20 MHz Selective RF amplifier

[Gofs]

Hello,

I don't have any experience in RF amplifiers and I need to design a 20MHz selective RF amplifier. After reading some information on the internet I designed it as following:



and the simulation results are as follows:



I would like to ask you for commenting this circuit. I am afraid that it will become a generator and it will generate frequency around 20MHz. I would like to know is this design is OK in general? Maybe I missed something and it won't be working at all? I want to amplify signal from antenna which will be AM modulated and receive amplified signal that will have an amplitude around tens of milivolts. I don't want to demodulate the signal, just amplify it. I have used L1, L2, C2, C6 as a filter. I found this solution somewhere in the internet and I don't have any equations for it but the simulation show that this is very good circuit. Please, any comments welcome.

Additional question: How I can match antenna impedance to the transistor impedance in this circuit. Is 50 ohm resistor a good solution? I want to make an antenna on the PCB if it will be possible.

 

[chuckey]

If you connect this circuit to an aerial, you will pick up all sorts of AM stations because the first two stages will be non-linear and have no filtering. If you are cunning, you will use a common base amplifier as your input stage. You can get the input impedance at the emitter to be 50 ohms and CB amp is inherently linear. Your tuned circuit is very unusual. L1 is shorting out the 300 ohm collector load, so there is no DC feed back to the first transistor.
Frank

 

[Gofs]

Thanks for your comment chuckey. Please take a look on this website: **broken link removed**

There is an AM receiver where two first transistor are connected in (I suppose) darlington type of connection. Does it count as two stages of amplification? Based on that design I designed my amplifier but I din't put the 1k resitor between emiter and base connection. Does it decreases the non-linearity? If the design showed on thos website is working than I suppose that this amplifier is somehow linear becasue the author is claiming that is working. The only thing that I don't understand in his design is the reason why he / she conencted the antenna in this way.

This is a good point about L1. I just copied this solution as I have found it in internet and I didn't notice the problem with L1 shorting the 300 ohm load. But I am sure that this is exact copy of some solution. I am trying to find it again and put it here so I can show you the source.

"f you connect this circuit to an aerial, you will pick up all sorts of AM stations" - is this a problem when anyway it will be filtered later?

I have seen solution with two transistor cascode one as a common emitter or common dren and second one as a common base or common gate but the simulation results we worse than this one...

 

[chuckey]

"- is this a problem when anyway it will be filtered later? ". yes its a problem, strong AM signals will cross modulate your wanted carrier with their modulation. Once this is done there is no way of getting rid off it. Proper filters for receivers have bandwidths in the order of a few KHZ, which is why the superhet receiver is used for communication receivers, unfortunately this sort of bandwidth is very hard to achieve at 20 MHZ. I would Google "AARL" or RSGB" or "TRF". The American "ham" society, the British "HAM" society, tuned radio frequency receivers.
Frank

- - - Updated - - -

Sorry I have just looked at you link. This is a regenerative receiver, which will work once trimmed, but I have doubts to its stability and linearity. It gets its selectivity by positive feed back acting as a Q multiplier on the coil, more Q more signal voltage across the tuned circuit. However the noise increases a lot and the question of stability, i.e. its gain will change with temperature and with supply voltage. its horses for courses, if you are trying to build a high quality RF voltmeter, do not use this. If you want to listen to a short wave station, then a good cost effective solution.
As an aside matching the aerial to your amplifier can be done with a PI network or a tapped tuned circuit - better then a 50 ohms, no power loss!
Frank

 

[Gofs]

Thank you a lot for this help. It seems that this is more complicated piece of design that I was thinking at the beginning. Now wounder why I can't find any proper information about this in the internet.

Maybe you could pass me a name of a good book about designing this type of circuits? I were searching for them but it seems that it is hard to judge is a particular book good enough for what I am trying to do.

You have made a good guess - I am trying to build something that will be measuring power of a signal with Analog Devices AD8319 Log Detector. I need something that will respond fast on signal changes.

 

[chuckey]

https://www.simplecircuitdiagram.com/2010/03/19/wideband-rf-amplifiers/ Seem to have various different RF amps.
Frank