RF transistor amplifier design, any suitable literature?

[Yansi]

Hello!

Let's say I have a bunch of these nice RF transistors: 2SC5773 Datasheet
and I want to build an RF amplifier for a say 145MHz band. I have only basic understanding of a LF common emitter amplifier, including some basic models and in/out impedance calculation. However these basics do not apply at RF frequencies well and also I need a 50ohm matching for the in- and output of the amplifier.

What steps do I need to design such transistor amplifier matched both sides to 50ohms, at such (or any other) frequency?
Can you please recommend any suitable literature or appnotes, related to this topic, how is this done?

Thank you for help or any recommendations.
Y

[dmills]

"Radio frequency transistors" by Dye and Granburg is a good one for this sort of thing.

Regards, Dan.

[Neomys Sapiens]

Advanced Techniques in RF Power Amplifier Design, Steve. C. Cripps, Artech House 2002

Solid-State Microwave High-Power Amplifiers, F.Sechi & M. Bujatti, Artech House 2009

Handbook of RF and Microwave Power Amplifiers, John L.B. Walker(Ed), Cambridge univ. Press 2012

RF Circuit Design Techniques for MF-UHF Applications, A. Eroglu, CRC Press 2013

Circuit Design for RF Transceivers, D. Leenaerts, J. van der Tang, C.S. Vaucher, Kluwer Acad.Publ. 2001

Feedback Linearization of RF Power Amplifiers, J.L.Dawson, T.H.Lee, Kluwer Acad.Publ., 2004

RF Power Amplifiers, Marian K. Kazimierczuk, Wiley   2008

Fundamentals of RF and Microwave Transistor Amplifiers, Inder J. Bahl, Wiley 2009

Radio Frequency Transistors - Principles and Practical Applications, N.Dye, H.Granberg,   Newnes 2001


Of course, some works on RF circuit design in general do also hold information pertaining to this area.

[rfeecs]

Some app notes:

"S-Parameter Techniques for Faster, More Accurate Network Design":
http://www.hparchive.com/Application_Notes/HP-AN-95-1.pdf

"S-parameter design":
http://www.sss-mag.com/pdf/AN154.pdf

You'll notice from the Renesas data sheet that this transistor has an S21 of more than 25dB at 145MHz.  So all you have to do is bias it and you will have a ton of gain in a 50 ohm system.

You'll also notice that the above app notes don't even tell you how to bias it.  They assume you can figure that out.  Your biggest problem with biasing will be keeping it from oscillating.

A book I had long ago and probably long out of print.  Not sure I would really recommend it:
"Design of Amplifiers and Oscillators by the S-parameter Method" by Vendelin:
https://www.amazon.com/Design-Amplifiers-Oscillators-S-parameter-Method/dp/0471092266

[KE5FX]

Don't forget Chris Bowick.

[CD4007UB]

Yes, Bowick is a very good book to start from. It covers the basics of S-parameter design and much else. If you're mathematically inclined, "Electromagnetic Waves & Antennas" by Orfanidis (available online) gives a more advanced treatment. "Radio Frequency Transistors" by Dye & Granberg is a good, general reference (mainly non-mathematical).

Instead of using a bipolar transistor, for an RF LNA you might consider using a pHEMT GaAs FET such as ATF-54143 - see Avago app notes on this, including #5057.

[T3sl4co1l]

Well, hell.  A pair of bias tees, and the conditions specified in the datasheet, will do just fine!

Look at the S-params, s21 > 5 below 700MHz, and s12 < 0.12 as well.  That's a lot of gain and more than enough isolation*.

*Maybe not unconditionally stable (you can always accidentally make a tuned-base oscillator), but at 50 ohms, you're pretty good to go.

50MHz is "DC" for this thing!  145MHz is barely into its "AC" range (h_fe about 60 there?).  Your problem is simply this: keep the system as close to 50 ohms as possible, up to 10GHz.  You aren't going to see where this sucker is oscillating, if it decides to run away from you.  It's just going to be out there, whistling, and you're only going to be suspicious because bias is wrong and IMD is high!

(I recently built up some amps with BFR92AWs, which are 5GHz transistors.  They were a bit ornery and took some extra dampening components to behave.  All I could see on my 1.5GHz spec was high-order modulation spurs, not the fundamental or harmonics.  The circuit was also a bit more complicated than a simple 50 ohm CE amp.)

