Gain block RF Amplifiers - Building and Testing [2/2] 

I'm happy you are enjoying my content! I will try to keep it up :D

Ha! I was going to suggest a cascode implementation until I read your comment! I second the cascode implementation 👍👍

I tried to give some details in my "4 year special" video ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-XWTfEEJtI58.html . I have no plans to do anything more detailed at the moment...

@@FesZElectronics thanks. Good luck with future videos.

There are a lot of videos on RU-vid about building BPFs 😉

The diagrams are directly obtained in the VNA control software - I used nanoVNA-saver. I connected the vna to my pc trough a USB cable, and then I control it with the program.

I used the BFU520AR; its the single transistor version in SOT23 package.

@@FesZElectronics Ah, gotcha. Thank you.

don't worry, I'm just happy you are enjoying the content! Also keep up your education, that is more important

Brother can you do a video on regenerative recievers, designing them and understanding their principles have always been a mystery and there are very less detailed videos on the internet regarding such regen recievers

Brother can you do a video on regenerative recievers, designing them and understanding their principles have always been a mystery and there are very less detailed videos on the internet regarding such regen recievers

Lately I have been playing around with CircuitMaker.

You can start with the generic formula for input and output impedance of the common emitter amplifier; but after that the finetuning is easiest done in the simulator - try multiple values until you get closer and closer to the desired ones.

@@FesZElectronicsdo you have such video of building a NPN common emitter amplifier? Thanks

I did actually make a series on both the various connections as well as biasing transistors - for the common emitter this is the first ep: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-D-76AL5fijE.html

Thanks @@FesZElectronics I'll work through an example or two at VHF.

Indeed, for the output impedance it was removed; I had to use the attenuator only for the amplifier input

With a 10dB attenuator and a nanovna you can still make usable measurements in this band. Just calibrate the analyzer with the attenuator in place and you'll get the graph out nicely

For this application I just used a random non-latching relay I had in my old box of parts; As long as the frequency is low enough it should not matter; anyway, the only modification I made to the relay though was to add some copper tape on its outside to act as a bit of shielding.

It seems to be "HZW". Out of curiosity, why do you ask?

@@FesZElectronics I want to try building such an amplifier and I am wondering if maybe I will find it in the analog TV tuners I have.

I would recommend that you take a whatever transistor you have and if you have a simulation model, just try to adjust the values in the circuit simulator to get the desired effect. I would not rely on the marking on small smd transistors since usually there is no clear rule that they follow... in the olden days if it wrote "BF200" you knew it was a BF200 transistor, but now, its just some random set of characters, and each manufacturer just uses something different...

By definition the Z11 and Z21 are obtained when the second port is left open; when measuring Zin, the second port will be connected to whatever load is in the circuit - I think this is why you are getting different results. For example, I simulated a basic 20db resistor attenuator (40.9; 10.1; 40.9 R) with the second port being connected to a 50R resistor which is the output port load. the Zin param is exactly 50R and Z11 is 51 -> for Z11 the ouput port is left floating (the 50R resistor in disconnected); but for Zin its left in the circuit.

@@FesZElectronics Thank you sir! I had a couple of dead "spf5189z" chinese LNA modules. Both was modified (by me) with protection diodes on input, current limiting resistor and TVS diode on power line but anyway just died from static when I touched something of my gear (not even LNAs directly). So I decided to use these boards to make my own LNAs. I used bfg591 transistors. I connected it like original ic: collector - to inductors, emitters - directly to ground and base - to input capacitor. Then I picked a resistor from collector to base to set dc bias current within 50-60 mA (for best perfomance according to datasheet). And that's it. Such solution worked well judging from what I saw in SDR software (at 70 cm there was 8-10 db of gain, very similar to "spf5189z" thing). But it worked well only with my homebrew BPF (I use it on 70 cm band). Without BPF it works very bad due to extreme gain around 100 MHz band. I don't have any attenuators for my nanovna to properly test S21. I didn't tried to make my own because I doubt that homebrew attenuator can work well at such high freq. So I tried to carefully measure S21 with low voltage supply. At ~100 MHz I got over 17 db gain with just 2V supply! So I decided to not go further to not blown up input of my nanovna. Then, thanks to your videos, I figured out that Zin (S11|Z|) of my LNA is too low. I didn't change original 100pF input-output capacitors. In LTSpice I found that I need 8.2 pF cap to get 50 ohm Zin at around 440MHz. But in reality I picked 4.7pF cap to get 46-47 ohm Zin on 70 cm band (actually nanovna measured 8 pF series C so it is close to simulation, those extra 3 pF are parasitic). I don't know how to measure Zout with nanovna (probably you can explain?) so I just picked 10 pF for output cap like in my simulation. Also I realized that instead of attenuator I can safely measure S21 of my LNA just in series with BPF because gain on 70cm band is not that big. Finally, after all, I got 14+ db gain at 70 cm! It is significally better than "spf5189z" suddenly dying thing. I can't measure noise floor with generic common SDR receiver but it seems like it is also a bit better or at least not worse. So I got really nice LNA with simple circuit. Thanks to channels like yours I begin to understand deeper sorts of magic. Thank you again!