1. Time for some upgrades in server hardware and software to enhance security and take AK to the next level. Please contribute what you can to sales@audiokarma.org at PayPal.com - Thanks from the AK Team
    Dismiss Notice

Allen Organ Model 90 Ultralinear Taps?? Any Ideas

Discussion in 'Tube Audio' started by Scripted, Oct 12, 2018.

  1. Scripted

    Scripted New Member

    Messages:
    8
    Location:
    Phoenix AZ
    I recently picked up a pair of allen 90's and now in a quandary. Do I refresh the existing circuit or make use of the ultralinear taps on the huge output transformer? I am still a novice and would value any feedback you have !!!

    Since I am planning on using these for my primary tube hifi system some of the questions I have:

    - Are there more advantages by incorporating the ultraliner (UL) taps?

    - What would be the best way to switch over to utilize the UL taps?

    - Would I need to drop the B+ voltage by about 100v (from 580v to 480v) to accomodate the anode and screens?

    - Does the bias voltage have to be changed from -35v to say -70v?
    Allen Organ Model 90.jpg
    The amps I have are similar to the schematic however they do not have a 5k pot in the the bias circuit and the bias circuit extends to what I think is the 6SN7 driver stage, right before the KT-88s (entering the 6SN7's by the ground and 510k resistors which ultimately bridge pins 1,4,6 and 3)

    These seem like they could be sweet amps but I cannot find any information let alone modification advice about them.
     

     

    Please register to disable this ad.

  2. kward

    kward AK Subscriber Subscriber

    Messages:
    3,512
    Location:
    Utah
    The Tung Sol 6550 datasheet says max screen voltage when screen is connected to a tap on output transformer is 500V, whereas the GE data sheet says 450V. I know of several commercial amps that use 6550's in UL mode that exceed that at 515V or so (5% to 15% overage). But exceeding it by 30%? I wouldn't do it.

    I'd leave the output stage topology as is. With choke input power supply filtering you've already got a very good regulated supply. Besides, to do it right, to lower the plate voltage enough to be in spec would require some decent engineering, probably a hefty series-pass power supply regulator stage would be required.
     
  3. Pio1980

    Pio1980 AK Member Subscriber

    Messages:
    27,036
    Location:
    Eastern Bamastan
    Lower screen voltages reduces likelihood of flashover.
     
  4. dcgillespie

    dcgillespie Fisher SA-100 Clone Subscriber

    Messages:
    9,781
    Location:
    Ball Ground, GA
    Agreed with both Pio and K. To change the design over to a UL based one as is would entail all that you have surmised, along with adjusting the NFB system as well. You've just got too much B+ voltage to consider UL if you're going to keep the existing plate transformer, and IF you stay with fixed bias. Therefore, a conversion really all starts with first measuring the primary impedance of the OPT. If it's about 5K (which it should be in this design), then UL operation with cathode bias could be very nice if you then also switch to KT88 tubes. That way, the tubes will easily handle the voltage, the transformer will (likely) be the right impedance for cathode biased UL operation of the tubes (doubtful that it is for fixed bias UL operation), and then you can tweak from there. As for how good the results will be depends on just how good the OPT really is. Might be worth a try though.........

    Dave
     
  5. Brice

    Brice AK Subscriber Subscriber

    Messages:
    1,507
    Location:
    Hoboken NJ
    If I recall correctly, the impedance on those OTs is low, something around 3k.
     
  6. Scripted

    Scripted New Member

    Messages:
    8
    Location:
    Phoenix AZ
    I am definitely going to run KT88's. Let me measure the OPT primary impedance and see what I get.
     

     

    Please register to disable this ad.

  7. Scripted

    Scripted New Member

    Messages:
    8
    Location:
    Phoenix AZ
    So I just tested the OPT impedance and you were right Brice about 3K.

    Any other options?

    I imagine if I was dead set on doing this I would have to lower the B+ voltage like Dave said.
    Could you take some windings off the PT secondary to lower the voltage? or would this begin to be a futile exercise.
    Maybe this is the reason there are so many more modifications to the model 75 over 90.
    It would certainly be interesting to pull this off given the power supply is all tube and incorporates 0D3's.
     
  8. Pio1980

    Pio1980 AK Member Subscriber

    Messages:
    27,036
    Location:
    Eastern Bamastan
    It would be much easier if the opts had separate independent screen windings instead of the percentage tap points. One could then feed them any screen voltages safely appropriate for the overall design and parameters.
     
  9. gadget73

    gadget73 junk junkie Subscriber

    Messages:
    37,033
    Location:
    Southern NJ
    honestly I'd leave it as-designed and test the performance. You may find it doesn't need any modifications.
     
    drtool, Bill Ferris and Pio1980 like this.
  10. Brice

    Brice AK Subscriber Subscriber

    Messages:
    1,507
    Location:
    Hoboken NJ
    I would agree with Gadget. I looked at it once and I didn't like the overall design.
    It is complicated, has a unusual PS, and 3 coupling cap points, so 4 stage amp.