Tim

[Yansi]

Thank all for the tip, will try to go through them and pick possibly what I can digest.

T3sl4co1l: Exactly! That was most like what I was looking for. Some shortcuts for a stupid RF n00b. Yes, 145MHz is a DC for it. I didn't want to jump straight into it. I want that amplifier at 1627MHz to be specific.  As an alternative to using quite expensive GaAs MMIC wide band gain blocks, which are nowhere in need for a fixed frequency application, where a single NPN or NMOS would do absolutely fine.

So I can imagine how to bias a common emitter stage, however don't know the rest.
First I'd begin with a supply voltage selection. As the NPN is only 6V Uceo, I'd start with a common 5V supply for the amp, should be enough I think.
Then what collector current should I use and based on what should I select that?
Should I use resistor load in the collector or is an RFC a better choice?

Maybe some very basic questions I am asking, but it will help me at least get started. Thanks!

[T3sl4co1l]

FWIW, this is probably one of those parts where they give the DC average value rating, not the peak (breakdown) rating.  This also shows in the current rating: 80mA is nowhere near what this thing is capable of sinking.  It would probably make a good high speed, low voltage substitute for a MMBT2369 switching transistor!  This can be seen in the hFE and fT plots: normally, fT starts to droop first, then hFE, then out at collector currents where hFE is 10 or less, is where the current limit is usually placed.  (There might be other reasons for choosing such a conservative current limit... but it seems unlikely that it would be something like wire bond limits at least, at only 80mA. )

Also, 80mA at 6V is 0.48W, just under the power rating.  It's made for CW operation!

The junction is also larger than, say, a BFR92A, which has a fraction of a pF Ccb, while this is 1.6pF (quite a lot for 10GHz, unless it's capable of the current ... which it is!).

Note also, 5V into 50 ohms is 100mA, so if you hook this thing up to a transmission line with a bias tee, you're basically set, no impedance matching necessary and you've got full power output.

In short, it's just begging to be let loose at 50 ohms.  It was made for it. 

As for coupling, you can use as little as (50R) / (2*pi*145MHz) = 55nH for an RFC, but a few times that will be more comfortable.  That sets LF cutoff there.  Likewise, a coupling cap of >= 22pF.  (Read up on what a bias tee, or bias network, or LC coupling network, is.  It's just a highpass filter stage to carry and block DC.)

Or you can use a resistor (from +10V), but you lose efficiency and halve gain.  You do get a stable output impedance, which makes the amp less sensitive to load SWR.

Same thing at the base side, bias tee.  This gets supplied by a resistor, and you want some kind of bias network or feedback to limit collector current so it's not just set by hFE (suicide bias).  Normally, you'd use a series collector resistor, then connect a resistor divider between C, B and E.  (You can still use a collector bias resistor with a bias tee, just design it to drop a few volts rather than half the supply.)

You may also consider some negative feedback, which is sometimes called neutralization but I don't mean the tuned kind, I just mean a B-C resistor (at AC, not just DC).  This reduces input and output impedances (shunt voltage feedback), so you may need to add a series input resistor to maintain 50 ohm input impedance.  (Oh, if you want 50 ohm input anyway, you need damping anyway, because s11 isn't exactly 0.  Probably some resistance in parallel will do.)

So, if you've been following along, you may've noticed that some resistors can be combined.  Like, if you have shunt feedback AND DC bias feedback, just do both with one, and set the other resistor values appropriately.  So AC gain (shunt R value) sets DC bias resistor (B-E).  Which sets AC input impedance.  But shunt also reduces Zin (which may be all you need, if the feedback amount is small), so you may need a resistor in series, in front of it.

Beware that CE amps tend to exhibit negative resistance, too (a consequence of Miller capacitance plus additional phase shift), so a series input resistor can be helpful, or necessary even, though it worsens your SNR a little bit.

And this just gets you the most bandwidth possible (including DC, if you use all resistors for bias!).  HF limit basically set by the transistor properties, and whatever parasitics you've accidentally left in.  You probably don't want that.  You probably don't want FM BCB interfering with your quiet 145MHz signals, or aircraft, or TV, or FRS radios for that matter, or anything else out there.  You may actually want a narrowband, tuned amplifier, centered on 145.  If you start with a nicely resistive, terminated amp (using all the resistors as above), you can simply tack on filters and be done.

