27 MHz Amplification Query

[Borber]

May be that your counter has not enough sensitivity or his input capacitance is too big and shows wrong results on other points than on generator.

Crystal for 27.125MHz is 3rd overtone and fundamental frequency is slightly above 9MHz.

 

[RMMK]

May be that your counter has not enough sensitivity or his input capacitance is too big and shows wrong results on other points than on generator.

Crystal for 27.125MHz is 3rd overtone and fundamental frequency is slightly above 9MHz.

Nah Sir....... Its been very rare or you may say the first time for me that wikipedia actually solved my problem. It was an article on Parasitic Oscillations in which it was written that:-
"Bypass capacitors may be put at power supply connections, to provide a low-impedance path for AC signals and prevent inter-stage coupling through the power supply"

I placed a 470pF capacitor across the +12V rail & ground in b/w the previous stage and the first stage on the pic i uploaded in my last post! Just now i counted 27.130MHz on my frequency counter over the collector of this stage......

Plz do continue to help me.... I will update on the results I get from the next two stages!
Thanks!

- - - Updated - - -

But sir I couldnt understand what you said in the last two lines....... could you please elaborate it for me?

 

[Borber]

Last line is pure info about crystal and has nothing to do with first line. First line is describing possible problem with counter.

 

[RMMK]

At-last I have been able to successfully tune the whole circuit to the last stage of 2N3866 Transistor PA....... I counted some 27.130MHz at the collector of this transistor!

One thing I wanted to ask is when i placed the capacitor at the collector and the other end of the capacitor was free; i counted the frequency by placing probe of freq. counter on that terminal of the capacitor and it read the same 27.130MHz but when I connected that free end of the capacitor to the pi-network and tried to measure again the frequency at the end where antenna was supposed to be or even at the collector of any transistor it counted some 2Mhz or 3Mhz or at time 9MHz....... is that okay??
But when I remove the capacitor originating from the collector from the next stage and cont the frequency it does give me a measurement of 27.130MHz....... why is it so??

 

[RMMK]

Why is it happening as stated above?? I had an old chinese flying heli toy... it works on a 40MHz system...... i placed the +ve probe of my frequency counter on the place from where the antenna was originating on the transmitter and the -ve probe of the freq. counter on the ground....... all i counted were some really weird values over it........
But the transmitter works and helicopter flys normally........
So why this anomaly or is it just normal........

But when i place the probe while disconnecting the next succeeding stage while the preceeding stages remain intact on the 27MHz transmitter, i just completed, its giving me a count of 27.130MHz........
i havent been able to verify whether the transmitter works b/c i havent built the receiver yet....... I will though upload a schematic of a receiver for verifying the design.......
So first plz tell me why is it happening???

 

[Borber]

To test RF amplifier or transmitter and tune it use RF diode detector and DVM: http://www.techlib.com/electronics/detect.htm

 

[betwixt]

It can be for many reasons, the most likely is that the signal you are looking for is masked by other signals at different frequencies. The input amplifier on the frequency counter is not selective (nor should it be) so any signal will be included in the count. It could be that the frequency you measure is simply the strongest one it can see or more likely, the combination of signals is causing some to block out others over some of their cycles.

I'm afraid the tester you need to verify this is called a "spectrum analyzer" and is very much more complicated than a frequency counter. Instead of measuring cycles per second at it's input, it sweeps the frequency spectrum (like tuning a radio receiver) and plots the amplitude of any signal at each frequency it tunes to. It gives a break down of each frequency so you can easily see the exact composition of the waveform being measured rather than just seeing the total.

Back to more practical things though. Are you using screened RF cable to connect your counter to the circuit and ensuring the ground connection and measurement probe are as short as possible?

Are you still applying bias to the second stage though a base resistor to a positive supply? The bias normally comes from the IC and would only be present during a data transmission so you have to 'fake' the data being high all the time to keep the transmitter running while you take measurements.

You should note also that the later amplifier stages run without DC bias so if there is insufficient RF to drive them they will stay dormant and actually reduce the output at the antenna.

Brian.

 

[RMMK]

It can be for many reasons, the most likely is that the signal you are looking for is masked by other signals at different frequencies. The input amplifier on the frequency counter is not selective (nor should it be) so any signal will be included in the count. It could be that the frequency you measure is simply the strongest one it can see or more likely, the combination of signals is causing some to block out others over some of their cycles.

I'm afraid the tester you need to verify this is called a "spectrum analyzer" and is very much more complicated than a frequency counter. Instead of measuring cycles per second at it's input, it sweeps the frequency spectrum (like tuning a radio receiver) and plots the amplitude of any signal at each frequency it tunes to. It gives a break down of each frequency so you can easily see the exact composition of the waveform being measured rather than just seeing the total.

Back to more practical things though. Are you using screened RF cable to connect your counter to the circuit and ensuring the ground connection and measurement probe are as short as possible?

Are you still applying bias to the second stage though a base resistor to a positive supply? The bias normally comes from the IC and would only be present during a data transmission so you have to 'fake' the data being high all the time to keep the transmitter running while you take measurements.

You should note also that the later amplifier stages run without DC bias so if there is insufficient RF to drive them they will stay dormant and actually reduce the output at the antenna.

Brian.

I will try to check the transmitter on a spectrum analyser........ I found it recently in a lab!

