Variable small signal Amplifier

Feels like you hit your word count more than just the schematics. Nonetheless, hope you proved something.

If you must know, as if you even care anyway, I only aim to make it easy for anyone to understand... So I don’t care if I wrote a lot, as long as someone else found it helpful.

Not the best topology for high frequency amplification. Common emitter amps suffer from HF limitation due to the Miller effect. A more reasonable solution would be to use a common base amplifier as the grounded base ,,shields'' the collector signal from feeding back to the input and affecting the high frequency gain

Very crude example, but something like this is more suitable - https://everycircuit.com/circuit/5860958508154880

Issacutt, l found your write-up very informative and we'll explained, l for one like to see info about the circuit, the more the better, you don't force anyone to read it if it's to long for them, to many put up a circuit with no text, that's probably why they get few views

Thanks Tedski, I appreciate that!

@thebugger, Ive actually never tried using common base in any of my designs before, but if the base is grounded, than where would the signal come in at? Anything fed into it would have to be 0.6-0.7v below ground in that case right? ...which I guess you could do that if you biased it through a capacitor to oscillate around ground so it would go negative, but I don’t know I’ve never tried that before, could you explain a little more about it?

Ok just checked out your circuit from the link, I think it looks like you are using one transistor to buffer the signal into the emitter of the second transistor, but I honestly have no idea how you get a 25v peak sine wave out of it, My only guess is very carefully chosen inductor/capacitor values... not too sure on that

No worries, let me explain. You're on the right track with biasing the emitter negatively if you ground the base. I'll give an example later on. The common base amplifier can be made linear, however with high frequencies, the most used topology is class C - the collector will conducts for a very small part of the cycle to preserve the efficiency high. You can later on restore the signal with a filter. Example coming.

P.S. The emitter follower helps with the impedance. Common base amps tend to have a relatively low input impedance, but don't suffer from the Miller Effect and can have significant voltage gain

https://everycircuit.com/circuit/5834769777295360

Ok I see what your saying, but in reality what about when you get into really high frequency circuits such as 400/600MHz to say even a few GHz ...would common base still work then??? Or do they just use very specialized transistors or do they use completely different designs?!? Also, one last thing... why exactly does the Miller effect only happen when feeding the signal at the base? Does the emitter follower suffer from the same thing or not really since the emitter isn’t directly grounded?

GHz circuits tend to use special transistors for low power amplification or tubes for high power amplification. I guess with a good design and a proper transistor you can use either common base or common emitter for amplification. It's just a bit more problematic with common emitter, because the Miller effect is an apparent amplification of the collector-base capacitance, which results in modification of the input impedance of the amp. Such circuits tend to feed back instabilities from the output back to the input, which is not desirable. For instance, if you're working on tuning your antenna, any changes in the output will be fed back to the input (and thus the generator) via this feedback. Common collector and common base amplifiers don't exhibit this effect, because the input and output terminals are shielded by the grounded base (common base) or the output is in-phase with the input and doesn't pull it down (emitter follower).

Common emitter vs common base 1GHz - https://everycircuit.com/circuit/5083433196584960

Sweet, thanks for explaining all that, just helps confirm what I was thinking..... By the way, how is it that every circuit you’ve gave me a link to somehow always puts out almost like twice the source voltage, you just picking the magic combination of inductors or something each time😄!?!

Yeah, with RF circuits it's all about resonance. Resonance can get you a much higher voltage than the DC source can actually supply.

Here's a resonance example that can get you hundreds of thousands of volts on the output with a 20V input

https://everycircuit.com/circuit/5774339553886208

Btw in reality RF circuits are tuned somewhat easier than here. You just need an RF peak detector and a steady hand when tweaking the resonance points.