    It is very odd that Allen Organ produced such amp. I wonder if it was prior or after the 75.

    When you get it running, I would be interested of your opinons.
     
  11. gadget73

    gadget73 junk junkie Subscriber

    Messages:
    37,033
    Location:
    Southern NJ
    Doesn't seem to be that unusual. front end is basically a Williamson, but it does have two extra sections of a 12AY7 in front of the power amp as a mixer / preamp. Could probably stand to bypass that and feed input to the grid of the first 6SN7. FRom the 6SN7 to the end its pretty much a Williamson, but using fixed bias pentode output tubes instead of the usual cathode bias triode or UL config. PS isn't all that odd either, other than the 5v heater winding's center tap for HV takeoff. That should produce less ripple though, so not a terrible idea. Filter caps in series with divider resistors for voltage reasons. Not a fan of how they drew the PS just because its cramped and hard to follow but I suppose that can be forgiven.
     

     

    Please register to disable this ad.

  12. Scripted

    Scripted New Member

    Messages:
    8
    Location:
    Phoenix AZ
    Thanks gadget, I definitely have plans to bypass the 12AY7 and will probably take the input through a 0.10uf NP cap followed by a 1M grid leak resistor then a 47k resistor into the grid (pin #1) of the 6SN7.

    Since I am relatively new and want to explore all options before diving in what about using the output tubes in triode mode? My guess is the 580v B+ going to be the killer
    I have heard of amplifiers that have pentode/ triode switches that allow you to change between the two.

    If I stick with the original design should I consider the addition of screen grid resistors (pin #4) on the output tubes, regardless of the configuration? Anywhere from 100- 470 I was thinking
     
    Last edited: Oct 15, 2018
  13. Scripted

    Scripted New Member

    Messages:
    8
    Location:
    Phoenix AZ
    So far I have the chassis stripped clean and boxed in walnut. I relocated all the connections to the back of the amp and will be adding a single led power switch to the front. The power and filament transformer are getting a black coat while the OPT is getting a grey one. The only color to remain original will be the gold colored chassis which will be refreshed.

    IMG_5973.JPG IMG_5971.JPG
     
  14. gadget73

    gadget73 junk junkie Subscriber

    Messages:
    37,033
    Location:
    Southern NJ
    Screen grid resistors would be a good idea.

    Not sure about triode mode either, sometimes the voltage ratings for triode operation are different from pentode operation. I don't have the datasheet in front of me to know.

    If the amp ends up not sensitive enough, you could put something with a single triode in front of it. Williamson amps often need closer to 2v to get to full power, and not all old preamps can do that. Modern ones usually aren't an issue.
     
  15. Pio1980

    Pio1980 AK Member Subscriber

    Messages:
    27,036
    Location:
    Eastern Bamastan
  16. gadget73

    gadget73 junk junkie Subscriber

    Messages:
    37,033
    Location:
    Southern NJ
    something I would add now that I see a better copy of the schematic is a 10 ohm resistor between each output tube cathode and ground to serve as a current monitor. Looks like one version has an adjustment and the other does not. I'd make sure it has an adjustment.

    With the regulated screen setup it has, it should perform pretty well honestly.
     

     

    Please register to disable this ad.

  17. Scripted

    Scripted New Member

    Messages:
    8
    Location:
    Phoenix AZ
    Agreed. My version does not have the adjustable 5k pot in the bias circuit but will have to add it in. I might even consider dual bias pots, one for each output tube.
    Today I was curious to know the voltage of the secondary's on the PT. The tap used for the bias circuit measured 60vac with my mains around 117vac and seemed normal however the others are running around 800-0-800 vac which seems simply scary to me. I think its close to the limits of the 5R4GY's. Other than that I am trying to get a list together of the parts needed. Thanks for all the discussion.
     
  18. gadget73

    gadget73 junk junkie Subscriber

    Messages:
    37,033
    Location:
    Southern NJ
    Should be a little lower when loaded, but high power tube amps are inherently scary. Lots of voltage and relatively high current is involved.
     
  19. dcgillespie

    dcgillespie Fisher SA-100 Clone Subscriber

    Messages:
    9,781
    Location:
    Ball Ground, GA
    Remember, you're dealing with a true choke input filter system here, so the normal PIV requirements are not nearly as great (relatively speaking) as when a cap input filter system is used.

    Dave
     
    Pio1980 likes this.
  20. Scripted

    Scripted New Member

    Messages:
    8
    Location:
    Phoenix AZ
    Looking over a couple other amplifier designs and was surprised to find the Lafayette KT-550 and the Harman Kardon Citation V are both pentode with fixed bias. I am guessing these designs are pre and post Citation II respectively. Would be nice to trial all three of them !!
     

Share This Page