Bicycle antenna for long wave 200 khz

[dr pepper]

I commute 20 miles a day, and want to pick up a distant but strong station on 200khz.
Such freq's are usually received by a magnetic loopstck in commercial radios, but they are directional, so I think I need an e field whip.
I have a couple of designs similar to the popular miniwhip, but I was wondering what to use for the receiving element, an actual wire whip or a plate, maybe a couple of pieces of brass or pcb board arranged like a dart flight.

 

[betwixt]

My first thought is to use a short 'active antenna' (one example at: https://www.ra0sms.ru/p/the-active-antenna-mini-whip-10-khz-30.html). Basically a short whip feeding a simple amplifier with high input resistance to compensate for the poor impedance match. If you have access to the existing Ferrite rod you can feed the output to a couple of turns around it, if not a few turns around the entire receiver will provide enough coupling.

Brian.

 

[dr pepper]

I lashed up a quick circuit with a Ta7642 (Zn414), a Trf receiver, with the ferrite as the tuned portion.
It worked Ok.
Then added just the fet fed by a 24" whip, I put the ferrite vertically to reduce magnetic wave reception, it works but theres a ton of noise, then I tried my miniwhip which is similar to the above circuit with 5 turns around the ferrite, it too works but very noisy.
So I guess selectivity is an issue, I have a old car radio round here somewhere, I'm going to take at look at that for inspiriation.
Could also look at getting an old metal cased analogue tuner with Lw off ebay, they have a antenna input.

 

[betwixt]

Selectivity and TA7642 don't go together. You simply don't get enough selectivity from a single tuned circuit and those devices are also prone to overloading when a strong signal is present.

You shouldn't get too much noise though, did you add the resistor between the whip and ground? Without it you will build up static charge on the whip and that might sound like random interference or noise.

What signal on 200KHz are you trying to receive? It isn't a standard broadcast frequency.

Brian.

 

[dr pepper]

Agreed on the first point.
Yep I tried various resistances to ground on the whip, 10meg, 1 meg and a 470k trimmer, 50k seems to give the best performance, but its still unlistenable.
198 khz is radio 4 here in the Uk, it is a standard frequency its locked to an atomic clock, but this time I just want to listen to the broadcast, an hours riding early hours gets boring.
I couldnt find a proper schem for thsi car radio, however 'cept for the front end it looks like a standard 2 IF am receiver, the front end is tuned by a variable inductor, and the ant goes to a variable transformer coupled to the local osc tuning coil, then to a Rf pre amp.
The Ta7642 rx hisses, hums & some weird 50hz mains modulated crackle, tried it in 3 locations, very similar, the caps in the car radio are all dried out, however its reception is better, the noise is there but much less.

 

[betwixt]

I thought it might be 198KHz rather than 200KHz. I'm in the UK too here but all LW stations are on 9KHz spacings.

For single station listening, I would suggest you simply use two tuned circuits with a small capacitor between them. Use the existing ferrite rod and/or whip for the first tuned circuit, couple through a capacitor (suggest 47pF) to a second tuned circuit, also at 198KHz and you should get adequate selectivity. For the second tuned circuit, don't use a ferrite rod, use a screened coil. You can make your own but the easiest method is to utilize an old IF transformer from an AM radio (typically tuned to 470KHz) and add an extra capacitor across it to drop its resonant frequency.

Are you using any additional AF amplification? If you are, please post a schematic so we can see how it is coupled.

Brian.

 

[dr pepper]

Radio 4 was 200 khz a long time ago, before we went to 9khz spacings.
I have some IF cans, old ones with 2 slugs and later japanese style, I think I have yellow & white.
Yes I have a single transistor amp after the '7642, then to a Lm380, Agc is down to the '7642.
I have a schem at home I'll do a scan later, to save a transistor could I put the IF trans in the whip Rf amps collector and couple its o/p to the ferrite, ie have the IF can before the ferrite.
I have a Vna so I can make an effort to impedance match.

