Maker Pro
Maker Pro

Radio transmitter, options for reducing harmonic emissions

Status
Not open for further replies.

plat

Sep 26, 2017
8
Joined
Sep 26, 2017
Messages
8
I'm looking for options to reduce harmonics in a small radio transmitter I've built.


I have tried a low-pass filter on the amplified output but it saps so much power that it's basically useless. I would try an active filter using an op-amp but I'm not sure if this could carry enough power through to the antenna coupling to be useful? I'm looking for 1 watt total power dissipation. A power MOSFET is the main switching element at the carrier frequency. I suspect this is the main problem as the square wave may introduce stronger harmonics than a sine wave? This is all a learning experience for me.


What other options might I try? This is a totally homebrew project so I'm not following a pre-designed schematic. Besides the harmonics, the transmitter works surprisingly well.
 
Last edited:

kellys_eye

Jun 25, 2010
6,514
Joined
Jun 25, 2010
Messages
6,514
You have to account for output filter losses when you design for a particular power output requirement. Designing a circuit and THEN expecting to add harmonic filtering without considering the losses it introduces is just.... wrong.

Can you post a schematic of the circuit you have to date?

There are legal requirements for the level of harmonics that have to be met - if your starting point is a squarewave you've already got 'problems' you don't need although even they can be over come - but at the cost of efficiency.
 

plat

Sep 26, 2017
8
Joined
Sep 26, 2017
Messages
8
This shows the low-pass and high-pass filters before the autotransformer. I haven't labeled components because I'm still experimenting with it. Audio to be transmitted is supplied to the large BJT.

IMG_0213.JPG
Should the filters be located where I have them or after the transformer?


I have a wideband reciever so I can make sure the harmonics and overall power are not problematic. Making every effort to comply with the regulations.


I think I need to experiment more with the filter. I have tried a high resistance(4.7k) and matching capacitance for the cutoff frequency I want, but it it wouldn't pass enough power.

And when I tried a low resistance, 2.5 ohms, and matching capacitor, the harmonics were still strong. Maybe I just need an intermediate value?
 

kellys_eye

Jun 25, 2010
6,514
Joined
Jun 25, 2010
Messages
6,514
An RC filter for RF???

What frequency are you actually hoping to operate on?

The oscillator (your astable mutivibrator) should be set to match the resonant frequency of the inductor/capacitor that forms the tuned circuit operating at the chosen frequency.

You don't have any suitable capacitance on the tuning coil (other than your RC filter) therefore you won't bring the arrangement to resonance and will just be radiating 'crap' everywhere - 'illegal' doesn't even begin to describe the way it's set up!
 

davenn

Moderator
Sep 5, 2009
14,254
Joined
Sep 5, 2009
Messages
14,254
I'm going to repeat this with lots of emphasis !!

You don't have any suitable capacitance on the tuning coil (other than your RC filter) therefore you won't bring the arrangement to resonance and will just be radiating 'crap' everywhere - 'illegal' doesn't even begin to describe the way it's set up!
 

davenn

Moderator
Sep 5, 2009
14,254
Joined
Sep 5, 2009
Messages
14,254
I suspect this is the main problem as the square wave may introduce stronger harmonics than a sine wave?

a square wave will ONLY produce harmonics

PLEASE I beg of you
If you really want to learn about radio transmitters and receivers, join a ham radio club and get you amateur radio licence
and do it properly with supervision till you understand what you are doing

To do what you are trying to do can cause grief to a lot of people and eventually yourself
If the FCC catch you out causing interference to commercial or other services eg emergency and amateur
you may find yourself in jail or at minimum heavy fines


Dave
 

kellys_eye

Jun 25, 2010
6,514
Joined
Jun 25, 2010
Messages
6,514
Get hold of a copy of the ARRL Handbook - their explanation of the theory behind RF transmission is excellent. They also have schematics for simple and LEGAL transmitter circuits, explaining the requirement for cleanliness of signal (i.e. low harmonics) etc.
 

plat

Sep 26, 2017
8
Joined
Sep 26, 2017
Messages
8
By radiating crap, do you mean the harmonics or other noise well outside of the intended frequency?


I don't see that behavior at all. There is no emission more than a couple of kHz above and below the intended frequency, it is very narrow.


I'm using the license-free band from 160-190kHz in the U.S.


