Harvard 5F10 Voltage and Bias Questions

Discussion in 'Shock Brother's DIY Amps' started by Skish, Jul 21, 2019.

Tags:
  1. Skish

    Skish TDPRI Member

    Age:
    45
    Posts:
    18
    Joined:
    Apr 20, 2019
    Location:
    New England
    I've built a Harvard 5F10 Clone that I'm quite happy with. I've been doing some measurements. I'm using the Mercury Magnets ToneClone Tranformers (FTWHARV-O and FTWHARV-P). I'm getting 376V at the Rectifier and 370V at the 6V6 plates. I've also measured a bias current of 18.6mA. Also, the voltage between the 220k bias circuit resistors is -35V.
    Comparing those voltages with the 5F10 schematic, they are quite different.
    Rectifier 305V, bias circuit -27V.

    Any thoughts about what that might have for an effect? Headroom differences maybe?

    Should I adjust that bias current higher?
     
    rjtwangs likes this.
  2. Skish

    Skish TDPRI Member

    Age:
    45
    Posts:
    18
    Joined:
    Apr 20, 2019
    Location:
    New England
    IMG_1686.jpg IMG_1683.jpg IMG_1684.jpg IMG_1685.jpg
     
    rjtwangs likes this.
  3. dunehunter

    dunehunter Tele-Holic Double Platinum Supporter

    Age:
    65
    Posts:
    570
    Joined:
    Jul 14, 2011
    Location:
    Larkspur, CO
    Wow. Build looks great! Sorry, I don't have any comments on your numbers, but am wondering if that's a 5F2 chassis, modded for the extra Harvard power tube and input socket.
     
  4. rjtwangs

    rjtwangs Tele-Afflicted Silver Supporter

    Age:
    71
    Posts:
    1,241
    Joined:
    Jan 18, 2010
    Location:
    Reston, Virginia
    It's a great looking amp! I wonder why more people aren't as taken with the Harvard as I am. Mine was built for me by Michael Clark. It has a beautiful 1959 P10Q in it and it sounds wonderful.
    20190316_235500.jpg
    It's a real joy to play.....and it has sparkling cleans.....


    RJ
     
    Uncle Daddy likes this.
  5. Wally

    Wally Telefied Ad Free Member

    Posts:
    32,227
    Joined:
    Mar 17, 2003
    Location:
    Lubbock, TX
    With Pv at 370vdc and 18.6ma of current draw, the amp is operating at about 49% of max plate dissipation. Your bias voltage is to the cool side of things. If you bias hotter, the B+ will come down some, your current draw will increase, and the plate dissipation will increase. Where you want to operate the amp is a personal decision, but I like things warmer than that.
    You will not get the B+ down to that schematic value due to variables.....wall voltage, power tranny specs.
    Good looking build!
     
    tubedood likes this.
  6. Skish

    Skish TDPRI Member

    Age:
    45
    Posts:
    18
    Joined:
    Apr 20, 2019
    Location:
    New England
    Yes, It's the Ted Weber 5F2 Chassis. Three additional holes and they all fit quite nicely. (power tube, preamp tube and extra input jack)
     
  7. Socrates01

    Socrates01 TDPRI Member

    Posts:
    35
    Joined:
    Jul 6, 2009
    Location:
    Chicago, USA
    Hi. Just for context, I am not an electrical engineer so please consider that when evaluating my comments. That being said, I have built around ten or so amps from scratch (not kits) including a number of 5F11-ish circuits. Lastly, my apologies if I am telling you things you already know.
    Looking at the spec sheet of your MM PT, it's secondary voltage is stated as 376VDC, so it's not too surprising that you're getting that at the rectifier rather than the schematic number, of 305VDC. (Note that wall voltages nowadays are often 122VAC -126VAC, so most vintage amps used today put out more secondary VDC than spec'd as wall voltages used to be more like 115VAC or so.)
    Next, the bias current alone doesn't tell the whole story. I am of the school of thought that you want to set your bias to a value that produces an acceptable level of tube dissipation. Therefore you would set your bias voltage to induce the desired tube dissipation. If you haven't already, check out Rob Robinette's wonderful website that explains biasing and offers a calculator to aid the process. Adjusting the bias to achieve 'proper' tube dissipation may alter your voltages and currents. Once biased, as long as your values are within the acceptable range for your given tubes, you should be all set. Note that there is a range of bias that is generally considered acceptable and one should use their ears too in this process - you might like the tone/response when it's a bit 'hotter' or 'colder'.
    Most importantly, be careful. Measuring current in an amp is risky business and touching leads to the wrong spot can cause very bad things to happen. Safety first.
     
    screefer and tubedood like this.
  8. Socrates01

    Socrates01 TDPRI Member

    Posts:
    35
    Joined:
    Jul 6, 2009
    Location:
    Chicago, USA
    Wally said it better than I did! :)
     
  9. Skish

    Skish TDPRI Member

    Age:
    45
    Posts:
    18
    Joined:
    Apr 20, 2019
    Location:
    New England
    Thanks for the input. I've messed around with Rob's calculator and the Ted Weber bias calculator. It looks like I would be able to go up to about 22 or even as high as 26. Especially if the voltage does drop some as I go up. There seems to be some differences between the two calculators. I will need to get a bias pot ordered and get this switched over to adjustable bias.

    What am I going to hear with a 'hotter' bias? Maybe more clarity, more chime? Maybe less headroom and easier to get into some overdrive?
     
  10. Wally

    Wally Telefied Ad Free Member

    Posts:
    32,227
    Joined:
    Mar 17, 2003
    Location:
    Lubbock, TX
    A warmer biasing brings a richer signal....more harmonic content....and yes...to my ears a better distortion. Extremely cold bias will yield a harsh crossover distortion. Where that is biased right now might do some well. I would want it warmer. Ommv.
     
