A small, single-ended amp with reverb and tremolo

Discussion in 'Shock Brother's DIY Amps' started by Snfoilhat, Mar 14, 2019.

  1. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,095
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    I began work on this amplifier yesterday. Cleaning and testing old parts. Walked into a store and bought one capacitor. You know, hitting the ground running :rolleyes:


    A cascode input for some flavor like a small-signal pentode (EF86) -> a cathode-follower stage for clipping and driving the tone stack -> a James type bass and treble control with some kind of mid-shift TBD -> a common anode mixing stage to mix the reverb and drive the power tube -> either a 12W 6BQ5/EL84 for 4-5W output, or a 6W 6DL5/EL95 for maybe 2W output.

    A medium-impedance reverb tank is driven by Merlin's SRPP reverb driver (no transformer) and recovered in a normal gain stage.

    So it's like a Vox AC4 Top Boost Reverb... ... ...o_O

    The 6DL5 appears to be even more sensitive to high voltages than the 6BQ5, so I'm not sure if the transformers I have available are going to work. Every time I think about building around that tube, this issue crops up. But I want to hear it someday. 1-2W power amps get a lot of attention around here, and I think the designs look and sound great, but they are all based on triodes. This would be a chance to hear a pentode and judge if it's worth the extra trouble.

    Single-ended-reverb-amp_II_6BQ5.png
    Single-ended-reverb-amp_II_6BQ5.png
     

    Attached Files:

    Last edited: Mar 14, 2019
    sds1, Treynor, RLee77 and 3 others like this.
  2. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,095
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    IMG_2640.JPG
     
    gigante likes this.
  3. gigante

    gigante TDPRI Member

    Posts:
    76
    Joined:
    Mar 7, 2016
    Location:
    Kansas City
    I'm excited to see where this goes.
     
    Snfoilhat likes this.
  4. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,095
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    IMG_2641.JPG
     
  5. tubeswell

    tubeswell Friend of Leo's

    Posts:
    2,368
    Joined:
    Jul 1, 2008
    Location:
    NZ
    Be aware that you have several gain stages in the dry-signal path and only 1 in the reverb recovery path, so you will need to think about the dry/wet mixing levels.
     
    Snfoilhat likes this.
  6. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,095
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    Yeah! If the tone stack isn't lossy enough, maybe a series resistor and a dedicated grid leak to form a voltage divider right before V5 to balance better with the wet channel.

    ***

    First draft of the controls and graphics
    IMG_2642.JPG
    Since I'm taking so many cues from vintage Vox, I've appropriated the old name of their bass cut control, 'brilliance.'
     
  7. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,095
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    Mini toggle switches are probably superior in every way to these clunky plastic and phenolic slide switches, but I love the 1960s Fender visual aesthetic.

    IMG_2643.JPG
    I'll have to pull these and shellac the front and rear control plates before going much further with the build, but I wanted to see how the graphics file matched up to the chassis. Not perfect but not bad.

    Since I'm doing chassis work today, I cut a new 1/2" by 7/16" oval opening in the rear for the fuse holder, closer to the spot the power cord strain relief sits. I don't like how the vintage Fender location is toward the center of the chassis and sits right above a tube socket.

    Thanks!
     
    D'tar and tubeswell like this.
  8. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,892
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Love it. Looks great.

    Did you say what type and where you got the chassis? Also, in broad strokes, how do you align your graphic layout with it? Impressive.

    I doubt you’re dying for suggestions, but if it were mine I’d try to visually center the birds 'n' text combo in the logo space at the right. Sorry, just my designer OCD inner self talking here... ignore at will, and carry on.


    Sent from my iPad using Tapatalk
     
    Snfoilhat likes this.
  9. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,095
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    Princeton Reverb repro chassis from Zach Hunter/that metal shop in Arkansas. This is either the first or second I bought a couple years back. Now all my 2017 amps have been parted out to try new ideas and save space and $.

    Since the original chassis work is so good (fitting w/ Mojotone's parts is part of their business model), I just confirm the placement of the centers of all the 3/8" holes with a ruler (measuring from the left = 0") and lay them out on a document 18.25" * 1.90". Add layers for the vertical stripes that go with black panel style and that type of knob, or the 1-10 labels for brown panel style with radio knobs. Controls are all along a single horizontal line which I think is 1.00" from the top. Since the document is built from layers (the white layer that will form the pinstripe, the rounded rectangle that can be any color background, the hole placements, and the labels and stripes and birds), it's easy to try out new stuff cheaply.