Alternately, you can integrate the filters as part of the amplifier itself, and also take advantage of the free impedance matching and high gain available to the reduced bandwidth (you could get probably 30 or 40dB gain, in a 10% bandwidth, with that thing).  That gets more and more complicated, of course, and more and more critical of the parasitics you introduce with tuning coils and caps.

Or you can shell out five bucks for the MMIC modules and a few more for the filter blocks and such.  It's up to you.

Tim

[CD4007UB]

These are good tips for building a VHF amplifier.

Working at UHF is a bit trickier. We're developing a replacement pre-amp for our radiotelescope at 1420MHz based on an ATF-58143 pHEMT operating as a common-source amp. This FET has low noise and high gain. Biasing is via a drain resistor (which also helps to stabilize the circuit), decoupled from the supply by a small inductor. An LC circuit on the input provides some impedance matching and helps the FET operate with low noise - the standard kind of circuit described in various Avago/Agilent/HP app notes. The RF S-parameter design procedure broadly follows that described in Bowick's book (and by Orfanidis).

We simulate the circuit first using MATLAB, which optimizes the circuit design for low noise and high gain at the design frequency without introducing instabillity (pro RF engineers would probably use a more powerful package from Agilent, etc). The figure below shows the theoretical performance. If we're lucky we may get somewhere near that in practice; we'll know when we get some PCBs made up.

So, 145MHz may be a good place to learn about RF amps before tackling 1600MHz. If you're unsure about biasing a BJT, you might find a FET easier to work with (and capable of lower noise). I think dual-gate FETs are quite popular at VHF, as they overcome the Miller effect and can be used in an AGC.

[Yansi]

I still need to figure out what Tim has written in his lengthy information dense post. Might take a while (as I need to work at work on very different things in between).

I am familiar with biasing AF transistor amplifiers, however some circuit structures might be good and some better at RF.  For example: Should I use an emitter resistor, or should I stabilize bias using the C-B connected resistor,  leaving emitter grounded solid? (Then comes into mind, that putting a very little inductance in the emitter circuit might increase stability margins, huh?)

What I am still unsure about, is the impedance matching stuff and to better understand all the S data in the datasheet.   
For example, if the 2SC5773 has about 11dB of gain at 1GHz, it seems from the S parameter tables, that it will have about 5dB gain at 1.6GHz. Just squared the S21 MAG value. (that is not the greatest gain, is it?).

I also can understand basic principles of how the noise works (as I work quite a lot with audio amplifiers), so for example can understand that using chokes for the bias tees instead of resistors will reduce noise (as the resistive component of the choke is rather small).

The main motivation to learn this, is to be able to build better amplifiers. I see at least a few advantages over the MMIC gain blocks:
+ way lower cost
+ smaller noise figures might be achievable
+ way higher power output levels available. (it is such a pain to find a MMIC gain block with more than 10dBm P1dB at few GHz frequencies!  For example almost impossible to find a MMIC gain block that will drive a +13dBm level MIXER at 6GHz).

[rfeecs]

This looks like a pretty good discussion of biasing:
http://www.qsl.net/va3iul/Bias/Bias_Circuits_for_RF_Devices.pdf

One issue for higher frequency circuits is modelling the components.  Vias act like inductors.  Capacitors have resonances that make them look like opens and shorts at some frequencies.  Resistors have inductance and capacitance.  Basically everything has inductance and/or capacitance and/or resonances.  These are often the culprit when your circuit oscillates or doesn't work as expected.  Once you put your circuit in a box, you can have cavity resonances and signals feeding back through the air.  As you go higher in frequency, everything acts like a transmission line.

Yes, your transistor kind of poops out by 1.6GHz.

Since my company makes MMICs, I can't agree with you on that subject.  I will agree that if you want to learn about RF circuits, it is good to play around with discrete circuits.

I tend to think that higher frequency stuff at say 6 GHz+ is beyond the reach of hobbyists.  Forget the oscilloscope.  You need signal generators, network analyzers, spectrum analyzers, power meters.  Even attenuators, couplers and all the little accessories cost a fortune.  Not meaning to discourage anyone.  I guess if you're interested you find a way.