That’s pretty nice, a couple weeks ago I built my own high voltage supply that put out around 40kv, (it arc’d about 1.3cm through the air), but that was done using about a 20v supply also (actually got that 20v from a 65watt laptop charger from Walmart). I built the oscillator using a Schmitt trigger that I had designed from a while back; interestingly enough, It achieved such high voltage without any resonance and I also made my own transformer. Although unfortunately, the transformer still sucked and it’s still no where near as good as your circuit because I had to use like 5 capacitive multiplier stages to get that high..... but because of the really short duty cycle I went with, at lower input voltages of 5-10v, the mosfet wouldn’t get hot and it could probably run for quite a while before ever having to worry about it overheating... but the arcs wouldn’t be as large, that’s ok though cause it won’t repetitively kill my mosfets with that design

Designing resonant circuits are like an art of its own so it seems, I haven’t actually tried building anything around resonance, but I would like to figure it out eventually

The beauty of the slayer exciter is that it finds its own resonance point without tweaking.

Just out of curiosity, how much do you know about designing wireless RF circuits?? ...Because I have recently been trying to build my own digital pwm transmitter and receiver lately... starting with making the transmitter, and simply using my oscilloscope with an antenna to try and receive something from only a few inches to a foot away, and I can’t pick up anything on the oscope! I’ve tried several different frequencies from 200/500kHz to 8MHz to even 40MHz all about 1-2v peak to peak and I still get nothing. But my fm radio can pick it up some at least, the oscope just continues to see nothing though.... I’m thinking I didn’t add the antenna properly, would you know how an antenna is supposed to be incorporated into the circuits, I can’t find any information online about it?? Or do you think it may be a different issue?

@issacutt There could be many reasons why it doesn't transmitt or receive properly. RF is an art on its own. Your "carrier" frequency seems pretty low, which means you need alot of output power and a pretty long antanna for proper transmission (lambda/2 or lambda/4). The higher you go in frequency the shorter your antenna get, but impedance matching and effects like reflections (Swr etc.) will be more important. Maybe you can briefly describe how your circuit looks like, what output signal you want to archive and what range it's supposed to work on. When using your scope to see if a signal is transmitted, there are some tricks but you should not expect a high amplitude or a consistent and reliable reading.

Ok yeah I can definitely provide a link to the circuit, honestly though it was mostly an experiment to start with, so all I had circuit wise was just the transmitting circuit which comes from a really simple 1 transistor oscillator which I inserted an antenna at the collector of the transistor, my plan was to figure out how to transmit and receive a signal first and then I’d start designing a pwm circuit that would interface with simple binary inputs.... here’s a link to the transmitter....

http://everycircuit.com/circuit/6179333125963776

In real life that base-emitter capacitor was not necessary and the inductor I used was actually 1uH instead.... and it produced oscillations somewhere between usually 37- 47MHz (that’s not to say that it was constantly varying in those two frequencies, but rather it occasionally would oscillate at a different frequency upon powering it on, most likely due to somewhat loose connections in the breadboard and thus different parasitic capacitances)

It's a rather crude oscillator so frequency shift is expected. 50MHz isn't especially high but it can be a little annoying on breadbords already. Just air wiring the legs of the components and soldering them usually still works fine at 100MHz. And maybe change the oscillator design for a collpits or sth similar.

Alright I guess I’ll give that a try, thanks

Sure no problem. Now I'm curious what scope you are using. What bandwidth does it have and what kind of probes do you use. I had a similar issue some time ago and figured out that most problems were created by my cheap scope probes. I have three post concerning this topic so if you have some time to read them take a look at : Alot of Discrepancy, Testing my Scope, Further scope investigations. The first one is about the oscillator but most of the conversation unfortunately is in german. Maybe you can still find some points that'll help you. The other two post are about testing my scopes bandwidth and the probes bandwidth.

