[ARC5] Rollerductor setting

Sat Apr 22 10:40:27 EDT 2017

I remember this subject coming up a few years  back, and searched my email archives.  The thread was something like  "ARC-5 antenna match back in July of 2013.  Here's what I posted on what I use here - the KJ4KV auto-transformer matching network that Walt described in his ER in Uniform column command set series of articles around1989-1990.  I think some one later posted the specific ER issue in the same thread:

....Its really pretty simple

Start by envisioning a parallel L/C tuned circuit.

Both antenna shield and a lead to TX chassis run to "cold" end of coil, along with one side of the variable cap of the tuned circuit.  I used a 200pf wide spaced transmitting cap to avoid flash-over.  
Antenna coax center conductor tapped on coil about 6 turns above ground/cold end of coil.
ARC-5 transmitter connected to tap about 3 turns above cold end of coil
through a fixed 200 pf series cap.   I use a small transmitting type mica cap.
Other side of variable cap connects to top "hot" end of parallel circuit coil, nothing else connected there.

Operating instructions:

With transmitter keyed, resonate the L/C circuit for max output on a wattmeter connected between the matching network and the antenna.

Peak transmitter output indicated on the wattmeter by adjusting the transmitter's roller inductor for best power transfer through it and the 200pf fixed series cap.

Back off on the transmitter coupling coil, then advance it until you find a "sweet spot" or reach output saturation.  You will like go into saturation somewhere past 50% of full travel, vs never maxing out when feeding a nominal 50 ohm unbalanced load directly into the transmitter.

I am getting about 50 watts out of a totally stock dynamotor powered system.  (Actually it's ARC-5 transmitters lined up to a SCR-274N BC-456 screen modulator/dynamotor unit, with the final amplifier screen resistor changed to the SCR-274 transmitter value). The screen resistor difference is the only significant electrical difference between the two types of transmitters, most likely differ due to plate modulated ARC-5 and screen modulated SCR-274 sets.....(end of  old posting)

I don't recall the coil  winding info from Walt's article.  The matching network I built and still use here has a pi network coil set from an old Eico 720 as the parallel L/C circuit inductor.  The 720 coil consists of an  LF section to get down to 80 meters of many close spaced turns, and a second section of much fewer wide spaced turns used on the higher bands.  I use both sections in series to resonate at 80 meters, and just the LF section to resonate on 40 meters, shunting the smaller coil section with a clip lead.
Since I originally posted the info back in 2013, I have become suspicious of what appears to be an overly optimistic Heathkit wattmeter here, the 50 watts noted may be a bit on the high side.  However, in comparison to any other matching method that I have tried  (series cap at TX output, series cap feeding a 1:4 W2AU broadband un/un transformer, internal roller inductor only), the KJ4KV circuit has always provided the highest measured output (50-60 watts indicated on the Heathkit meter vs 35 or less otherwise) to my  low SWR 80 and 40 meter inverted vee dipoles and to a Bird 500W dummy load (unfortunately  the internal wattmeter of it is non-functional).
Also, IIRC, Mike, WU2D did an article in ER a relatively few years back that compared results of using various types of output matching circuits with the command sets.

    On Saturday, April 22, 2017 8:11 AM, Michael Hanz <aaf-radio-1 at aafradio.org> wrote:

  All good stuff.  It may also be useful to remember that things can go south in a hurry when changing frequencies, depending on how close to a quarter wave your antenna is.  See the impedance curves in Chart 2 at http://aafradio.org/docs/Aircraft_Antenna_Design.html for example.  I'm not familiar with the MFJ tuner, but the active tuners that were used for HF in postwar military aircraft (AN/ARC-21, etc.) had only one job to do, and that was to condition the antenna to represent a purely resistive load for the transmission line at the operating frequency.  That made tuning the transmitter much less complicated.  Perhaps the MFJ doesn't have the parameters necessary, but I think I would be considering ways to modify *it* rather than the command transmitter.  That way you ought to be able to run a 4:1 or 9:1 unun by itself between the tuner and the command transmitter.

           73,
  - Mike  KC4TOS

 On 4/22/2017 3:42 AM, Brian wrote:

   OK folks,   Time to get back to basic principles.   A pair of 1625s in parallel probably requires a resistive load of about 1.2 k Ohm. The tank coil has about 20 turns – the lower frequency sets have more turns and the 7 – 9.1 MHz set has fewer. The variable coupling coil has 4 turns. So, the turns ratio is about 5:1. Therefore, the impedance ratio is about 25:1. Hence, the output resistance required, if we follow the Jacobi theorem, should be 1.2 k / 25 = 48 Ohm. By altering the degree of coupling using the ANT COUPLING knob, matching down to 5 Ohm should be easy. But the coupling coil doesn’t reach full coupling, unless you open the transmitter  top cover and reset the coupling gear on the coupling coil shaft; so, 48 Ohm matching is not possible. The simpler way is to lift the grounded end of the coupling wiring and put that end in series adding with a turn or two of insulated wire wound round the bottom end of the tank coil. An extra turn gets you an impedance ratio at full coupling of about 16, thus giving an output resistance of 75 Ohm, variable by changing coupling. With extra turns, the lowest antenna resistance you can match will increase. So, now you can see that resistance matching was done with varying the coupling.   An average HF antenna on an aircraft would stand off the fuselage say about a metre, possibly be 3 mm diameter and might be 10 m long. The radiation resistance would be 1 to 2 Ohm, depending on frequency. The capacitance between the antenna wire and the fuselage would be about 256 pF. You can use the roller inductor to cancel out this capacitance. At 7.15 MHz, you will require about 2 uH. However, if the coupling range is not sufficient to match the resistive component, then you can use a capacitor between the ANT terminal and antenna feed-point. The original supplied with the BC-442 or the RE-2 was 75 pF. What you now have is a capacitive divider. The roller inductor now needs to cancel out a smaller capacitance – two capacitors in series. So,  you will need more inductance. The notion that you start with half or half of half of the roller inductance is relatively meaningless unless you know the specific radiation resistance and capacitance.   Now, say your home antenna is also 3 mm diameter wire, but 10 m above ground and 15 m long. The capacitance will now be about 284 pF. And the radiation resistance will be higher. At the higher frequency of the 40 m set, if your antenna is 0.18 of a wavelength above ground (7.2 m), its radiation resistance will be about 50 Ohm, and its capacitance will be 295 pf. The likelihood of matching to an original 40 m Command Tx is close to zero. So, adding a turn to the coupling system and possibly using a series capacitor should get you there. Using a UNUN is irrelevant and may introduce losses.   Say you claim to get 40 W into a 50 Ohm antenna from one of these rigs. The Voltage at the input to the antenna will be about 45 V. Not very bitey. But if you use a series capacitor to achieve your impedance match, the Voltage at the rig’s antenna terminal can be 450 V if your impedance transformation is 10:1. The average broadcast band tuning capacitor may flash over. I wouldn’t recommend varying the capacitor with a bare hand. You can also see that with an aircraft’s much lower radiation resistance, the Voltage at the Tx antenna post would be much, much higher – which is why a 10 kV vacuum capacitor was supplied.   73 de Brian, VK2GCE.

 ______________________________________________________________
ARC5 mailing list
Home: http://mailman.qth.net/mailman/listinfo/arc5
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:ARC5 at mailman.qth.net

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.qth.net/pipermail/arc5/attachments/20170422/959c64b1/attachment-0001.html>