ECP Torpedo III [indexed in first post]

Discussion in 'Headphone Amplifiers and Combo (DAC/Amp) Units' started by FlySweep, Nov 2, 2015.

  1. bazelio

    bazelio Friend

    Pyrate
    Joined:
    Nov 16, 2015
    Likes Received:
    3,417
    Trophy Points:
    113
    Interesting... Yeah, I guess there is no shielding around the transformer. I was considering going a similar route to MortenB - minus the 18 awg silver wire - but may have to reconsider. If I do go with a different pot, then short jumper wires from its terminals to the through holes might just be the best option.
     
  2. MortenB

    MortenB Facebook Friend

    Joined:
    May 3, 2016
    Likes Received:
    111
    Trophy Points:
    43
    Location:
    Denmark
    You beat me to it :) With everything except the input wire being fully balanced the input wire running close to the mains transformers EM field will be the critical point for sure.

    I considered running the twisted wires under the PCB to get under the ground plane, but I would have to drill holes in the beautiful pcb, and I could not make myself do that. If there is any hum from the transformer with this set up I will introduce a new ground plane over the wires instead - potential problem solved. The braiding is ready to go on the wires if needed.

    [​IMG]
     
    Last edited: May 17, 2016
  3. MortenB

    MortenB Facebook Friend

    Joined:
    May 3, 2016
    Likes Received:
    111
    Trophy Points:
    43
    Location:
    Denmark
    If you are very worried about this you can just use shielded cable and connect the shield to the ground plane at one end. This will give you the best possible shielding from the transformer
     
  4. Jun

    Jun Friend

    Pyrate
    Joined:
    Feb 28, 2016
    Likes Received:
    169
    Trophy Points:
    43
    Someday when I get the bored or have time I'll shield the Input pcb traces from the power transformer with electrical tape and grounded aluminum foil.
     
  5. MortenB

    MortenB Facebook Friend

    Joined:
    May 3, 2016
    Likes Received:
    111
    Trophy Points:
    43
    Location:
    Denmark
    Or you could shield the mains transformer in mu metal :cool:
     
  6. MortenB

    MortenB Facebook Friend

    Joined:
    May 3, 2016
    Likes Received:
    111
    Trophy Points:
    43
    Location:
    Denmark
    The Mundorfs are in... I added a bit of glue to the bottom/side of the caps to hold them. I'm not really comfortable with big components ''dangling'' by the component legs, it tend to stress the solderings over time.

    Now I'm just waiting for the 100K ohm Takman resistors to arrive.

    Also I will do some experiments with a tiny choke instead of R2 in the power supply. Chokes are really effective at reducing ripple, but there is no room for a proper power supply choke (they are as big as the power transformer). I have ordered a special small choke, it will be interesting to do measurements to check, if they are any good. I will share some measurements...

    [​IMG]
     
  7. bazelio

    bazelio Friend

    Pyrate
    Joined:
    Nov 16, 2015
    Likes Received:
    3,417
    Trophy Points:
    113
    You're going to pop in the kit resistors and give her a listen in the meantime, though, right? :)
     
  8. MortenB

    MortenB Facebook Friend

    Joined:
    May 3, 2016
    Likes Received:
    111
    Trophy Points:
    43
    Location:
    Denmark
    I'm waiting. Don't have more time anyway as tonight is the weekly practice session with the band. So in an hour I will get live music instead :)
     
    Last edited: May 17, 2016
  9. MortenB

    MortenB Facebook Friend

    Joined:
    May 3, 2016
    Likes Received:
    111
    Trophy Points:
    43
    Location:
    Denmark
    Time to start some interesting measurements to check the effect of a choke instead of R2 in the power supply.


    [​IMG]



    This first measurement is made across R2 where CH1 is at the C1/R2 junction and CH2 is at the R2/C2 junction. Notice that CH1 is 200mV/div and CH2 is 50mV/div. Channel two is the ripple we have before the regulator (Q1).

    [​IMG]




    This next picture has CH1 at the R2/C2 junction and CH2 is now at R23 away from the plate of the tube. So basically the measurement is across Q1, and CH2 is now at 2mV/div - this is the limit of the scope and probes I have.

    WOW..! The regulator really does an amazing job. Well done dsavitsk :)


    [​IMG]
     
  10. MortenB

    MortenB Facebook Friend

    Joined:
    May 3, 2016
    Likes Received:
    111
    Trophy Points:
    43
    Location:
    Denmark
    And now to the same measurements with R2 replaced with a nice choke...

    [​IMG]
     
  11. MortenB

    MortenB Facebook Friend

    Joined:
    May 3, 2016
    Likes Received:
    111
    Trophy Points:
    43
    Location:
    Denmark
    And here we have the measurements across the choke (across where R2 used to be) with the choke... Compare this with the first close up picture of the screen above, and notice that both CH1 and CH2 are at the same V/div as in the picture above. Chokes are effective..!

    [​IMG]


    Of course the interesting part is, how the ripple looks after Q1..!? Here is the result, again measured at R23 - away from the tube. A little bit better with the choke, but not by much...

