Smaller tubes with a lot of glow?

Discussion in 'Amateur Radio Amplifiers' started by KC8QVO, Dec 26, 2019.

ad: L-HROutlet
ad: l-rl
ad: L-MFJ
ad: abrind-2
ad: Left-2
ad: Subscribe
ad: Left-3
  1. N2EY

    N2EY XML Subscriber QRZ Page

    Hold on a moment there.

    750 volts at 90 mA is 67.5 watts input. Yet the claim is 72 watts output? NOPE.

    What you're seeing there is the input for one tube but the output for two tubes.
  2. N2EY

    N2EY XML Subscriber QRZ Page

    The main problem with using tube rectifiers is the need for suitable transformers and chokes.

    Say you want a 1000 volt 250 mA power supply, using a pair of 866As and a choke-input filter.

    To get 1000 volts, you need a plate transformer that is about 1200 volts each side of the center tap, or 2400 volts CT, at about 300 mA. Also a filament transformer of 2.5 volts at 10 amps, with at least 2500 volt insulation.

    The filter choke should be of the "swinging" type, meaning its inductance changes (intentionally) with current. A suitable choke might be 25 henries at 40 mA and 5 henries at 300 mA. With choke-input filters and 60 Hz full-wave rectifiers, the choke must always have a value in henries at least equal to the load resistance in kilohms.

    And of course a filter capacitor - say, 10 uF at 1500 volts.

    The simple capacitor filters used with silicon rectifiers cannot be used with mercury-vapor rectifiers such as the 866A (or its Xenon equivalent, the 3B28) because they cannot handle the high peak currents encountered.

    The W8CDX (actually NO3M) transmitter uses techniques common in the 1920s and 1930s. Such techniques were largely abandoned after WW2 for a number of reasons, ranging from electrical safety to RFI to stability.

    Covering "all HF bands" is a tall order because what is a good design for one end of the HF spectrum often isn't for the other end. There's also the issues of unwanted stray resonances in the various parts.

    There's a reason most tubes suitable for HF transmitting above the QRP level have plate caps.

    What you describe is possible, but not easy. To go from 5 watts to 250 watts is a power gain of 17 dB. That's doable, but it pretty much rules out grounded-grid circuits and triodes. So you wind up choosing a high-gain screen-grid tube design that is grid-driven. Such designs require a lot of careful design to be stable and usable, plus there's usually a need for grid bias and screen supplies. (This is why I suggested the single-811A design).

    Do you have any experience with HF transmitters or power amplifiers? The reason I ask is that your questions indicate that you are a newcomer to RF design, and don't realize how many ways such projects can go wrong. Not that what you want can't be done - it can - it's just not simple, and it's not like audio. It's very easy to make very expensive mistakes.
  3. N2EY

    N2EY XML Subscriber QRZ Page

  4. N8FVJ

    N8FVJ Ham Member QRZ Page

    140ma (typo)
  5. N8FVJ

    N8FVJ Ham Member QRZ Page

    Corrected 807 tube spec:

    RF, AB1 Class
    VA: 750
    Vg2: 300
    Vg1: -35
    Ia: 15-140
    Ig2: 3ma
    Power Out: 72 watts
    Drive Power: 2.5 watts[/QUOTE]
  6. N2EY

    N2EY XML Subscriber QRZ Page

    That's for 2 tubes.
  7. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    I'd have to agree.

    I've never seen an 807 deliver 72W output power. It's rated 25W ICAS plate dissipation (16.5W CCS) and normally to remain somewhat linear nobody designs an amp to run more than 2x rated Pd for output power, and even that would be a PEP rating.
  8. N8FVJ

    N8FVJ Ham Member QRZ Page

    I guess bad spec sheets I quoted from normally trusted sites. I know 807 produces about 50 watts each PEP output.
  9. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    Yeah, I've used 807s.

    Then, I also have a 6-pack of them in the fridge.:p
  10. KC8QVO

    KC8QVO Ham Member QRZ Page

    I had not come across that design consideration, but that appears to be an issue. I don't have that voltage available. I have selections in the 800-1600v range (DC from a solid state rectifier with a filter cap) from a complete secondary, no center tap. So that would appear to require a different transformer. That isn't out of the question, but another part I would have to acquire vs using what I have.

    OK on the current vs. filter cap capabilities.

    I was going to correct the call in my post referencing that but by the time I went back to do it the "edit" option was gone. OK on the issues with that design/construction style.

    OK on the variances in the frequency spectrum. I have some ideas about that - but yes I am aware that can be a stretch.

    As far as direct experience - I have never built a transmitter or amp from scratch. The closest thing that may relate is I built an Elecraft K2/10. It was from a predefined kit with procedures already laid out for design theory, construction, tuning, and set up. From a design perspective - I started an amp project about 10 years ago. Back then I had worked with some elmers looking at schematics and learning about circuits in amps.

    At the end of the day - the project idea here would be (and is already) a giant learning experience. That is why I ask a lot of the questions I do. An idea is one thing, but as you note - there are a lot of ways something may not work out. Sharing ideas is cheaper than frying parts. To that point - I appreciate your, and everyone else's, input.

    Going back to the capacitance between tubes and multiple tubes - can that be measured when not in-use? Or is it one of those "imaginary" values that you only find when the circuit is in operation?

Share This Page