My new old amp project: SE46 monoblocks

Discussion in 'DIY' started by Redboy, May 28, 2014.

  1. Redboy

    Redboy a few good watts Subscriber

    Messages:
    11,147
    Location:
    Twin Cities, MN
    If you search for Gordon Rankin's Bugle 45 amplifier, you'll have what you need. Mine are simply adapted to use different power transformers and to get a slightly lower operating point for the 46. The rest is Gordon's circuit, really.
     
  2. Alan0354

    Alan0354 Super Member

    Messages:
    1,665
    Location:
    Bay Area
    When I google, quite a few schematic popped up and they are quite different. Can you give me a link for the schematic that you use?

    Thanks
     
  3. Redboy

    Redboy a few good watts Subscriber

    Messages:
    11,147
    Location:
    Twin Cities, MN
    Here's the original article:
    http://www.wavelengthaudio.com/bugle.pdf
     
    Alan0354 likes this.
  4. Alan0354

    Alan0354 Super Member

    Messages:
    1,665
    Location:
    Bay Area
    Thanks

    This is a really simple circuit. One SRPP stage driving the SE triode tube. No negative feedback at all.

    I just print out the file, let me read a little, I am sure I have questions for you. I am not that familiar with tubes, so it's going to take a little bit of time.

    What is the output power of your amp? What kind of speaker you are using?
     
    Last edited: May 31, 2017
  5. Redboy

    Redboy a few good watts Subscriber

    Messages:
    11,147
    Location:
    Twin Cities, MN
    Yeah, it doesn't get much simpler than that!

    Power output is better measured in milliwatts... maximum output of my amps is ~1 watt per channel on a good day, but the spec'd 45 would give you a smidgen more power than that. You need very efficient speakers. Mine are single driver, full range field coils from Feastrex.
     
    asilker likes this.
  6. Hull Rust

    Hull Rust AK Subscriber Subscriber

    Messages:
    660
    I like my Pilot, but this is the direction I want to head down the road.
     
    Redboy likes this.
  7. Alan0354

    Alan0354 Super Member

    Messages:
    1,665
    Location:
    Bay Area
    I think the power is too low for me. I was expecting like 5 to 7W like a normal SE amp.

    Thanks anyway.
     
  8. asilker

    asilker Bible Reader Subscriber

    Messages:
    2,124
    Location:
    Chicago IL
    Many normal SET amps re between 1 and 3 watts. With a 300b or a pentode, the output can rise above 5wpc
     
  9. Alan0354

    Alan0354 Super Member

    Messages:
    1,665
    Location:
    Bay Area
    Is SET means single ended tube?

    I have the idea of parallel two big tubes to get the necessary power. My speaker is not that efficient. Not too many high end speakers are efficient and they usually are 4ohm.
     
  10. asilker

    asilker Bible Reader Subscriber

    Messages:
    2,124
    Location:
    Chicago IL
    Single ended triode

    There are some nice high efficiency options out there. It's kind of a niche.

    By the way SEP would be single ended pentode (el, kt, 6l6 etc)
     
  11. dowto1000

    dowto1000 New Member

    Messages:
    38

    No No No, just get a 98 dB or higher speaker, and then you can do audio, and audio amps, right.

    Vintage ALTEC comes to mind, 604s, A7s, maybe even Valencia types, basically a 15 inch ALTEC woofer, 811 horn, and an 802D. Crossover about 800 HZ.to as high as 1500 HZ.
     
    Last edited: Jun 16, 2017
  12. dowto1000

    dowto1000 New Member

    Messages:
    38
    P1010008.JPG
    - - - - - - - -

    TYPE 46 : 250 or 260 VDC P-K and 22 -23 mA is about 1.3 Watts output.

    The really HUGE problem with Gordon's circuit is his power supply. Its so sad to me, people copying that design, never get to hear what a Type 46 can do.

    For chokes, his schematic is showing a 30 HY / 40 mA. inductor, ( 157G ) for each channel. He was also specing a Hammond 157L in the schematic, which is 14 HY at, as I recall 75 mA.

    Anyone's use of either of these inductors RUINS that amp's dynamics immediately, due to ( many-fold ) too high resistances, DC resistances.

    The 157G measures 595 Ohms, and the 157L measures 429 Ohms in DCR. Chokes filtering the Finals, need to be about 20 Ohms maximum, or preferably less in DCR, to play a loud speaker properly .

    In 2017, my recently-completed Type 46 DIY design uses 6 Ohm DCR chokes. Being low in DCR, and also, conversely, low in Henries, calls for the use of two equally constructed Ls in series as in L1/C1/L2/C2 . So, I have in my 46 amp, only 12 Ohms total of series resistances to the Finals, versus 420 Ohms or 595 Ohms in a Bugle. This low series DCR to the Finals is mandatory.

    A second very serious design problem with the Bugle is that the audio circuit is just a bit too complex.

    It uses a capacitor-couple, to couple the first stage into the Finals stage. This is a HUGE design error.

    There are three ways we can couple tube stages, capacitor, transformer, and direct. A capacitor couple, used in the Bugle, is sonically, THE absolute worst way to couple two stages. The direct couple is THE absolute best way to couple two stages.

    My own Type 46 amp, pictured herein uses half a 12AX7, directly coupled, to the grid of a Type 46.

    I employed similar circuits in 2015 -2016, for my two Type 45 amps, and find the Type 46 tube is markedly superior in my implementations. I have sold all the high-testing ST 45s ever owned, about 60 tubes, every one, on eBay.

