Hello

Since many people are looking for low power radio transmitter plans, I’ve decided to post the plan of the most simple radio transmitter I know. It is a little AM transmitter, a wireless microphone really, for the Medium Wave band (but it can be used for low power broadcast on the Short Wave band to).

What is attractive in this circuit is its simplicity.You can’t expect much from  such a little project, but, it’s a good and especially easy start with low power broadcasting.  It won't cost more than 5-10 USD, the components are everywhere readily available (or, they can be cannibalized from any old AM portable radio). It can be constructed and tested in less than 2 hours.

Diag. 1 (PNP type tranzistor)

Description

The transmitting power is subject to the transistor type one will use and the voltage of power supply, but it’s somewhere around 50mW, as far as I can tell.

I found that my construction works best on the 800-1200 kHz range, but with changes to the inductor L or the variable capacitor C1 it can transmit to the whole range of the AM dial.

The range depends mainly on antenna length and grounding. With a simple 75cm telescopic antenna (or just a short wire) and no grounding, it’s between 20-50m. With a full L/4 single wire antenna and good grounding the signal can be heard several hundred of metres away on an ordinary radio - 300-500m most likely - even further with better radios. Not much, but with that range it can reach many people in an densely populated urban are, or as in my case it can cover a whole village. The audio fidelity one gets from this little transmitter is good for AM standards. I mean that.

The current from the power supply is subject to transistor type and the voltage and is between 1-6mA, very low indeed! Running it from a battery (3-4.5V) it will work for a long time before the battery is exhausted.

Construction

It can be constructed with only 7 components. One transistor (T), two resistors (R1 & R2), three capacitors (C1, C2 & C3) and an inductor (L). The transistor can be of any type. In the above diagram (Diag. 1) it’s an PNP transistor but one can use an NPN transistor (as in Diag. 3) but notice that in this case the polarity of the power supply (battery etc.) is reversed. There is no need for a Printed Circuit Board (PCB), though if you want you can use one (see Diag. 2), it is always giving better result with RF projects.


Diag. 2 Printed Circuit Board (optional)

Component list

R1= 100 KOhm  1/2W
R2= 500 Ohm  1/2W
C1= 50-350 pF variable capacitor
C2= 250 pF
L= 66 turns of 0.35mm enameled wire around a 8X140mm ferrite rod with a connection point made at the 33rd turn
C3= 150 pF
T= AF 126 (but also AF 127, AF 139, 2G108/109 or any equivalent)

The transistor (T) can be of any type. It doesn’t even need to be a high frequency one. I've used so far the AF 126 and AF 127 (AF 139) PNP type both, and the NPN type AC 127. Any low power PNP type (like the AF126, AF127, AF139, 2G108, 2G109 etc.) will do. If you use an NPN transistor like my AC 127 (or the 2N1711, 2N708, 2N930, 2N1893, BC170 etc. ... or any equivalent) you’ll have to be careful with the polarity of the power supply.

LPR editor's note:

The AF 127 was apparently an older germanium PNP, 32 V, 10ma 75MHz unit, and the AF 126 was probably an NPN.

US Silicon models with similar characteristics would probably be the 2N3904, 2N2907, 2N2222 and their derivatives.

It might be necessary to adjust the values of R2, depending on the transistor used.

(M= It’s an optional «charcoal» microphone (like those used in telephone sets a few years ago). Instead, you can put there a headphone jack and feed the transmitter with audio from an audio mixer/a stereo unit/CD/cassette player, or whatever audio source you wish...)


Diag. 3 (NPN type tranzistor)

Component variations

I‘ve used a 1000 ohm resistor as R2 (instead of 500 ohm). I don’t remember why I did that but the unit works OK as is.

Sometimes instead of the variable capacitor of 50-350 pF as C1 I use a fixed 150pF ceramic capacitor. With that the transmitting frequency is around 850-855 kHz but the stability suffers a bit.

The ferrite rod I currently use is just 8cm long (because the, very fragile, original 14X0.8cm broke to pieces, I used the bigger fragment). I believe that a longer ferrite rod can increase the range somewhat.

I initially used as T an AF 126 transistor with a power voltage of 3-4.5V. I fed it with more voltage to increase the range, but it finally blew after working sometime at 9V. (Tip: don’t use more than 4,5 V as a power supply with the above PNP transistors, or you are going to blew them. If you do want to experiment with higher voltages - up to 9V/12V - have some spare transistors at hand and prepare to use them).

I later used the AF 127 (2G type) instead. With that, the range was a bit reduced, and they were too sensitive to voltage increases: don’t use more than 6V with those (but feel free to experiment with higher voltages -7.5/9V - only don’t forget those spares!). With more than 6V this transistor is overheated, so a heatsink can save it, for sometime, with 7.5 or 9V.

The NPN I currently use, the AC127, it’s absolutely immune to higher voltages. In fact with less than 9V the range is very limited (just a few metres), but with 12V or more (I use an 18V power supply at the moment, for best results).

My next project will be to increase the voltage to 24V and even 35V if the transistor survives! Only, I don’t have a spare AC127, so I won’t try anything until I get a couple. I believe that (with a decent antenna) 24V will give me an effective range of 1 Km and even more with 35V.