 

[dr pepper]

Well I changed the npn transistor in the Rf pre amp to a fet, with a 220k to ground on the ant, and theres now a yellow 1st If trans in the drain circuit instead of a resistor, it seemed to work best with a 150pf cap to the ferrite ant, maybe this would be better with a winding on the ferrite.
The ferrite doent seem to pick up stations, the whip amp must be damping it, with a 24" whip though if I turn off most of the appliances in the house the radio sounds Ok, however anything electrical wipes it out.
So I dont think the Ta7642 is going to be any good for this project, but I will try the receiver in a couple of different places.

 

[betwixt]

I would remove the resistor in series with the whip and add one (~220K) from the base of the whip to ground. Its purpose being to stop static charge building up and zapping the first transistor or the coupling capacitor.

However, the problem seems to be more related to the tuning, I would say 150pF is far too high and you have effectively placed two tuned circuits across each other. I would forget the secondary of the IF transformer, go back to no more than 10pF and couple it from the drain/collector of the first amplifier. You should be able to tune the first stage for a very noticeable increase in signal. It is quite likely that the secondary winding on the transformer only has a few turns as it would be for impedance matching to an IF amplifier stage rather than peaking at 470KHz like the primary side.

Trick: if you have a capacitance meter, try isolating the capacitor in the base of the transformer and measure its value. From that, and knowing it was originally tuned to 470KHz, you can work out the inductance of the coil. The you can calculate how much extra capacitance is needed to drop the resonant frequency down to 198KHz.
If you have an inductance meter its even easier, just isolate the capacitor and measure it directly.

Brian.

 

[dr pepper]

I deleted the cap and 1k inline with the gate, these were for the bipolar tranny I had anyway, so now there is a 220k to ground on the whip, I cycle past a military transmitter on one of my rides so I'll put a pair on 1n914's to ground as well.
My yellow If can doesnt have a parallel cap its vintage, I measured it at 920 uH.
I tried various things including tapping the primary hi z side instead of the lo z, but then realised instead of connecting to the drain I was connecting to the source of the fet.
Correcting this and replacing the 150pf for a 10pf the yellow If trans does peak although its fairly wide, and the radio works 100x better, reception during the day at least is good, hardly any noise, I'm now having issues overloading the '7642, I can deal with that though, I also note theres 2 points on the tuning cap across the ferrite where I get radio4, maybe something is clipping.

 

[betwixt]

With 920uH inductance you need to add 703pF of capacitance across it (680pf & 22pF in parallel) to make it resonate at 198KHz.

Probably the best way to reduce overload is to reduce the 10pF even more -or- to use say two 10pF capacitors in series and from their junction add a pair of parallel back-to-back Schottky diodes (BAT85 or 1N6263) to ground. The less coupling capacitance you use, the higher the 'Q' of both tuned circuits will become so you not only reduce overload but increase selectivity as well.

Brian.

 

[dr pepper]

Well I have 470 and 150pf in parallel, there is some adjustment on the trans, and I can adjust it through a very gentle peak using a sig gen's o/p close to the ant.
I guess a gimmick capacitor would give me <10pf and some adjustability.
I took the rx out in the back yard, poor performance, took it out on a ride, 2 places kinda Ok, and 1 poor performance, so I have both too much and too little signal.
It seems sensitivity is an issue now, I was thinking put another fet between the ferrite and the '7642, the latters impedance is <100k, a fet could provide 1 meg ohm, increasing Q of the ferrite, and giving me some more db's.
I dont know why my 'shop has a good signal and not anywhere else, my house is below street level, another thing my bicycle has a hub alternator for lighting, one terminal is gounded to the frame, I hope thats not going to be a problem.
I was thinking this'd be an 'easy' project, thats always a mistake.

 

[betwixt]

I think the mistake, if that's how you perceive it is to use the TA7642. It was designed for super-low cost pocket receivers and assumed good signal levels were present at its input.

