S.E.X buzzing

greyhamster · 18671

0 Members and 1 Guest are viewing this topic.

Offline mcandmar

  • Hero Member
  • *****
    • Posts: 1599
  • Not all engineers are civil
Reply #60 on: January 31, 2014, 02:23:48 AM
I get the same results as Mauro, T10 to IEC/chassis ground is .1ohm,  but T10 to T1 or T2 has no continuity?

Are we talking about the right terminals here, T1 and T2 being the 240v mains side?

M.McCandless


Offline J. Mauro

  • Full Member
  • ***
    • Posts: 77
Reply #61 on: January 31, 2014, 03:19:46 AM
I do not know the differences between the 240v mains and 120v mains versions. I am going to try to diagnose the problem tonight with the 47 Ohm resistors. Thanks again.

Jeremy Mauro

So long, and thanks for all the fish.


Offline Paul Joppa

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 5822
Reply #62 on: February 03, 2014, 05:46:27 PM
Crap!  I apologize, guys, I screwed up the terminal numbering. I'll go through and try to fix my posts.

The transformer terminal numbers are embossed on the bobbin. They go 1-2-3-4-5 across the top (the back edge of the amp) and 6-7-8-9-10 across the bottom. (Every other electronic component goes clockwise from below; transformer bobbins are non-standard).

Terminals 4 and 5 are the 6.3v heater winding, and terminal 10 is the center tap on that winding. You should have continuity from T10 to T4 and from T10 to T5.

Again, sorry - my bad.

Paul Joppa


Offline J. Mauro

  • Full Member
  • ***
    • Posts: 77
Reply #63 on: February 04, 2014, 08:50:52 AM
Ok, that makes sense. I got nothing when I connected the resistors between 1, 10 and 2, 10. I will try with the proper terminals after I check the internal connection. Thanks for the clarification Paul.

Jeremy Mauro

So long, and thanks for all the fish.


Offline J. Mauro

  • Full Member
  • ***
    • Posts: 77
Reply #64 on: February 05, 2014, 03:32:48 AM
Well, I thought we had the problem found but no such luck. There is continuity between T4 and T10 and T5 and T10. Basically 0 Ohms resistance. Back to square one as they say. Ground issue probably but I don't know where to go from here. Thanks in advance.

Jeremy Mauro

So long, and thanks for all the fish.


Offline J. Mauro

  • Full Member
  • ***
    • Posts: 77
Reply #65 on: February 08, 2014, 04:28:44 AM
I plan on working on this tomorrow but does anybody have some advice on where to start, if there are certain spots that should be looked at first, resoldered, reflowed, anything? Thanks again.

Jeremy Mauro

So long, and thanks for all the fish.


Offline mcandmar

  • Hero Member
  • *****
    • Posts: 1599
  • Not all engineers are civil
Reply #66 on: February 08, 2014, 06:15:33 AM
Just reading back through your posts i noticed you have a couple of sets of tubes.  I cant believe you got coin base Sylvanias with the kit, you lucky bar stool!  The black plate Sylvanias are the best tubes IMO.

Of the sets you have, pick the best of them and let them run for a couple of days.  I have roughly a dozen sets and i have found most of them start of very noisy to the point of being unusable but after a day or so constant use they quieten down.  Some never do, but its very rare to find a pair that are quite right out of the box.

Its hard to say if you have just reached the natural noise floor of the amp, just have noisy tubes, or if you have a problem.  I found that differentiation difficult at the beginning.  The voltages you posted seem so high its either noisy tubes or a problem IMO, so as above leave a set of tubes burning in to rule out the tubes.

If you are re soldering i would double check all the grounding posts as somebody mentioned before the fact they are connected to the chassis plate makes them a great conductor of heat so they are hard to get a good solder on. Specifically posts 3,13,23.  I would also double check all the caps on 1-5 and 6-7, and 16-17 for the power supply section. I also found the plastic insulation on the wires really annoying as it melts so easily and if your not careful you can end up with a bit of it in your solder contaminating the joint. If its not shiny i would suck it out and re solder rather than just re flowing.

Going beyond the stock build, adding the two 100ohm resistors in the heater supply that Paul suggested a few pages back made a big difference to mine so its worth trying. Basically run a resistor from C1&C2 to C3 creating a grounded center point.  That would probably have been enough for my HD650's but with sensitive 32ohm cans i had to put 120ohm resistors into the headphone socket. Basically wire them in series to the left/right channel. i.e. audio signal passes through them.  Ideally i wouldn't use them, but i really had no choice to get away from the noise floor.

If you get that far and your still not happy i suggest reading through my thread as i made a few other modifications that all helped in their own small way, but for now just stick to the above and get it working as it should be. Its IS worth it though when all said and done, my Grados and this amp are a match made in heaven.

M.McCandless


Offline J. Mauro

  • Full Member
  • ***
    • Posts: 77
Reply #67 on: February 09, 2014, 07:53:07 AM
Thanks very much for the advice. Unfortunately, those coin base Sylvanias are the worst set of tubes I have. Granted, the other 2 sets are full base GE, but I do have 3 more sets coming from The Tube Store. I like the way the full base tubes click into the sockets, seems more solid almost. Anyway, I have the SEX amp running right now with some low level tunes and some old crappy Sony headphones. I may add the resistors but I do not have any 100 Ohm until tomorrow. I will report back with the results when I am finished. Cheers.
« Last Edit: February 09, 2014, 11:32:47 AM by J. Mauro »

Jeremy Mauro

So long, and thanks for all the fish.


