Low-Power AM Broadcast Band Transmitter

[VA6GDG]

Hi Brownout;

Very much appreciate the response and I do hope you can help me get this running. The kids I'm helping have an impending deadline that I'd really like to help them meet.

The coil is 220 uH according to the NEC poly-bag from which it came and the red-red-brown dots seem to confirm that. The article was not all that specific about any of the parts, even the coil. It says one could scavenge the Variable Cap and Coil from an old MW radio. I don't have an old radio to cannibalize, but I suppose I could also try winding a coil on a 35 mil plastic film canister and I do have a 365 pF variable cap from an old transistor radio in my junk box. Thoughts on this?

Also; the following is an extract from the original article and I've attached the original drawing to which the comments pertain. This should allow one to see if I've made any errors relative to the original or whether it must be the specific coil and cap.

"Broadcast Band AM Transmitter

This circuit is deliberately limited in power output, but will provide amplitude modulation (AM) of voice over the medium wave band.

The circuit is in two halves, an audio amplifier and an RF oscillator.

The oscillator is built around Q1 and associated components. The tank circuit L1 and VC1 is tuneable from about 500kHz to 1600KHz. These components can be used from an old MW radio, if available. Q1 needs regenerative feedback to oscillate and this is achieved by connecting the base and collector of Q1 to opposite ends of the tank circuit. The 1nF capacitor C7, couples signals from the base to the top of L1, and C2, 100pF ensures that the oscillation is passed from collector, to the emitter, and via the internal base emitter resistance of the transistor, back to the base again. Resistor R2 has an important role in this circuit. It ensures that the oscillation will not be shunted to ground via the very low internal emitter resistance, re of Q1, and also increases the input impedance so that the modulation signal will not be shunted. Oscillation frequency is adjusted with VC1.

Q2 is wired as a common emitter amplifier, C5 decoupling the emitter resistor and realizing full gain of this stage. The microphone is an electret condenser mic and the amount of AM modulation is adjusted with the 4.7k preset resistor P1. An antenna is not needed, but 30cm of wire may be used at the collector to increase transmitter range."

 

[BrownOut]

I built an oscillator similar to this one, and it worked, so I know the layout is good. Have you tried tuning your receiver across the entire AM band? If you get a point where the radio gets real quiet, you might be transmitting but may not have the audio modulating the signal. Give that a try, also might try the tuner cap you have. I'll try to think of another way to test it.

 

[VA6GDG]

I have a little Multi-band AM receiver that goes from 440 kHz to 1690 kHz. I tried tuning across the AM Band without success. I wish I knew it was even oscillating. Then it would only be a matter of where in the band.

I'll go dig out the transistor radio tuning capacitor I salvaged years ago. I'm pretty sure it's 365 pF at the high end, but don't remember the bottom end figure. If the bottom is 15 pF and the coil is working and the circuit is OK, it should have a range of 560 to 2770 kHz or so. Lot of IF's.

Hope you can think of something. The only other thing is the coil and winding one just to diagnose a problem is a heroic effort.

 

[audioguru]

The oscillator is supposed to be common-base. Then C7 should connect to ground, not to the supply voltage.
The supply voltage needs a 10nf capacitor to ground like all battery-powered circuits.
The electret mic should be powered from a 10k resistor, not a 1k resistor R1.

The antenna length for this low frequency transmitter should be half a block long plus an earth ground.

 

[VA6GDG]

Hi Audioguru.

I really appreciate the comments. Given that you fellows have made a number of improvements on the FM transmitter side and discussed it thoroughly, I'm all ears and hope to learn something from it.

As I'd never built this particular circuit before, I just followed the plans from Andy as closely as I could. I assumed that the published circuit worked. Maybe not a good assumption.

I've looked at a few other schematics of the FM variety and they do have the C7 cap from the base of the oscillator transistor to ground. What you said about the common base configuration makes sense and has been bugging me since our last chat on the oscillator. Wonder why Andy had it where he did. I can try your fix by changing one jumper (thank goodness for an easy one). That seems like it would be the biggest potential cause for no oscillation. Correct?