What also helped me to get into rf is some pretty cheap gear. Non of that stuff is real lab gear but for their price I think they do pretty amazing as long as you are aware of some hickups. First on the list and always useful to have is a SDR (software defined radio). I got myself the NeSDR smart (~30$) and a HackRF(~400$). The NeSDR doesn't have such a huge bandwidth and is only able to receive but not transmitt, therefore the input sensitivity is rather amazing. The HackRF isn't something I'd buy right at the beginning of doing RF stuff tho. The second item is a PLL based signal generator built around a ADF4531. It'll help you to create signals and sweeps up to 4.4GHz. There are some different types of those evaluation boards, some have to be programmed with a PC and some have a uC and a Touchscreen. But you could also do a board yourself. Last bit of equipment would be the nanoVNA. It's a cheap handheld vector network analyzer. It's quite capable in my mind and I'm not bothered to type everything about it on my phone now. As for all of those things, before you buy them you should definitely check out some videos and reviews to see it they actually meet your needs. At this price they simply can't compete with real lab gear and might be the wrong things sometimes. Anyway, if you have any questions about the oscillator or other RF stuff you can try to catch me on discord. I'm also still learning and happy to talk to someone about it.

Wow that all sounds like some pretty crazy advanced equipment... Don’t know that I’d wanna explore that side just yet, I would really wanna do some successful experiments with creating and detecting RF first so I can grasp a better understanding of its nature before I invest into that kinda equipment. Currently I have a Siglent Oscilloscope with a bandwidth of 200MHz, it is a 1 GSa/a scope, actually got it from Amazon cause it’s my first scope, I’ve had it for about 7months now I think.... but I would definitely like to get one with higher bandwidth when I can afford to do so, I’m interested in possibly getting a 1-5GHz bandwidth scope. As for the probes I have, also not too special just what came with it 1x/10x and designed for up to 200MHz. But unfortunately, I quite possibly may have ruined them after taking some measurements on a couple high voltage arc projects; 😬😅 up to about 1,700 volts from a transformer a made, which went through a few capacitor multiplier stages and bumped up to about 40,000 volts at best for making some cool arcs through the air. I designed the oscillator that drove the transformer, it was pretty sweet. But those were just some side projects to keep me busy when I was a little bored.... I don’t usually focus on high voltage, but rather am usually trying to come up with new oscillator designs and occasionally do some digital builds; although I usually have a rough time with ICs so I usually focus on the transistor level.

What is discord by the way? (never heard of it before). Also even after those high voltage measurements the probes still work fine, but even since I first got them: every time I get into the several hundred kilohertz and the megahertz range, my scopes show that even a floating resistor of a value above 1kohm drops a lot of voltage and that worsens especially with higher values ... and that in particular is very frustrating, but I don’t know if that is a probe issue or a scope bandwidth issue or maybe even just a breadboard issue!?

Hah I skipped some of the discussion, but yeah - the thing I noticed and agree on is that RF circuits are an art. Usually output stages are designed via some formulas - I can share some of them with you - you can work out how to change the variables based on what piece of data you have

Usually the output of a class C amp has a lower impedance than the antenna. In some examples it's between 5 and 15 ohm, whereas an ideal load is 50ohm or 75ohm (more on that later). By applying the power transfer theorem we get that most power with least waste is transferred when the impedances match. That's why you keep noticing higher outputs on the load that the supply voltage - the circuit is resonating to provide power to thw load in a most efficient way.

In order to match these impedances, a typical stage requires a capacitor divider network, and the center tap is fed either to the antenna (or through additional filtering)

Now as for the antenna part, as I mentioned the best power transfer is obtained when source and load impedances match. Since we've already tweaked the source impedance with the capacitor network to match 50ohm, now we must make sure that the load is also matching 50ohm. There are 50ohm and 75ohm coaxial cables, so it's good to choose a matching one as well. As for the antenna, that's where shit gets real. A well tuned antenna would present only an active impedance to the source, and should have no reactive impedance (L/C). Best tuning occurs at the wavelength of the transmitter frequency and its divisors (/2,/3,/4 etc). For instance if you look into this, a half wave dipole antenna will exhibit a 75ohm purely active resistance, whereas a half wave monopole antenna will exhibit 50ohms. Once you have that matched you're good to go.

