Modified 5f10 Layout - check over?

Discussion in 'Shock Brother's DIY Amps' started by jsnwhite619, Jan 11, 2020.

  1. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    35
    Posts:
    2,772
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    I was messing around with this on paper and wanted some more experienced eyes to look it over. The only real changes are adding an extra eyelet on the board for the extra V2 grid stopper, and I moved & changed the NFB. My idea is doing it like a Princeton Reverb and making the bypass cap switchable, the ground side being on the switch.

    The plan would be regular Harvard clean, then the gain boost on a switch or single button footswitch & RCA jack, like a lead boost. But, keeping NFB in the circuit - I've tried it with no NFB and it's just painfully bright and too rowdy.

    Thoughts? 5f10 modified.jpg
     
  2. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    35
    Posts:
    2,772
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    Stock layout.

    harvard_5f10_layout.gif
     
  3. chas.wahl

    chas.wahl Tele-Meister

    Posts:
    108
    Joined:
    Dec 8, 2012
    Location:
    NYC
    I'm no expert by any means, but there doesn't seem to be any problem with what you're proposing. Years ago, a switchable bypass was part of Dave Hunter's design for the "Two-Stroke" in one of his books -- 5F2-A with single or twin (in parallel) 6{K|V|L}6s*.

    Note that while Fender often (usually?) used 25 uF cathode bypass capacitors for "fully bypassed" gain stages, it's possible to use a lower value to bypass only frequencies over a shelf value, for more treble gain, less bass (cf. Merlin Blencowe's article on this in his Preamps book). Hunter's example has a switch in there for either 25 uF or 0.47 uF. In fact, you could use a 3-position spdt on-off-on switch, following @robrob Robinette's switching mantra, to have one of two bypass values, or no bypass (center).

    Screen Shot 2020-01-12-0926.jpg
    * note that while the layouts I can find online for Dave Hunter's amp show only one 6V6, the one in his Guitar Amp Handbook shows two, and he even says that other similar octal tubes can be used.

    references for Merlin Blencowe's info:
    http://valvewizard.co.uk/ChoosingBypassCaps.pdf
    http://www.valvewizard.co.uk/Common_Gain_Stage.pdf (useful graph on p. 26)
     
    Last edited: Jan 12, 2020
    jsnwhite619 likes this.
  4. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    35
    Posts:
    2,772
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    Thanks. Yeah, my pencil layout wasn't really clear, but I did put down 2-5uf on that V2 cap instead of 25uf for the reasons you mentioned. I have that Hunter book, but haven't taken it off the shelf in a while. Will have to look at it again.

    Sent from my moto g(6) using Tapatalk
     
  5. chas.wahl

    chas.wahl Tele-Meister

    Posts:
    108
    Joined:
    Dec 8, 2012
    Location:
    NYC
    Take a look at the last link I posted, p. 26 -- there's hardly any difference between 25 uF and either 2 uF or 5 uF. The 2 uF is "half-bypassed" at about 65 Hz, and 5 uF at about 25 Hz.
     
    jsnwhite619 likes this.
  6. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    35
    Posts:
    2,772
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    I finally had some time to get back to this project. In the meantime I've had time to think about a few things with it.

    • I think I'm going to finally try the paralleled triode for V1 with the spare half of the 12ax7. Basing it on what I see with Lite IIb and one of the Matchless amps (I think), with normal shorting jacks I should have one channel normal & one parallel without any switches to add.
    • With that in mind, I'll probably play with the V1 & V2 bypass values. Initially, I was going to go with the normal Fender 25uf V1, then go small on the modified V2 stage for a treble boost -- .68uf-2uf range maybe to start. Now, I wonder if I should take the opposite approach and with the parallel V1, start with a smaller value bypass there to keep it cleaner up front, then hit it with a 5uf-25uf bypass on V2. Also, 12ax7 & 12ay7 are both on the table for V1 depending on what works best.
    • From advice in another thread that was active at the time I started this one, I've dropped the P.I. input cap from .02uf to .01uf in hopes of cleaning up the blocking distortion that is always present when any of these style amps are cranked or humbuckers. I already have a dedicated eyelet on the board to experiment with a grid stopper there. It's not anything I particularly notice in person, but when I record it, I can always hear it in there.
    • Kinda limited room on the chassis for switches, so I think I'm going to forget the "boost cap" switch on V2, and use a mini switch for a NFB options. I liked how that Princeton Reverb ended up last year, and since I'm stealing the NFB style from it, I think the same 3-way mini-switch will be cool there.
    • Up in the air on 5y3 rectifier or 5u4. I'm leaning toward 5u4 so that maybe I can tweak the circuit more for the desired effect and know that I'm really hearing the preamp differences and not taking the rectifier into account. Maybe tune it for a 5u4, then figure the 5y3 to be a "low power" option.
    Any thoughts & feedback are welcome.
     