    Skish and D'tar like this.
  11. Skish

    Skish TDPRI Member

    Age:
    45
    Posts:
    18
    Joined:
    Apr 20, 2019
    Location:
    New England
    Updated it with an adjustable bias last night and it really did seem to come to life! More chime and like you said more hamonics that seem to ring out with more clarity. I think easier access to the overdrive as well. Really happy I made the change. Thanks for the advice.
    Will have to mess around a little dialing it in. It was an immediate difference going to about 22.7mA. But then seemed to be less noticeable differences going all the way up to 26.5.
    Also, I had been thinking 12W 6V6 but noticed JJ6V6S on Rob’s bias calculator, and it’s 14W. Then rechecking tubes online, many 6V6’s say they handle at least 14W. Guess we can bias quite a bit higher than I was thinking.
     
    King Fan and Wally like this.
  12. Wally

    Wally Telefied Ad Free Member

    Posts:
    32,227
    Joined:
    Mar 17, 2003
    Location:
    Lubbock, TX
    Skish, your experience mirrors what I find. There is a point where the difference becomes less discernible......the payoff is not worth the increase in tube wear. How much did the voltage drop with the adjustments? And....14 watts is proper for almost any 6V6 that we will come across. Ime, 26ma of current draw is pushing a 6V6 in fixed biased....depending on the plate voltage, of course.
     
    King Fan likes this.
  13. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,787
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    What Wally said, of course. For bias calculation, Rob's page is sort of an operating standard here. FWIW I think you'll find it useful, and easier for others to understand, to discuss bias in terms of % dissipation.

    If it isn't completely clear (you seem to know your way around amps) Rob's calculator wants *plate-to-cathode* (not plate-to-ground) voltage and tube type in the first section. The P-K voltage will vary as you change the bias, so re-enter and hit Calculate again. For tube type, I just set it to 6V6GT and leave it there. `

    JJs are indeed max 14W by design (many folks point out they aren't true 6V6 in construction; some folks feel they aren't true 6V6 in sound either :D). Old high-quality 6V6 (RCA especially) are notorious for surviving > 12W max despite their spec. But at the same time, old high-quality 6V6 (RCA especially) aren't cheap, so running them at or under 12W may be smart.

    *If* your plate-to-cathode voltage stayed at 370 (not really likely), your 22.7mA would be exactly 70%, a safe logical max for standard 6V6 in this amp.

    OTOH, to make your 27mA 'safe', you'd have to drop P-K voltage to < 315V. You were super-smart to put in a bias adjust pot. You're smart to adjust the final bias by ear. IMHO, you may want to use your pot to select a dissipation somewhere 60 to 70% range.
     
  14. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,070
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    @Skish , would you be willing to include the Vpk along with your currents so that 1) we can see how the voltage supply sags with greater current draws and 2) begin to link your sonic impressions to plate dissipation (W), which might be more generalizable or translatable to other folks w/ push-pull 6V6GT amps but who are working w/ different B+?
    Cheers
     
  15. Wally

    Wally Telefied Ad Free Member

    Posts:
    32,227
    Joined:
    Mar 17, 2003
    Location:
    Lubbock, TX
    Plate voltage measurement to ground suffices. I have seen comparisons of the two methods, and there is not enough of a difference to warrant any concern.
     
  16. D'tar

    D'tar Friend of Leo's

    Posts:
    2,090
    Joined:
    Jan 11, 2013
    Location:
    WNY
    Fixed bias correct? Plate to ground would be = to plate to cathode. I know.... 1r resistors and all.... no worries.

    Cathode bias. Sure... the ~20v on the cathode will scue results a bit. So plate to cathode in that situation.
     
    Wally likes this.
  17. Wally

    Wally Telefied Ad Free Member

    Posts:
    32,227
    Joined:
    Mar 17, 2003
    Location:
    Lubbock, TX
    +1 with D’tar’s observation. In my dull mental process today...allergy meds, I was not thinking completely. Some people are concerned with considering the screen current in the equation, and that was the point of my observation...not much of a difference with vs. without that consideration. And....in cathode, I always have the plate voltage to ground from which to subtract that cathode voltage to ground.
     
  18. Skish

    Skish TDPRI Member

    Age:
    45
    Posts:
    18
    Joined:
    Apr 20, 2019
    Location:
    New England
    so this is pretty embarrassing, but I broke a JJ Tube when I was removing it to swap the Hoffman bias checker that I’m using!
    I put in an old matched pair of Ruby tubes that a friend had while I order a replacement pair. Because of that I dialed it back to 23.4mA. That gave me plate voltages of 367V. Only a 3v drop from when I was running cool at 18.6mA. Of course now I have different tubes so it’s not necessarily apples to apples.

    Thinking of getting a pair of Tung Sol tubes instead of JJs. Thoughts?
     
    Last edited: Jul 24, 2019
  19. Wally

    Wally Telefied Ad Free Member

    Posts:
    32,227
    Joined:
    Mar 17, 2003
    Location:
    Lubbock, TX
    Fwiw, you can order a matched tube for that JJ to make a pair.
     
  20. Skish

    Skish TDPRI Member

    Age:
    45
    Posts:
    18
    Joined:
    Apr 20, 2019
    Location:
    New England
    Got a matched pair of Tung Sol today and did some more bias adjustments and was surprised how much the Voltage Dropped
    Bias Current: 25.2mA
    Plate Voltages: 352V each 6V6
    Rectifier: 360V

    I’ll keep playing around and listening!

    (I used that spare JJ 6V6 for a 5F2A build that I just finished).
     
    Wally likes this.
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.