    It's the 'custom' parts of the chassis that I cut that are tricky, because the holes I drill are never exactly where they should be:eek:

    Princeton Reverb has six controls with spacing of 1.75". When I added an additional control my drill bit wandered about 1/16", as you can see by the crescent of white on the Intensity control. I didn't correct for that in the graphic and it's going to look a little out of center. If I want to get it closer to perfect, I can use the face plate as a guide and ovalize that hole with a round file, nudging the pot closer to center. Which now that I've typed it I suppose I should do:D. Alternatively, for this unique chassis with the mis-drilled hole, I could just open the graphics file and nudge the Intensity control and all its associated text 1/16" to the left.

    IMG_2644.JPG
    Power switch always moved to the front beside the indicator light, Rivera-era style.

    New fuse placement. I build my cabs with a simple horizontal upper back door, not the (admittedly cool) bow-tie shape that Fender used, so the extreme left and right of the rear chassis aren't obscured.
    IMG_2645.JPG

    Reusing parts blues. Leads are too short, tube sockets all have to be carefully desoldered. Can't all the parts of an amp just be modular and snapped together with little quick connectors?:twisted:

    Thanks!
     
    Last edited: Mar 15, 2019
    BigDaddy23 and King Fan like this.
  10. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,095
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    I'm dying for suggestions!

    Here is an example of how the 6DL5/EL95 dream keeps slipping away from me with all these 6V6-centric parts I have on hand.
    Single-ended-reverb-amp_II_6DL5_working_01.png
    360VDC is such a sweet place for B+ to be for a 6V6. A Vp-k of something like 330V. Great. But the 6DL5 wants to see something like 275V. Here's a question for y'all. This is a single-ended power amp that is going to draw current pretty constantly, right? So if I can handle the waste heat, can I just drop 85V across some huge resistor and not impact the performance of the amp? Is this no different from a string of zener diodes constantly shedding power out of a OT center-tap, or does it pose some problem I'm not seeing?

    The 6BQ5/EL84 can handle a little more voltage, but I'm still going to have to drop a bunch from B+1 if I want to use this transformer. And I do. Worst-case I could use a 6V6GT, but it wouldn't be as cool!:rolleyes::rolleyes::rolleyes:

    ***

    Eagle-eyed schematic readers may note that I've shown how I hope to make the Brilliance bass cut work - a switchable partial/full 'screen' bypass capacitor on the cascode.
     
    King Fan likes this.
  11. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,892
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Thanks for all the detail about the chassis and esp. the faceplate. Yeah, drifting holes are my lot in life too. My kingdom for a drill press.

    As far as resistor v. zeners, if I understand, always doubtful, both waste heat, but zeners don’t add sag. A resistor is super easy to try out anyway.
     
  12. tubeswell

    tubeswell Friend of Leo's

    Posts:
    2,368
    Joined:
    Jul 1, 2008
    Location:
    NZ
    Looks great - and practical
     
    Snfoilhat likes this.
  13. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,095
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    Thanks!!



    Well, it wouldn't be a tinfoilhat build without a bunch of dubious running part substitutions...

    I need a handful of 1M linear pots and have instead audio tapers.

    IMG_2647.JPG
    I want to humbly suggest to all our kit-building shock siblings that kit suppliers don't necessarily include every part you might use, and for example that this fuse holder lead dress is safer than what I have seen again and again, for the cost of one 2 cm length of shrink tubing. One less thing to bump your multimeter probe (or finger) onto while cavorting in your amp chassis.
    IMG_2648.JPG
    IMG_2649.JPG
    Speaking of safety, that enormous bolt that my rectifier sub-board is currently attached to is temporary for getting the power supply up and running. It will be permanently anchored with nylon hardware eventually.

    Things are shaping up.
     
    BigDaddy23 likes this.
  14. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,095
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    Power supply tests OK so far. I think the yellow control plate needs a new color jewel light, though. Very important detail. Apologies to UCLA, UC-Berkeley, University of Michigan, Murray Middle School (Martin County Florida!) and anyone else. I'm not into blue and gold:p

    IMG_2653.JPG

    For a little cleaner layout, I moved the heater elevation circuit from B+1 to B+4. You can see the resistors and the small blue electrolytic capacitor on a terminal strip. I'm seeing about 30 VDC on the heater center tap now, which ought to be just fine. I did a few more calculations of what I expect the different tubes to draw for current and settled on a 1.5k - 5W resistor to drop voltage between B+1 (unused) and B+2 (OT primary to power tube plate). With the additional voltage drop provided by the OT itself I'm hoping to get a manageable plate voltage.

    I don't really enjoy the next step. I have to decide if I want to build point-to-point (ugh) or on a piece of fiberboard with eyelets (ugh). PtP is currently leading in my mind, but I need to lay the parts out physically and get a better idea of just how messy it is going to be. Eyelet boards sure do look better after they are done, but all the flying leads are their own kind of mess.