    Have FUN, I do. ............................... Jeff

    P1010005.JPG P1010004.JPG
     
    Last edited: Jun 15, 2017
  13. Alan0354

    Alan0354 Super Member

    Messages:
    1,665
    Location:
    Bay Area
    I am trying to study up to design switching power supply. For single end, it's too critical as there is no common mode rejection.

    You can never filter out the 60Hz rectified ripple no matter what. I designed a SS amp, I have 40,000uF -R-C pi filter and I still have some ripple. It's very critical to have a clean rail.

    I think insisting on use rectify tube is not the wisest thing to do just to say it's all tubes. You have to go with what's best out there as long as you keep the signal path all tubes.
     
  14. dowto1000

    dowto1000 New Member

    Messages:
    38
    Alan,

    You are designing for the WRONG parameter. Low ripple, at least to the Finals, is the LEAST most important, and the least needed thing to have. It is, to use your term, uncritical, I have found.

    In a great SET, the low mu Finals can have anywheres from 200 mVAC to 800 mVAC and be perfectly fine, as good as ANYONE has ever heard. However, I find the front end needs to be shunt regulated, and no more than 2 or so mVAC.

    In a DC amp, the use of a tube rectifier is the best possible compromise. ( Asano, Shishido, etc ) Can you figure out why, besides it sounding good ??

    Jeff
     
    Last edited: Jun 15, 2017
  15. Alan0354

    Alan0354 Super Member

    Messages:
    1,665
    Location:
    Bay Area
    No why don't you tell me.

    Why is it wrong parameter to avoid ripple on the +B? Single end is known to not be able to cancel out the power supply ripple like the push pull tube amp. In push pull, the same ripple present to the plate of both tubes and become common mode and will not show up on the secondary of the OPT. SE amp cannot do that.
     
  16. dowto1000

    dowto1000 New Member

    Messages:
    38

    I avoid ripple to the front end tube, 2 mVAC, which is where low ripple counts.

    Since I am "opposite" you, ( and have been designing and building 2A3 amps, etc, push pull and SET since 1982 ) why don't YOU think some, independently, take a day or two, about what I have told you, about

    (1) ripple to the low mu Finals ( 200 to 800 mVAC ) and

    (2) rectifier tubes in a DC SET amp

    and lets see what you come up with !! Hint: The answer is given in the question, each question.

    Have fun, I do.


    Jeff .
     
    Last edited: Jun 15, 2017
  17. Redboy

    Redboy a few good watts Subscriber

    Messages:
    11,147
    Location:
    Twin Cities, MN
    Cripes Jeff, this old tune again?! Maybe you should start your own thread about your own amazing amps that you built THE RIGHT WAY instead of crapping on everyone else's threads and pointing out their HUGE PROBLEMS and their WRONG PARAMETERS...

    I'm guessing your amps sound very nice. A single ended 46 can do that. Please tell us more about them... in another thread.
     
  18. Alan0354

    Alan0354 Super Member

    Messages:
    1,665
    Location:
    Bay Area
    I don't know what you mean you are "opposite" me.

    I am thinking about in terms of equivalent circuit and noise. Here is what I think according to my understanding:

    OPT is used to match the load ( speaker impedance) to the PLATE resistance of the output tube. I draw the typical circuit of SE triode and the equivalent voltage divider:

    SE power supply.jpg

    The left side is the triode with plate going to the OPT, the other side of OPT to +B( shown as battery). The turn ratio is N1 to N2. The OPT will present an impedance equals to the plate resistance rp to the plate of the triode.

    Now look at the right drawing, I just draw the ripple voltage Vr. Since the impedance of the primary is equal to the plate resistance, the ripple voltage ACROSS the primary of the OPT is 1/2 X Vr. The ripple voltage at the load end will be (1/2 X Vr X N1)/N2 as shown in the writing.

    Now lets put in some numbers.

    Let Vr=1V
    N1:N2 = 100:1

    So the ripple voltage across the primary of the transformer is 1/2 X Vr= 0.5V

    Since N1:N2 is 100:1, ripple voltage at the load is 0.5V/100= 5mV. That's a lot of ripple noise in my book.

    I am not familiar with tubes and transformer, you can correct me and put in some real numbers and see what is the output noise.

    Of cause it is understood it's more critical for preamp tubes as it's going to be amplified by the output stage.
     
  19. Alan0354

    Alan0354 Super Member

    Messages:
    1,665
    Location:
    Bay Area
    I really love to do a SE amp, I am just afraid that it might not loud enough for me. I already have the speaker, it's not like I can go buy another pair. I think 7 or 8W will be good enough, not 1 or 2W.

    I am not giving up yet, I am thinking about paralleling tubes, but then the problem is how much DC current I can realistic pass through the OPT without saturating the transformer and destroy the sound. I read something about using transistor to remove part of the bias current, but then, it's not a true tube amp anymore. I am still at the thinking stage.
     
  20. dowto1000

    dowto1000 New Member

    Messages:
    38
    \

    Hi redboy,

    We have never met to my knowledge. I wasn't trying to promote my recent implementation, but rather, address to that one novice poster, areas that IMHO need improvement as far as him, or anyone else, copying the Bugle.

    I don't know how you build, DCR of Ls, etc, but the Bugle schematic reference you posted for us, was plain to see, and it begs to be commented on in its most deficient design areas.

    I hope I have not offended you inadvertently. Sorry if I did.


    Jeff
     
    Last edited: Jun 16, 2017

Share This Page