Tuning a loopstick to 3rd harmonic

I'm revisiting an old project.
I want to pick up 100khz loranc on a loopstick.
I can tune said loopstick to 33 khz 66khz and I sorta get a recognizable loran signal but messed up with local higher power stations, 100khz is completely wiped out by a local transmission.
So before I try some eleborate and therefore non functional filter circuits I was thinking of trying to tune said loopstick to 300khz and see what happens, 300khz is quiet on a comms receiver here.
Any better ideas?

By loopstick, I assume that you mean the item that we call a "ferrite rod antenna".

If you want to receive 100kHz, then the antenna should be resonant at 100kHz.
If it is resonant at some other frequency, the wanted signal will be attenuated.

dr pepper said:

I was thinking of trying to tune said loopstick to 300khz and see what happens

Click to expand...

Sounds like a truly bad idea.

If strong signals at some frequency other than 100kHz are causing problems, then you need some extra filtering.
A 100kHz bandpass filter perhaps.

JimB

That was my first thought, only 100 khz is totally swamped by 80kc which is much more powerfull, and from the same direction.
I can get a reasonable signal at 2/3 the frequency, but even that contains harmonics from radio4 on Lw.
In order to filter it I'd have to have the filter in the front end as any kind of gain just makes the front end clip.
The only other thing I thought of was a Q multiplier, however regen circuits are not particularly reliable over long periods.

I am thinking that the 100KHz Loran system has been decommissioned?

Many LORAN transmitter stations have been closed, but not all of them.

LORAN is currently rattling away to itself on 100kHz.
I have just been listening to it.

JimB

Loran was put back in commission at Anthorn a couple of years ago, I'm pretty sure its not so much for navigation but as an alternative to Gps as the latter is easily open to attack.
There are still a load of Fs700 frequency standards around that are compatible with loran, accuracy is excellent, the way it works compensates for reception phase errors.

I ordered some 100 kc crystals, I'll see if I can make a filter to remove everything I dont want.
Not sure if this is going to work, as I'm looking for the 3rd zero cross of the 100kc signal, the xtal may damp this out.
One thing about a loopstick is that theres always some noise at its tuned frequency.

dr pepper said:

I'm revisiting an old project.
I want to pick up 100khz loranc on a loopstick.
I can tune said loopstick to 33 khz 66khz and I sorta get a recognizable loran signal but messed up with local higher power stations, 100khz is completely wiped out by a local transmission.
So before I try some eleborate and therefore non functional filter circuits I was thinking of trying to tune said loopstick to 300khz and see what happens, 300khz is quiet on a comms receiver here.
Any better ideas?

Click to expand...

if the loran transmitters are working properly, you won't pick up harmonics of the signals on 300khz. don't confuse the pulsed modulation of the transmitters for the RF being anything other than a sine wave.

dr pepper said:

Loran was put back in commission at Anthorn a couple of years ago, I'm pretty sure its not so much for navigation but as an alternative to Gps as the latter is easily open to attack.

Click to expand...

after several public examples of GPS spoofing were revealed, they decided to reactivate LORAN stations. unfortunately, there are some that had already been bulldozed. from what i've read about it, i'm not sure if they intend to rebuild them, or just leave them out of the system. the position accuracy isn't as good (plus there's no altitude measurement with LORAN), but is sufficient to keep ships from colliding in the trade routes.

Loran from Anthorn only has one slave instead of the usual 3, and I heard it comes from the same mast as the master, how much use that will be for navigation I dont know.

If I tune the loopstick to 3rd harmonic below 100kc I get some signal, and if I tune to 66.6kc I get a good signal, not sure whats happening there.
At 100kc I cannot see anything but the submarine transmission from Skelton from the simple amplified loopstick setup.
My Sdr picks it up Ok, the issue is too wide a bandwidth using the loopstick.
I guess if I add some complexity I could go superhet and lock the local osc to the incomming signal.

you might want to put a variable cap across that loopstick so you can tune it. for 100khz, you might want to use 2 sections of an AM tuning cap to get about 600pf.

dr pepper said:

If I tune the loopstick to 3rd harmonic below 100kc I get some signal, and if I tune to 66.6kc I get a good signal, not sure whats happening there.

