EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: Mechatrommer on May 06, 2016, 12:39:45 pm
-
i have the China RF Transmitter (1st pic)... it produces RF signal as shown in 2nd pic (2.2Vpp @ 12Vcc). i want to amplify it. I found method from the net (3rd pic), beside the black magic amplifier shown that i have no stock avail, it has RLC network on the output that is beyond me, i'll appreciate if someone can explain it, but thats not the main question. why i cant simply amplify the signal using 4th pic? no black magic RLC network on the output, can i?
-
I don't see why you shouldn't be able to use the config. in #4, but with a circuit that simple, I think I'd start by building it up on some protoboard and see if it works. It's only like, 7 parts or so. Maybe you could add a choke to the voltage supply on the op amp.
-
I don't see why you shouldn't be able to use the config. in #4, but with a circuit that simple, I think I'd start by building it up on some protoboard and see if it works. It's only like, 7 parts or so. Maybe you could add a choke to the voltage supply on the op amp.
2 ghz gbw opamp on a breadboard, this guy has jokes. :-DD
As far as the 3rd mmic amp, the rlc network is a power filter. The r-c filters the supply coming in, and the L let's dc pass to power the amp, while keeping RF from bleeding back out to the power rail. The 1nf on the input and output are just a.c. coupling because the amplifier will have a dc bias on each one.
There's no additional components for matching, so the amplifier is probably internally matched to 50 ohms in and out. Probably the simplest solution short of buying a connectorized amplifier from minicircuts.
-
I don't see why you shouldn't be able to use the config. in #4, but with a circuit that simple, I think I'd start by building it up on some protoboard and see if it works. It's only like, 7 parts or so. Maybe you could add a choke to the voltage supply on the op amp.
Not sure what you mean by protoboard, but at this frequency I can't recommend regular stripboard/breadboard techniques.
To the OP, there are plenty of designs but most use tuned and matched discrete transistor stages or brick modules (both offer out of band filtering) or MMICs rather than op amps. Taming high speed opamps, even those a decade or two slower than the one you cite, is an art in itself.
I also don't know what your local regs are, but generally the band is shared globally and secondary/ISM users must accept interference from primary users. Bear in mind those primary users may well be applications with military, radar or space use, and they won't take very kindly to interference from unlicensed or secondary users.
The third harmonic is often used by aviation radionavigation, and although they model out interferers it won't take much for your signal to appear, and bearing in mind the perceived safety-of-life aspect you need to make sure that whatever you come up with isn't going to upset someone.
You should check your local regs if you haven't already.
The network on the output is trying to match the impedance of the transistor output stage to the antenna to obtain maximum power transfer. Remember maximum power transfer theorem at DC? This is the same, but at AC. It will also have some filtering effect, but not a whole lot judging from the circuit. You must bear in mind that at this frequency parasitic inductive and capacitive components come into play so layout and construction techniques become critical. With the schematic as shown, it's as much about what's shown as what isn't.
To match one complex impedance to another, it's common to use a combination of an inductor and a capacitor. There are online calculators to do this, but in the old days we used Smith Charts.
http://www.daycounter.com/Calculators/L-Matching-Network-Calculator.phtml (http://www.daycounter.com/Calculators/L-Matching-Network-Calculator.phtml)
https://home.sandiego.edu/~ekim/e194rfs01/jwmatcher/matcher2.html (https://home.sandiego.edu/~ekim/e194rfs01/jwmatcher/matcher2.html)
+ many others.
-
I'm no ham, but my very first reaction to "[what is the] Simplest Method [of] Amplifying [a] 433MHz Radio" is "a better antenna."
So the alternative question is, "what is your goal?" If my assumption is correct and it's get more usable distance, and if you can afford to be directional on at least one end, I think you can get a lot of gain passively with a better antenna.
Unfortunately that's the end of my knowledge in this area so I'll stop. :)
-
Not sure what you mean by protoboard
Sorry, I was unclear, I meant manhattaning it on copper clad pcb. For 433 MHz that should (i think) be kinda sufficient (better than breadboard :D)
a better antenna.
Aren't there more strict rules regarding antennas with gain? Or is that only a WiFi thing?
