Page

molar-endocrine