I will build a very simple antenna array (1MHz...5MHz adjustable oscillator=>emitter follower=>some adjustable signal delay mechanism=>antennas) with two antennas of varying phase angles between them to produce different EM waveforms (superposition of sinewaves).

The phase delay mechanism will probably be simply, some coax cables of different lengths.

For some theoretical and practical reasons, I will have to keep the distance between the antennas no more than 1 meter.
The phase delay caused by the EM wave velocity between the antennas will be about 0.3 nanoseconds (for 1 to 5 MHz, that means a phase angle of 0.36 to 1.8 degrees).

To be able to keep track of what is exactly going on between the antennas and to make sure that the phase delay mechanism (length difference of two coax cables) is set right, I will have to use some kind of phase angle measurement method or device of great precision and which wouldn't cost an arm and a leg.

So far, I could only find the first method mentioned in this article (http://eeshop.unl.edu/pdf/OscilloscopeTutorial--PhaseMeasurement.pdf), requiring an oscilloscope. I suppose I will already need to buy a humble oscilloscope to measure the frequencies and amplitudes. However, to be able to see a 0.3 nanosecond phase delay, maybe an oscilloscope of about 1 GHz or more bandwidth would be needed. That would be way higher than my budget (300-400 euros at most)

I've heard of vector network analysers and spectrum analysers but I can't make sure if a reasonably cheap one could do the job.

In short, is there any means (device or method) of measuring a 0.3 nanosecond phase shift with a low budget?
Or am I asking too much for such a budget and should reconsider a different antenna array design?

Any ideas would be welcome.

 

I would suggest moving from 1-5 MHz to > 300 MHz. And mount antennas on adjustable distance mounts.

In this manner.......the phase may be adjusted to any ratio with antenna separation distance.

Once you see how physical distance can change phase.........then it's easy to discern different feed line lengths for same effect.

 

I will build a very simple antenna array (1MHz...5MHz adjustable oscillator=>emitter follower=>some adjustable signal delay mechanism=>antennas) with two antennas of varying phase angles between them to produce different EM waveforms (superposition of sinewaves).

The phase delay mechanism will probably be simply, some coax cables of different lengths.

For some theoretical and practical reasons, I will have to keep the distance between the antennas no more than 1 meter.
The phase delay caused by the EM wave velocity between the antennas will be about 0.3 nanoseconds (for 1 to 5 MHz, that means a phase angle of 0.36 to 1.8 degrees).

To be able to keep track of what is exactly going on between the antennas and to make sure that the phase delay mechanism (length difference of two coax cables) is set right, I will have to use some kind of phase angle measurement method or device of great precision and which wouldn't cost an arm and a leg.

So far, I could only find the first method mentioned in this article (http://eeshop.unl.edu/pdf/OscilloscopeTutorial--PhaseMeasurement.pdf), requiring an oscilloscope. I suppose I will already need to buy a humble oscilloscope to measure the frequencies and amplitudes. However, to be able to see a 0.3 nanosecond phase delay, maybe an oscilloscope of about 1 GHz or more bandwidth would be needed. That would be way higher than my budget (300-400 euros at most)

I've heard of vector network analysers and spectrum analysers but I can't make sure if a reasonably cheap one could do the job.

In short, is there any means (device or method) of measuring a 0.3 nanosecond phase shift with a low budget?
Or am I asking too much for such a budget and should reconsider a different antenna array design?

Any ideas would be welcome.

For a test and instrumentation solution; the best I can come up with is the avalanche transistor nS pulse generator. The ZTX415 is made for the job, some people get away with selected samples of the 2N2369.

The technique was common in GHz bandwidth sampling CROs.

 

I believe there are digital scopes----300-400 dollars.......that can measure that.

Where are your measuring points? The feed-points of the antennas?

I am old school taught. I would think that you need at least 1/4 wavelength separation between antennas for phasing.

That's gonna be hard to do with 1 MHz and only a meter separation.

 

Sinan tannsever wrote in post #1:
"...with two antennas of varying phase angles between them to produce different EM waveforms (superposition of sinewaves)"

Remember: superposition of two sine waves of the same frequency gets you a sine wave, not another waveform (taken literally).