-
As far as the 3rd mmic amp, the rlc network is a power filter. The r-c filters the supply coming in, and the L let's dc pass to power the amp, while keeping RF from bleeding back out to the power rail. The 1nf on the input and output are just a.c. coupling because the amplifier will have a dc bias on each one.
thanks for the info, i learnt something...
I also don't know what your local regs are, but generally the band is shared globally and secondary/ISM users must accept interference from primary users. Bear in mind those primary users may well be applications with military, radar or space use, and they won't take very kindly to interference from unlicensed or secondary users.
The third harmonic is often used by aviation radionavigation, and although they model out interferers it won't take much for your signal to appear, and bearing in mind the perceived safety-of-life aspect you need to make sure that whatever you come up with isn't going to upset someone.
You should check your local regs if you haven't already.
thanks for the friendly reminder. i'll keep this in mind..
The network on the output is trying to match the impedance of the transistor output stage to the antenna to obtain maximum power transfer. Remember maximum power transfer theorem at DC? This is the same, but at AC. It will also have some filtering effect, but not a whole lot judging from the circuit...
thanks for the info, i learnt something...
To match one complex impedance to another, it's common to use a combination of an inductor and a capacitor. There are online calculators to do this, but in the old days we used Smith Charts.
http://www.daycounter.com/Calculators/L-Matching-Network-Calculator.phtml (http://www.daycounter.com/Calculators/L-Matching-Network-Calculator.phtml)
https://home.sandiego.edu/~ekim/e194rfs01/jwmatcher/matcher2.html (https://home.sandiego.edu/~ekim/e194rfs01/jwmatcher/matcher2.html)
+ many others.
these are the black magic voodoo i was talking about. i'll catch up later, i'm suspecting reading a thick rf book to become so. thanks.
I'm no ham, but my very first reaction to "[what is the] Simplest Method [of] Amplifying [a] 433MHz Radio" is "a better antenna."
So the alternative question is, "what is your goal?" If my assumption is correct and it's get more usable distance, and if you can afford to be directional on at least one end, I think you can get a lot of gain passively with a better antenna.
Unfortunately that's the end of my knowledge in this area so I'll stop. :)
agreed, and you are right about my goal... i have here another and more expensive 2.4GHz transmitter, range spec is 2KM, nobody complaining about interference or legality. the above china cheapo 433MHz only rated at 400M the best. i thought lower band can penetrate more into the space, i mean if the lower watt and legit 2.4GHz can go further, i'm not sure why the 433MHz cant, 2.2Vpp @ 433MHz is big already i think.
the drawback with the 2.4GHz Tx is that its less programmability friendly, and its not expendable. the 433Mhz is jellybean easy if i connect it to a pic or avr, as i worked with them before (at lesser range) hence i bought 10 pairs of them (Tx and Rx modules). so i will do whatever necessarily to my knowledge and skill level to get more distance out of it. that including using the above proposed gain amplifier, and then later, the better antenna. i think "better antenna" is the easier part since it involves only cutting, glueing, bending, on passive mechanical part, and then range testing and redo many times until maximum possible range or satisfaction or exhaustion is achieved. but at least no involving black magic or voodoo math. (ps: no spectrum analyzer in the lab here sigh anyone care to donate? i'll pay shipping ;D).
I don't see why you shouldn't be able to use the config.
thanks for the confirmation.. ;D
-
I am a ham. first though is if you can stand some directional / polarization That is often the cheapest way to get distance but not all applications will allow it. Home built antennas are often made "long" and then trimmed down to resonance or best 50ohm match depending on test equipment available. If 1/2 the ham guys i have met can do it should be a piece of cake.
On the amplifier side, there are plenty of linear amplifier projects floating around on the net. Most will be ether tuned for HF but I have seen some around UHF. at the power output level you are looking at just a simple transistor and low pass filter would work for you. I was going to build one for 220 MHz about a year back, if I find my bookmarks I will edit this message.
I think a class A transistor amp would be much easier to tame that a fast op amp. You will likely need a low pass on the output to keep spurious emissions down.
-
at the power output level you are looking at just a simple transistor and low pass filter would work for you.