I also plan to use a 2N3055 tranzistor and see what happens!

Diag. 4 (A variation of the plan with PNP tranzistor, C2=C1 and C1=C2 in the component list)

Other Possible modifications (Haven’t tried them myself except #1)

There are a number of possible modifications that can improve the performance of this little transmitter.

1. You can use an electrolytic capacitor 25-50 microF/25V at the input of the power supply to eliminate any interference (50/60 Hz "hum") from the power supply, if you are on mains (no need for that when using a battery).

2. A matching transformer like that old AD9014 or any equivalent, at the input of the Audio feed might improve further the audio fidelity, though as I said above it’s already good enough as it is.

3. The use of a power transistor with a heatsink as transistor T will definitely improve the range, as it will allow higher voltages.

4. I think that a lengthier ferrite rod (one about 19cm long) will also improve the radiating power and, hence, the range.

Adjustments

When your construction is finished earth (ground) the unit, connect the antenna, switch on your audio source, tune an AM radio on a clear channel between 800-1200 kHz, make sure that the polarity of the power supply to the transmitter is the right one and then switch on the transmitter. Adjust the C1 var. capacitor until you find the your signal on the radio.

Performance and Use

If battery operated, with the simple «charcoal» microphone as an audio source (as in the diagrams), and a short (50cm or so) helical wounded wire for an antenna this little unit can be used as a «wireless microphone» with a range between 5-15m. A 75cm telescopic antenna is better if someone want to use them as AM walkie talkies along with small portable AM radios. The construction can be put inside an emptied plastic 4.5V torch case for such a use. The range will be somewhere between 20-100m.

If used for low power broadcasts instead of a microphone, connect the unit to the phone jack of a stereo unit, an audio mixer or a PC sound card(!). I obtained best results when taking the audio feed directly form the loudspeaker wires of my home stereo unit, but from a «line out» or a «phone» output jacks the results are quite similar. Use a longer antenna if you want to be heard in some distance.

Pic. 1 (possible construction, sorry for the quality of the pic)

Antennas and Grounding

AM transmitters require long antennas and good grounding to perform at their best. If you want maximum performance use an L/4 single wire antenna, calculated to your desired frequency (i.e. the antenna length (ANT) should be equal to the Wavelength of the desired frequency (L) divided by 4). A 5/16 wavelength antenna might be even better but it will be more lengthy. You can calculate the particular wave length (L) of any given frequency (f) if you divide the light speed C in metres (300 * 10^6 m/sec) with the frequency f in Hertz. Practically 300/F in MHz will give you the wavelength in metres.

Example:

If your chosen frequency is 1250 kHz then, Wavelength L=300/1.250=240 metres. Thus the optimal antenna length (ANT) in that case will be ANT=L/4=240/4=60 metres of 0.3-1mm single strand wire, usually suspended with insulators between two buildings, or trees (except if you live on a building 60m tall. In that (unlikely) case, just hang down the wire out of your window, to have a true 60m vertical antenna!)

Grounding must also be good for best results. There are several simple ways to «earth» your transmitter.

One way is to connect it with a lead to a metallic water pipe or a tap. Another is to connect it with the «ground» of an electrical socket but avoid that option, especially, if you are unfamiliar with electrical wiring. Mains voltage can kill. A third way is to take a wire from your transmitter’s ground and connect it to a single or a number of metallic rods stuck in - preferably moisturized - soil. For those perfectionists out there, there are more elaborate (much more effective) ways to earth their TX units. Grounding frames, grounding radials etc. A number of ground radials of L/4 length (2,4,6...16 etc.) is one such a way, but if you take the trouble to construct such a grounding system, do yourself a favor and get a more powerful transmitter! For more info on AM/MW antennas see related web sites.

Range

That's the most important aspect in transmitters, is it not? Well, as I said, don’t expect much from such a simple transmitter. To get the maximum out of it, experiment with the higher voltage possible for the transistor you are using and use an appropriate antenna for your chosen frequency.

The chart below will give you an idea from my experience («L/4 ant» is the optimal antenna length for the chosen frequency, and «antenna» is the actual antenna length I used. In the AF126 case the 6m antenna was of 1mm braided insulated wire. In all other cases was of 0.3 single stranded insulated wire).
 
 

TR type	Voltage (DC)	Current	Power	Freq	Wavelength	L/4 ant	antenna	range (MW)	range (SW harmonic
AC127	18V	6mA	~50mW	1215 kHz	247	61.75m	30m	250m
AC127	18V	6mA	~50mW	1250 kHz	240m	60m	15m	150m	200m 3750 khz
AF127	6V	~4mA	~15mW	1080 kHz	278m	70m	6m	50m	100m 9720khz
AF127	4.5V	~3mA	~10mW	1080 kHz	278m	70m	2m	20m	50m 9720khz
AF127	3V			1080 kHz	278m	70m	6m	20m	50m 9720khz
AF127	3V			1080 kHz	278m	70m	2m	5-10m	20m 9720khz
AF126	9V			~1200 kHz	250m	62.5m	6m	100m

I estimate that an AC127 with 18V at 1250-1251 kHz with a 60m antenna will have an effective range of 500m (for reception with an ordinary portable AM radio) or 700m (for better AM receivers).