@ian field I have tried many small signal NPN transistors in collector avalanche mode and I don't remember any of them not working. The 2N3904 was much faster than the 2N2369's that I tested. And the '3904 was a lot cheaper and easier to get my hands on too.

 

I will build a very simple antenna array (1MHz...5MHz adjustable oscillator=>emitter follower=>some adjustable signal delay mechanism=>antennas) with two antennas of varying phase angles between them to produce different EM waveforms (superposition of sinewaves).

The phase delay mechanism will probably be simply, some coax cables of different lengths.

For some theoretical and practical reasons, I will have to keep the distance between the antennas no more than 1 meter.
The phase delay caused by the EM wave velocity between the antennas will be about 0.3 nanoseconds (for 1 to 5 MHz, that means a phase angle of 0.36 to 1.8 degrees).

To be able to keep track of what is exactly going on between the antennas and to make sure that the phase delay mechanism (length difference of two coax cables) is set right, I will have to use some kind of phase angle measurement method or device of great precision and which wouldn't cost an arm and a leg.

So far, I could only find the first method mentioned in this article (http://eeshop.unl.edu/pdf/OscilloscopeTutorial--PhaseMeasurement.pdf), requiring an oscilloscope. I suppose I will already need to buy a humble oscilloscope to measure the frequencies and amplitudes. However, to be able to see a 0.3 nanosecond phase delay, maybe an oscilloscope of about 1 GHz or more bandwidth would be needed. That would be way higher than my budget (300-400 euros at most)

I've heard of vector network analysers and spectrum analysers but I can't make sure if a reasonably cheap one could do the job.

In short, is there any means (device or method) of measuring a 0.3 nanosecond phase shift with a low budget?
Or am I asking too much for such a budget and should reconsider a different antenna array design?

Any ideas would be welcome.

Build two identical receivers with all cable lengths exactly the same. Then place one antenna about 4" further from the source than the base antenna. There is your 0.3nSecond delay.

 

The period of 1 MHz is 1 usec or 1000 nsec. 90 degrees is 250 nsec.
1 degree is 2.77 nsec. 1 nsec ~ 1 ft. (11"). A little less than 1 meter will give about 1 degree shift.

But that's all that it gets you. Your radiation pattern is poo poo.

The idea of phasing antennas is not so that you can measure a phase shift. The phase shift, working with antenna element separation, is to direct or guide the radiation pattern. That pattern is the superposition result.

 

If you drive two spaced antennas with the same frequency you could perhaps use a receiver tuned to that frequency to detect the angle, relative to the line joining the antennas, at which the received signal is a maximum. That angle would be a measure of the phase difference between the two antenna signals.

 

I would suggest moving from 1-5 MHz to > 300 MHz. And mount antennas on adjustable distance mounts.

In this manner.......the phase may be adjusted to any ratio with antenna separation distance.

Once you see how physical distance can change phase.........then it's easy to discern different feed line lengths for same effect.

I could adjust the interference of the waves smoothly by adjusting the distance. And without a horde of cables. Thanks for the great idea. I wish I could work in the 300 MHz range but the oscilloscope price would be a problem there.

 

The period of 1 MHz is 1 usec or 1000 nsec. 90 degrees is 250 nsec.
1 degree is 2.77 nsec. 1 nsec ~ 1 ft. (11"). A little less than 1 meter will give about 1 degree shift.

But that's all that it gets you. Your radiation pattern is poo poo.

The idea of phasing antennas is not so that you can measure a phase shift. The phase shift, working with antenna element separation, is to direct or guide the radiation pattern. That pattern is the superposition result.

I think I see your point. For the antennas working at 1 to 5 MHz, a 0,3 ns delay would have negligible effects on te resulting pattern. However, I will be doing some experiments with this setup, it would be ideal if I could calculate the resulting radiation pattern as accurately as possible.

 

For a test and instrumentation solution; the best I can come up with is the avalanche transistor nS pulse generator. The ZTX415 is made for the job, some people get away with selected samples of the 2N2369.