  7. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    35
    Posts:
    2,772
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    Shot of the board.[​IMG]

    Sent from my moto g(6) using Tapatalk
     
    milocj likes this.
  8. cobaltu

    cobaltu Tele-Meister

    Posts:
    149
    Joined:
    Feb 19, 2010
    Location:
    New York, NY
    I would go with this route. In my experience, it is better to cut bass early on and then let-it-out as you get to later stages. This keeps the bass from over compressing and fatiguing the gain stages with excessive bass that just muddies the sound. However, as the amp compresses and distorts, the bass/fullness will creep back in and you never know you lost it. But the amp will sound clearer.

    I was the guy who gave Klobasa that advice. Personally, I would go even lower than 10 nF. I would try anywhere b/w 5-1nF. The input impedance of a cathodyne PI is so high that even with a smaller(22nF) cap, it's very slow to recover, causing the blocking. and evern smaller on is still flat in in EQ bu recovers more quickely.
    Case and point: I recently built a 5e3 with a 2.2nF PI input cap. This was he only mod I pre-incorporated into the the build as I was trying to see how the amp really was before I decided I didn't like it. I tried this mod on a PRRI with a 1nF cap and really liked the improvement. still sounded like a Princeton, but just less flubby.The point is even with the smaller 2.2nF cap, the 5e3 is still REALLY bassy and saggy. I have not desire to see what it's like with the "correct" 22nF cap.
     
    jsnwhite619 likes this.
  9. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    35
    Posts:
    2,772
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    I'm glad you chimed in! I filled out the board and made some changes on it. Split the inputs like they should be for the single/parallel inputs. Have a 820r parallel bias resistor on V1 with a 2uf bypass. Going with a fixed NFB - which I had to drill and add to the board - don't want a switch taking up room and causing problems or noise. Starting with the '68 Custom PR level for now.

    I think that's all the changes I made tonight. I left a couple things off the sketch that I accounted for - even a couple holes I had to drill.

    Sent from my moto g(6) using Tapatalk
     
  10. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    35
    Posts:
    2,772
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    Here is the latest shot of the board.

    Sent from my moto g(6) using Tapatalk
    [​IMG]
     
    Mr Ridesglide likes this.
  11. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    35
    Posts:
    2,772
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    So, I added the second stage bypass cap to it, 25uf, and finished the wire leads off the board. I also realized that I didn't account for two separate inputs and switched the 68k grid resistors to 33k each. I'm using the 18 watt Marshall style inputs, the 18 Lite IIb anyway - I have no historical knowledge of Marshall amps. The best I can tell, it's wired like a Hi/Lo Fender, except they each use their own half of the tube, and #1 will have paralleled inputs.

    I've been up in the air on the transformers. I know I'm using a 290AX for the PT, but I've been torn between 8k Tweed Deluxe OT and 6.6k Deluxe Reverb OT. I know the 6.6k can give amazing clean tones in the original Harvard/Vibrolux circuit, and gives a tighter sound. But, I'm wanting a mean rocker on this. I'm leaning toward the 8k OT. I've used it several times and the Princeton Reverb I built last year had it, and with a 5u4 rectifier, it gave similar clean tones as the Harvard with a 6.6k did with a 5y3 rectifier. And it really surprised me with how much clean headroom it actually has when used beyond Tweed circuits. The PR had amazing cleans & headroom with a C-Rex. So, I'm thinking that if it comes out too muddy with the 8k, between adjustable bias, a 3A rectifier option, and circuit tweaks, I should be able to find a happy medium.
     
  12. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    35
    Posts:
    2,772
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    Only update is that I got the 290AX as mentioned and the 1750E 15 watt 8k OT. I thought about it tonight and the Princeton Reverb last year with the same transformer setup sounded like angels screaming with a Les Paul cranked up, and had plenty of punch. So, I'm not worried about the transformers being to blame. If it's mushy or muddy, I can tweak the circuit, but I know the transformers can deliver the goods. They'll be here Monday, so should have something to listen to sometime next week.
     
    Mr Ridesglide likes this.
  13. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    35
    Posts:
    2,772
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    I'm having thoughts about the power section & dropping resistors. The 5f10 & 5f11 both used a 470R resistor in the first spot, then a 22k going to the preamp node. But, both at considerably lower voltages than I'll be using. Should I increase the 470R to a 1k or anything to lower the screen voltage? I currently plan on a 5u4 rectifier, so I'll be in Princeton Reverb neighborhood for expected voltages.
     
  14. Lowerleftcoast

    Lowerleftcoast Tele-Holic

    Posts:
    507
    Joined:
    Dec 30, 2019
    Location:
    california
    The harvard used 470R but the 5e3 used 5k. I'd change it higher to 2k - 5k. It would help the filtering. Anywhere up to 10k should be fine.

    As a general rule, I put 470R on my screens like every other more modern amp.

    You are making good progress.
     
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.