I think a class A transistor amp would be much easier to tame that a fast op amp. You will likely need a low pass on the output to keep spurious emissions down.
if i go nuts, i will improvise higher voltage amplifier, say 30V or 60V? ;D and go for transistor solution and learn how to make good filter, i'll keep that in mind thanks..
-
nobody complaining about interference or legality
I think a more complete statement would be "nobody has yet complained to me about interference or legality".
-
nobody complaining about interference or legality
I think a more complete statement would be "nobody has yet complained to me about interference or legality".
i mean, the 2.4GHz Tx is pretty much mainstream (used by Turnigy rc remote controller) in rc community used by hundreds if not thousands, and no serious report i'm aware of in the net regarding that... although, it has modern tech like freq hopping or what not that is also beyond me..
-
nobody complaining about interference or legality
I think a more complete statement would be "nobody has yet complained to me about interference or legality".
i mean, the 2.4GHz Tx is pretty much mainstream (used by Turnigy rc remote controller) in rc community used by hundreds if not thousands, and no serious report i'm aware of in the net regarding that... although, it has modern tech like freq hopping or what not that is also beyond me..
None of which changes the validity of the modified "complete" statement.
You have been told by others in this thread a few of the things that can go wrong and how you can foul up other people. Listen to them.
Sanity check: if, knowing the components' IP2 and IP3 specification, you cannot predict the output spectrum of the transmitted signal, then you don't know enough to predict whether you will cause problems. And once you can predict it, you will have enough knowledge to realise that you must experimentally test the transmitter and transmissions.
As for "no serious report i'm aware of in the net"... That is about as sensible as "I haven't seen any serious reports on the net about people that have been injured by walking into a road without looking", with the implication that it is OK to do so! It also sounds like you are, to coin a UK phrase, "turning a Nelsonian eye"[1]. Look harder.
[1] FFI http://www.answers.com/Q/What_does_to_turn_a_Nelson%27s_eye_mean (http://www.answers.com/Q/What_does_to_turn_a_Nelson%27s_eye_mean) but I don't think you have as much knowledge and wisdom as Nelson!
-
Sanity check: if, knowing the components' IP2 and IP3 specification, you cannot predict the output spectrum of the transmitted signal, then you don't know enough to predict whether you will cause problems. And once you can predict it, you will have enough knowledge to realise that you must experimentally test the transmitter and transmissions.
i'll need a specturm analyzer for that..
"I haven't seen any serious reports on the net about people that have been injured by walking into a road without looking", with the implication that it is OK to do so!
since we are at it... its not ok to do so, its all in the net, and its certainly not ok to do so with a smartphone. if you (viewers, not tggzz) cant bear to watch people got smashed to death, dont hit play button..
https://www.youtube.com/watch?v=Z_sAY_BHPNI (https://www.youtube.com/watch?v=Z_sAY_BHPNI)
-
Natural selection at work there.
I believe from first hand experience there's a severe disparity in capability in various people, which makes most man-made rules inappropriate since they treat everybody as being equal and as such they are both too strict for some and not enough for others. As we've drifted onto the road topic, among the people I know some scare me at very turn while driving even if driving slowly and being fully concentrated, while others can be talking on the phone, playing with the radio and be 30% above the speed limit all at the same time yet they never had an accident in 40 years of driving.
I've learnt to adapt, and in most cases throw away what you're told is the way things are supposed to be. Self-regulation works pretty well.
-
What's the point of the 50 Ohm resistor?
All it will do is attenuate the signal.
The source impedance should be as low as possible, as long as the cable is terminated by its characteristic impedance, it'll be fine.
-
I think a class A transistor amp would be much easier to tame that a fast op amp. You will likely need a low pass on the output to keep spurious emissions down.
I think you mean class C amplifier, which is the more common linear amplifier configuration for RF. It is far more efficient than a class A amplifier.
-
I think a class A transistor amp would be much easier to tame that a fast op amp. You will likely need a low pass on the output to keep spurious emissions down.
I think you mean class C amplifier, which is the more common linear amplifier configuration for RF. It is far more efficient than a class A amplifier.