[dmills]

I tend to think that higher frequency stuff at say 6 GHz+ is beyond the reach of hobbyists.  Forget the oscilloscope.  You need signal generators, network analyzers, spectrum analyzers, power meters.  Even attenuators, couplers and all the little accessories cost a fortune. 

Really? When hams have been playing in the 10, 24 and 47Ghz bands for years, and 110GHz is starting to see a little action these days.

Coupler? You mean a couple of bits of waveguide with a slot or two, brazed together? No reason for that to be expensive, just fiddly. Duroid is not the sourcing pain it once was, and to 10GHz or so, modern parts have enough gain that even FR4 will work if you dont need state of the art noise performance.

A SA with a harmonic mixer will get you to well north of 20 GHZ for a few grand on the surplus market (Much less if you are prepared to wait and know what you are looking at). 

73 Dan.

[KE5FX]

Amateur Radio on 122 GHz and 241 GHz using a mixture of home made and commercial bits.

Not to mention https://www.cv.nrao.edu/~demerson/bose/bose.html ...

[G0HZU]

Really? When hams have been playing in the 10, 24 and 47Ghz bands for years, and 110GHz is starting to see a little action these days.

Just look at some of the ropey/dodgy replies on here and then also consider that most radio hams just buy or modify equipment. The ones that do mess about with stuff above 1GHz will probably adapt commercial kit to suit their needs or build a kit designed by someone else. That's a lot different to learning/mastering amplifier design at microwave frequencies. There will be exceptions but most of the exceptions will probably have a professional background in electronics anyway.

Not many hobbyists will have sat down and designed/simulated and analysed a BJT based RF amplifier at 1670MHz. Plenty of people on a forum will talk a good game but ask them to be honest about how many amplifiers like this they have actually designed and (properly) analysed themselves. Then up the ante to 6GHz. How many hobbyists have actually designed/simulated/analysed an amplifier at 6GHz? Note that I don't just mean 'build and use' an amplifier at 6GHz. It isn't that difficult to design stuff with a 6GHz 50 ohm MMIC amplifier for example.

[Yansi]

With all due respect gents, would you mind leaving your hate out of this thread and create one yourself for that?

If you think the above suggestions are garbage, then help better. Otherwise don't criticize others, that are willing to help.

[KE5FX]

Not many hobbyists will have sat down and designed/simulated and analysed a BJT based RF amplifier at 1670MHz. Plenty of people on a forum will talk a good game but ask them to be honest about how many amplifiers like this they have actually designed and (properly) analysed themselves. Then up the ante to 6GHz. How many hobbyists have actually designed/simulated/analysed an amplifier at 6GHz? Note that I don't just mean 'build and use' an amplifier at 6GHz. It isn't that difficult to design stuff with a 6GHz 50 ohm MMIC amplifier for example.

It's worth considering why the packaged MMICs are so successful in the first place.  For many years they were noisier than properly-designed amplifier stages built with discrete transistors.  That is still true to some extent but the margin is now relatively small, with more cheap sub-1 dB NF MMICs becoming available. 

IMHO almost no one working at 50 ohms should be doing discrete small-signal microwave amplifier designs anymore.  The industry apparently agrees.  New MMICs are being released all the time in both packaged and die form, while we can hardly go a week between EOL announcements of classic RF/microwave transistors.  NXP in particular has been taking a chain saw to their product line, and thanks to the last few waves of consolidation they have much more than their fair share of the best stuff.  Put it this way, if your design includes any type numbers that start with "B," don't get too attached to it.

Parenthetically, I don't think I'd describe most of the microwave hams as your typical appliance operators.  Try to catch one of our Microwave Update conferences over here sometime. 

[T3sl4co1l]

With all due respect gents, would you mind leaving your hate out of this thread and create one yourself for that?

If you think the above suggestions are garbage, then help better. Otherwise don't criticize others, that are willing to help.

I read it more as a call for skepticism.  A sorely needed gift* that many people miss out on!

*And curse. 

