1. Win a Broadcaster or one of 3 Teles! The annual Supporting Member Giveaway is on. To enter Click Here. To see all the prizes and full details Click Here. To view the thread about the giveaway Click Here.

UL Bassman Mods

Discussion in 'Shock Brother's DIY Amps' started by Fred Mertz, Jun 17, 2016.

  1. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    8,344
    Joined:
    Dec 29, 2012
    Location:
    United States
    Yes. For the lower "Double Input Jacks" you need to connect the right jack's G terminal to the ground bus too.
     
  2. Commodore 64

    Commodore 64 Friend of Leo's

    Age:
    45
    Posts:
    4,292
    Joined:
    Mar 1, 2010
    Location:
    Kent, OH
    OK, the amp is together.

    1. The resistors to ground from the rectifier bridge roasted. So that method of dropping voltage may not be ideal; however, to rule it out...
    2. I wired the bridge rectifier directly to ground.

    Turned it on, and tubes started red plating right quick. Best measurements observations I could get before turning it back to standby:

    1. 450V on the plates, and decreasing steadily
    2. 250MV across the 1-ohm cathode resistor to ground
    3. Without tubes in, the lightbulb limiter works as it should.
    4. With tubes in, as soon as you hit the standby switch, the bulb glows brightly.
    5. Bias voltage is -55 before the bias pot, and maximum I can at the grid get is -19V. (Lar Mar seems to give me about 1V of range, BTW)

    So I'm drawing too much current, for sure. Could it be as simple as not enough bias voltage? And here's where I get to show my complete noobish...the resistor in series with my bias pot is 18k. Do I need to raise that or lower it?

    [​IMG]
     
  3. Commodore 64

    Commodore 64 Friend of Leo's

    Age:
    45
    Posts:
    4,292
    Joined:
    Mar 1, 2010
    Location:
    Kent, OH
  4. Wally

    Wally Telefied Ad Free Member

    Posts:
    39,020
    Joined:
    Mar 17, 2003
    Location:
    Lubbock, TX
    It matters to some extent what voltage you have before the bias pot, but you really want to know what voltage is being applied to pins 5, right? What is coming out of the pot? What is hitting the bias feed resistors? What is at pin 5?
     
  5. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    8,344
    Joined:
    Dec 29, 2012
    Location:
    United States
    Yes. You need more negative voltage on the grids.
     
  6. Commodore 64

    Commodore 64 Friend of Leo's

    Age:
    45
    Posts:
    4,292
    Joined:
    Mar 1, 2010
    Location:
    Kent, OH
    Wally, -18-19V on the grids.

    Also, all I have is switched mono jacks. I used them for the speaker jacks. Do I just wire the common to both the ground and the shunt?
     
    Last edited: Apr 19, 2017
  7. Wally

    Wally Telefied Ad Free Member

    Posts:
    39,020
    Joined:
    Mar 17, 2003
    Location:
    Lubbock, TX
    I don't think you want to use two shorting jacks. I would suggest wiring up only one Jack until you get a non-shorting Jack to use for the Ext Spkr Jack.
    Re: the bias voltage....you might want more negative voltage prior to that bias pot. That would entail changing the first resistor in that circuit....not the 18k. Then, that series resistance can be manipulated if needed.
     
  8. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    8,344
    Joined:
    Dec 29, 2012
    Location:
    United States
    Leave one of the two jacks without the jumper from shunt to ground. Use that one as the aux speaker jack.
     
    Wally likes this.
  9. Commodore 64

    Commodore 64 Friend of Leo's

    Age:
    45
    Posts:
    4,292
    Joined:
    Mar 1, 2010
    Location:
    Kent, OH
    I've got it running. Need to continue troubleshooting. But won't have time until tomorrow. Couple of notes.

    1. I get sound on both channels. Not very loud, and very distorted, with a lot of fizz.
    2. FM's B+ method does work, but the resistors get very, very hot. I suspect this isn't going to work long term.
    3. I will probably need to look at running ~500V if I don't drop voltage. B+ is 520 unloaded without the resistor to ground off the diode bridge.
     
  10. Commodore 64

    Commodore 64 Friend of Leo's

    Age:
    45
    Posts:
    4,292
    Joined:
    Mar 1, 2010
    Location:
    Kent, OH
    OK, grid stopper/coupling cap network on the 3rd gain stage for each channel (the ones I attempted to design myself) are screwed. Grid stoppers weren't referenced to ground, and the coupling caps weren't the right spot. Think I got it sorted, but I'll have to wait til tomorrow to make some circuit changes.
     