Is the 10 nF to ground in addition to the C6 of 10 uF to ground or a replacement for C6? If it is in addition, I'll have to scrounge for one (unless its value is not super critical). (Remember we discussed C6 and R8 a bit. What would it be for? C6 and R8 have a cutoff of about 15 Hz which seems pretty good for blocking AC from entering the mic and preamp.)

I can swap out the R1 (1k) with a 10 K, but wouldn't that be something that would affect the modulation level, rather than the oscillation? It needs to be done, but in an effort to not change too much at once I perhaps should wait 'til I've got a signal going out? Reasonable approach?

Also; please let me know if you see anything else that is incorrect in the circuit.

Sorry for the long post...

 

[audioguru]

Is the 10 nF to ground in addition to the C6 of 10 uF to ground or a replacement for C6?

C6 is a filter for the supply voltage to the preamp transistor and mic. It has no effect on the oscillator transistor that also needs its own supply bypass capacitor. Use 10nF which is also marked 0.01uf.

I can swap out the R1 (1k) with a 10 K, but wouldn't that be something that would affect the modulation level, rather than the oscillation? It needs to be done, but in an effort to not change too much at once I perhaps should wait 'til I've got a signal going out? Reasonable approach?

Change R1 later if you want.

 

[BrownOut]

I've looked at a few other schematics of the FM variety and they do have the C7 cap from the base of the oscillator transistor to ground. What you said about the common base configuration makes sense and has been bugging me since our last chat on the oscillator. Wonder why Andy had it where he did. I can try your fix by changing one jumper (thank goodness for an easy one). That seems like it would be the biggest potential cause for no oscillation. Correct?

Probably won't make any difference.

 

[audioguru]

The battery has an internal resistance that might be high enough to stop the oscillator if its base is not bypassed to ground.

 

[BrownOut]

If the oscillator is that sensitive to resistance, then I'd say there is something else wrong. If reconnecting the cap makes it oscillate, then you probably need higher Q tank components. Also, measure the terminal voltages on your transistor. You should get about 5 1/2 volts on the Emitter, about 6V on the Base and probably the full 9V on the collector. If these voltages are way off, you've got a bad baising component or miswire.

 

[VA6GDG]

IT LIVES !!! IT LIVES !!!

Switching C7 to link the base of Q1 to Ground made the difference.
I've now got to search for a 10nF disc and a 10k resistor.
Hi Ho, Hi Ho, off to my parts box I go!

I've not only got carrier, but also have some so-so modulation. The addition of the remaining components (1K resistor and the 10nF cap) should improve the sound.

THANK YOU AUDIOGURU!!! 2 Thumbs Up!

Let me know if you spot any other items that are odd or wrong, or have any other suggestions.

 

[VA6GDG]

Fellows;

I checked the voltages and the wiring. It looked good at the time. That said, I'll probably work through the load line etc. to see whether the published resistor values are close to what the calcs say. But I'll do this once the young ones have met their deadline. (P.S. Working with the curves from the datasheets is a pain. Wish I had a curve tracer. Oh well... Next Christmas.)

As for the Q of the tank.... I doubt it's anything to write home about. Given the short time the young fellows had and the difficulty in getting the quality of components that I probably should have had, this has turned out not too badly.

For future thought: This circuit would likely benefit from a regulated 9VDC supply. I know that the frequency will shift in this simple oscillator as the battery runs down.

 

[BrownOut]

I have a very similar oscillator working with the capacitor connected as in the original circuit. But as all parts are different, you're coils were probably the weak link, and you were operating just below the threshold. Glad you have it working, but I think it might not start reliably, so hold your breath

 

[VA6GDG]

Brownout;

You may be correct. I'll see if there are better quality ones around, but that may need to wait. As long as it works, it will do for now.

If I might ask... Are there better sources for transistor data than the typical manufacturers datasheet curves? I mean short of running your own curves. I find the ones in the datasheets are nearly impossible to use effectively.