The technique was common in GHz bandwidth sampling CROs.

Thank you for the suggestion. As I could understand, avalanche transistors operate by pushing the collector-base diode to past its reverse biased breakdown voltage, reaching relatively high currents in less than nanoseconds. It really sounds very handy but wouldn't it produce square waves? And did you recommend it as some means of measurement or producing some kind of high precision indicator?

 

Sinan tannsever wrote in post #1:
"...with two antennas of varying phase angles between them to produce different EM waveforms (superposition of sinewaves)"

Remember: superposition of two sine waves of the same frequency gets you a sine wave, not another waveform (taken literally).

@ian field I have tried many small signal NPN transistors in collector avalanche mode and I don't remember any of them not working. The 2N3904 was much faster than the 2N2369's that I tested. And the '3904 was a lot cheaper and easier to get my hands on too.

Thanks for the early warning. That was a misconception. I mean, I intend to produce variable forms of sine waves, by adjusting the phase angle between the two interfering sine waves.

But as I mentioned in the above reply, I couldn't understand, how would an avalanche transistor pulse generator be used here?

 

I do not recommend you try this approach because the electrical design of it would be very difficult, and verification of the performance would be technically challenging, even if you had the necessary equipment, but since you asked, here it is for background:


The circuit above and related information can be found at
http://www.microwavejournal.com/art...nications-and-radar-part-i-uwb-communications
The input signal is presented to the sampling circuit and complimentary narrow sampling pulses are applied to J51 and J53, causing a voltage to be stored on C4 and C5 that is representative of the input signal during the period of the sampling pulse. The pulses can be much shorter than a nanosecond and by changing the timing of the pulses with respect to the phase of the input signal over a number of input signal cycles an image of the input signal can be constructed. Since the circuit deals with sub-nanosecond pulses, it would be difficult to design and calibrate.

Other meghods such as the phasing methods mentioned above in this thread are probably more suitable for your needs.

 

Thank you for the suggestion. As I could understand, avalanche transistors operate by pushing the collector-base diode to past its reverse biased breakdown voltage, reaching relatively high currents in less than nanoseconds. It really sounds very handy but wouldn't it produce square waves? And did you recommend it as some means of measurement or producing some kind of high precision indicator?

You tune the pulse width by connecting a coaxial stub to the collector. In a sense its the timing capacitor - but more importantly; its a transmission line with predictable characteristics.

AFAICR: the end is open - so you get a reflected wave.

 

I will build a very simple antenna array (1MHz...5MHz adjustable oscillator=>emitter follower=>some adjustable signal delay mechanism=>antennas) with two antennas of varying phase angles between them to produce different EM waveforms (superposition of sinewaves).

The phase delay mechanism will probably be simply, some coax cables of different lengths.

For some theoretical and practical reasons, I will have to keep the distance between the antennas no more than 1 meter.
The phase delay caused by the EM wave velocity between the antennas will be about 0.3 nanoseconds (for 1 to 5 MHz, that means a phase angle of 0.36 to 1.8 degrees).

To be able to keep track of what is exactly going on between the antennas and to make sure that the phase delay mechanism (length difference of two coax cables) is set right, I will have to use some kind of phase angle measurement method or device of great precision and which wouldn't cost an arm and a leg.

So far, I could only find the first method mentioned in this article (http://eeshop.unl.edu/pdf/OscilloscopeTutorial--PhaseMeasurement.pdf), requiring an oscilloscope. I suppose I will already need to buy a humble oscilloscope to measure the frequencies and amplitudes. However, to be able to see a 0.3 nanosecond phase delay, maybe an oscilloscope of about 1 GHz or more bandwidth would be needed. That would be way higher than my budget (300-400 euros at most)

I've heard of vector network analysers and spectrum analysers but I can't make sure if a reasonably cheap one could do the job.

In short, is there any means (device or method) of measuring a 0.3 nanosecond phase shift with a low budget?
Or am I asking too much for such a budget and should reconsider a different antenna array design?

Any ideas would be welcome.

Why not just use a lecher line and a micrometer?