  11. Commodore 64

    Commodore 64 Friend of Leo's

    Age:
    45
    Posts:
    4,292
    Joined:
    Mar 1, 2010
    Location:
    Kent, OH
    I couldn't quite figure out how to use the RGKeen mosfet method since I'm using a diode bridge rectifier. The centertap of the PT is taped off, not connected. It's 180-0-180 PT that Fender ran at 520V B+ in the original circuit. Fender used a voltage multiplying circuit by running the CT to the junction of the series filter caps in the first filter stage. http://www.geofex.com/Article_Folders/mosfet_folly/mosfetfolly.htm

    I'm open to suggestions on how to use a 180-0-180 PT to get ~450V B+. I don't see how the FM's resistor-to-ground-bridge-feedback network can withstand the current draw of the amp which is probably going to be over 200 mA when it's all said and done.

    I've been abusing the hell out of a pair of Tung Sol resissue 7581s trying to get the bias situation sorted. SO without thFM resistor, my loaded B+ is 486V. With my bias at -44V, I get 120 mA per tube which puts me around 60W dissipation. Max bias I can get is -44V with a 33k resistor off my 10k bias pot. Probably need to change that value, or lower the 18k bias dropping resistor. Lotta distortion, not much volume, but signal does appear to be getting all the way through the circuit and out to the speakers.

    I'm assuming that 486V is sagging quite a bit, drawing 120mA per power tube. Not sure what kinda B+ I'll be looking at once I get the bias sorted.

    But I don't think the 7581 can take being redplated too much longer, so I better figure out what the heck I'm doing fast, Heh.
     
  12. Commodore 64

    Commodore 64 Friend of Leo's

    Age:
    45
    Posts:
    4,292
    Joined:
    Mar 1, 2010
    Location:
    Kent, OH
    -58V on bias will give me about 14watts of plate dissipation at 500V plates. So that's biased pretty cool (48-50%)

    Until any signal goes through. Soon as I crack a vol pot and put a signal in, the tubes redplate. And B+ drops to around 467.

    So I'm drawing huge current I guess.

    So on the mosfet follies, I assume I could just do it on the center tap of the diode bridge. So would a 1/2w zener be OK to use: http://www.mouser.com/ProductDetail/Nexperia/BZX79-B43113/?qs=sGAEpiMZZM...

    And, I don't know what to search for as far as a 75W mosfet.

    for reference: http://www.geofex.com/Article_Folders/mosfet_folly/mosfetfolly.htm
     
    Last edited: Apr 25, 2017
  13. Commodore 64

    Commodore 64 Friend of Leo's

    Age:
    45
    Posts:
    4,292
    Joined:
    Mar 1, 2010
    Location:
    Kent, OH
    Yes, indeed. I went and rewired the bias into a Hoffman:

    [​IMG]

    That sorted it.

    So I biased it at 9W per tube. At 30%. 503V on the plates and screens. The AA864 channel is so loud at 1 on the volume knob that the neighbor came out and had words with me. After 1 strum. :D LOL. It is so fricken loud, I'm afraid I just blew the G12 Gold Celly with one big open G. Holy crap. Seriously. I think the bypass caps are going to have to come off of a couple of stages.

    And the Hiwatt (CP103) Channel needs work. it's a distorted mess, high pitched squealing and feedback, but the signal is getting through.

    So now I can start the fun part. Tweaking the tone.

    edit: realized the cab had a 15 inch Emi legend in it, so it's fine. And I don't even think the amp was at 1. Good news is, the Lar Mar works really well.
     
    Last edited: Apr 26, 2017
    jonrpick and robrob like this.
  14. Commodore 64

    Commodore 64 Friend of Leo's

    Age:
    45
    Posts:
    4,292
    Joined:
    Mar 1, 2010
    Location:
    Kent, OH
    Voltages:

    B+ (A): 508V
    B V4: 472V
    C V3 V2B: 432V
    D V2A, V1: 396V

    V4
    1 265
    3 120
    6 251
    8 120

    V3
    1 285
    3 2.1
    6 233
    8 1.7

    V2
    1 282
    3 2.0
    6 231
    8 1.9

    V1
    1 262
    3 2.0
    6 257
    8 2.0

    I'm worried about Hk on V4, 120V seems awfully high on the cathodes. I think a bigger dropping resistor between A and B is in order. Bring all the preamp voltages down a fair bit won't hurt.

    No real clean to be had on Channel 1, so I think I'm sending square waves pretty early. Will hook up Signal Generator and Oscilloscope tomorrow.

    Channel 2 is just high pitched squealing
     
  15. Commodore 64

    Commodore 64 Friend of Leo's

    Age:
    45
    Posts:
    4,292
    Joined:
    Mar 1, 2010
    Location:
    Kent, OH
    Good god what a nightmare.