To be a bit pedantic, Class C is not going to give a linear response. Having said that, in this application which I presume is OOK, it would make sense with appropriate filtering. I also realise that the term "linear amplifier" is often, if not always correctly, used to describe any RF power amplifier in the amateur community.
-
What's the point of the 50 Ohm resistor?
for opamp protection from capacitive load? and possible short? and from my basic lesson 101 here in eevblog few years ago... that a properly 50ohm terminated at cable end must be from 50 ohm source, to make the whole transmission line 50 ohm impedance flat, and maximum (half) power transfer, no? i'm no ham and no rf guy, so i'm ready to be educated.
-
The "easiest", going back to the original question, is to use an RF module. These offer 50 ohm in and out, so other than a few coupling/decoupling caps and a biasing circuit (usually just a DC voltage regulator), all you need is a suitable double sided board and a heatsink. Probably the simplest method if you have no test gear other than, say, a power meter. I would caution though because the harmonic power is likely to need filtering out, and for that you'll want some test gear.
These RF modules offer high gain and rarely self-oscillate thanks to their design. Some examples are here:
https://www.rfparts.com/module.html (https://www.rfparts.com/module.html)
Second simplest is a mmic, but you will usually need to provide matching as they are not 50 ohm in/out. The vendors provide S parameters, the most important of which for matching purposes will be S11 (input) and S22 (output) typically in polar format. You convert these polar values into a cartesian complex impedance for the frequency of interest, plug them into your favorite L match calculator, and miraculously appropriate L and C matching values will drop out the end. Just keep in mind that the parasitic C and L of your board construction are usually of a similar order of magnitude to the matching components. That is the hard part.
For example, http://www.analog.com/ADL5324 (http://www.analog.com/ADL5324)
Edit: here's a calculator that takes you from S parameter polar characteristics to L match component values in one go http://www.daycounter.com/Calculators/Complete-RF-Amplifier-Design-Analysis-Calculator.phtml (http://www.daycounter.com/Calculators/Complete-RF-Amplifier-Design-Analysis-Calculator.phtml)
-
What's the point of the 50 Ohm resistor?
for opamp protection from capacitive load? and possible short? and from my basic lesson 101 here in eevblog few years ago... that a properly 50ohm terminated at cable end must be from 50 ohm source, to make the whole transmission line 50 ohm impedance flat, and maximum (half) power transfer, no? i'm no ham and no rf guy, so i'm ready to be educated.
The answer is the same for high-speed analogue and medium-speed digital circuits.
Absolutely any ham radio or basic RF textbook will tell you in great detail what would happen and how to avoid it.
It would take a long time anyone here to generate information to educate you. And they would do a poor job compared with the vast amount of high-quality information that is readily available at your fingertips. You demonstrably even know the relevant search terms!
It will be best if you actively educate yourself, rather than passively expect to be educated.
I expect you need the answers fairly quickly, in which your best solution will be to use a cheap, readily available, predesigned and pre-certified off-the-shelf module. Then all you have to do is measure the complete output spectrum (i.e. including the 5th harmonic) and demonstrate that (when combined with the antenna) it complies with local laws. Even that isn't trivial.
Don't reinvent the wheel, especially if you don't know that it needs to be circular and without eccentricity :)
-
When trying to extend the range of 433MHz fitting better antennas on the receivers on electric gates I have got from RF Solutions to be quite effective.
http://www.rfsolutions.co.uk/acatalog/Antenna.html (http://www.rfsolutions.co.uk/acatalog/Antenna.html)
-
It would take a long time anyone here to generate information to educate you. And they would do a poor job compared with the vast amount of high-quality information that is readily available at your fingertips. You demonstrably even know the relevant search terms!
It will be best if you actively educate yourself, rather than passively expect to be educated.
a quick tip might help. i will educate myself when thing get serious. for now i have many things to do than the time i have, i'll take it slow and steady. furthermore, i believe even though after i mastered the voodoo maths and concepts, empirical study is inevitable, meaning that i need the tool to verify, meaning that i need to gather resources to get one, meaning that getting resources first is quite a better idea to do for now, imho. but a quick free tip and advice on my mistake here might give me some idea.. thanks.
