Page 14 of 15 FirstFirst ... 4 12 13 14 15 LastLast
Results 131 to 140 of 144

Thread: Studer 928 Story...

  1. #131
    Join Date
    Feb 2009
    Location
    northampton uk home of Dr Who and Blackstar Amps!
    Posts
    9,799
    Thanks
    1
    Thanked 653 Times in 617 Posts
    Rep Power
    9266934
    Sign in to disable this ad
    Quote Originally Posted by michael1991 View Post
    The TDK document is very general on capacitor properties. In figure 4 is drawn the nonlinear relation between voltage and current "The oxide layer constitutes a non-linear voltage-dependent resistance that causes the DC current to increase more steeply as the voltage increases. A characteristic curve as shown in figure 4 is obtained."

    for audio signals this leads to an non linear characteristic curve. the higher the voltage rating is, the more left the signal will be in relative to the drawn curve, where the characteristics are more linear.

    You are right, the distortions don't add up in a sum. but at every new non linear stage, the new generated frequency content (by intermodulation) of a previous stage acts as new signal, that intermodulates again with the other signals and so on. (here a wikipedia link to intermodulation Intermodulation - Wikipedia )
    I can not quantify the effect right now, and don't want to state that there are no other sources of non linearities in the audio signalflow, but this effect contributes to it.
    The fact that DC leakage current increases sharply at some voltage around the WV limit has been known for decades...Shoot! Budget concious TV mnfcts use the property as an HT surge limiter and used 275V capacitors when the peak mains was at least 339V! I think it is a huge stretch to sya that large (very!) DC effect is related to the minute amount of distortion produced at low frequencies and at very much reduced voltages.

    In case others have lost the plot a bit or dozed off! My position is...Yes, electrolytic capacitors DO produce measurable harmonic distortion IF there is a significant voltage across them and as far as my information goes that distortion is at much the same level regardless of capacitor brand or working voltage. As I say, if I see AC distortion plots that contradict that I am willing be be proved wrong.

    I am also stating for the record that the distortion is SO small that at normal signal levels it is audibly insignificant even when dozens of capacitors are involved. I also think you will have to search a long time and pay a serious wedge to find a speaker that can put out 90dBSPL at less than 1% THD!

    Dave.

  2. #132
    Join Date
    Mar 2019
    Posts
    26
    Thanks
    0
    Thanked 2 Times in 2 Posts
    Rep Power
    0
    Quote Originally Posted by ecc83 View Post
    The fact that DC leakage current increases sharply at some voltage around the WV limit has been known for decades...
    The curve is getting steeper, but its not linear until voltage xy, it is only getting even more non linear at higher voltages.
    This is the same effect for ac signals because the characteristic curve does not care if there is a given dc voltage or if that voltage is a momentary value in an ac signal. the only difference between ac and dc is that dc is at frequency zero.

    shure, the signals that are traveling through the capacitor are not in the right area of the characteristic curve but in the left side, where the non linearities are not that huge. but still, if your voltage rating moves up, the signal will move in the equivalent ratio even further to the left, which leads to even less non linearities.

    I don't want to argue with you if this effect is audible or not, if you think this is usefull, if loudspeakers are linear in the same range. The effect exists.

  3. #133
    Join Date
    Feb 2009
    Location
    northampton uk home of Dr Who and Blackstar Amps!
    Posts
    9,799
    Thanks
    1
    Thanked 653 Times in 617 Posts
    Rep Power
    9266934
    Quote Originally Posted by michael1991 View Post
    The curve is getting steeper, but its not linear until voltage xy, it is only getting even more non linear at higher voltages.
    This is the same effect for ac signals because the characteristic curve does not care if there is a given dc voltage or if that voltage is a momentary value in an ac signal. the only difference between ac and dc is that dc is at frequency zero.

    shure, the signals that are traveling through the capacitor are not in the right area of the characteristic curve but in the left side, where the non linearities are not that huge. but still, if your voltage rating moves up, the signal will move in the equivalent ratio even further to the left, which leads to even less non linearities.

