How can I create a high frequency signal generator?

[kersleywyy]

Creating a signal generator

Hi all,

i want to create a high frequency signal generator. However i have no clue how to create a wave. I am looking in crystal oscillator which have pretty precise frequency and having a good stability. I also look in pierce circuit oscillators but i still have no clue how should i do it in practical and create one in real :P. Any suggestions are welcomed and thanks in advance

 

[davenn]

hi kersley
welcome to PF

firstly, what is your definition of high frequency ?
2) Do you want a fixed freq or a variable range ?
3) if variable ... what range ?
4) what level of output?
5) do you want that level fixed or variable ?
6) 3) if variable ... what range ?
7) if its an audio generator, do you want sine, square and triangle waveform outputs or just one of those?
8) if an RF generator its most likely to be a sine waveform

cheers
Dave

 

[kersleywyy]

Hi Dave,

im planning to have a range of 20-50Mhz if possible with an output of 1A 50V which is pretty huge but i can use an amplifier to do that.

Hmm I am still a bit new on that stuff but I don't know the difference between audio generator and RF generator... and I need a minimum of a sine waveform but would be prefect for square waveform too.

 

[AlephZero]

Make sure you check out your country's regulations on operating radio transmitters before you get too far down this road.

If you don't really know what you are doing (which seems a reasonable deduction from your questions) A 50 watt square wave oscillator running at tens of MHz could easily interfere with radio reception over a wide frequency range - not only commercial radio and TV broadcasts, but also things like air traffic control and the emergency services.

 

[kersleywyy]

http://www.ebay.com/itm/50Mhz-Dual-Channel-Sine-Wave-DDS-Signal-Generator-with-software-control-via-PC-/221197917267

 

[Bobbywhy]

Since you want to generate a sine wave signal within the range from 20 through 50 MHz, what frequency crystal oscillator would you choose "for stability"?

 

[kersleywyy]

I think a 125Mhz crystal oscillator would work

 

[Bobbywhy]

OK, kersleywyy. Do you have a design in mind for using that 125 MHz crystal oscillator? Do you have a "conceptual block diagram" of how you would connect that in your circuit to act as a reference? Show us what you have so far, please.

 

[kersleywyy]

I don't have nothing I just researched a lot and most of the signal generator which i actually got on ebay are using the 125Mhz crystal oscillator. I have no clue how to design one. I just want a head start how to know which circuit to choose. I have Pierce circuit which are pretty common for crystal

 

[Bobbywhy]

Crystal oscillators are manufactured for oscillation over a wide range of frequencies, from a few kilohertz up to several hundred megahertz. Their outputs remain fixed and can be used as a reference frequency.

Using frequency dividers, frequency multipliers and phase locked loop circuits, it is practical to derive a wide range of specific frequencies from one crystal reference frequency. But they cannot be used as a function generator that requires ALL frequencies within the range, in your case, 20 to 50 MHz.

Here are two examples of units you could buy to give you the signal you’re asking for:

http://www.berkeleynucleonics.com/products/model_645.html
http://siglent.en.ecplaza.net/function-generator-waveform-generator-sdg1000--216995-2358000.html

May I suggest you read this Guidebook to get a good overview of this subject?

“Function & Arbitrary Waveform Generator Guidebook”
http://www.bkprecision.com/support/...d-arbitrary-waveform-generator-guidebook.html

Now, if you decide to design and build a function generator, know that will be difficult for a beginner. But at least you will have some idea of what manufacturers already make. If your budget is limited it may be simpler to look for a simple used unit.

 

[kersleywyy]

I was given this project to create a signal generator! :P

This is my final year and I know how to use a signal generator but i would like to know more abt signal generator itself inside them(schematics) which i can't find

THX

 

[jim mcnamara]

Google images has this:
https://www.google.com/search?q=sig...6CtDsoASWhoDYBw&ved=0CDkQsAQ&biw=1066&bih=526

 

[the_emi_guy]

I was given this project to create a signal generator! :P

This is my final year and I know how to use a signal generator but i would like to know more abt signal generator itself inside them(schematics) which i can't find

HELP PLZ!

THX

What exactly are the project requirements? Do you need this exact frequency range? Does it require a crystal? Can you use whatever parts you want?

Can you use the AD9850 DDS used in the link you provided? If so, the complete schematic and parts list is in the datasheet, you don't have to design anything.

It would be difficult for you to "roll your own" DDS system so you would be better off not using a crystal that is locked to a single frequency, but using a pullable oscillator.

If you can make your frequency range 1Hz to 100KHz then you can use Bob Pease's LM331. And there are other VFCs that go up to a few MHz (but not 50MHz).

Crystek, RFMD, and ZCOMM make voltage controlled oscillators (VCO) that are typically used within PLL circuits. However, they don't have parts that go as low as 20MHz typically, but if higher frequency is ok, just connect a potentiometer to tuning pin (this may be sort of cheating, they designed the synthesizer, you are just powering it up and using it).

If you are expected to be designing this at the transistor level, then using a varicap or a tunable inductor (Coilcraft) would be easiest.

Just a few random thoughts...

 

[berkeman]

Hi Dave,

im planning to have a range of 20-50Mhz if possible with an output of 1A 50V which is pretty huge but i can use an amplifier to do that.

Hmm I am still a bit new on that stuff but I don't know the difference between audio generator and RF generator... and I need a minimum of a sine waveform but would be prefect for square waveform too.

50MHz at 50V is a very strong transmitter. You've told me in a PM that this is for medical research -- what kind of safety and regulatory issues have you been dicsussing with your project advisor? Your device will create illegal RF interference if you do not do a number of things correctly in your device design and use in the lab...