Here's a late thought; I used BC107s instead of the specified BC109Cs. That could have had an effect on the circuit operation. Hmmm...It might be a good exercise for the boys to work a set of load line calcs, if there is time.

 

[audioguru]

What is its range? 10 feet? How long is its antenna? Does it have an earth ground?

I don't think its frequency changes much when the battery voltage drops because the capacitance of the transistor is almost nothing at 1MHz. It is 100 times as much at the 100MHz of the FM broadcast band.

The specified BC109C has a very high gain of 420 to 800. Your BC107 has a low but wide range of gain from 110 to 450.

 

[VA6GDG]

Hi Audioguru;

The range without an antenna seems to be about 10 feet. I haven't tried it with an antenna or earth ground yet. I just sat it on the workbench and put a small radio at the other end of the bench for testing purposes. Once I get the replacement resistor and additional capacitor in place, I should have a cleaner signal. At that point I'll add an antenna and ground. It'll be interesting to see how that helps.

Andy sent me the following:

"Output power in the coil (tank circuit) is low and around 20mW. However using a simple telescopic whip or 3 feet of wire as an antenna and some of the energy will be lost. Having said that you should get at least 30 feet (10 metres) or so depending on receiver. To weaken the signal increase R2 to say 1.5k or for more range try 470 ohms. "

By the way, I looked at the transistor cases and they had the part number as BC107C. The letter is the grade of the unit and tends to reflect the Beta. While I couldn't find a datasheet showing a grade C BC107, I suspect the gain ought to be more like that of the BC109C than the plain-jane BC107. This is part of why I was asking Brownout about a better source of transistor data. I'd like to see the specs and check the values of the biasing components.

It's good to hear that there shouldn't be much shift based on battery level. That will make it easier for the young fellows to demonstrate it at their science fair booth.

 

[audioguru]

Philips and a few other manufacturers do not list a BC107C on the datasheet and digikey does not sell any. Maybe you have a counterfeit.

My 1968 (yes your transistor is very old) Philips Silicon Transistors databook (my first job was with Philips) also does not list a BC107C.

 

[BrownOut]

I wouldn't fret too much over the difference in beta. As I recall, the gain of the common base is the base impeadance divided by the small signal internal emitter resistance. That emitter resistance is, if I remember correctly, equal to beta/(beta+1)*gm, for any reasonable value of beta, the equation simplifies to simply 1/gm, which gives the gain as Xc/gm (Xc here is the impeadnce of the tank circuit). The gain will be more influenced by the collector reactances, and that's why I harp on the quality of those components. However, what might be more important is the transistor's transistion frequency spec, Ft. If your transistor has a low transistion frequency, you might have a problem. Sorry, I'm too lazy to look it up

 

[VA6GDG]

Some Examples of BC107C Datasheets

I really appreciate your advice, fellows.

By the way; I've attached a couple of datasheets I found from firms that manufactured a BC107C. Yes; the BC107/108/109 series hasn't been around for a while, but It looks like it was a likely legitimate part. SEME and Continental Device India both made one.

Wonder what would be a good standard replacement for the transistor? I may try this again, or make one of the FM transmitters. If I know what to get, I'll grab them when I come across them. Sure as heck don't want to try to find more of the BC107s or BC109s. Took some digging to get them.

 

[audioguru]

I have never seen a transistor from India. I have heard that they still are using very old parts.

The BC547 is the same as an old BC107 but is made by many manufacturers and is in a modern epoxy case. Its pins are the opposit to the 2n3904 American transistor.

 

[VA6GDG]

Well, I put in the extra capacitor and changed the one resistor to 10k.
It is VERY touchy to tune and the sweet-spot is VERY narrow.
The modulation level control seems to have little effect as well.
Also there isn't much tuning range at all, nor is there a great deal of tuning range.
I suspect the tank is in a sort-of "Oscillate or Not" state.
I take it the circuit is sound, so I may have to replace the tank components.

Must have a good "think" on this. I built an AM transmitter many years ago using a fixed cap and a variable loopstick. It worked just fine and had lots of range. This one is a big disappointment.