UK Vintage Radio Repair and Restoration Discussion Forum - View Single Post

kalee20 · 15th Mar 2008, 4:42 pm

Hi Dom and Mike,

First, if you are using a short length (and at 750kHz this would be less than 50 metres), you can forget about it being a transmission line, and think of it as being a capacitance. If it is 'lossy' coax, then it will be a capacitor with a resistor in series. You won't need to worry about matching to it - I'm assuming you haven't yet wires it around your house, Dom.

Coax cable is lossy due to the series resistance of the conductors, and the leakage due to the insulation not being perfect. So, power is lost as heat. If the ratio of these two loss resistances is correct (and the value of the ratio is related to the value of capacitance per metre and inductance per metre), the cable will have a constant impedance with frequency - even though what you get out is less than you put in. And, if you have an infinitely long length, you can measure this impedance with an AVO at the available end, no matter what the remote end is conected to!

When people talk about 'leaky feeders' they mean that there's an additional loss due to the cable radiating some of its signal. I doubt whether any sensible-sized coax would radiate anything at 750kHz, but in post 68 above, Dom's already got the idea of a scheme to ensure a bit of local radiation in each of his rooms.

Now, as to practice, Dom I'd certainly remove your dummy load when it's time to try radiating. You need your wire aerial from the centre of the coax, and a good earth (or a similar length wire) from the coax outer. And, you'll need a lot more voltage! The turns ratio down from the tank is only to match down to your coax (in anticipation of your full-house-length). At the remote end, you need to transform back up again, exactly as you've thought in your post 68.

Mike, I'm not convinced that matching for maximum power transfer from the tank circuit is the way to go. I'd certainly agree with you that the higher the Q of the loaded tank, the more power is lost in the tank itself due to coil losses. But, in principle, you can get around that by making a better coil (which will increase the Q of the loaded circuit), then increasing the coupling to the aerial (which will bring Q back down again). And now, you've got more power in the aerial! If you match for maximum power transfer with a given tank, then without doing any sums I'd predict that for maximum power transfer you'd have half the power dissipated in the tank, and half in the aerial. Not the best for efficiency. (After all, very few audio amplifiers are matched for maximum power transfer - most are either matched with a lower resistance, such as pentode output, or with a higher resistance, such as triodes or emitter-followers.)

The reason that maximum power transfer is often a red herring, is that there is the extra degree of freedom in that the drive is (in principle) adjustable. If you slug the output too much, you just increase the drive. And the limiting factors are distortion-causing mechanisms such as bottoming the valve (or even letting the anode swing negative, as Dom's reported).

Both of you - you are challanging my thoughts and I'm really appreciative!

15th Mar 2008, 4:42 pm	#81
kalee20 Dekatron Join Date: Feb 2007 Location: Lynton, N. Devon, UK. Posts: 7,088	Re: AM Micro Transmitter Hi Dom and Mike, First, if you are using a short length (and at 750kHz this would be less than 50 metres), you can forget about it being a transmission line, and think of it as being a capacitance. If it is 'lossy' coax, then it will be a capacitor with a resistor in series. You won't need to worry about matching to it - I'm assuming you haven't yet wires it around your house, Dom. Coax cable is lossy due to the series resistance of the conductors, and the leakage due to the insulation not being perfect. So, power is lost as heat. If the ratio of these two loss resistances is correct (and the value of the ratio is related to the value of capacitance per metre and inductance per metre), the cable will have a constant impedance with frequency - even though what you get out is less than you put in. And, if you have an infinitely long length, you can measure this impedance with an AVO at the available end, no matter what the remote end is conected to! When people talk about 'leaky feeders' they mean that there's an additional loss due to the cable radiating some of its signal. I doubt whether any sensible-sized coax would radiate anything at 750kHz, but in post 68 above, Dom's already got the idea of a scheme to ensure a bit of local radiation in each of his rooms. Now, as to practice, Dom I'd certainly remove your dummy load when it's time to try radiating. You need your wire aerial from the centre of the coax, and a good earth (or a similar length wire) from the coax outer. And, you'll need a lot more voltage! The turns ratio down from the tank is only to match down to your coax (in anticipation of your full-house-length). At the remote end, you need to transform back up again, exactly as you've thought in your post 68. Mike, I'm not convinced that matching for maximum power transfer from the tank circuit is the way to go. I'd certainly agree with you that the higher the Q of the loaded tank, the more power is lost in the tank itself due to coil losses. But, in principle, you can get around that by making a better coil (which will increase the Q of the loaded circuit), then increasing the coupling to the aerial (which will bring Q back down again). And now, you've got more power in the aerial! If you match for maximum power transfer with a given tank, then without doing any sums I'd predict that for maximum power transfer you'd have half the power dissipated in the tank, and half in the aerial. Not the best for efficiency. (After all, very few audio amplifiers are matched for maximum power transfer - most are either matched with a lower resistance, such as pentode output, or with a higher resistance, such as triodes or emitter-followers.) The reason that maximum power transfer is often a red herring, is that there is the extra degree of freedom in that the drive is (in principle) adjustable. If you slug the output too much, you just increase the drive. And the limiting factors are distortion-causing mechanisms such as bottoming the valve (or even letting the anode swing negative, as Dom's reported). Both of you - you are challanging my thoughts and I'm really appreciative!