A friend, Bill Harbour, KD4PWB was enquiring about various aspects of the design of CW QRP transmitters. In particular, his interest centred around the amateur radio band of 30 metres (10.000 - 10.150 Mhz).
Various topics and design principles had been previously discussed and in the end I decided to present an experimental transmitter for CW (morse code) QRP (low power less than 5W) for operation on the 30M amateur radio band. Bill's original project revolved around a crystal oscillator using a TTL 7400 Quad 2-input NAND gate. The original project is described on Harry Lythall's - SM0VPO pages.
Bill has on hand some T37-6 toroids so we will try and use them but I will suggest other types.
For our purposes, as a guide to demonstrate design principles, I decided to go back to basics and use a conventional crystal oscillator.
Although in my pages on crystal oscillators I stress the need for a buffer amplifier I decided for sake of simplicity not to use one here (oh heresy!).
Here is our 30M crystal oscillator. There is nothing special about this one. You will note wherever possible I will try and make many components the same value e.g. by-pass capacitors 0.1 uF. My reasoning here is it is a lot more convenient to keep re-using same values, less storage, often cheaper buying.
Figure 1. - a 30 metre (10Mhz) crystal oscillator for the QRP Transmitter
The transistor can be any convenient NPN general purpose type transistor such as BC108, 2N2222A, 2N3904 types.
The transformer T1 and associated variable capacitor I will discuss later when we start to work backwards in our design. Now on to the driver amplifier stage.
I'm going to do the keying in the driver stage and I hope that doesn't produce any problems. For the driver transistor I'm selecting the 2N3053 transistor which is also cheap and readily available. I want about 1.5W out of this stage.
Figure 2. - a 2N3053 driver stage for our 30M QRP CW transmitter
Now that's a pretty bog standard driver (or even output) circuit you will find in 1,000's of places. Once again the critical components R1, C1, C2, L1 and RFC we will discuss later.
Because we want 5W output, the limit of QRP power and, we are using 2N3053 transistors which also happen to be rated at 5W, I going to use a pair to minimise the stresses on one single transistor.
Here is our power output stage for our 30M QRP CW transmitter in figure 3.
Figure 3. - Power output stage for 30M QRP CW Transmitter
Let's get through some pretty basic stuff here. We have T2 a centre tapped secondary transformer. Note the CT is at ground. Similarly we have two 10 ohm (10R) resistors going from base to ground of each transistor. Also in the collectors of bother transistors are 36V Zener Diodes for protection. Also in both collector lines going to the primary of T4 we have in both instances two 0.1 uF capacitors in parallel, this isn't terribly significant.
Starting from the 12V power supply input we have respectively a polarized 22 uF / 25V electrolytic capacitor, a 22 uH choke, a 0.1 uf by-pass capacitor and a 470 pF by-pass capacitor. Again none of those values are especially significant although the choke ideally would be capable of carrying one ampere of current. The polarised capacitor should not be less than 25V rating.
Having dispensed with the preliminaries and bearing in mind there is absolutely nothing remarkable about any of these circuits we will now discuss the "meat" which rarely gets discussed elsewhere - the how-to-fors.
These following principles more or less apply at any frequency, even though this design is for 30 metres I will constantly talk in reactances. In my opinion this is the only way to proceed with any designn and a jolly good habit to get into.
We require 5W output from a 12V supply and feeding a 50 ohm load. Firstly let us be clear all our calculations are based on that 12V figure and a presumed genuine load of 50 ohms. If either should vary then the results you achieve vary accordingly, just had to get that off my chest because it is rarely mentioned elsewhere.
Now I assume you know something about low pass filters and why we use them. Here it is used as an aid to suppress harmonics of our fundamental signal which in this example is around 10 Mhz. We don't want to radiate signals on 20 Mhz, 30 Mhz..... and so on. This particular filter is simply two "L" network LC circuits back to back forming a "T" network, designed for 50 ohms in and out. Now how good or sharp a filter do we want? This is called "Q" or more correctly loaded "Q". Let's use an arbitrary number of four, this is at the higher end of the range encountered in amateur radio work.
If our "Q" is set at 4 and our impedance is 50 ohms then our reactances are 4 X 50 = 200 ohms. Terribly difficult isn't it? Sometimes you might want to fiddle with that loaded "Q" number purely to accomodate components you might have on hand. The number 4 is NOT sacred, it could have been 2, 3, or 3.7634 or whatever you think is kewl! I would not recommend going beyong 5. Now I said these were simply two "L" type networks. Look at figure 4 below.
Figure 4. - two "L" network LC circuits back to back forming a "T" network
Note here that L2 and L3 remain at 200 ohms, the two left hand capacitors when added in parallel reduce to a net reactance of 100 ohms. If we make our cut off point around 10.5 Mhz the XL = 200 and L = 200 / ( 2 X Pi X Fc ) or 200 / ( 6.2832 X 10.5 ) or 3.03 uH and a capacitance having a reactance of 100 ohms at 10.5 Mhz is about 152 pF. Here I'd be inclined to use a 120 pF fixed capacitor in parallel with a 50 pF varable for adjustment purposes if you have the necessary test equipment. If not then your stuck with a 150 pF fixed capacitor. Terribly easy. BTW for a T37-6 toroid (AL 30) a 3 uH inductor is around 32 turns, on a T50-6 toroid (AL 46) it is 26 turns. Other frequencies of course produce different inductances and capacitances for 200 and 100 ohms respectively.
If we want 5W output from a 12V power supply then the collector to collector load, RL is going to be Vcc2 / (1/2 X Po). That means RL = [(12 X 12) / ( 1/2 X 5W )] or 144 / 2.5 = 57.6 ohms. Got that?