To be honest, in reality you don't really need any complex pieces of equipment such as oscilloscopes to tweak these kinds of circuits. You just need an RF peak detector, such as this one, and you can fine tune the amplifier based on the DC reading on the meter. A personal example is that with good tuning, I've maxed out a 30V meter with a 12V power supply.

https://everycircuit.com/circuit/4740638804017152

Oh, just one more thing I failed to note - if improper matching is obtained, a standing wave will reflect back to the amplifier, which is a form of a power loss which is dissipated in the output transistor. High powered solid state amps are usually destroyed so fast by such an occurrence, that they employ safety mechanisms that down-step the power when they sense an impedance mismatch to prevent failure

Here's a fine tuned amp @100MHz with the peak detector - https://everycircuit.com/circuit/6452222463049728

Dang, so your saying I can’t even obtain like a reception distance of a couple feet unless the impedances are matched?! I tried using the same LC values of the oscillator in the transmitter once for the receiving circuit, but I still couldn’t see anything on the oscope... but it was only a resistor capacitor and inductor and the antenna, I was trying to pick it up on the scope so that I could then figure out what the required dc offset would be in order to amplify it through a transistor.... but the scope wouldn’t pick up anything even in the millivolt range within 6 inches, it wasn’t until I removed everything from the probe and just plate the probe itself a few mm away from the antnenna that I actually picked up a little something. But my biggest struggle here, is where the antenna would even go on an amplifying receiver circuit, should it be right at the base of the transistor amplifying it or should it be in between the inductor and capacitor... and if so should one side of the LC tank be grounded or floating or should it be a series LC, I have no idea, and I’ve tried looking at several different schematics before as an example and I really don’t see any relation; ever rf receiving circuit I’ve found has the antenna in many different various spots, I don’t see how any current, or voltage field I guess flows through the antenna 😕

No no, there's leakage EMI, so even with 5-10mW leakage, you should pick it up with a good receiver up to 10-15m. Low power FM transmitters usually operate at 5-10mW and they also don't have an external antenna. Keep in mind that RF power is some serious thing. It's actually more important to have a good receiver, rather than a good transmitter. As an example, imagine an earth-moon-earth bounce. At best, you have around 250dB attenuation, which is incredibly high. NASA has successfully transmitted and received a 2mW signal with this attenuation - all thanks to some super bad ass receiver. My early experiments with tubes held some success in this area. Since tubes work at higher voltages, a tank circuit usually dissipates more power than a low voltage solid state transmitter. Without antennas, I was able to pick up my signal over 10m, just off the tank circuit EMI

Wow thats freaking amazing, so just one watt of power can go a long way! That’s crazy even just 2 mW was detected from earth and back.... Man if it’s all in the receiver then like where do I even begin learning how they’re designed? Is there maybe a book which focuses on the component level I could get that you’d recommend perhaps?

As I mentioned, receivers are way more complex. Simple receivers are easy to make - I've even made one by mistake - just a class A darlington pair picked up quite well on some AM radios. Quality receivers use heterodyne detection, and are way more complex

Ok so I guess no one ever builds them, they just always buy already made modules for rf applications huh

I think so - even with VHF transmitters, most hobbyists only buy the synthesizer, and build only the power amp

Man that’s crazy, makes me wonder how in the world they ever get improved upon if no one ever focuses on the component level anymore

Well that's mostly true for hobbyists but not for commercial use or scientific research. The equipment needed to built state of the art tec is just too much for most people at home but there are still some who do that. In industry the focus these days is mostly on integrated circuits which are printed on silicon wavers. Replicating this at home is down right impossible if you don't have the funds and a huge knowledgebase to back this. Also semiconductors having been changing alot with for example other elements they are doped with. Still I'm sure there are some electronic engineers out there who cook up new circuit concepts at home.

And now a days, the focus is probably even leaning more towards quantum and qubit, I have no doubt it may completely replace semiconductors as soon as they figure out how to make it more practical and scaleable, I don’t know how I’ll be able to keep up, everything will surely skyrocket after they develope full quantum computers

Man even Keysight technologies is advertising testing equipment for quantum research and development now it’s crazy