    I don't want to argue with you if this effect is audible or not, if you think this is usefull, if loudspeakers are linear in the same range. The effect exists.
    EVERYTHING is non linear to some degree even resistors, you can measure it for the old carbon comps. My points are still valid and unanswered. Where are the actual, real signal level graphs that show the effect is anything to be concerned about and why worry about such a tiny effect that can be all but eliminated anyway with a larger value of capacitor than the required LF turnover would indicate, always keeping an eye out for any untoward effects of course. As for "room for larger components". I will bet you bag of ten that 100uf 22V caps are now about the same size as 50uf 25V were thirty years ago...And! Better in every degree.

    Chasing the fifth decimal place for distortion is an excercise that is only done for the reviewers and specc' sheets. The transducers "each end" have a hell of along way to go before we need to worry about a couple of dozen electos in the signal path. OR! just go DC and servos.

    Dave.

  4. #134
    Join Date
    Mar 2019
    Posts
    26
    Thanks
    0
    Thanked 2 Times in 2 Posts
    Rep Power
    0
    Quote Originally Posted by ecc83 View Post
    Where are the actual, real signal level graphs that show the effect is anything to be concerned about
    grab some caps, rebuilt the test circuit you posted before and measure it up, if you don't believe it...
    some colleague made this kind of tests, but i'm sorry, i dont have the plots here.

    For new designs, sure, built a dc coupled design, and get rid of all electrolytics possible, is for sure not the worst idea. but in this given design, i have to deal with them and if i have to replace them because they are dried out (and no, there is nothing "inside the specs" if all caps i measured (about 10 or sth like that) were in the range of 60-80 uF instead of 80-120 uF...), increase the voltage rating (in this amount of caps at no additional cost) to bring the non linearities more down than they are, and even go a step further with bypassing them with foil caps.

    I am curious if the effects that are audible occur and can be pictured in standard measurements. I do not know so far. But to state that they are negligible and not audible if you have not calculated, measured or heard it, is not a qualified statement.

  5. #135
    Join Date
    Apr 2008
    Location
    Pacific Northwest, USA
    Posts
    8,533
    Thanks
    217
    Thanked 217 Times in 197 Posts
    Rep Power
    3753124
    I’m taking a break from the interesting but hair-splitting conversation to reflect again at just how great my console sounded in my subjective listening session, and to post these updates:

    I got to the bottom of why my Master A module has a significantly diminished output on the left channel...using the block diagram and the module’s own features I narrowed the issue down to somewhere in the output amp...pulled out the scope and the schematic, injected tone to the module and started probing, first at the inputs of the four 5532 opamps that drive the two balanced outputs...all good there...then the outputs of those same opamps...same story, looked good...hm...then the inputs of the output transformers, and A-OK there...then the outputs of the transformers...not good there...scratched my head...powered down the console and pulled the module, got out a good light and my magnifier...solder joints all looked good, flipped it over and then finally see what’s been out for all to see all along:

    f3af1fff-3c78-4b9e-8532-70229c3741a5-jpeg

    See it? The transformer on the left is the left output transformer...see the broken winding? Duuuuuuuude. So at some point the master module was grossly mishandled and was dragged across or has some thing dragged across it tearing up the transformer. Don’t think that’s going to be repairable...it looks like other sections of winding have the insulation layer abraded so who knows if there are shorted windings, etc. fortunately I can order this transformer NOS from the Netherlands, but it’s probably going to cost me about $100 shipped. Bummer.

    On a more fun note, I have barely any TT patch cables and I’m going to need a mess of them for the onboard patchbay and two to three 96-point patchbay that will be mounted in an outboard rack. TT patch cables are spendy! BUT...I found, via two separate auctions, a total of 18 ADC cables for a total of something like $35 shipped, all in good condition. They were probably cheap because they are short (16” tip to tip), but these are PERFECT for the onboard patchbay. And they are grey like the Studer! I’m planning on getting some different colors of heat shrink to make it easier to ID cables. But look...just the right size.

