Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Transmitter with class-e amplifier attached not working

Status
Not open for further replies.

mik3ca

Member
Its been forever since making a good FM transmitter but I couldn't figure it out unless LTSpice is garbage software, I don't know. I don't own an oscilloscope and maybe now I need one in the worst way, however I want to avoid that as it costs big money.

Anyways, If I strip the extensions from this circuit starting at wire n002 (aka the class-e amplifier), I could pull off a basic transmitter by making fine adjustments to R2 and R1 and C6 (under R1), but the problem is, the transmission range is rather too weak. I'll be lucky to pull off 50 feet. I'm looking for one that can transmit at least 200-400 feet.

I'm trying to transmit at about 433Mhz. I believe C5, C2, and L1 make a colpitts oscillator that determines this frequency. I'm willing to change the values of C5 and C2 later on if necessary.

capacitor C1 (located under n002 label) is tricky. I was told somewhere it represents part of a high-pass filter, but if I use a higher value I feel more signal can go through yet at the same time, it causes the overall transmission frequency to change. can someone verify this? I also tried connecting C1 to emitter without success.

I used a 1m resistor as an attempt to feed part of the input to the final stage but that made things worse.

I need to fix this so I can transmit at least 200 feet, however there are restrictions since I am adding this to another project with microcontrollers:

1. The source voltage coming into the transmitter is 5VDC which is fed through a 7805 voltage regulator. This regulator currently shares its output with everything else (including the microcontrollers).

2. The input signal (when transmitter is 100% done) is a fixed 32Khz square wave with output coming from the output pin of a 74HC86 XOR gate. (that will be connected to IN and the 20Khz signal will be removed). However, I want to test this thing first, so the input will be audio (between 0 and 20Khz) from my computer at normal volume.

3. The antenna will be a 6-inch shielded 22awg wire with about 1 inch stripped from the end.

4. I need to use the fewest tuning elements possible (preferably 1, the capacitor across the inductor would be the only thing I'd like to tune).

and of course...

5. I must not break any law. which means if I have to select a different frequency, then I will.

So what can I do to fix this?

and if the simple answer is to increase voltage, would 5VDC sourcing a charge-pump circuit (made of diodes and capacitors) be sufficient to power this circuit and produce long range?

circuit.png
 
I must not break any law. which means if I have to select a different frequency, then I will.?

If you do not want to break any law, then abandon this attempt to build this circuit.
What is shown in the schematic will not work at 433MHz.
The 2N3904 has a gain-bandwidth product of 300MHz, which means that it will not amplify at 433MHz. Hence Q1 will not oscillate and Q2 will not amplify.
A quick and dirty calculation shows that L1 and C2 resonate at about 350MHz. Capacitance from other circuit elements will only lower the frequency (if it worked, that is).

What were you intending to use as a receiver?

So what can I do to fix this?
I suggest that you use one of the many 433MHz licence free modules which are available from many sources.

Note: "Licence free" low power radio systems, do not give carte blanche to make any radio transmitter you like and just get on with it to the annoyance of your RF neighbours.

JimB
 
I made a super regen receiver that can pick up many stations. So then maybe LTSpice is dumb after all then. I was originally going to use PN3563 transistors instead of the 2N3904 which has a gain bandwidth product of at least 600Mhz. And how was 350Mhz calculated? Does the colpitts equation with 2pF, 5pF and 0.1uH not work here?. Also, buying things on the internet can take a long time to arrive, especially from ebay.
 
LTspice correctly showed that your circuit does not work. The unshielded part of the antenna is only 1" long which is too short. Q2 has almost no base current if its base resistor is 1M but you have it as 1m (o.oo1 ohms) which is almost a dead short.

My FM trasnsmitter does not work in LTspice when its RF amplifier is connected to the collector of the oscillator trasnsistor but it works fine (and performs very well) when connected to the emitter.
 
LTspice correctly showed that your circuit does not work. The unshielded part of the antenna is only 1" long which is too short. Q2 has almost no base current if its base resistor is 1M but you have it as 1m (o.oo1 ohms) which is almost a dead short.
the software is odd because I meant 1 megaohm, not 1 milliohm and it automatically makes the M lower-case. ok, so I guess I need the full 6 inches unshielded.