From the outset you could have used a single transistor regenerative receiver which would give high gain and good selectivity in a very simple circuit. The drawback with that kind of receiver is having two controls, one to tune and one to set the feedback but you are using a single fixed frequency so once set they shouldn't need adjustment.

For interest, I've built single stage receivers followed by an audio amp that can pick up MW stations from the USA across the Atlantic with just a few metres of hook up wire for the antenna.

Brian.

 

[dr pepper]

I have built a couple of receivers with this device and they work, however not with a whip ant your probably right I'm expecting too much from such a device.
I never liked the idea of a regen, having to fiddle with the regen control, a bit like fiddling with the agc control on the '7642 I spose.
So a different approach, I've been looking at schems for old car radios, the ant seems to go straight to a tuned transformer, I tried that with the '7642 just to see what would happen, the peak is much sharper and mains borne noise much less, the Q obviously much higher looking into the whip one side and the fet gate the other.
Another trick I see on old car sets is a choke inline with the ant socket, looks like maybe a load or compensation, using a calc for long wave I'd need 100mH that seems a lot.
So I'm looking for new ideas, I have a vintage analogue hifi tuner, it has Long wave, and an unbalanced ant i/p.

 

[dr pepper]

Went to my old mans house, he's a mechanic, found an old radiocassette, been in the damp for a while, the tape deck, amp, power supply & Fm were all destroyed, but I managed to get the Am chip to work (La3800 series), on Lw too, interesting the stereo decoder is a seperate chip.
The receiver does have some noise however it works with a larger ant, which makes it more useable, looks like agc is definately a problem with the 7642, anything more than a poor noisy signal is too much.
Something learned there.

 

[dr pepper]

I found the source of noise, my outside floodlights, they have a switching supply for the Led, probably a buck reg.
And they must be a near harmonic of 198khz, a loopstick doesnt seem to be too sensitive to it.
So I'm going to see if street lights cause simiar noise, if so then a different tack is required, maybe a pair of receivers with loopstick ants 90 degrees rotated and a circuit to select the best signal receiver.

 

[dr pepper]

Using a battery powered Vlf receiver that covers 198khz with a whip ant at various places en route reception was good most of the time, the only bad spots were those markers at traffic islands, and a couple of times close to houses.
More than good enough to hear the news, so I'll continue with the project.

 

[dr pepper]

I dissasembled the radiomobile car radio, hacked the board so it just has the La1130 am section, and hard wired it for Lw, then fixed a nut/bolt to the variometer tuning device and set it to 198kc.
I added a J113 pre amp to a 24" brass whip aerial and a Lm380 audio power amp, the rx quiescent pulls < 10ma.
A simple bridge powers it from the hub dynamo, I put a 220n across the ac from the dynamo to suppress Rf buzz, it works Ok.
I have a 90s car speakerphone, might use that to house the speaker.
The only issue is the lights, the front one pulls the power down to 12v, not a problem, but when the rear flashes it kills the radio as its Led forward voltage is only 6v, so I'm going to replace it with a 'grow light', bright red with a Fv of 12v, must have 6 leds in series x2.

 

[dr pepper]

Replacing the rear red led module for a bright red grow light increased the bus voltage from 6v to around 11v the rx doesnt cut out now, and the rear light is still bright.
The rx had an annoying buzz that changed with road speed, it was hard to find, turned out to be the inductance of the dynamo resonating with the capacitance of itself and the wiring, judicous placing of 100nf caps reduced the freq of this way down below 198 khz and now the radio works well.
Its interesting to see out on the road what causes intereference and the effect it has, unlike hf/vhf it isnt always that good signals are had high up on open ground.

 

[dr pepper]

OK I figured out coax impedance.
Coax is a transmission line and it seems that the physical layout affects the impedance, max efficiency for tx occurs around 30r, and rx 77r, so 50r is inbetween.
I made the antenna itself plug in on 4mm binding posts, I had to make them a little tighter so's not to loose them in the hedge but its a good way to test different ones.
An 80 x 35mm tin plate made from a can of contact cleaner seems to work the best, good signal and just a tiny bit of dynamo noise.