Click to expand...

a) what are you using for a radio? b) "3rd harmonic", and "below" are mutually exclusive, as a harmonic of 100khz will be a multiple of 100khz, i.e. 200khz, 300khz, etc... every once in a while there are textbooks that mention "subharmonics", but there don't seem to be any distortion mechanisms in practice that actually generate them, except a PLL seriously out of alignment. c) what you are seeing is likely to be an image in the SDR. it would help to know the sample rate and center frequency when you see the image at 66.6khz. theres a phenomenon with SDR radios known as "wrap around", where signals from outside of the passband show up. sometimes they are caused by strong signals generating IMD, or they have some mathematical relation to the sample rate, and center frequency. if you do effective filtering of the AM broadcast band, i think most of the images you are seeing will go away. also, SDR radios with 8-bit ADCs have a limited dynamic range, and the real cheap ones don't have baseband filters, so images and wraparound effects are more common with them. i have a regular RTL-SDR, a HackRF, and an SDRPlay radio. the RTL-SDR and the HackRF have 8 bit ADCs, and the RTL-SDR definitely has images and wraparound. The HackRF has baseband filters, and it's performance is better, but i can definitely tell that it has problems with images under certain conditions. the SDRPlay has 14 bit ADCs, baseband, and broadcast band filters, and i don't think i have seen any problems with images, except when the gain is cranked up too high. under normal usage it performs pretty well. what you should try is when you think you might be seeing an image, step the center frequency up 10 khz at a time. images will move the opposite direction in the waterfall from other signals. wraparound is more difficult to detect, but if you know what frequency a signal should be on, and you see it somewhere else, it's wraparound.

I have a loopstick with a 400pf polyprop cap and a 900pf polyvaricon across it, to get it to resonate at 100kc, I tested resonance both with an Rf gen, and by connecting the loopstick and its caps to my Lc resonance oscillator.
The ant goes to a untuned amplifier circuit comprising an initial fet amplifier, then 2 bipolar trannys as wideband amps, to give enough gain to get about 1000mV of signal, I've tried various gate resistances to control the Q of the loopstick.
My next step without anyones input would be to Lc tune the first stage of the amplifier, then if that not enough integrate a crystal filter into it, this would be a pain with the required impedance matching transformer & transistor.

Sorry incorrect use of terminology, but it sounds like you know what I meant.

The Sdr setup it completely seperate I only mentioned that as I used it to make sure loran was actually on air, the Sdr has its own antenna (miniwhip), the sdr doesnt pick up any subharms or harms from loran, that I can tell.
The Sdr doesnt show the usual 'lines' you get when imd sets in.
Reception of the loran signal with my sdr lloks 'normal', and comms receivers sounds 'normal'.

The more I think about it the more I suspect the amplifier is the cause of my woe's.

dr pepper said:

The more I think about it the more I suspect the amplifier is the cause of my woe's.

Click to expand...

Could easily be.

To be brutally honest, this idea of tuning to some weird (sub)multiple of the required frequency is a bit of a non-starter in my opinion.
Have you investigated the "Q" of the loopstick antenna?
What you need is some selectivity, so that when tuned to 100kHz for LORAN, the strong unwanted signals from Inskip (81kHz) and Anthorne (60kHz) are well down the slope of the response curve of the tuned antenna.

At my location, near Peterhead, I have Crimond on 52kHz about 10 miles away.
I have a home made VLF frequency converter using an SA602 which converts 0-500kHz up to 10.0 to 10.5MHz.
The input circuit is a 500kHz lowpass filter, directly in to the SA602.
Using 100feet or so of wire as an antenna, I have no problems with overloading and cross-modulation in the SA602 mixer.

dr pepper said:

The only other thing I thought of was a Q multiplier, however regen circuits are not particularly reliable over long periods.

Click to expand...

I stumbled across this:
http://theradioboard.com/rb/viewtopic.php?f=2&t=6660
which discusses applying Q multiplier techniques to a crystal receiver.
Worth trying?

JimB

Maybe your right.
I 'thought' it would work with a simple amp as I have nearly the same setup receiving 60 khz and it works fine, and inskip is only 1 khz closer to loran than 60 khz Msf.
I have a ferrite ant setup I made a while back from an epe article, if I can find it I'll try that, it has a q multiplier that can act as a bandpass or notch.