    So, high pitched squealing on Channel 2 was a bad tube.

    Caused some smoke. 2x. Once fried my humdinger. The other time, well, signal went away. No signal on plates. Horrific mess on the phase inverter on the scope. So I pulled the OT, built an RG Keen short tester, and the OT was still good (tested 2 other suspect trannies while I was at it, they are bad.)

    Replaced entire lar mar assembly, took the opportunity to use shielded rg174. No change.

    Found 1 Khz square waves on my power tube cathodes...checked the 1-0hm resistors to ground. 1 was open. The other was 500k. That explains why the power tubes weren't matching very well, LOL.

    So now, I've got it all sorted. Waiting for a new humdinger pot, got the heaters to 100ohm resistors to ground for now.

    I believe there is WAY too much gain. My first channel has 3 preamp triodes, all fully bypassed. So I'm thinking that needs to change. Some pics you may or may not find interesting:

    The transformer tester. Pushbutton and neon bulb.
    [​IMG]

    Input signal 60mV peak to peak. 10x Probe.
    Gain Stage 2 (1 Khz tone)
    [​IMG]

    Gain Stage 3 (1 Khz tone)
    [​IMG]

    Phase Inverter Output (1Khz tone)
    [​IMG]

    That's all with the volume on 1, Lar Mar dimed. 12AY7 in all preamp stages except the phase inverter (12AT7).

    I don't mind if my Hiwatt Channel looks like this, but I believe I'd like the Fender Channel (this one) to feed the power tubes something a little less orthogonal.
     
    Last edited: May 1, 2017
  16. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    8,344
    Joined:
    Dec 29, 2012
    Location:
    United States
    Consider turning your high gain channel's second gain stage into an unbypassed cold clipper like the Marhall 2204. You'll get less gain but add sweetness to the overdrive tone. Also note how they follow that cold biased stage with a hot bias stage to keep the asymmetrical distortion going:

    [​IMG]
     
  17. Commodore 64

    Commodore 64 Friend of Leo's

    Age:
    45
    Posts:
    4,292
    Joined:
    Mar 1, 2010
    Location:
    Kent, OH
    That's pretty interesting. Simply swapping out a larger cathode resistor on stage 2 and a lower one in stage 3? Seems easy enough. So, I might leave a bypass cap on one of those stages since I'm not using a cathode follower into my TMB, I could probably use the extra signal. Both channels on my amp basically go from the first triode, into the tone stack.

    Your Marshall schem there, puts the tone shaping after all the preamp stages. Wonder what the practical tonal effects of this are.
     
    Whatizitman likes this.
  18. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    8,344
    Joined:
    Dec 29, 2012
    Location:
    United States
    Leave the 10k cold clipper unbypassed.

    Having the tone stack late after all the preamp distortion is generated allows the tone stack to affect all the harmonics created during clipping. High gain amps need some type of late tone control like this, or a presence control or Vox style cut control. I like to add a combined Vox Cut/Master Volume control:

    https://robrobinette.com/Generic_Tube_Amp_Mods.htm#Vox_Cut_Control

    [​IMG]
     
  19. Commodore 64

    Commodore 64 Friend of Leo's

    Age:
    45
    Posts:
    4,292
    Joined:
    Mar 1, 2010
    Location:
    Kent, OH
    So, recapped the Heathkit signal generator (was having issues). Now, feeding a 200mV signal, I've got 8V coming out of gain stage 1. Gain of 40. Seems about right.

    Waves seem to clip and max at about 90V, which is easy to achieve in gain stage 2.

    There are 3 gain stages, + the LTP PI. I think I've got way too much signal. Even with the variations of the plate driven TMB on both channels.

    My LP copy, with humbuckers will not do a shred of clean on either channel, at any setting.

    Tubes biased at 63% ([email protected]).

    There's no NFB on this amp, running the UL OT, I figured I'd start with no NFB, but perhaps some is needed?

    Should I just clip all the bypass caps in the preamp?
     
  20. Commodore 64

    Commodore 64 Friend of Leo's

    Age:
    45
    Posts:
    4,292
    Joined:
    Mar 1, 2010
    Location:
    Kent, OH
    I clipped the bypass caps on the 3rd gain stage on each channel. They shouldn't have been there. Now it's behaving better. Still a hugely powerful amp though. But I think it's sorted enough that I can put it all back together and put some hours on it to see how it behaves.

    However; i might have 1 more go at Fred's voltage dropping scheme, now that I seem to have the power draw issues sorted.
     
    Last edited: May 12, 2017
    jonrpick likes this.
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.