Similarly, one must remember what matters:
- What counts as reliable information?  Is it a random forum post?
- Is it repeated the most often?
- Does it come from an authority?
- Is it scientifically confirmed?

Only ONE of these is correct: if it meets the specs, by measurement and analysis, by the scientific method, it is correct.

Repeated information is not reliable.  Authorities are not reliable (though you at least stand a better chance of getting it there).  Only when it is proven, is it correct, reliable.

And even then, you must check carefully that the assumptions that went into that proof (there almost always are assumptions) are reasonable, or themselves proven through a similar method.

We have an unusual privilege, in the domain of electronics, that every single judgement we make, can be scientifically proven.  While it's true that many proofs are inconveniently lengthy (provable computing, for instance), or that expensive test equipment is required to easily and accurately make those verifications (microwave equipment isn't exactly cheap), you must not wholly deprive yourself of this unique and wonderful feature!

And yet, I see it all the time here.  Not accusing you of this, no -- but it's the kind of advice that must be repeated so often that those too lazy to seek it out themselves, have no choice but to see and, hopefully, eventually, understand it as well.

Tim

[Yansi]

So lets just analayze  an amplifier from some piece of equipment for educational purposes. It is a part of a quad PLLVCO board. Schematic I reverse engineered is here:



Unfortunately, both of my meters fail measuring the capacitors in circuit, one fails completely, the other one shows values I would not expect to see in that part of circuit, so I did not put any values in the schematic.  The amps are powered 5V from a LP2951 regulator. I have measured also all bias voltages in the circuit, as the biasing is also quite interesting:

The bias for each amp stage goes through a resistor (100 or 180ohm), but there is none decoupling cap after that resistor. So every current change from the collector circuit propagates to the base bias. Strange. I would expect both biased completely separated. No idea why they have done that.

The first stage is BFP181, second stage is BFP182 Also interesting to note the first stage works with a Vce of 2.92V, the second stage works with 1.86V. Quite low voltages, huh? Would expect them to be way higher, as it seems the more Vce, the more fT (as seen in the device datasheet, maybe due to the parasitic capacitances inside being lower at higher voltage)
Also the first stage works at roughly 8mA, second stage 14mA.

There is a about 7dB pad on the input (the values however doesn't match 50ohms exactly), then there is a kind of shunt RL impedance matching circuit.  Guessing the transistor input is rather capacitive impedance, so they used an inductor to resonate it. The series resistor might be (?) to lower the Q, so the match may be more broad band. The VCO in the circuit tunes around 1070MHz (1055 to 1080 to be exact) - so not exactly broad band, but whatever.

The output of the amplifier also has some shunt RL matching network. I am quite curious to know, how the values of these could be determined.

(Note: It also seems these transistors would make a decent 1.6GHz amplifier, about 11dB of gain.)

Datasheet of the BFP181 here:
Infineon BFP181
Siemens BFP181, more detailed
Datasheet of the BFP182 here:
Infineon BFP182
Siemens BFP182, more detailed

[T3sl4co1l]

Oh good, they've got a little collector feedback as well as emitter degeneration.  That's more stable than either alone.

In effect, the emitter resistor turns a sloppy Vbe into a more well-behaved resistor-plus-offset.  This makes it reasonable to bias the transistor with a base voltage divider.

Alternately, you can use a resistor divider that sets collector voltage as a multiple of Vbe.  This isn't as stable because of Vbe tempco, but that's usually fine.

If you combine the two, you are effectively setting collector voltage as a multiple of that resistor-plus-offset Vbe that the emitter resistor affords.  This can give more stability than either one alone (for a given voltage overhead), or alternately, less collector resistance can be used, so that the transistor Vce is closer to full +V, while preserving the same stability.

Looks like they've got L and LR collector loads, which will have to do with the desired frequency response and matching of each stage, and little to do with DC bias (because of how small the extra resistance is).

Tim

[bd139]

Interesting. Looks exactly like the simple RF amps I build (phew!)

Confused with one thing though - do those transistors have two emitters or is it just the package has two emitter pins? I am aware of this used in TTL but nothing in the RF space so something to learn here methinks.