I got it in the end, all it needed was some scots shortbread.
Namely a shortbread tin grounded up near the loopstick to shield it from the strong 80 khz, the signal is so strong it booms through even when the loopstick is tuned for loran.

dr pepper said:

I got it in the end, all it needed was some scots shortbread.
Namely a shortbread tin grounded up near the loopstick to shield it from the strong 80 khz, the signal is so strong it booms through even when the loopstick is tuned for loran.

Click to expand...

having a close 80khz source definitely would cause problems difficult to deal with. most loopsticks are wound with very small litz wire, so they don't have enough Q to get a really narrow bandwidth. selectivity in communication receivers usually requires more than one RF stage, and multiple IF stages (which is where crystal filters are usually placed, so that there's a lot of gain before the filter, and the loss through the filter at that point is easily made up for). i've never seen a crystal filter in the RF portion of a receiver, even one operating on a single frequency. LORAN requires a fairly wide receive bandwidth in order to get clean pulse waveforms, and a crystal filter might be a bit too narrow for a LORAN receiver. it might not be too narrow if all you are trying to get is the 100khz signal itself to phase lock an oscillator to. as a matter of fact, you could have the 100khz come in, and feed the phase detector directly, and the phase locked oscillator would be a very narrow range VCO. the output of the phase detector would be the error signal for the VCO as well as an indication of how well the oscillator is tracking the LORAN station.

Agreed.
I used a 60khz crystal on my Msf (60khz) version of this circuit, it also is just a Trf receiver, it works well, although the xtal is well damped to prevent it from ringing, and to widen the bandwidth so tuning it isnt so critical.
Yes the loopstick has a basket woven coil normally intended for Lw, it has a Q of around 300.
The rx has an unusal effect of being 'pulled' onto the 80khz signal, the e field screen just about makes it possible to get a good signal at the 100khz, pulse shape isnt great compared to what it should be.
I'm not concerned with the 100 khz, the 3rd zero cross of each pulse is accurately timed, and is less affected by sky wave reflections because of the gaps between pulses, so grey line effect at dusk/dawn doesnt mess up accuracy, like it does with my other off air circuits.
You might have gathered that I'm going to use a processor to pick up on the zero crosses to control an oscillator, rather than a analogue Pll.
However as an interim play around I've been breadboarding a circuit that locks a Pll onto the 100hz portion of the signal, which is created by the repetition of pulses (not the Gri).
Locking the 100khz would be trickly as they invert the phase on certain pulses and theres phase mod, you'd have to detect the first pulse of the sequence and control phase comparison at certain times, I dont think even a long long loop filter would work as the 100khz isnt there long enough in the same phase over time, the 100hz repitition is though.

NTSC video locks the local subcarrier generator to 8 cycles of 3.58Mhz color burst. you can do pretty much the same if you lock to the main pulse of the LORAN station. of course the loop filter is going to use a longer time constant than you would need if the 100khz signal were continuous. since it's a gated reference signal, the trick is to use a sample/hold as part of the loop filter. when i was in the army, our frequency standard was locked to a WWVB receiver, and whenever the system was set up and restarted (we went on field exercises frequently) it usually took 72 hours for the system to settle completely, and be used for calibration.

Yes.
I did something remotely similar on my Msf standard, which is the same as Wwvb was pre 2012, 'cept the carrier doesnt reduce its commutated.
I used a tri state buffer, the output enable was switched with the 60 khz carrier, so sync was only attempted when the carrier was present.
Not quite sample & hold, more hold & sample.
10hz and 1hz content was pretty high so the loop filter is real slow, not quite 72 hours, 15 mins for less than 1 hz in 10 mhz, probably a few hours for max res.

It and my bench is a right mess, however I have a working protoype.
A 74hc00 as a VCXO divided down to 100hz with 3 74hc390's, the phase comparator is one of the 'hc00 gates, and the 100hz comes from a Ne567 tone decoder with raw loran Rf being fed into it from the loopstick amp.
Comparing it to one of my other frequency standards its pretty good, no phase drift over 15 to 20 mins, I dare say if I blow on the xtal it would drift.
Dont know whether to prototype it or start the code for the 'proper' version.