My FM trasnsmitter does not work in LTspice when its RF amplifier is connected to the collector of the oscillator trasnsistor but it works fine (and performs very well) when connected to the emitter.

I'm gonna see what I can do about that even though LTspice doesn't have Pn3563 in their list of transistors
 
2N904 only works to 300MHz 1M is far too high for the second transistor. It should be less than 50k and come from the 5v rail
 
Does the colpitts equation with 2pF, 5pF and 0.1uH not work here?
Depending on the exact configuration of the two capacitors, no, it does not work here.

And how was 350Mhz calculated?
I used the short form of the resonance equation for the parallel pair of L1 and C2.

F = 160/sqrt(L.C) where L is the inductance in uH, and C is the capacitance in pF.

JimB
 
I used the short form of the resonance equation for the parallel pair of L1 and C2.

F = 160/sqrt(L.C) where L is the inductance in uH, and C is the capacitance in pF.

JimB

Thats different. I thought the equation is 1/(2pi times sqrt(LC))
 
I thought the equation is 1/(2pi times sqrt(LC))
Indeed it is.
In that equation, F is in Hertz, C is in Farads and L is in Henries.

If you kick the numbers around a bit, so that F is in MHz, C in pF and L in uH, you get F= 159.13/(sqrt(L.C)) which for most practical purposes can be rounded to F = 160/(sqrt(L.C))

JimB
 
I came across one here: **broken link removed** but it looks like the designers cheated by using a crystal at the transmitting frequency. The reason I want to make one is because later I want to be able to change the station in case in the future the station I'm using is no longer allowed by law, and if I buy modules, then I'll have to buy modules again if I want to change the station. That's a pain.

I do want to admit tho that I have better luck (getting waveforms) when I change the inductors to at least 1uH instead of 0.1uH and when I raise the first emitter resistor to 1K, however I'll have a harder time getting 433Mhz
 
There is no crystal at 315MHz. I use the 315MHz version in a project. It is pointless making your own when you can buy them for one dollar.
 
I'm still going to make my own. and besides I can learn from it too. In fact I'm already learning (from experimentation). Judging by this waveform, I think I'm capable of making a wonderful transmitter of up to about 199Mhz.
circuit.png


Is there some math I can use on this circuit? I want to make the inductor value lower than 1uH and still get this wonderful waveform so that I can create a tank circuit that gives me 433Mhz. It seemed that adding a capacitor between emitter and collector of the last transistor helped but question is by how much.
 
You don't seem to realise. It takes an enormous amount of skill to make a 433MHz transmitter. You are going to get sidetones and splash all over the place. You need a 1GH CRO and a field meter to see what you are producing. If it was easy, there would be hundreds of different kits on the market.
You have no idea what you are doing. The transistor 2N5210 is a general purpose transistor of 30MHz.
It is pointless continuing this operation.
 
You don't seem to realise. It takes an enormous amount of skill to make a 433MHz transmitter. You are going to get sidetones and splash all over the place. You need a 1GH CRO and a field meter to see what you are producing. If it was easy, there would be hundreds of different kits on the market.
You have no idea what you are doing. The transistor 2N5210 is a general purpose transistor of 30MHz.
It is pointless continuing this operation.

Back in the day I managed to make a transistor transmitter that allowed me to transmit audio for approximately half a kilometer. I just forgot what the exact schematic was. Yes I understand it takes time which is why I'm using software to verify things are ok. In my real circuit, I want to use Pn3563 since it works to 600Mhz. Perhaps the software has a better transistor model for me to use?. What I am not familiar with is spice commands themselves.

Thanks to the spice software, I'm gonna have to check the datasheet of each transistor one by one now.
 
Yes. You made a 88MHz transmitter. I sold over 50,000 of these kits and they are very easy to get working. But working at 300MHz is a completely different story. I have used thousands of PN3563 You will not get much at 300MHz. You need a 1GHz transistor that is properly designed for high frequency. It will take hundreds of hours to get this type of circuit working.
 
I am talking about the transmitter. You always design the transmitter first.
At 400MHz the Tx coil is 100mil wide track less than 5cm long and folded
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top