-
It would take a long time anyone here to generate information to educate you. And they would do a poor job compared with the vast amount of high-quality information that is readily available at your fingertips. You demonstrably even know the relevant search terms!
It will be best if you actively educate yourself, rather than passively expect to be educated.
a quick tip might help. i will educate myself when thing get serious. for now i have many things to do than the time i have, i'll take it slow and steady. furthermore, i believe even though after i mastered the voodoo maths and concepts, empirical study is inevitable, meaning that i need the tool to verify, meaning that i need to gather resources to get one, meaning that getting resources first is quite a better idea to do for now, imho. but a quick free tip and advice on my mistake here might give me some idea.. thanks.
OK. Here is a good starting point (http://lmgtfy.com/?q=transmission+line+rf+termination).
It would have been polite to mention that you aren't "serious" (whattever that might mean) about the answer in your original post.
-
It would have been polite to mention that you aren't "serious" (whattever that might mean) about the answer in your original post.
the OP question is serious, at that particular skill level. simply can or cannot, and a quick explanation why it cant might help. thanks for the link its broken, but i get your idea. the basic 101 transmission line i learnt back then was about coax cable and end to end points on a pcb. rf is different, we are sending signal to an unterminated dead end with a particular none grounded coupled length (antenna), wrong antenna = blown amplifier stage, what?. and the fact it involves passives on the output like some miscellaneous network of inductors, bogged me down, there are many kind of networks and i dont know which one applied to which...i'm not trolling into deep depth of rf tech and mastery, i only need the answer for my particular case above, with parts that i currently have. but if people kind enough to provide materials here, i'll appreciate it, just dont expect i can digest them all in this particular thread, my capability is rotting. thanks.
@Howardlong. thanks for the great info! i'll try to catch up later, if i cant for now.
-
thanks for the link its broken, but i get your idea.
Sorry about that; I've fixed it in the original. I always have difficulty remembering the syntax, and stupidly forgot to check before posting.
the basic 101 transmission line i learnt back then was about coax cable and end to end points on a pcb. rf is different,
It is actually exactly the same, although the components are different. The electrons don't care why they are moving!
my capability is rotting. thanks.
Join the club :(
-
my capability is rotting. thanks.
Join the club :(
:-+
-
RF Circuit Design 2nd edition by Chris Bowick is a very practical and approachable text. None of the unnecessary mathematical dick swinging that's all too common in academic texts, there's just enough to explain it, perfect for engineers IMHO.
While I have the Kindle edition, there are PDF editions around that I find more a more readable format than the Kindle one.
-
None of the unnecessary mathematical dick swinging that's all too common in academic texts, there's just enough to explain it, perfect for engineers IMHO.
Somebody once said something to the effect that the best mathematical result is that you don't need to use it - because the result also proved there are valid simplifications.
-
RF Circuit Design 2nd edition by Chris Bowick
golden! downloading...
-
None of the unnecessary mathematical dick swinging that's all too common in academic texts, there's just enough to explain it, perfect for engineers IMHO.
Somebody once said something to the effect that the best mathematical result is that you don't need to use it - because the result also proved there are valid simplifications.
When I first got into DSP about 15 years ago there was a paper published written by a friend of mine who, as well as being an expert in the field with a ton of academic credentials, also worked for one of those three letter agencies in the US. It wasn't a long paper, about a dozen pages, it was about analytically solving image rejection in real time in software on unbalanced quatrature demodulators. At that time it was an important text because thus far only numerical methods had been used.
I read it every day, twice a day, on my short commute in and out of work, for two weeks. I just could not work out how he'd made one step, an "it can be shown that" step from one equation to the next, fundamental to the whole paper.
In the end I plucked up the courage to ask him how derived equation B from equation A, and, er, well... oh dear. The whole paper was pretty much worthless. I left it at that, we've never mentioned that episode since, but I do think back wryly of it nonetheless.
-
None of the unnecessary mathematical dick swinging that's all too common in academic texts, there's just enough to explain it, perfect for engineers IMHO.
Somebody once said something to the effect that the best mathematical result is that you don't need to use it - because the result also proved there are valid simplifications.