Therefore T4 needs to transform, backwards, 50 ohm to 57.6 ohm. Here you have a couple of options. The output "T" network will introduce some losses and to make up for this you could shoot for marginally high power output to compensate. In that case you could make T4 a straight 50 ohm CT to 50 ohm transformer, on the other hand you may elect to make T4 strictly according to "Hoyle". An impedance ratio of 57.6 to 50 ohms is a turns ratio of the square root of 57.6 / 50 = 1.073:1
If we make T4's reactance (it's untuned) a nominal 200 ohms, we would at mid-band (10.000 - 10.150 Mhz) have a required inductance of 3.16 uH. If we use a T37-6 toroid then that equates to around 32.5 turns for the secondary and with a turns ratio of 1.073:1 the primary needs to be 35T, however as it is centre-tapped we'll use an even number 34. So there you go, use 34T CT primary and 32T secondary or simply make them both the same around those numbers.
Different toroids, different frequencies, different values by calculation.
Transformer T3 (fig3) is often called a "balanced collector choke". Here again we can use a reactance of about 4 times the 57.6 ohm load or 230 ohms. At mid-band again we get a required inductance of 3.63 uH. A T37-6 toroid requires 35 turns bi-filar wound to achieve this HOWEVER we need a wire gauge of sufficient size to carry the required current. Each transistor is going to draw around 5W / 12V = 416 mA! We definitely don't need power drops! You need to use at least #26 wire, perhaps preferably #24. T37-6 is capable of only 31T of #26 wire and 23T of #24. You could reduce turns by either using a larger toroid e.g. T50-6 or use a ferrite toroid such as T37-43. A T50-6 iron powder toroid requires 30 turns instead of the 35 for T37-6. Using a T50-6 toroid you still couldn't satisfactorily wind that number of turns bifilar.
However you can consider using a little known trick. The AL value of any toroid is directly proportional to it's thickness, that's a mathematical fact. Stacking two toroids together almost doubles its AL value. Two T37-6 toroids should have an AL value a bit below 2 X 30 = 60, let's say 90% of that value. In this event 3.63 uH would then require around 25 turns. Worth a thought!
A bi-filar winding of 3.63 uH wound on a ferrite T37-43 toroid would require only 3 turns (dramatic difference). Using this toroid I think I'd wind at least 5 bifilar turns using #24 wire. The comments about wire sizes also applies to transformer T4 and inductors L2 and L3. BTW I've never used a toroid smaller than T50 size for these reasons.
Transformer T2 (fig3) needs to transform our nominal 50 ohms input to a nominal 10 ohm base impedance. Again using 4 X 50 oms = 200 we get at mid-band an inductance of 3.16 uH and our T37-6 toroid requires about 32T primary. An impedance transformation of 50 / 10 is a turns ratio of 2.24:1 making secondary around 14T centre-tapped.
NOTE: I would ensure the 10 ohm base resistors were 1W rating.
Nothing particularly flash here in figure 2. You have the output low pass filter which will help determine the output power of the driver stage. I'm looking for around 1.44W watts from this stage. Happily a 50 ohm load will provide this nicely. If the driver stage provides 1.44W and the final output is 5W then we are only asking the final stage to provide 5W / 1.44W = 3.47 or a modest 5.4 dB of power gain.
The output filter C1, L1, C2 could be designed for whatever loaded "Q" you wanted. Just for variety let's use a QL of 2. This means our reactances are all 2 X 50 ohms = 100. At around Fc 10.5 Mhz C1 = C2 = 150 pF and L1 = 1.5 uH or 22T on a T37-6 toroid.
The RFC in figure 2 needs to have a reactance of around 5 to 10 times the load resistance (50 ohms here). In this event something between 250 ohms and 500 ohms or around 3.9 uH to 7.9 uH. Ensure a suitable gauge wire to reduce power drops.
This stage is keyed (hope it works) and the PNP transistor can be any general purpose type such as 2N2907A, 2N4126 or whatever you can get your hands on.
Again a pretty straight forward circuit here in figure 1. Here all we need concern ourselves with is T1 and the associated variable capacitor. Again we'll make the transformer reactance around 200 ohms i.e. 3.16 uH or 32T on a T37-6 and the secondary is usually around 10% of this or 3T. Resonating capacitance required is about 79 pF, I'd use 56 pF with a 50 pF variable in parallel.
NEW! - How to link directly to this page
Want to create a page link to me from your site? It couldn't be easier. No HTML knowledge required; even the technophobes can do it. All you need to do is copy and paste, the following code. All links are greatly appreciated; I sincerely thank you for your support.
Copy and paste the following code for a text link:
and it should appear like this:
href="https://www.electronics-tutorials.com/transmitters/30m-qrp-transmitter.htm" target="_top">visit VK2TIP's 30M QRP CW Transmitter Page</a>
visit VK2TIP's 30M QRP CW Transmitter Page
and it should appear like this:
the author Ian C. Purdie, VK2TIP of www.electronics-tutorials.com asserts the moral right to be identified as the author of this web site and all contents herein. Copyright © 2000, all rights reserved. See copying and links. These electronic tutorials are provided for individual private use and the author assumes no liability whatsoever for the application, use, misuse, of any of these projects or electronics tutorials that may result in the direct or indirect damage or loss that comes from these projects or tutorials. All materials are provided for free private and public use.
Commercial use prohibited without prior written permission from www.electronics-tutorials.com.
Copyright © 2000 - 2001, all rights reserved. URL - https://www.electronics-tutorials.com/transmitters/30m-qrp-transmitter.htm
Updated 7th March, 2001