In testing this I am limiting the power to well below 100 miliwatts for the sake of not causing interference.
 

davenn

Moderator
Sep 5, 2009
14,254
Joined
Sep 5, 2009
Messages
14,254
By radiating crap, do you mean the harmonics or other noise well outside of the intended frequency?

yes, as well as your intended frequency if you are transmitting on a frequency allocated to another service

You MUST NOT use a square wave generator


I don't see that behavior at all. There is no emission more than a couple of kHz above and below the intended frequency, it is very narrow.

and how did you determine that ?

your particular circuit has almost zero frequency control, therefore it will be wandering all over that part of the spectrum

In testing this I am limiting the power to well below 100 miliwatts for the sake of not causing interference.

you will be surprised how far 100mW goes on 170kHz
 

kellys_eye

Jun 25, 2010
6,514
Joined
Jun 25, 2010
Messages
6,514
Yes, harmonics. A 'wideband radio' is not a suitable test measurement device. You need a spectrum analyser to properly observe the 'crap' you'll be putting out and, believe me, crap is what's happening.
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
I do not see any DC path to provide power to the fet.

There are NO harmonics in a sine wave.

The fet drain should go to a parallel tuned circuit connected to the positive supply. It should be designed to give a Q of between 10 and 15. Then make a matching circuit to drive the aerial

The audio modulation will affect the fet RF drive. The drive looks to provide 50% on time and 50% off time. For efficiency the on time should be small. With no modulation, the npn modulator will not be turned on.

Note the the oscillator should not be provided with more than 5V or the transistors will be damaged.
 

plat

Sep 26, 2017
8
Joined
Sep 26, 2017
Messages
8
My reciever can tune from 150KHz to 180MHz, AM and FM. My signal is coherent, such that when I tune it to 180KHz, it is recieved well at 180KHz and not recieved at all at 177KHz or 183KHz. This reciever is sensitive enough to pick up just the oscillator running with no amplification at all, so I felt confident using it for testing. Does this sound reasonable?


It is perfectly legal to transmit at 160-190KHz, at 1 watt or less, and an antenna no longer than 15 meters.


But I think I'll drop this and maybe try again in the future with a proper circuit and no square waves. Thank you for the advice everyone, I honestly want to be in compliance with all applicable regulations, or I won't do it.


Nonetheless, this has been a fantastic learning experience.
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
Don't you understand about harmonics?
If the two transistors produce a perfect squarewave then all the harmonics will be high level and will be odd numbers. If the fundamental frequency is 180kHz then the first odd harmonic is the third which is at 180 x 3= 540kHz on the AM broadcast band. It is very illegal to cause interference on a broadcast band. The next odd numbered harmonics are the fifth and seventh which are also on the AM broadcast band.

But the two transistors will not match perfectly so they will also produce even-numbered harmonics also on the AM broadcast band.
 

plat

Sep 26, 2017
8
Joined
Sep 26, 2017
Messages
8
I didn't know that exact mechanism was the cause of harmonics in square waves, but everything else, yes. But thank you for the explanation of the harmonics cause.

Which is why I asked here about filtering those harmonics so as to not cause interance.

Nevertheless, it is legal to transmit on the AM band at 100 miliwatts or less, which is why I never ran my transmitter above that power level, to guarantee that the harmonics it did produce were not in violation.
 
Last edited:

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
If the old granny in your neighbourhood is listening to a weak distant AM radio station like she does every day but your 100mW or less interference is heard and she complains then you will be in big trouble.

An RC filter is simple and cannot do much filtering to the first few harmonics and they are the strongest. You need a sinewave generator (it won't be perfect and will produce some harmonics) and an LC filter that has a high Q so that it passes the fundamental frequency but not the harmonics.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
If you can find someone with a spectrum analyser you can quickly see what your transmitted spectrum looks like.

With just a carrier you ideally see a single spike, with a square wave, you'll see all the odd harmonics too.

Even with a nice pure carrier, if you overmodulate, you'll also see a lot wider and nastier signal.
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
You are totally misinterpreting the FCC's 100mW regulation. While it permits unlicensed radio transmission it does NOT absolve the unlicensed or even licensed transmission of any interference on any band or any frequency.

Chris
 

davenn

Moderator
Sep 5, 2009
14,254
Joined
Sep 5, 2009
Messages
14,254
and on those wise words

time for this thread to close
 
Status
Not open for further replies.
Top