Fender Super Tweed Build

Discussion in 'The Squawk Box' started by Cjsinla, Sep 23, 2017.

  1. Leña_Costoso

    Leña_Costoso Senior Member

    Messages:
    734
    Likes Received:
    692
    Joined:
    Aug 28, 2017
    I'd be interested to see how that turns out since what's feeding the phase inverter is already patted quite a bit. The use of a grid stopper will not usually help the low end at all..... So if anybody tries it please report back!
     
  2. D'tar

    D'tar Senior Member

    Messages:
    362
    Likes Received:
    170
    Joined:
    Sep 23, 2014
    I thought the concern was phase inverter double frequency blocking distortion.

    For low end flub, Lena's suggestions are spot on. It may sound silly but how about turning down the bass pot?
     
  3. Splattle101

    Splattle101 V.I.P. Member

    Messages:
    8,645
    Likes Received:
    4,167
    Joined:
    May 30, 2008
    I put a large grid stopper on the PI of my Super. I built it that way to begin with, having done the same mod to my 5E3 clone. The effect in the 5E3 was dramatic. Killed the blocking distortion dead. It was a winner, so I had no compunction about using it sight-unseen in the Super, too.
     
    Cjsinla likes this.
  4. Cjsinla

    Cjsinla Senior Member

    Messages:
    3,638
    Likes Received:
    2,922
    Joined:
    Jul 7, 2014
    Do you have a diagram of what you did with the 470k resistor in your Super?
     

    Attached Files:

    Last edited: Dec 29, 2017
  5. Leña_Costoso

    Leña_Costoso Senior Member

    Messages:
    734
    Likes Received:
    692
    Joined:
    Aug 28, 2017
    2. The 470k grid stopper also couples with the 1M grid leak to form a voltage divider that reduces the input signal by a third.

    Looking at things... I think the key information about the "470k grid stopper" is contained in that sentence. What he has done is create a voltage divider which reduces the voltage going into the phase inverter.

    Whether or not "blocking distortion" exists at the phase inverter is unclear to me. Pushing a cathodyne inverter to draw grid current... I dunno gents. First its class A operation, second is that its cathode biased as well. So there are self limiting features "built into" the design that prevent blocking distortion.

    Not to say some nasty sounding distortion may occur in the 5e3 for a variety of reasons. One of those being the whole setup of the cathode biased output (remember, you look up cheapbastard in Websters, it shows Leo's picture). You have to bias those things damn near class A to get 'em to sound good, in my experience. And... they have no negative feedback to put the k'bosh on some of the little nastiness that generally occurs when we push things over the top, as we are addicted to do.

    So, the suggestion there for the 5e3 may well be a great one, just not for the reasons stated.

    If you look at the tone controls on the 5F4....

    The signal is fed from the cathode of the 2nd half the 2nd tube. Cathode driven means lower impedance. An easy way to think of impedance is high Z = higher voltage, lower current while low Z = higher current, lower voltage. Also its notable to point out that the 5E3 phase inverter is not driven by a cathode, but by a plate, which is high Z, although both are capacitor coupled.

    Back to 5F4 tone controls. In my little bit of experience, that tone control setup allows ungodly amounts of low end through. Its the nature of the design, and it could well have been a good one when used with the speakers Fender chose back in the day. Originally Jensen C10R's, fairly poor in their low end response. Each of the Supers I've put together have had gobs of low end. They run best in open back cabinets, btw. The 4x10 "Bassman" (5F4) I put together was interesting, because the speakers were 10ALK's, perhaps the best modern speaker for that design. The new Eminence Legend which replaced the 10ALK ... I have no experience with, sorry.

    If you look at the tone controls, the signal splits after it leaves the cathode. One direction treble, the other bass (or really "all the rest of the signal"). The treble control works by increasing resistance between the .00025 capacitor and decreasing it to ground through a .01 capacitor. In the extreme "treble" setting, you get all the signal that passes through the small capacitor, and hardly any getting shorted to ground through the big capacitor. That signal does the famous football "end around" play, and bypasses the bass control.

    The other side of the split, which is for the bass control, has an interesting signal reduction scheme, which is the negative feedback, and padding to ground though the 220k resistor. Its as if Leo and the boys said "rut-roh, this things over the top" and realized they had to pad it down some. Interesting to note, is that every "old" schematic you see, has a weird looking 4.7meg notation on that feedback resistor. Never seen a "clean" schematic, except those which have been redrawn totally, and are obviously more modern. Ever wondered what that was? Could it be that Fender played with different values and the draftsman wasn't about to redraw the entire thing for the value of one resistor? Another side note.... drafting film, back in the day, could have been double side coated acetate, which could be erased and redrawn many times, and drawn on the back side (in reverse) when the front "tooth" wore out. Mylar films weren't available until the 60's.

    Anyway, the signal gets padded a little, then gets voltage divided by the 100k resistor and the bass pot, and isolated from any remnant treble with the 220k resistor.

    You can see how those controls not only effect the tone response, but also greatly effect the overall gain of the amp. I think you'll find that the breakup qualities are hugely different when tone controls are set in their lower ranges, versus higher ranges. More so than on a Baxandall tone control used on more modern amps, which can "zero out" signals when turned fully down, but after a certain point, you don't seem to get much more "strength" to the signal, just shaping. I've found that tone control settings useful for loud "bedroom" playing are totally inappropriate as the volume is cranked up some. That's my finding at least.

    Not mentioned, yet, is the .1 coupling cap for "all the rest of the signal" that the bass control is taming...

    Question -- What do you think would happen if you lowered the value of that capacitor to say .01 or so? Or to .0047 ?
    Answer -- Why, you'd limit the low end going through to the phase inverter, by golly!

    Just sayin'
     
    Last edited: Dec 24, 2017
    Splattle101 likes this.
  6. Leña_Costoso

    Leña_Costoso Senior Member

    Messages:
    734
    Likes Received:
    692
    Joined:
    Aug 28, 2017
    I've said this before, but I'm gonna add it again, if you look at the old Fender blackface schematics, you'll find variation on the coupling capacitor to the phase inverter. Yes... they were long tail pairs, but even so, among those, the coupling changed quite a bit - depending on the response Fender sought.

    Case in point. I once got a really nice shape FEI marked blackface Tremolux head. I had a 2x10 open back combo cabinet made for it. It sounded tinny as all hell. I went through the amp with a curry comb. It was as it ought to have been. WTF is with that tinny sound, very little bass response. Look at the coupling cap on a Tremolux schematic. Fairly small value. Why? It was meant to be played with a closed back cabinet, which had a lot more bass.

    Not quite a "d'oh!" moment, but I changed the value to a larger one, and the low end returned. Funny how that works.

    I remember when my old HS friend Dom got his Twin Reverb. He already had a LP custom, but was playing it though an old Univox 2x12 piggback at about 40watts. When that Twin got on board, I can still remember him saying "holy shit, these things have so much bass!". His assumption was that it was because of the "trebly" tone of Fender guitars (I was playing a 72 Tele at the time). I think... its just the way Fender did things. Maybe gave you more than what you needed.

    Another observation, is that I've seen players set their amp "by the numbers" rather than by their ear. The remark being "this amp sounds like crap". Why? Well listen to it, I've got all the knobs set where I usually set them, and it sounds terrible. Jeeze, ever thought about using OTHER settings? Hmmm? Seriously... I've experienced that first hand (and been guilty of it when I was a kid).
     
    D'tar likes this.
  7. Leña_Costoso

    Leña_Costoso Senior Member

    Messages:
    734
    Likes Received:
    692
    Joined:
    Aug 28, 2017
    Just because its Christmas Eve.... some pictures for your amplification enjoyment

    Found a picture of that Tremolux
    [​IMG]
    [​IMG]

    Starting one of the Supers, some years ago
    [​IMG]

    And the "5F4 Bassman"
    [​IMG]
     
    Splattle101 likes this.
  8. Splattle101

    Splattle101 V.I.P. Member

    Messages:
    8,645
    Likes Received:
    4,167
    Joined:
    May 30, 2008
    My thinking on this was guided by Merlin Blencowe's book on preamp design. He spends a chapter talking about the cathodyne PI: the fixed and cathode biased versions of it, and how the cathode biased version is inherently balanced...up to the point that it distorts, after which all bets are off.

    I should probably get it off the shelf and clean the cobwebs out of the old brain case. Sounds like a job for next week, after the immediate effects of Christmas have receded! ;-)

    On a different but related point, I speculated somewhere that perfect balance of the cathode biased cathodyne PI probably contributes to the nice clean sound of the tweed Super. One of the things that really struck me about the sound of it was that the cleans were really good, and that the fixed bias output stage made them hard enough to really dig in. I can play funk on it, FFS! Using the neck pickup of my Standard!:shock:
     
  9. Leña_Costoso

    Leña_Costoso Senior Member

    Messages:
    734
    Likes Received:
    692
    Joined:
    Aug 28, 2017
    It could be, I suppose. My own thinking is that since cathode bias is self-limiting, it would do just that, well before being at the point of drawing grid current.... and getting the nasty blat blat blat of blocking distortion.
    The other thought that comes to mind is that there are plenty of very high gain preamps the show no amount of blocking distortion even though they are capacitor coupled.
    Never too old to learn.... But, blocking distortion hasn't been a problem I've had to deal with. All kinds of other crap like parasitics and run away.... But, not blocking distortion.
    I would really like to hear an audio sample of the super suffering in such a manner. The design has been one that's fairly well proven over decades. Turning to blocking distortion as the cause for audio woes should be ruled out after elimination of other possible issues.

    (edit in)
    After reading the suggested chapter... I'm still not convinced there is any sort of inherent issue with blocking distortion, but it seems easy enough to check for with a scope, even with the amp run into a dummy load. The one immediate thing I noticed was discussion of a capacitor at the grid, along with the resistor, which would not have the effect of limiting blocking at higher frequency, and going on to say that lower frequencies are the issue. I agree with that.

    Going back to using a smaller coupling capacitor for "all the rest of the signal". I'll call it the Festofusignal capacitor, as in Festofusignal for the Rest-of-the-Signal.... I'm just gonna blurt out, that if blocking were an issue, that huge ass coupling cap would only tend to make it worse. Limiting the value, will make the RC charge/discharge time faster _and_ limit the current hungry low end.

    Any thoughts on that?
     
    Last edited: Dec 25, 2017
  10. Cjsinla

    Cjsinla Senior Member

    Messages:
    3,638
    Likes Received:
    2,922
    Joined:
    Jul 7, 2014
    I’ve finally been putting some time in with the Super. I didn’t put any time in it until this week due to the problems I was having with the input jacks. I also had no faith in the Weber PT. So, after cleaning the jacks and putting in the Mercury Mag PT, I put a number hours on it yesterday with an ES335. Today I pushed it hard pounding out chords on my Strat. I swear I heard the speakers break in. It was sounding pretty aggressive until today. After 3 hours of punishment it seemed to smooth out. Is that possible? All I know is that my Tele sounded incredible through it this evening.
     
    D'tar likes this.
  11. Leña_Costoso

    Leña_Costoso Senior Member

    Messages:
    734
    Likes Received:
    692
    Joined:
    Aug 28, 2017
    Depending on speaker...yup.
     
    D'tar likes this.
  12. D'tar

    D'tar Senior Member

    Messages:
    362
    Likes Received:
    170
    Joined:
    Sep 23, 2014
    I put a Weber in my 5e3. I had them break it in and requested twice.... It sounded good right out of the box. I heard the same thing after about 3-5 hours of play at stage volume. It was like a sigh of relief... Aahhhhhh! There it is!!!
     
  13. Leña_Costoso

    Leña_Costoso Senior Member

    Messages:
    734
    Likes Received:
    692
    Joined:
    Aug 28, 2017
    Back in a previous life.... I would do the "blanket trick", which is to run my audio generator at about 50 cycles at 1 watt into speakers to be put into amps that went to other people. Put 'em on the carpet, throw a blanket over 'em, and let the speakers sit for a few hours (aka, while I was at the day job). The Celestions lose their characteristic "brittle" top end quite nicely, and the 10ALKs would just become sweetness personified.

    The whole idea is to loosen up the spider and surround. Cones don't flex all that much, but surrounds sure do, spiders do too. Celestion speakers especially... very tight when new, they take volume and low notes to get all limbered up.
     
  14. Cjsinla

    Cjsinla Senior Member

    Messages:
    3,638
    Likes Received:
    2,922
    Joined:
    Jul 7, 2014
    I’m using 3 Weber 10a125 alnico speakers. I like them so much I just ordered a 30 watt 12a125a for my Deluxe. The MOD 12/50 ferrite speaker that came with the kit emphasizes bass but the Weber alnico is supposed to have a tight low end and smoother breakup and high end. We’ll see.
     
  15. Leña_Costoso

    Leña_Costoso Senior Member

    Messages:
    734
    Likes Received:
    692
    Joined:
    Aug 28, 2017
    WTF alert....

    I was playing my Super this afternoon, low volume, quiet volume even for bedroom play. And the volume started goin away... I thought it was the Tone Corset compressor. Nope. Volume knob needs a tweak? Nope. Cable connection? Nope.

    Cranked up the volume and got fizzle, so somethin's wrong. Lookin at the back of the amp and one 6L6 has gone cherry red at the top of the plate. Flipped standby off, waited a few minutes, flipped on, all is well until the plate started to glow again. Weird.

    So, off came the back. One tube sitting there at 39.8 ma at idle. The other at 90ma. WTF? Bias voltage the same on each -41 volts DC

    Bum tube. That sucks. Had maybe 1 hour of heavy volume play, maybe 10 hours of bedroom play, and most of that was biased really really cold. My guess is that it was bound to "go" but didn't when it was biased real cold. Once I got it to 40ma, then cranked it up a bit.... it pushed it over the edge. Could be it has a warped anode (plate), I've seen that occur, or there's an issue with the screen grids. I didn't check, but will.

    Just tossin' all that out there.... sigh.

    I suppose I'll pull the chassis later on, slot the bias pot shaft and run it down cool until I can get another set of 6L6s in there. @#$@#$!!!!
     
  16. Leña_Costoso

    Leña_Costoso Senior Member

    Messages:
    734
    Likes Received:
    692
    Joined:
    Aug 28, 2017
    Well, good news at least. Shamefully, I did what I should have done. I swapped tubes in their sockets. The problem stayed with the socket. So, its not the tube.

    Some thought... what could cause high amperage on a tube that should be negatively biased?
    a. Could be the grid wire is not soldered well.
    b. Could be the grid wire is shorted to ground under the board (I hope not!)
    c. Could be the 220k resistor has gone kerfluey. Not likely
    d. Could be the grid stopper has opened up.

    Ding Ding Ding!

    Its "D", the grid stopper has opened up! WTF.... never ever seen this. Infact, the wire lead has basically "pulled out" of the body of the resistor. New one on me at least. At least its a CHEAP fix. A bit of a pain in the rectum to solder, but that's amps for ya.
     
    Splattle101 likes this.
  17. Leña_Costoso

    Leña_Costoso Senior Member

    Messages:
    734
    Likes Received:
    692
    Joined:
    Aug 28, 2017
    More on those grid stoppers.... I got into things with the magnifying glass today. Sad, but at my age, its a must have. My B&L and Harbor Freight "goggles" are at work, and this bein' the NYear's weekend...

    Anyway, I think I know the cause of the failure. Or causes, plural. Again, I am shamed. I got lazy. I put in that bias adjustment circuit, and its a good one, but with the position of the pot, I wasn't able to turn it easily, and I didn't slot it. The chassis was already in the cabinet and, with low hung head admitting, I was too damn lazy to pull it out and slot the shaft for a screwdriver.

    How'd that blow resistors. Well, I must have had the original setting "too hot" and it was, pulling upper 50's for milliamps at idle. Apparently, the grids started to draw some current, and those resistors got hot.
    And, we just had a discussion regarding grid stoppers and grid current and such..... yikes.

    But, the one resistor failed when I put the tube back in its socket. There was little "extra" in the resistor lead, the tabs on the socket were drawn closer than they should be, and the pins on the tube tried to "straighten" those socket connections, and "stretch" the resistor. Couple that with a hot resistor... "poof".
    The remaining grid stopper "works" but has a dark streak on its edge. Heat. Those are 1/2 watt carbon comp resistors and should be good for any normal use. I used them abnormally.

    I am sure a "little" (as in short time) of over current wouldn't have been too bad. But, I over did it.

    Both tubes seem to be ok for the abuse, thankfully.

    New grid stoppers ordered. Plus a few odds and ends. Should be in by Monday or so.

    The replacements will be carbon comp, "just because", but this time I'm going to dress the leads so there is some "spring room" to account for socket shake, and heat and such.

    The reason those grid stoppers are where they are is simple. They are there to cut off RF. If the leads to the grids were "feet" away, we could put those resistors "inches" away, since a few inches between grid and resistor is unlikely to pickup RF. In this case,... you want them close, just so there is no way to introduce RF "after the fact" of putting the k'bosh on 'em. Simplistic terms... and we'll sidestep the math, and the Miller capacitance and such. You just want to make them as close to the grid pin as possible, which Leo did by jumpering the two sockets.

    I'm amp-less again, and it aint fun.
     
  18. Cjsinla

    Cjsinla Senior Member

    Messages:
    3,638
    Likes Received:
    2,922
    Joined:
    Jul 7, 2014
    @Leña_Costoso
    How could you be amp-less with all the building you have done over the years?
     
  19. Leña_Costoso

    Leña_Costoso Senior Member

    Messages:
    734
    Likes Received:
    692
    Joined:
    Aug 28, 2017
    Hey thought I'd revive this to give some correction and closure.

    A page or so ago, I showed a bias adjust mod, which uses a 250k AUDIO taper pot and a resistor - in parallel with the 56k resistor on the bias supply on a 5F4.

    I used a 8.2k resistor, and while the range was ok, the limit of the range allowed a much higher idle than one would want. As fate would have it, during my initial testing, the idle was very high (and since I couldn't easily turn the added pot), I burnt out a grid stopper. Sigh.

    Ok, all fixed. New grid stoppers, (hey do both, ya know) and I experimented with resistors to add to the pot. 22k seemed to do just fine, giving (with my tubes) a range of about 20ma at idle to a high of about 65ma at idle. Both ends are beyond "normal", but a short time at 65ma while adjusting wont hurt a thing. As it was, my 6L6's didn't suffer a bit running a bit cherry during the initial mishap.

    And yes... the solid shaft got a nice slot cut in it, to make adjustment easy.

    And its good to have the Super purrin' along well again!
     
  20. Cjsinla

    Cjsinla Senior Member

    Messages:
    3,638
    Likes Received:
    2,922
    Joined:
    Jul 7, 2014
    Any pics of your handiwork?
     

Share This Page