Umm..... nopes I wasnt using any screened RF cable for connecting the counter with the circuit....... but yes they are pretty small in length!

Yup! I havent removed that 33Kohm bias on the 1st stage as of yet..... havent placed the TX-2 IC in its place........ am only trying to look at the carrier wave!

Sir the thing is when i disconnect the pi-network with the collector of the last stage (i.e., of 2N3866 transistor); the freq. counter does show a solid 27.130MHz count on the collector of the same transistor with all the preceding stages intact......

 

[RMMK]

I just checked it and weirdly even when the connection b/w the last stage of PA and pi-network is open; still i get a reading of 3.065MHz on the freq. counter whereas on the same condition with the connection still broken i get a count of a solid 27.13MHz at the collector of the last transistor (with all preceding stages intact).....
here is the schematic of the actual circuit (without the oscillator and the 1st stage)......

 

[betwixt]

I wouldn't get fixated about the frequency at the output stage at the moment. The most likely scenario is the 27.130MHz signal there is very weak and the pi-network is dropping it below the sensitivity threshold of the frequency counter. Although unlikely, it could also be that the pi nework is resonating at another frequency and acting as a receiver, you could be seeing a signal it is picking up.

Try with that spectrum analyzer and let us know what you see. There should be a single 'spike' in the display at 27.130MHz and much smaller ones at multiples of it.

You really should use an RF grade co-axial cable to couple the signal to the counter. Don't use audio grade or microphone cable as these are likely to have large capacitance or very high loss at high frequencies. Make sure the screen is grounded close to the input stage of the counter and keep the probe/ground connections as short as possible.

Borber's idea is a good one, build an RF probe (aka a 'sniffer' or Field Stregth Meter) and see if you can measure the signal voltage on a DVM. The simple parallel circuit should work best.

Brian.

 

[Borber]

Just forget your counter. It is useless when dealing with amplifiers.

 

[RMMK]

Alright sir........ Its Eid Holidays here...... i will come with results from the spectrum analyser by next week!
On the Diode probe; i have a 1N34A Germanium Diode....... is it suitable for this operation i.e., is it fast enough???

 

[betwixt]

Plenty fast enough - they are good to several hundred MHz!

Most small diodes will work, ones for RF work are best, small Schottky types are probably next best followed by Silicon signal diodes. The ones to really avoid are power rectifiers which sometimes can't handle more than 500Hz!

Brian.

 

[RMMK]

Thanks! I will come up with results as soon as possible!
But what up with this 10Mohm voltmeter in the below link? Can i use my Digital multimeter set at VDC for this purpose??

http://www.techlib.com/electronics/detect.htm

 

[Borber]

Just one important remark on your circuit. At least one component in LC tank must be adjustable usually C (trimmer capacitor) or you will not be able to tune amplifier to work. The same is valid for output filter where coupling output capacitor and inductor has to be adjustable.

Almost all DVM has 10MOhm input resistance...

For safety reasons do not use spectrum analyzer because you can damage it...

 

[RMMK]

Just one important remark on your circuit. At least one component in LC tank must be adjustable usually C (trimmer capacitor) or you will not be able to tune amplifier to work. The same is valid for output filter where coupling output capacitor and inductor has to be adjustable.

Almost all DVM has 10MOhm input resistance...

For safety reasons do not use spectrum analyzer because you can damage it...

Dont worry about the damage sir...... there are lab attendants to help me over that! Besides I wont learn anything if i start worrying about damages! :evil:

Just forget your counter. It is useless when dealing with amplifiers.

Please dont phrase it like that....... it took me quite a lot of working hours which i could have surfed on studying my academics while building this frequency counter! It hurts.... hahaha

 

[betwixt]

I think what Borber meant was a frequency counter is useless for testing amplifiers. It has no concept of amplitude so it would (should!) give the same reading at the input and output points.

Similarly, the warning about damaging spectrum analyzers is saying they are sensitive receivers so don't push several Watts of power into them and fry their input stages! As with all test equipment, you need to read the instructions and respect their limitations.

You can use any meter to measure the voltage, it doesn't specifically have to be 10M input resistance but with such a potentially weak source of RF you don't want to load it more than necessary or the reading will become lower. As most DVM have an input resistance of 10M or more, they are ideal for this application. You will only measure a few mV so don't expect high voltages!

Brian.

 

[chuckey]

As you have class C stages, as the RF passes through them they take current. So if you had any tuning elements you could tune for maximum supply current. It is impossible to build a transmitter for this frequency without variable tuning of the tuned circuits because the stray capacity (which is unknown) will be larger then your 27 PF fixed tuning capacitors.
Frank

 

[Borber]

First two lines in my post you quoted are essential. Have you understood what I am saying?

OK, chuckey was faster and additionally explained the problem.

 

[RMMK]

Okay here are the results!
I placed the diode probe +ve into the output from the pi-network and the -ve to the battery negative terminal and measured some 0.59Volts!
But when i disconnected the collector capacitor of the last stage from pi-network and connected the collector capacitor to the diode probe it showed some 19.76Volts!
What does it mean?? why this large voltage spike??

- - - Updated - - -

i used this diode probe circuit:-

the diode i used for the above is 1N34A
But i didnt place any 50 or 75 or 100ohm resistor across the input of the diode probe!