When I first got into DSP about 15 years ago there was a paper published written by a friend of mine who, as well as being an expert in the field with a ton of academic credentials, also worked for one of those three letter agencies in the US. It wasn't a long paper, about a dozen pages, it was about analytically solving image rejection in real time in software on unbalanced quatrature demodulators. At that time it was an important text because thus far only numerical methods had been used.
I read it every day, twice a day, on my short commute in and out of work, for two weeks. I just could not work out how he'd made one step, an "it can be shown that" step from one equation to the next, fundamental to the whole paper.
In the end I plucked up the courage to ask him how derived equation B from equation A, and, er, well... oh dear. The whole paper was pretty much worthless. I left it at that, we've never mentioned that episode since, but I do think back wryly of it nonetheless.
That, to use a mathematician's terminology, is "trivial": it is an unmathematical result :)
-
i want to amplify it.
Depending your local laws and regulations this might be illegal to do. Lisence free bands usually have limits how much power, noise, etc... you can put out there...
Check your regulations before some very pissed off engineer comes kicking down your door and write you nice ticket to pay... And hes pissed off because of you, someone else made complaint of noise in 433MHz band and he had to leave hes confty office to investigate... Engineers hate sitting in damn van full of gear needed to locate radio noise sources... 8)
-
What's the point of the 50 Ohm resistor?
for opamp protection from capacitive load? and possible short? and from my basic lesson 101 here in eevblog few years ago... that a properly 50ohm terminated at cable end must be from 50 ohm source, to make the whole transmission line 50 ohm impedance flat, and maximum (half) power transfer, no? i'm no ham and no rf guy, so i'm ready to be educated.
No, you want it to be as efficient as possible. If the load resistance is equal to the source, the efficiency is 50% at most. The source should have as lower impedance as possible and the load should be matched to the transmission line.
-
The source should have as lower impedance as possible and the load should be matched to the transmission line.
what is the load? its a dangling wire dead end to me. i'm guessing if any, the load will come from reflection or the sort...
-
The source should have as lower impedance as possible and the load should be matched to the transmission line.
what is the load? its a dangling wire dead end to me. i'm guessing if any, the load will come from reflection or the sort...
Your 50 ohm antenna is the load. What you can do is use an L match consisting of a C and L which will be much less lossy than a resistor, but you will need to know the output impedance of your opamp circuit at 433MHz to calculate the values.
Having said that, the opamp is a broadband device and I can imagine it not liking such a load particularly if harmonic content is high: it strongly suggests therefore that an opamp is probably the wrong tool for the job.
-
ok thank you. i need to study more on this... reading the earlier said book is probably the wisest thing to do for now, then finding the resources to buy stuffs later. thanks.
-
ok thank you. i need to study more on this... reading the earlier said book is probably the wisest thing to do for now, then finding the resources to buy stuffs later. thanks.
Just keep in mind that nobody these days works out filters and L matches from scratch on paper. I usually just use one of the many on line calculator tools. There's not much you can do in RF without knowing how to do a L match. The most important chapter for you for now is the matching chapter, and then (I think it's the one after) where they take a transistor as a gain block and match it. In the end it's just building blocks like any other area of electronics, or Lego for that matter. The expertise comes in when knowing how to prioritise several simultaneous design requirements, and what's reasonable and what's not.
-
how poor i am after reading the rf book and this thread... i realized there are so much i dont know that i dont know, it changed how i look at things, esp the antenna, as an electrical (rf) component (load) rather than just a dangling wire with some integer division of a wavelength. another important aspect is the source (amp or transmitter) to load (antenna) matching, this was the black magic i was talking about, even it got mentioned in that very book. the OP is just... :palm: but look at the bright side, puzzles are sorted out more and more now, less blank in the gap of mind. thanks to people involved esp Howardlong... i would also like to give credit to w2aew for making the video to make things much clearer on antenna tuning... maybe this concluded my confusion and blind attempt in this thread...
https://www.youtube.com/watch?v=eYN7dhdt1Dw (https://www.youtube.com/watch?v=eYN7dhdt1Dw)
-
You could always use one of these:
http://rfbayinc.com/products_pdf/product_399.pdf (http://rfbayinc.com/products_pdf/product_399.pdf)
Or it could be a bit overkill...