Problems

Not any particular problems will occur with this transmitter. Attention should be paid to the careful construction of the inductor L, to the right connection of the transistor legs and the battery polarity. If there is "hum" in the audio, check your mains adaptor, choose a grounded one to avoid the "hum". Sometimes, when I use a portable CD player for audio source odd things happen. The CD disk stops or starts going backwards. To avoid such problems I connect the CD player to an Amplifier and take the audio from there, however I haven't been able to resolve the problem permanently. My solution some times works an other times fails.

Pic. 2 (the coil for the inductor L).

Interference

No annoying spurious signals are likely to be detected in other positions of the AM dial. In my experience even on a radio sitting just next to the operating transmitter only a couple of a faint signals can be detected few 100s kHz above and below my chosen frequency, and those disappear completely 2-3 metres away from the transmitter. So it is unlikely that any neighbors trying to listen to their favorite AM station will complain for interference from you if you choose an unoccupied channel to transmit. Splatter is also unlikely to cause interference to adjacent channel reception for more than a few metres from the unit, but if someone nearby is trying to listen to Short wave, well there might be a problem because of harmonics...

Harmonics

Harmonics are inevitable and this transmitter is no exception. In fact this little goblin is a true garden for harmonics! I’ve found out that some of the harmonics produced can be received from greater distances than those in the basic frequency. So, if you are going to use this transmitter in a highly populated area (especially in the night hours) use a low pass filter to filter out the harmonics because it can cause interference to Short Wave reception. The 3rd harmonic and oddly the 9th seem to be the more powerful ones. There are also sub-harmonics that can be received on the Long wave band. When transmitting somewhere between 1300-1400 kHz I also received my signal on Long wave 320-340 kHz.

Odd behavior: A bonus?

When setting the transmitting frequency to 1080 kHz I found out that the 9th harmonic at 9720 kHz on Short wave was so powerful that it could be received in distances 3 times greater (150-300m and more!) than the MW signal in 1080 kHz (50-100m), (6 to 15m antenna scenario). I don’t know why this happens with this harmonic but one might consider it an additional bonus if one wants to transmit on the SW band. In that case just calculate your antenna for the 31m band and your signal will be heard much further than the MW! The same applies for the Long wave sub-harmonics. Use an appropriate antenna for LW and broadcast in the LW band if you weirdly desire so! However if you just want to transmit on MW do filter your signal in order to avoid causing troubles to nearby DXers who try to log on a distant LW or SW station just where your harmonics are.

X-Radio Files

Strange things have occurred during my experiments with this transmitter (TX). If switched on and no audio fed to it I could turn the C1 knob and hear in the radio, coming through my transmitter (?!), on my preset frequency other AM radio stations with their signals amplified! It was working as a AM/RF amplifier & tuner!? But the most weird thing happened when, once, I switched on the audio before switching on the transmitter. Even though the transmitter was switched off, the audio was heard from the radio, weak but audible! Startled I checked the power source and confirmed that it was disconnected indeed! Ghost transmissions? Well, I figured out that the electrolytic capacitor I was using to filter out the 50Hz mains interference was discharging slowly and was behaving as a battery! As the current going through this transmitter could be as low as 1mA, even a charged electrolytic capacitor could emulate a battery!!!

Summary of transmitter characteristics

Characteristic	Recommended	Possible
Modulation type	AM	Transmissions on the LW, MW, SW bands are feasible with none...
Frequency range	700-1600 kHz (actual)	... or only minor modifications to the LC tank and/or the antenna
Working voltage	3-4 VDC (AF126/127 case)	tested up to 18V (with AC127, depends on tranzistor type)
Maximum current	6mA	0.3-10mA (depends on tranzistor type)
Output power	50mW	ranges from 10 to more than 50 mW
Audio fidelity	good for 9kHz channels	better in the case of 10kHz channels
Stability	good	is affected by the quality of the C1 var. capacitor
Range	150m max with 1/10 antenna	more than 500m with a full L/4 0.3-1mm single stranded wire antenna

Final comments

Perhaps that was a too exhaustive article for such an insignificant AM transmitter. However many things included here also apply to better and more powerful AM transmitters. And one can gain useful «broadcasting» experience with it before moving on to higher powers! I first saw the diagram of this little transmitter in a, Greek, primary school circulated magazine, in 1975, when I was 10. I was immediately hooked to it and started having dreams about making my own broadcasts.

I remembered that little AM radio-microphone of my childhood days and decided to build it just to have the pleasure of hearing my voice on the MW/AM band too. The results of that project are described in the above article. I dig out that little AM «bug», with it’s AF 126 transistor «burned». I quickly replace it with an AF 127 (and later with the present AC 127 when all my three AFs where sacrificed to experiments), an I was «playing» with the airwaves again on the AM dial!!!

Christos Z. Konstas
former operator of «FM radio Corsair» in Thessalonica, Macedonia-Greece
send comments at alt.radio.pirate
or at:
granazis@hotmail.com