Offline Karl5150

  • Sr. Member
  • ****
    • Posts: 629
Reply #68 on: February 10, 2014, 08:28:21 AM
I don't know if this was covered earlier in this or other threads but on page 22 of the S.E.X. manual a ground wire is attached from 17L to 22L with instructions to not solder. I didn't see instructions to solder 17L anywhere and found it loose on the 3rd or fourth check looking for the 1 ch hum/static I had.
Karl

Karl
Downstairs: Planar3>PH-16>Stereomour II>OB Betsy+
Upstairs: RP1>Eros/CD5004>Seductor (2x Monoblocks)>FH3
Office: Modi Uber 2/Sirius>SEX2.1.1>µFonken FF85WK + DC160 subs
BR: FiiO M6>SEX3.0.1>ScanSpeak 10F + TangBand W6 (Mono)/DT770Pro
Garage: X12 streamer>Quicksand>Minimus 77


Offline J. Mauro

  • Full Member
  • ***
    • Posts: 77
Reply #69 on: February 10, 2014, 06:11:55 PM
Question for Paul. Is it necessary to use 100 Ohm resistors for the C1 and C2 to C3 grounded center point or can one use similar values there?

Also, I have about 30 hours on this set of tubes now and there is no difference whatsoever to the buzz. Still measures at 3.7 and 3.7 mV.

As for the 120 Ohm resistors on the headphone jack, can someone explain which wires they go in series with? I don't want to mess anything up.

Thanks again.

Jeremy Mauro

So long, and thanks for all the fish.


Offline Paul Joppa

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 5822
Reply #70 on: February 10, 2014, 06:30:42 PM
Question for Paul. Is it necessary to use 100 Ohm resistors for the C1 and C2 to C3 grounded center point or can one use similar values there? ...
The value is not very critical. A 1/2 watt rating is adequate above 68 ohms, and I'd guess 200 ohms is too much. I've not done any experiments myself, so all this is "educated guessing".

3.7mV is way too much, and the fact that it's the same on both channels gives me some doubt whether it's the tubes. So far this is still a puzzler. You say have new tubes coming, and you can do the resistor test - if neither of those does the trick we'll have to dig a little deeper.

Paul Joppa


Offline mcandmar

  • Hero Member
  • *****
    • Posts: 1599
  • Not all engineers are civil
Reply #71 on: February 11, 2014, 03:25:30 AM
I agree, highly unlikely you would get the same noise level from two different tubes after that much use so i think we can rule out the tubes as a source.  There has to be something fundamentally wrong, looks like were back to basics on this one..

M.McCandless


Offline J. Mauro

  • Full Member
  • ***
    • Posts: 77
Reply #72 on: February 14, 2014, 03:00:40 PM
So my amp is now silent, but playing beautiful music as I type this. I finally took it to the one electronics guy in my city, who worked on my turntable and NAD 3020 last year. He and I did some troubleshooting together and connected it to an oscilloscope. Wow, was there ever a lot of ripple on this beast. So, instead, of doing the grounded center point with the 100 Ohm resistors, we tweaked the power supply by adding capacitance and resistance and balancing it between the two legs. Each leg now has 10000 uf  and .1 Ohm resistance. The voltages levelled out to about positive and negative 3.12 v for the filaments and there is now very little ripple. With my junky sony headphones, there was no hum or buzz at all. With my heresies, the same. Unfortunately, with my D5000, there is some hum, not horrible, but I will add the 120 Ohm resistors in series to mitigate that soon. I will also post a couple pictures and the updated schematic for what we tweaked in the next day or so. Thanks again, everyone, for the help and great advice. Cheers.

Jeremy Mauro

So long, and thanks for all the fish.


Offline Paul Joppa

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 5822
Reply #73 on: February 14, 2014, 05:26:39 PM
Very interesting, and a contribution to our knowledge base - thank you very much! I look forward to the schemo posting.

This has been a real puzzler; with this knowledge in hand (plus a few other recent posts) it is at least clear that the heater supply noise is more of a contributor that we thought, and/or than it was ten years ago.  Perhaps power lines are getting noisier, or the available tubes are getting to be of lower noise immunity, or perhaps something that hasn't occurred to us yet.

Paul Joppa


Offline mcandmar

  • Hero Member
  • *****
    • Posts: 1599
  • Not all engineers are civil
Reply #74 on: February 15, 2014, 06:59:37 AM
Looking forward to seeing this schematic.

While testing i found the added capacitance made no difference to the output noise, stock had something like 1.1v ripple vs .6v when i doubled it. In the end i put a 22000uf Panasonic TS-UP cap in there anyway.

Centering the rails via resistors did make a big difference, and with a variety of different tubes. So i believe this is the key element here, sounds to me like your guy just did it a different way by using caps between each terminal and the center ground.  Please post a few pics too so we can see how he did it..

M.McCandless