First Amp Build (Electronics portion)

Discussion in 'Amp Central Station' started by MrUnderwood08, May 21, 2019.

  1. MrUnderwood08

    MrUnderwood08 TDPRI Member

    Age:
    33
    Posts:
    26
    Joined:
    Mar 30, 2017
    Location:
    Prescott, Arizona
    I am finishing up the electronics portion of my first amp build... my take on a 6G3 Brown Deluxe with elements borrowed from Allen Amps, Rob Robinette, Deluxe Reverb, 6G11 Vibrolux and Cutthroat Down Brownie.
    IMG_4608.JPG IMG_4613.JPG
    Transformers:
    Power: Classictone 40-18028 (http://www.tdpri.com/threads/6g3-power-transformer-options-brown-deluxe.934472/)
    Output: Hammond 1750H
    Classictone Reverb Transformer
    Heybour Small Fender Choke

    Faceplate was off of ebay and the chassis was from Hoffman amps when they were having there closeout sale before they stopped selling chassis. I don't have a backplate and will either have to have one made or make one. Silkscreening isn't currently in my wheelhouse.

    Carbon Comp Resistors thoughout. 1 watt resistors are old stock allen bradley. All others are from Hoffman Amps. Capacitors are a mix of ceramic (Cera-mite if I could get it), Orange Drop polyesters (6PS)/polypropylene (715s), Jupiter Vintage and Sozo. Filter caps are F&T and bypass caps are Sprague atoms. Wire is push back cloth from Hoffman...


    IMG_4620.JPG

    TIP for others building an amp for the first time... buy extra wire. Alot of extra wire and in all the colors. It sucks feeling like your going to come up short or having to change your color coding method half way through because you don't have enough. You won't regret having extra and being able to leave leads long when building the board. I didn't have enough and used every inch of what I had.
    Features: I've added 3 tubes and rearranged things a bit.

    I designed my own layout combining portions of Fender amps and tayloring to what I wanted. I built the boards using hoffman eyelets and board stock. I went with black epoxy board for looks but If I had to do it over again I would go with the natural epoxy board material from hoffman and would strongly consider turrets. Since I was combonining portions of Fender layouts instead of engineering my own I went with eyelets (turrets tend to layout differently). The natural boards are semi-transparent which means leads running under the board are easier to trace. In the end natural looks just as sharp as black IMHO.

    I also cut the board stock down to make it narrower since I was planning on adding at least 1 tube in the front of the chassis. The board cut fine on a table saw with a carbide tipped blade. I located and drilled the holes on a milling machine but thats just because I had access. A drill press with a good eye and a good hand would do okay. If your building your own board and cramming alot into small spaces check with components to see how things are going to fit together before making the final product. I used cardstock for this. Also be careful how close you plan any 2 eyelets. In one spot I was too close together and had to flip an eyelet to the backside to prevent the top lips from touching... oops.

    If your doing alot of modifications and/or combining elements of different amps... have the authentic schematics and layouts handy during assembly as well as a hand drawn or computer generated layout that reflects your plan. I had a handful of errors in my plans that were sorted out as I assembled but it meant unsoldering connections. I'd much prefer to solder ONCE and be done.

    Double check you have EVERYTHING before you start. I didn't think about RCA jacks to go into my reverb tank for instance. Luckily Hoffman ships quick. I should have gotten wire then too. I had literally JUST enough... very uncomfortable.

    IMG_4625.JPG

    As stated the amp was roughly based on 6g3 but I also wanted some blackface and some tweed. Here is how I accomplished this:

    Screen Shot 2019-05-21 at 9.55.23 PM.png
    In the 6g3 V2A is a gain stage for both channels with mixing resistors before. V2B is a tremolo oscillator. I Instead I used V1 for channel 1 and V2 for channel 2 separating this gain stage. I changed the tremolo to the 6G11 which uses the second half of the tube as a cathode follower. Still a great bias vary tremolo.

    I prefer a bias vary tremolo instead of an optocoupler because of the sinusoidal wave form of the oscillation. The optocoupler is nearly a square wave in how it shunts the signal and doesn't sing the same to my ears. I did modify this tremolo circuit by changing the values on the board to .02-.01-.02 to slow the tremolo down. I subbed an LED for the 4.7K cathode resistor AND the 25uf bypass cap. In picture above its a red LED. Red has the lowest voltage and oscillates the strongest. It was too strong for my purposes and was causing some problems. I changed it for a blue and found that much more acceptable. I still want to try a clear.

    Screen Shot 2019-05-21 at 10.01.29 PM.png

    Channel 1 is the 6G3 bright channel. On my V1B split plate load I changed the 15K to a 27K to dump less gain to ground. I did this since I was using a 10K dropping resistor in the doghouse instead of 27K. I didn't want the lower voltages on both channels. I used a 10K grid stopper on V1B but could probably step up to a larger value to simulate the mixing resistors that would normally be in the circuit for the 6G3. V1A has 3k cathode resistor since I split the cathodes from the schematic and V1B shares a cathode resistor at 820ohms with V2B.