    5eecc1c9-ed6a-4782-bcdc-e9aedea66855-jpeg

  6. #136
    Join Date
    Feb 2009
    Location
    northampton uk home of Dr Who and Blackstar Amps!
    Posts
    9,799
    Thanks
    1
    Thanked 653 Times in 617 Posts
    Rep Power
    9266934
    Hair splitting! Good one Mr S and I shall have no more of it (funny how one gets "sucked in"?)

    Bummer on the traff front but you might take some comfort from the knowledge that a Home Grown top of the range Jensen 1:1 line drive transformer would set you back $108?

    Jensen also do a frame only PCB style that could be a drop in but I guess you want to keep the Studer original.

    Dave.

  7. #137
    Join Date
    Apr 2008
    Location
    Pacific Northwest, USA
    Posts
    8,533
    Thanks
    217
    Thanked 217 Times in 197 Posts
    Rep Power
    3753124
    I do enjoy getting sucked into the discussion though...I just needed to come up for air.

    And regarding the Studer transformer, yes I’d like to keep it Studer (and if I changed one to something non-Studer I’d have to change the other channel too), but the real rub here is it is a fairly specialized transformer with a third winding that is incorporated into the feedback loop of the four opamps. It’s not a straight-forward opamp output —> transformer primary. So the Studer transformer is really the easiest solution on top of being the preferred solution. They sound nice. And when you push them the program gradually morphs...there is a shift in the character to almost, like, a subtle increase in midrange clarity (louder without overpowering) but then this gradual increase in edge or bite to the sound...like the transient attack elements in the program come out front a bit. It’s neat. So...yah. I want the Studer iron. It’s about $43USD but the shipping is the killer.

  8. #138
    Join Date
    Mar 2019
    Posts
    26
    Thanks
    0
    Thanked 2 Times in 2 Posts
    Rep Power
    0
    Sad to hear your transformer is broken.

    Yesterday I did some THD measurements. With that I checked the calibration instructions to calibrate the vca for minimum distortions. Its kind of tricky and sensible, but with some patience the results are quite pleasing.
    If you want I can post the plots, but they have to be read carefully, since the analyzer draw some warnings in some parts of the measurements. (Multiple measurements differ more than 2,5 % that were set down as confidence interval) So probably this measurements don't show the "real and only truth" but point in the direction

  9. #139
    Join Date
    Mar 2019
    Posts
    26
    Thanks
    0
    Thanked 2 Times in 2 Posts
    Rep Power
    0
    Hi,
    I finally got started with the mono/mic channels. I treated the unencapsulated poti with DeosIT D5 to solve the dirt and DeoxIT F5 after that which contains some amount of lube. It moves smoothly and i can not hear any scratching noise etc.
    Greetings
    Michael

  10. #140
    Join Date
    Apr 2008
    Location
    Pacific Northwest, USA
    Posts
    8,533
    Thanks
    217
    Thanked 217 Times in 197 Posts
    Rep Power
    3753124
    You want to avoid using F5 on metal-to-metal contacts *because* of the lubrication. D5 is the correct product to use. If you applied F5 to the mic trim range rotary switches I’d re-flush with the D5.

    Someday I’ll check the VCA calibration. I have a Tektronix AA501A analyzer for that.

    I finally got the replacement output transformer for my Master A module...



    abdd0617-3265-4ecc-a552-80ba2a9c1d22-jpeg

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. studer b-67
    By stugazz in forum Analog Recording & Mixing - Tape & Gear
    Replies: 0
    Last Post: 01-02-2010, 23:48
  2. Studer A827 Story...
    By Beck in forum Analog Recording & Mixing - Tape & Gear
    Replies: 7
    Last Post: 06-15-2009, 22:19
  3. studer b67, pin 2 hot?
    By ethyrvalve in forum Analog Recording & Mixing - Tape & Gear
    Replies: 3
    Last Post: 05-16-2008, 01:48
  4. studer 827
    By Rocket Boy in forum Analog Recording & Mixing - Tape & Gear
    Replies: 5
    Last Post: 08-20-2003, 07:46

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •