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.

Harvard 5F10 / 5E3 Deluxe hybrid spawn.

Discussion in 'Shock Brother's DIY Amps' started by Bendyha, Feb 24, 2018.

  1. Bendyha

    Bendyha Friend of Leo's Silver Supporter

    Posts:
    2,633
    Joined:
    Mar 26, 2014
    Location:
    Northern Germany
    I've always wanted to just build a straight 5E3 clone....but don't think I could keep myself limited to the task long enough, and to not elaborate on the original design.
    Then again, maybe a Harvard would be better.
    So, I have everything I need lying around for a tweedy styled 6V6 amp.
    Power transformer from a 1960 Clavioline, Ca. 350V HT, 5Y3 rectifier, plenty of filament ampere.
    Output transformer from my Super Champ with 8K5 plate load from 8 Ohm.....maybe an 8K or even a 6K6 would be better, but I will use what I have.
    I think I have all the necessary switches, resistors, caps pots etc. that I will need, as well as a novel cab idea, so lets get on with thinking about the schematic.


    The Harvard 5F10 & 5E3 Deluxe will definitely stand as model for my new amp, and although I will add several modifications to the 1950's concept. I will limit myself to one input and channel, and use only two knobs; volume and tone........BUT.........I will be making both those pots push/pull switching, to be able to achieve a variety of tonal variation.


    So what are the differences and the similarities between the 5F10 & 5E3 from an electrical standpoint ?

    The input stage has the obvious difference of tube. The two halves of the 12AY7 vs. the 6AT6 (similar to half a 5751) This was also alternatively fitted with a 6AV6 (similar to half a 12AX7(ignoring the diode sections)).
    So, one channel instead of two.

    The input resistors relate to each other as shown here below. Obviously the Deluxe has no middle input, but both channels have the same arrangement for the top and bottom input.
    That the Harvard has no ground reference for the grid, other than through whatever is plugged in, is strange.


    upload_2018-2-24_23-21-11.png


    The differing tube stages are followed by a coupling capacitor leading to the identical tone and volume controls.

    In the 5E3, this is 0.1µ, whereas the 5F10 uses a much smaller 0.02µ cap.
    upload_2018-2-24_23-24-10.png


    The phase inverter stage identical in the two amps, other than that the driving stage of the 5F10 has an unbypassed cathode resistor, necessitated by the fact that this would ground out the negative feedback that is fed in at this point, unlike on the 5E3 which has no NFB coming from the output transformers secondary side.

    upload_2018-2-24_23-25-22.png

    The push-pull 6V6 power amp section of the two amps differs in that the 5E3 is self/cathode biased, whereas the 5F10 has a fixed-bias arrangement, along with the negative supply this necessitates.

    The 5F10 transformers:
    O.T. - 125 AIA 6.6K : 8 Ohms
    P.T. - 125 PIA from Woodward-Schumacher. I could find no additional information regarding this transformer, other than the schematic showing the low voltage of 305V at the O.T. centre-tap (315V on the 6G2). The 125PIB was possibly also used, this would have voltages more like the 360V or so we might expect to find in a 5E3.

    The 5E3 transformers:
    O.T. - 8K : 8 Ohm
    P.T. - 6452-E from Triad 325-0-325V.

    Then getting to the output, the 5E3 has a 12" speaker and no NFB, whereas the 5F10 used a 10" speaker, and does have NFB.

    So, similar in some ways....very different in others.

    Could one bring both these amps together in one amp? Maybe not, but that is not going to stop me being inspired by both, and utilizing aspects from both.

    Let's therefore call it a "HARVALUXE"

    The amp will have various modifications that are not unusual to see in modded 5e3's.....plus some others maybe.

    Opting for a 12AY7 input tube, I have decided to run the two halves parallel, but each slightly differently, in the hope of a more colourful sound. While one half is more mid balanced, the other is somewhat brighter, with the addition of a switchable mid-cut option. (tone pot pull switch)

    These are fed into a standard tone/volume stack, with a somewhat smaller treble bleed cap than 5E3 specs.

    The phase inverter driver stage, is taken, like the output transformer and NFB loop, from the Super Champ.

    Here we have the luxury of a bypass cap boost for the stage, as well as a moderate feedback. The NFB is minimized with the volume pot pull switch.

    The phase inverter has an added grid resistor, as do the output tubes, in order to suppress unwanted phase-inverter distortion.

    An adjustable negative voltage supply has been tapped off the H.T., and the option of switching between fixed and self/cathode bias is made possible through the second part of the volume pull switch.

    This way I get fixed bias with NFB like the Harvard, and self bias without, like the Deluxe.

    The addition of the small filter-choke after the screen supply, was chosen to allow screen-grid sagging to happen, whilst dropping the preamp tubes voltage as little as possible, but still supplying sufficient filtering.

    This is because the 12AY7's work best with a higher voltage than the 5E3 normally puts on them.


    So here is my schematic as it stands at the moment. Any suggestions as to alterations would be of interest. Any finalization of component values will first be possible after getting the thing up and running, as will knowing what the actual running voltages are.

    upload_2018-2-24_23-32-37.png

    Now off to do some metal-work on the chassis, and finding the bits for building the power supply.
     
    Last edited: Feb 24, 2018
    KPAE, dr_tom, dan40 and 1 other person like this.
  2. dan40

    dan40 Friend of Leo's

    Posts:
    2,186
    Joined:
    Aug 19, 2015
    Location:
    Richmond Va
    Sounds like a killer idea! I just finished a mashup utilizing a 5e3's cathode bias output with a single bright channel from the 6g3 circuit. I did go with the 6g3 phase inverter also. It ended up sounding really good but i'm still tweaking a few values.
     
    Bendyha likes this.
  3. Bendyha

    Bendyha Friend of Leo's Silver Supporter

    Posts:
    2,633
    Joined:
    Mar 26, 2014
    Location:
    Northern Germany
    The 6G3 normal channel was the inspiration for the input tubes plate load bypassed with a bleeder cap in my schem.

    Just spotted a mistake in the schematic, strange, 'cause I recall working out the ratio and voltage drop, but somehow..?

    The HT dropping, voltage divider resistor feeding the NB is shown as 47...this should be 180 Ohm, to get down into the 1/10th HT area.

    With 180 feeding a fixed tail 10 and variable pot 10, and lets say a voltage of 350V, I should be able to vary between about 18V and 34V...this should cover the range I need.
     
  4. etype

    etype Tele-Afflicted

    Posts:
    1,734
    Joined:
    Sep 23, 2014
    Location:
    Dallas
    I just saw this today. Check out the 12:40 mark.

     
    Bendyha likes this.
  5. Bendyha

    Bendyha Friend of Leo's Silver Supporter

    Posts:
    2,633
    Joined:
    Mar 26, 2014
    Location:
    Northern Germany
    Thanks for posting, it points out another mistake that I have in my schematic, one that comes from the uncritical methodology of copy-drag-drop. The labeling of the NB diode cathode and anode with + & - is wrong....as he points out in the clip at the 6:50 mark. The diode is drawn in the correct way around though.
     
    D'tar likes this.
  6. Bendyha

    Bendyha Friend of Leo's Silver Supporter

    Posts:
    2,633
    Joined:
    Mar 26, 2014
    Location:
    Northern Germany
    I'm not fond of doing metal-work, but this time I'm not building the whole chassis, just modifying a bit of discarded Russian laboratory equipment that I picked-up. None the less, there is still drilling, sawing, and filing that needs to be done. Because the power transformer will be protruding from the chassis back panel, I had to get a bell cover to cap it off. One of the other guitarists in my band is the workshop manager in a galvanising plant, so getting a highly polished plating done is never a problem, ;)or expense.
    ( The other guitarist is an oral and Maxillofacial Surgeon, who looks after my teeth,:D and is just in the middle of doing me a crown.) ( The drummer and bass player both have secondary uses as well:) )



    upload_2018-2-26_21-1-45.png upload_2018-2-26_21-2-25.png



    I didn't fancy spray painting the chassis, but the grey paint that is on there now was looking a bit ratty, so I have just brushed over a gloss paint....not as chic as it could be, but good enough for me.
    Then there was a few bits of aluminium that needed cutting-out and drilling, and then the tube sockets rivitting into place.


    upload_2018-2-26_21-5-26.png


    I have gathered together all the bits I need for building the power supply into this bottom chassis.

    The rest of the amp comes into a separate casing that will be bolted onto the top of this one.

    But first I have to sit down and watch the paint dry...................................................
     