    For Channel 2 I intended to duplicate the 6G11 preamp. However the tone stack proved difficult with how I was running my pots. I'm using concentric pots (more on this in a moment) and a tapped pot wasn't possible... so I ended up changing the tone stack to a standard blackface tone stack. The rest of the values are inline with the 6G11 preamp (which looks alot like a blackface preamp).

    The 6G3 has 6 pots in front. Volume and tone for each channel and then tremolo speed and intensity. I didn't want to drill more holes and wanted the amp to look like a 6G3 (at least to casual observer). So I ended up using 3 DPDT mini pots and 3 concentric pots. I was excited to discover concentric pots and for those that don"t know a concentric pot two pots stacked on top of each other with independent control. A real space saver. Unfortunately the knobs are limited and rather ugly for my application and their is not a great assortment of values.

    I bought 3 CTS concentric pots as well CTS pots in the value I needed the concentric to be. Don't try to use a Alpha pot wafer in a CTS pot by the way. Alpha pots cannot be taken apart like a CTS can and even if you do manage to separate the wafer they won't work. Trust me I tried. Like a bonehead.

    Here is another thread with disassembled pictures http://www.tdpri.com/threads/fender-cts-concentric-pot-anatomy.436859/

    I bent tabs and disassembled the concentric pots. I removed the wafers and swapped out for the value that I needed. The wafers are the same but the concentric pots have a larger center hole in the wafer so drill out the materal carefully. Reassemble and test. Mine worked great. Love CTS pots.

    I also shortened the primary shaft that operates the bottom pot (the one that would use a white knob later). I only took .07" off each shaft on the lathe. The first one I wasn't paying attention and cut short but it worked out. Once reassembled I bushed the ends with nylon standoffs that I drilled out to be a interference fit on the shaft.

    IMG_4653.JPG


    I modified tweed amp knobs (chicken head) to operate the bottom portion of the concentric pots which again confusingly is the top pot on the inside. To do this I turned off as much material as I could on the lathe in a mandrel. This got me to the height I wanted. The rest was removed with a belt sander and then hand files and sand paper. Then buffed on a harbor freight buffing wheel to polish the plastic back to a muted shine.

    IMG_4652.JPG

    The pointer is below the surface so the recessed portion of the white knob sits down on top of it and then the little pointer can be used to turn the other pot.

    IMG_4655.JPG

    Under the speed pot for the tremolo is a 250K RAW control. This pot is wired as a variable resistor in series with the 6.8k mid resistor and increases the resistance to ground of the blackface tone stack inchannel 1 giving more tradition brownface 6G3 type tones when desired.

    Back to the circuit.
    After V1B and V2B each channel gets its own coupling cap and then the signals combine using 220k mixing resistors before going to AB763 reverb circuit. So two more tubes. 6 Tubes in total and they don't fit in the back of the chassis... and that is why I moved the V1 tube up to the front near the channel 1 inputs. This circuit is straight forward from the AB763 except I subbed a 1M resistor for the 470K attenuation resistor that acts as a voltage divider with the 220K to ground

    Screen Shot 2019-05-21 at 10.39.21 PM.png

    A 1M shunts 14% more signal to ground. Stock 470K is 68%, 1M is 82%. Not my math... got that from RobRobinnette's wonderful website. I did this because I've never found myself playing with reverb set past 5 or so.

    The reverb circuit does change the tone of the amp overall. It adds a gain stage and even with the reverb signal shunted to ground with the pedal there is additional signal handling going on that changes the amps tone. I decided to add a switch on the volume pot of channel 2 to remove the reverb circuit entirely from the signal path. The switch sends the signal from the 220k mixing resistors after V1b/V2b to either the 500 pf cap that leads into the reverb (reverb on, knob in) or to the input of the phase inverter (reverb off, knob out). On the other end the phase inverter switches between the 220k mixing resistors and the .1uf cap that ends the reverb circuit. When the reverb circuit is removed the 500pf is connected to ground through a 100k resistor. I'll take a photo of this switch and get it into the post. I forgot a jumper between pins on the switch at first so wait for the correct photo if your interested. The switch is up front and the actual reverb pot is on the back of the amp. I tend to set reverb to a level and then don't touch the knob unless its to turn it all the way down. The switch accomplishes that so I put the pot on the back.

    Phase inverter and power tubes are wired following the 6g11. The 6G11 has a .05uf input capacitor at the phase inverter instead of .01uf (6G3) and changes the shunt resistor to a 4.7k instead of a 1.5k. The negative feedback resistor and this resistor to ground are key to determining the amount of negative feedback. It is the ratio between the two that matters. I used a negative feedback switch to change between the 100k 6G11 resistor and 170k which approximates the negative feedback of the 6G3 when you account for the 4.7k shunt resistor. (Interestingly 82k would be black face value... 820ohm/47ohm = 82k/4.7k).
     