    [​IMG]
     
  12. MortenB

    MortenB Facebook Friend

    Joined:
    May 3, 2016
    Likes Received:
    111
    Trophy Points:
    43
    Location:
    Denmark
    Next up will be to do the same measurements with the small TentLab choke I have ordered. It should fit as a replacement of R2. If it get's anywhere near the ripple rejection of the big choke used here for the experiment I will be very happy.

    And of course it will be interesting to check, if the small TentLab choke has a positive impact on the sound.

    Still waiting for the parts to arrive, here you have a link to the mini electronic choke. I ordered the MEC100 as I'm afraid that the inrush current charging C2 will kill the MEC50 http://www.tentlabs.com/Components/Tubeamp/page36/page36.html
     
    Last edited: May 18, 2016
  13. MortenB

    MortenB Facebook Friend

    Joined:
    May 3, 2016
    Likes Received:
    111
    Trophy Points:
    43
    Location:
    Denmark
    And there is sound. I think that Bowie would be pleased with the T3 :)

    No hum even with the volume turned up full and no music playing, this is with my 102dB sensitive Beyerdynamic DT-1770 PRO, so the implementation of the RCA's an wiring seems to work as intended. I might try screening the wires where they pass the transformer to check if it can get even better, but I'm perfectly happy with this.

    [​IMG]
     
  14. bazelio

    bazelio Friend

    Pyrate
    Joined:
    Nov 16, 2015
    Likes Received:
    3,417
    Trophy Points:
    113
    Nice job, MortenB. That hookup wire doesn't look like it was easy to work with. And I think you've paved the way for others. I still wonder about the severing of the RCA to Pot board traces though... They should be unconnected to anything anyhow, right? The point to point wiring just leaves them floating on both ends, doesn't it? Maybe I'm really brain farting here.
     
  15. MortenB

    MortenB Facebook Friend

    Joined:
    May 3, 2016
    Likes Received:
    111
    Trophy Points:
    43
    Location:
    Denmark
    The thick solid core hook up wire is very stiff, so once the 3 wires are twisted and the wires are bent into the proper position - they stay there - so actually very easy to work with. It takes a bit of time bending them in the proper shape of course.

    Ha ha... You are correct on the pcb traces. I did not need to cut them, so the brain fart was on me ;) The point is, that they must be floating at both ends avoiding them to work as antennas picking up hum as they could do if connected at one end. But you are right, they are not connected when using the wires - I forgot to think...
     
    Last edited: May 18, 2016
  16. Jun

    Jun Friend

    Pyrate
    Joined:
    Feb 28, 2016
    Likes Received:
    169
    Trophy Points:
    43
    Holy shit! the bass from the torpedo iii, it made me check to see if i had the bass eq on and nope the EQ was off. The brimar 6060 yellow tail are awesome!
     
  17. dpump

    dpump New

    Joined:
    Oct 26, 2015
    Likes Received:
    19
    Trophy Points:
    8
    dsavitsk and TomB: Asked this a while back. Can the Cinemag transformers be used in the T1? If so, would it be a worthwhile upgrade? TomB said before he thought the Cinemags in the T1 might be close to a T3 with standard transformers but he wasn't sure.
     
  18. Jun

    Jun Friend

    Pyrate
    Joined:
    Feb 28, 2016
    Likes Received:
    169
    Trophy Points:
    43
  19. dsavitsk

    dsavitsk Friend

    Pyrate
    Joined:
    Oct 2, 2015
    Likes Received:
    1,616
    Trophy Points:
    93
    Home Page:
    They should work just fine. I can't comment on the sound or the comparison as I have not tried it.
     
  20. MortenB

    MortenB Facebook Friend

    Joined:
    May 3, 2016
    Likes Received:
    111
    Trophy Points:
    43
    Location:
    Denmark
    The small miracle electronic choke has arrived, and it seems like TentLabs are on to something with this one..!

    [​IMG]



    This picture shows the ripple at the R2 / C2 junction on CH2 at 50mV/div. The ripple is around 125mV. CH1 is at R23 away from the tube and the setting is 2mV/div. Again this is the limit of my scope and probes, so it's quite hard to get a solid reading. What you see on the yellow trace is more or less noise from this crude measurement set-up at the dining table. I've measured over a longer time span this time at 1s/div to get a better understanding.

    [​IMG]


    This is the similar measurement with the small electronic choke... Check the difference on the blue curve for CH2 - WOW :) The ripple is more or less gone. Looking at the yellow curve we also see, that the ripple is reduced.

    I wish I had the measuring gear to get solid readings of these small signals... But since the regulator reduces ripple by a certain factor it seems realistic, that when reducing ripple this drastically before the regulator, there should be a reduction of (maybe) the same factor after the regulator.

    I hope for a noticeable improvement of the sound quality since this reduces the noise/ripple at the plates of the tubes. But hearing is believing, so time will tell...

    [​IMG]
     
    Last edited: May 20, 2016

Share This Page