    I have to anticipate some number of revisions since this design is unproven. I'm confident in the broad outlines since I built http://www.tdpri.com/threads/falco-columbarius-amp-build-reimagined-ac-4.903575/ last year and it's working like a charm, but I imagine I'll be tuning the gain and loss between stages and the tone stack for a few weeks, changing out parts.
     
    Last edited: Mar 18, 2019
  15. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,095
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    If anyone new to amp building is silently following along, I'll point out that this is how and when I do input wiring--before the chassis gets any more full of stuff.

    1M grid leak soldered to the Input 1 jack and Input 2 tied to Input 1 just like a vintage Fender layout, but I run two separate lengths of shielded cable from the jacks to a small terminal strip at V1. From there, two separate grid stopper resistors soldered right to pin 2. I don't really know if there is any advantage to this over tying the grid stoppers to the jacks and running one piece of shielded cable straight to pin 2, but it's how I do it.

    Then I insert and instrument cable into each input jack and measure resistance tip to sleeve and tip to pin 2. If all four measurements come out OK, then I don't worry about faulty input jacks for the rest of the build.

    Since the input jacks are my one and only place in the circuit where the circuit ground is bonded to the chassis, I'll add a dedicated short length of wire from the jack sleeve to a ring terminal to a small bolt on the chassis, instead of relying on the jack's own threads to give a good electrical connection throughout its service life. Technically that makes three places where the circuit touches the chassis, but they are all very close to one another, connected by very short leads, and shouldn't be carrying any current, and so I believe it is highly unlikely any kind of force will make ground-loop current flow through and cause noise.
     
    Last edited: Mar 18, 2019
  16. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,095
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    Instead of sharing the in-progress terminal strip disaster, here is a peek a future amp TBD. The 'bright' switch cutout was easier this time, and is almost square:D
    This chassis was a 6G15 two years ago, then a 6G15 with a built-in power amp, then a different built-in power amp, and floor is swiss cheese from so many mounting holes for different parts. Maybe something simple...

    IMG_2654.JPG


    Thanks!
     
  17. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,095
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    Maybe a minor course correction. Been playing with Duncan Tone Stack Calculator and am thinking about this. Maybe too subtle, but maybe not. A mid-shift control for a Fender-like tone stack.
    DuncanTSC_Fender-mid-shift-control.png

    IMG_2657.JPG
     
  18. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,095
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    Point to point progress
    IMG_2660.JPG
    Only 2 triodes left to wire, the LFO and reverb recovery at V4. Plenty of space. I decided to try the modded Fender tone stack (above) instead of the James stack w/ incorrect pots. My tastes have swung back toward scooped EQ over the last year and I am a loyal TDPRIer, poring over Leo's schematics by red jewel light :p
     
    thechad, gigante and Nickadermis like this.
  19. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,095
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    Speaking of Fender, I've been slowly working on my own contribution to the miniaturization of the classics. I'm not ready to commit to building, but I've been thinking about @robrob 's work in removing second channels and reducing power amps. Here is what I'm thinking:

    The 12AY7 in V1 seems to me like an important part of the gain structure that a lot of builders overlooked during that period when they were both out of style for being lower gain and were they also out of production for a few decades?

    I don't want a second channel so that frees up V1b. I'd rather have a LFO for tremolo than a cascaded or parallel gain stage, but an LFO requires a high mu triode. So I go back to 12AX7 for V1 but am thinking of a split plate load in order to match the gain of a 12AY7.

    Ampbooks cathode bypass cap gain calculator tells me to expect the 12AY7 to have a gain in the original 5F6A circuit of about 34, compared to 59 with a 12AX7 in the same spot.

    So the split load works out to 42k:58k in place of the 100k plate load resistor.

    Immediately after the volume control there is the matter of the channel mixing voltage divider, which should send something like 25-50% of signal to ground depending on how channel volume 2 is set. That's a small enough amount that I'm thinking maybe eliminate that voltage divider, eliminate the extra loading on the gain stage, and just let the ordinary 1M-A volume pot have a little more gain on tap in its upper reaches than a true 12AY7 equipped Bassman, a little less than a 12AX7 subbed Bassman.

    I keep saying Bassman, Fender, Bassman, 5F6, (let's not forget the JTM45 w/ trem!) and then show a schematic where the power tubes are cathode biased:eek::eek::eek:

    Sorry, it's the drawing I had on hand. Biasing TBD. I have the transformer set so it really might happen. Thanks!!

    Bassman_Tremolo_Amp_6BQ5s.png

    IMG_2532.JPG
     
    FenderLover and robrob like this.
  20. thechad

    thechad Tele-Meister

    Age:
    33
    Posts:
    265
    Joined:
    Dec 30, 2018
    Location:
    Kamloops
    Looks good. I am amazed how tight those heater wire turns are! I've often wondered how people keep track of what side goes to which pin when they use the same color for both runs....
     
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.


  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.