    Last edited: May 22, 2019
  2. Nickfl

    Nickfl Tele-Afflicted

    Posts:
    1,406
    Joined:
    May 24, 2016
    Location:
    Florida
    My goodness, that's some first build! Nice work.
     
    ecoast likes this.
  3. MrUnderwood08

    MrUnderwood08 TDPRI Member

    Age:
    33
    Posts:
    26
    Joined:
    Mar 30, 2017
    Location:
    Prescott, Arizona
    I moved the shunt resistor off the board and onto a presence pot wired as a voltage divider shunting the signal to ground through a .1uf cap This pot is the black knob under the intensity knob.

    Screen Shot 2019-05-21 at 11.15.38 PM.png


    Finally I added 470ohm 1w screen resistors and 1.5k grid stoppers to the power tubes.

    On the power supply side I added a Fender Small Choke and changed the filter cap board to the deluxe reverb/6G11. As previously stated I used 10K dropping resistors (6G11/Deluxe Reverb). The fixed bias supply is running off a dedicated center tap with 470ohm bias range resistor and a 47uf/100v cap to ground. I added a 10KL bias pot with a 22k resistor to ground. I am still experimenting with the bias and maybe a little cold. My el cheapo hand me down multimeter blew a fuse and went on the fritz for awhile but when it was working its resolution was not enough to determine an accurate plate current reading so right now I am biased at 22mA cathode current.

    Finally... and the part that gave me the most trouble thus far was the Cathode/Fixed bias switch.
    The first time I switched to the cathode bias side things went sour and I got a loud oscillation/ squeal that I had difficulty diagnosing. It was a combination of different things. Reverb tank under the chassis was too close to the transformer and humming loudly, the tremolo signal was oscillating to strongly and the bias was way too hot on the cathode bias side of the switch. Oh and my primaries on the output transformer were in the wrong polarity causing positive feedback. Add to this a output jack that was intermittently shunting to ground (switch and tip were bent too close together and the signal would drop) and it took me some time to get myself sorted.

    Once I swapped in the blue LED for the red, swapped the primaries, found the source of the signal interruption and figured out the other bugs I was able to get the cathode side of things to make a good sound. It sounded great too. I sat and played for 5-10 minutes happy to finally have tone on the cathode side... then I noticed one of the power tubes was starting to redplate. DOH... I hadn't checked bias for the cathode side yet. 270ohm with a 25uf/50v cap is where I started and that turned out to be 125% plate dissipation. I upped it to 300 which got me 114%. I know I am running a pair of 250 and 270 ohm in series (520 total) for 84% dissipation. I think I need to be around 400ohms and I'll have to order it but for now at least I'm not going to break anything.



    I'm still sorting bugs out so I don't want to show more detailed photos in case someone wanted to trace out connections that may not be right yet. I changed resistor and connection on the bias pot for instance from what I have pictured above. I am also working out the bias for the cathode side.

    As is right now the switches are working as intended. Sounds good but I'm still feeling it out and refining.

    I am getting 430v at the powertube plates. Higher than I anticipated. The power transformer is a Classictone 40-18028 wired with 300-0-300 primary voltage. It measured at 318-0-318 unloaded at 119v from the wall. Rectifier is a Genalex Gold Lion. I may try a different GZ-34 when I get a chance. I was hoping for closer to 400v. On the other hand when switched to cathode bias I am getting 360v plate to cathode and when biased hot it sounded stellar.

    My soldering isn't perfect. It was better but mistakes were made and components had to be unsoldered/changed as the learning curve caught up with me. I'm using a Hakko FX-888D soldering station which I HIGHLY recommend. Great iron. For me it was that or a weller and I found this video helpful in deciding...



    Oh and something that I was not 100% confident on... the tremolo is working on both fixed and cathode bias. The wire from the tremolo intensity that normally goes to the bias pot goes to the switch first and gets grounded in cathode bias mode removing the fixed bias supply from the circuit. It does have some noise that I am still chasing (regardless of fixed or cathode bias) but it maybe as good as it gets right now. The nature of bias vary tremolo. In cathode bias mode the intensity has to fight the cathode so the pot has to be turned to a higher setting than the fixed bias side but its working so I'll take it.

    I need to hear the reverb with the tank and chassis in the cabinet but it sounds alittle weak right now. I may switch the attenuation resistor down to 680k or even the stock 470k value. Speaking of cabinet I posted the cabinet build last week. http://www.tdpri.com/threads/first-cabinet-build-for-first-amp-done-hide-glue.949776/

    I'm leaving for trip in a week so I'll get done what I can before but I'll revisit this thread with the final details. Schematic, layout and pictures once its all sorted.

    IMG_4640 copy.jpg
     
    Last edited: May 22, 2019
  4. ecoast

    ecoast Tele-Holic

    Posts:
    711
    Joined:
    Jan 6, 2017
    Location:
    NWNJ
    Good job fella
     
  5. thechad

    thechad Tele-Meister

    Age:
    33
    Posts:
    229
    Joined:
    Dec 30, 2018
    Location:
    Kamloops
    Your first build puts all my builds to shame! Nice work! Looks like a great amp you have made yourself!
     
  6. printer2

    printer2 Poster Extraordinaire

    Posts:
    6,245
    Joined:
    May 24, 2010
    Location:
    Canada
    Your second will have a flux capacitor among other things I presume. ;)
     
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.