2x 813 gr grid 750 w ??

Discussion in 'Amateur Radio Amplifiers' started by W9WQA, Nov 14, 2019.

ad: L-HROutlet
ad: l-rl
ad: Left-2
ad: L-MFJ
ad: Left-3
ad: abrind-2
ad: Subscribe
  1. W9WQA

    W9WQA Ham Member QRZ Page

    i was curious about my uncalibrated output meter so i scoped the output of a 100 watt rig and then out of my amp running 3000 v on 2 813's
    i used rtty mode for a solid carrier.
    the 100 watt rig read 200 v pp which works out to 100 w
    the amp with 3000 v plate,apx 700 ma put out 550 v pp which is appx 750 w using same math.
    i thought it would be higher.
    the meter built in was caled yrs ago with a good watt meter reads over 800 but i trust the scope.you guys who know amps,does that sound right?

    and, any guesses what the plate ac volts would be??
    i have 5000 volt probes but im reluctrant!
    maybe half that power?

    and btw i tried tuned inputs on the amp and rig swr looked worse. didnt see any improvement what should i see?
  2. G3YRO

    G3YRO Ham Member QRZ Page

    Well according to your figures you had 2.1 kW DC Input power when you had 750W output . . . I would have expected over 1200W of RF output.

    May your Pi-Tank isn't providing the optimum matching, and it's badly loaded?

    What band was this on?

    Roger G3YRO
  3. W9WQA

    W9WQA Ham Member QRZ Page

    yeah it tunes and loads but im not sure i get max efficiency from the pi net parts. ive moved taps on the coil. its silver plated #10 i think from heath coil stock.
    i use mostly 40 mtrs.

    btw, not certain of plate current. meter was calibrated yrs ago. but the 550 v pp on 50 ohms i trust...
  4. KU3X

    KU3X Ham Member QRZ Page

    Years ago , one of my amps that I built had extra taps on that tank coil. To prove a point, I loaded the amp with the Q of 8, used anothr tap for a Q of 10 and moved a tap again for a Q of 12. It was just a test. Results....the output remained exactly the same.
    Focus on your tuned inputs. They are a key factor in efficiency. That is if you know that you are using good tubes.
  5. K7TRF

    K7TRF Premium Subscriber QRZ Page

    How much confidence do you have in the DC plate voltage and plate current meter readings?

    If you really have 700mA at 3kV that's 2100 watts DC plate power as Roger posted above. Assuming a good 50 ohm load your output power measured via scope voltage is as you say around 750 watts. That implies roughly 1350 watts being dissipated in the plates of the two tubes. 813's are pretty tough tubes but 625 watts of plate dissipation per tube will cook them in no time.

    I'd start by double checking the HV metering divider accuracy in a power down and discharged state and also validate the plate metering accuracy by injecting a known DC current through a good DMM in current metering mode to the current shunt resistor. If those meters are both accurate then something is seriously wrong in your output matching if 2100 watts of DC input power only yields 750 watts of RF output power which implies an efficiency of approximately 36%.

    You might also revisit biasing as it sounds like you could be biasing this much too close to Class A if you really are running close to 36% efficiency. What's your no signal plate current with the amp keyed?
    Last edited: Nov 16, 2019
    KD2ACO likes this.
  6. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    The "scope" voltage would only be meaningful if you really have a 50 Ohm resistive load.

    Is that what you're using, and not an antenna?
    K7TRF likes this.
  7. K7TRF

    K7TRF Premium Subscriber QRZ Page

    This is very easy to check assuming you have a decent DMM that can measure current up to an amp or so.

    Take either a low voltage variable bench supply or something like a AAA battery wired in series with a resistor to yield the current you want to test. Run that in series through your DMM in current metering mode right into the currrent shunt resistor in the amp. Compare the amp's meter reading to the DMM meter reading and try it for different currents throughout the meter's range either by varying the bench supply or changing the current limiting resistor in series with your battery.

    It only takes a few minutes to test the accuracy of your amp's plate current metering and is worth the effort.

    If you're really only getting 750 watts RF output for 2100 watts DC input your 813's plates should be glowing cherry red even for relatively short key down periods.

Share This Page