 

[kersleywyy]

Requirements:
-Frequency range of max 50Mhz

I can use any sort of device already made.

I was looking the AD9850 DDS but no idea how to continue from that there are so many pins and i couldnot see schematics before i buy it how would i know if it would work?

What exactly are the project requirements? Do you need this exact frequency range? Does it require a crystal? Can you use whatever parts you want?

Can you use the AD9850 DDS used in the link you provided? If so, the complete schematic and parts list is in the datasheet, you don't have to design anything.

It would be difficult for you to "roll your own" DDS system so you would be better off not using a crystal that is locked to a single frequency, but using a pullable oscillator.

If you can make your frequency range 1Hz to 100KHz then you can use Bob Pease's LM331. And there are other VFCs that go up to a few MHz (but not 50MHz).

Crystek, RFMD, and ZCOMM make voltage controlled oscillators (VCO) that are typically used within PLL circuits. However, they don't have parts that go as low as 20MHz typically, but if higher frequency is ok, just connect a potentiometer to tuning pin (this may be sort of cheating, they designed the synthesizer, you are just powering it up and using it).

If you are expected to be designing this at the transistor level, then using a varicap or a tunable inductor (Coilcraft) would be easiest.

Just a few random thoughts...

 

[kersleywyy]

50MHz at 50V is a very strong transmitter. You've told me in a PM that this is for medical research -- what kind of safety and regulatory issues have you been dicsussing with your project advisor? Your device will create illegal RF interference if you do not do a number of things correctly in your device design and use in the lab...

People are already using a signal generator and an amplifier for a treatment in lung cancer. My job is to create a signal generator that create a sine waveform of 50Mhz.

 

[the_emi_guy]

So this is not a school project?

 

[kersleywyy]

So this is not a school project?

It is a university project for electrical and electronic engineering.

And the equipment are already being used and there should not have anything illegal

 

[berkeman]

People are already using a signal generator and an amplifier for a treatment in lung cancer. My job is to create a signal generator that create a sine waveform of 50Mhz.

It is a university project for electrical and electronic engineering.

And the equipment are already being used and there should not have anything illegal

Do you use this apparatus in a shielded room? What is the effectiveness of that shielding? What EM level do you measure outside of the shielded room? Does that level comply with FCC regulations for your frequencies (in the US -- if you are in another country, it will be a different government agency that regulates the EM spectrum there).

What antenna structure or other coupling means are you using to couple the RF signal to the patient (Pt)? What is its antenna factor across the frequencies that you are driving into it?

Is the existing equipment being used in a hospital in a shielded room, and you are in an unshielded university lab developing the new equipment? Do you know what local public safety bands use the frequencies that you are driving?

 

[the_emi_guy]

kersleywyy,

Berkeman and others have mentioned concerns that your device may interference with other radio services.

Something else I wanted to make you aware of: Devices that contain radio transmitters are required to comply with human exposure limits (look at the fine print on the last page of every wireless router or cell phone manual). These are either "maximum permissible exposure" (MPE) or "specific absorption rate" (SAR) depending on the distances involved.

Your 50W/50MHz may exceed these limits depending on how you are exposing the body to this power (if that is your intent). Just to give you an idea, cell phone manufacturers do a significant amount of research in order to meet SAR limits with 2W power against the head.

I imagine that these requirements are waived in the case of intrusive cancer treatment on real patients.

Are you the cancer patient?

 

[kersleywyy]

i think you guys have been confused...

Im not doing anything signal transmitters in the air or some sort of... I creating a signal generator provided in those usual lab which are frequently used by so many people... I just want to "sample" a sine waveform that you can see on the oscilloscope... i just neeed to make the big generator to small ones which are available on ebay...(DDS)

Is that what you guys meant? Or if it is a risk, I might ask question to my supervisor but I am pretty sure there is nothing related to those signal in the air(wireless)... And I am not a cancer patient

 

[berkeman]

i think you guys have been confused...

Im not doing anything signal transmitters in the air or some sort of... I creating a signal generator provided in those usual lab which are frequently used by so many people... I just want to "sample" a sine waveform that you can see on the oscilloscope... i just neeed to make the big generator to small ones which are available on ebay...(DDS)

Is that what you guys meant? Or if it is a risk, I might ask question to my supervisor but I am pretty sure there is nothing related to those signal in the air(wireless)... And I am not a cancer patient

The key is what happens at the end of the coax cable where it gets to the patient. If the RF energy can couple to anything that looks like a resonant structure, that will cause significant radiation to be launched into the air.

If you are in a shielded treatment room, that may be okay, as long as the other equipment in the room can tolerate that level of RF interference. If you are in an unshielded clinic room, then you could cause significant RF interference with local radio receivers and other electronic equipment.

How is the RF coupled to the patient's body? Is the RF energy interacting with injected conductive particles? I believe that is an active area of research in cancer treatments...

 

[kersleywyy]

The key is what happens at the end of the coax cable where it gets to the patient. If the RF energy can couple to anything that looks like a resonant structure, that will cause significant radiation to be launched into the air.

If you are in a shielded treatment room, that may be okay, as long as the other equipment in the room can tolerate that level of RF interference. If you are in an unshielded clinic room, then you could cause significant RF interference with local radio receivers and other electronic equipment.

How is the RF coupled to the patient's body? Is the RF energy interacting with injected conductive particles? I believe that is an active area of research in cancer treatments...

But I am not doing any with the process of the treatment. I am doing the signal generator that is it for me

 

[berkeman]

But I am not doing any with the process of the treatment. I am doing the signal generator that is it for me

You should still be involved in the whole project -- that is part of learning to be on an engineering team. And each member of the team should be responsible for safety and regulatory compliance. Please discuss these issues with your advisor, to see how those issues are being addressed.