    Fiat_cc likes this.
  7. Bendyha

    Bendyha Friend of Leo's Silver Supporter

    Posts:
    2,633
    Joined:
    Mar 26, 2014
    Location:
    Northern Germany
    So, paint has dried, and found a little free time to be getting on with the power supply section.

    First to see how much space I have, and how the various components could be best distributed.

    Screw the transformer, voltage select switch, power-switch, pilot lamp and mains plug in.

    Mount the aluminium L-section with the rectifier tubes socket. Screw down the chassis earthing point, and see what space we have left.

    upload_2018-3-12_0-0-27.png

    Although not originally intended, I decided to keep the voltage selector switch with its built-in fuse, so now the amp could tour if I ever wanted it to. I mainly kept it for its looks.

    Then to cut-out a board for the filter caps, resistors, choke and bias supply to fit onto. It turned out to be not to cramped, and easy enough to mount and solder up. The bias adjust trimmer went conveniently onto the bottom of the board, so that adjustment can be made by poking a trimmer stick up through the base grid-work.
    upload_2018-3-12_0-1-46.png upload_2018-3-12_0-2-43.png

    The original pilot light was a neon NE51 fed by the HT through a limiting resistor, which was not very bright, so I changed it for a 6.3V lamp to feed with the filament supply.
    I haven't done a complete check-up on all the nodes, but a brief turn-on-&-off, to check, showed the pilot now to be very bright. This is probably partly due to the fact that there is no load on the filament circuit other than this bulb, and it is probably running at well over 6.3V.

    upload_2018-3-12_0-4-16.png upload_2018-3-12_0-8-8.png


    The fact that the transformer was originally from a small organ, and was powering quiet a few preamp tubes, means that I might have to do something to drop this voltage, once I see how much it sags with the present amplifiers load requirements.

    Rather than using a resistor to transfer some of the excess voltage into heat, here I have the advantage of being able to see the filament winding on the transformer, 27 turns of heavy duty copper-wire. This will enable me to carefully scrape of a bit of the insulating varnish from the wires surface, and let me short-out however many turns I require to get the voltage down to the right level. Each turn will drop about 0.25V, so I should be able to get very near to the perfect level.

    upload_2018-3-12_0-6-26.png upload_2018-3-12_0-7-5.png

    Other than this observation, a small wisp of smoke from the bias voltage's first dropping resistor made me flip the power-switch. A 5W resistor that was hot to touch. A quick measurement let me see that the resistor is still okay, but this also revealed the problem. I have a factor of "K" mistake.
    Instead of a 180K resistor, I had put a 180 Ohm resistor in. I hope the two 10µ caps are still okay.

    Anyway, that was all I had time for this time. Now I will have to wait until another day.

    Here are some photos of today's progress.
     
    D'tar and KPAE like this.
  8. Bobby Marshall

    Bobby Marshall TDPRI Member

    Posts:
    8
    Joined:
    Dec 5, 2007
    Location:
    Chicago
    How did this end up?
     
  9. Bendyha

    Bendyha Friend of Leo's Silver Supporter

    Posts:
    2,633
    Joined:
    Mar 26, 2014
    Location:
    Northern Germany
    Thanks for your intrest...What happened? I had a set-back, in that the power transformer headed south in the early testing stages, but a new one is now there. Actually, I did a little work on it last weekend, getting it back to the stage I had before, with a few slight modifications to the original concept....So it is not a dead project, just a sleeping one that I intend to carry on posting about again soon, detailing the progress, and possible new set-backs.
     
    Last edited: Nov 14, 2018
  10. Bobby Marshall

    Bobby Marshall TDPRI Member

    Posts:
    8
    Joined:
    Dec 5, 2007
    Location:
    Chicago
    Look forward to hearing about the corrections and the ultimate performance of this modified circuit. May need to get this done myself.
     
  11. printer2

    printer2 Poster Extraordinaire

    Posts:
    7,295
    Joined:
    May 24, 2010
    Location:
    Canada
    Would be good to put some resistance between the 180k grid resistors and the fixed bias cap so when it is in cathode bias mode the capacitor is not fully shorted. But if you do not plan on changing the position of this switch with the amp powered up it should not matter.
     
  12. Bendyha

    Bendyha Friend of Leo's Silver Supporter

    Posts:
    2,633
    Joined:
    Mar 26, 2014
    Location:
    Northern Germany
    Good point, I'll think about it.....
     
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.