[Yansi]

Understand the biasing, but if I take a device like the BFP18x or even the mentioned 2SC5773, by what design processes can one design the matching networks for the transistors so one can make a 50ohm matched amplifier?
(Interesting to note, the 2SC5773 has a very detailed datasheet including some S figures, however the BFP18x devices seems very "under characterized" so to speak)

bd139: They are SOT143 package, so have two emitter legs on the package, but I think physically they are both connected to the same spot on the package, therefore interchangeable. They most probably aren't multiple emitter devices as in the TTL gate input stages.

[bd139]

Thanks for the explanation - appreciated

[rfeecs]

Understand the biasing, but if I take a device like the BFP18x or even the mentioned 2SC5773, by what design processes can one design the matching networks for the transistors so one can make a 50ohm matched amplifier?
(Interesting to note, the 2SC5773 has a very detailed datasheet including some S figures, however the BFP18x devices seems very "under characterized" so to speak)

The easiest, most intuitive way to do impedance matching is with the Smith chart.  Once you get the feel of it, it makes narrow band (single frequency) matching very easy.  Even for broad band matching, it can be done just by playing with the Smith chart which allows you to visualize the impedance versus frequency.

This looks like a decent explanation:
http://www.ece.ucsb.edu/Faculty/rodwell/Classes/ece218c/tutorials_etc/Impedance_Matching.pdf

You'll want to use a program to do this.  Forget using paper and a ruler and compass.  The RF simulators have Smith charts built in, plus tuners so you can tweek your component values with sliders and watch your match improve.

Here is a really basic Smith chart program:
http://iowahills.com/9SmithChartPage.html

I'm sure there are a lot of other resources on the web.

One thing you realize when you get into RF, especially microwave:  the thinking is a little different.  Everything is referenced to Z0 or 50 ohms.  You think in terms of voltage magnitude and phase; reflection coefficients and power; not so much voltage and current.

bd139: They are SOT143 package, so have two emitter legs on the package, but I think physically they are both connected to the same spot on the package, therefore interchangeable. They most probably aren't multiple emitter devices as in the TTL gate input stages.

You are right.  So don't draw it that way.  It is just confusing.

[Yansi]

Exactly, currently am trying to get a better grasp of the Smith chart and impedance matching. Found some useful appnotes for that already, for example this one:  http://pdfserv.maximintegrated.com/en/an/AN742.pdf

Thank you for a tip for the software, will look into it. 

I know already some RF basics, but want to further advance my knowledge, so I will be better able to analyze or likely design some stuff.

From what I have found already, the impedance matching of a simple discrete transistor amplifier is done based on the S11 and S22 parameters, that just happens to describe the input and output port behavior.

[G0HZU]

It's worth considering why the packaged MMICs are so successful in the first place.  For many years they were noisier than properly-designed amplifier stages built with discrete transistors.  That is still true to some extent but the margin is now relatively small, with more cheap sub-1 dB NF MMICs becoming available. 

IMHO almost no one working at 50 ohms should be doing discrete small-signal microwave amplifier designs anymore.  The industry apparently agrees. 

Yes, the way RF stuff is designed these days is changing in many ways. It's becoming more and more like LEGO

Yansi : + way higher power output levels available. (it is such a pain to find a MMIC gain block with more than 10dBm P1dB at few GHz frequencies!  For example almost impossible to find a MMIC gain block that will drive a +13dBm level MIXER at 6GHz).

There are quite a few MMICs that operate up at medium or high power levels at 6GHz. I had a look here at home in my fairly modest stash of MMICs and found an old eval board for the Hititte HMC788A and the Macom MAAM28000-A1. The MACOM part is old and quite expensive but there are MMICs from various manufacturers that cover a wide range of power levels up to 6GHz.  eg there are wideband MMICs that can go to >+40dBm at 6GHz.

[Yansi]

HMC788 at 20$+ a pop.  Maybe  good  LEGO for a Company, not good enough for an individual hobbyist (pricewise), let alone a student. Maybe I could just shut up, if it would be possible to sample one or two of these for a project, but as it is not possible, screw it! Transistors can do it for a tenth of a cost (and they did it for the last decade), if one knows how to use them, wisely - which I unfortunately don't. It is not possible to sample even anything cheaper, like a decent mixer, or a VCOPLL.  And then I read how someone sampled a whole lot of RF EVAL boards and made something out of them, using a spectacularly craptacular bodginess. But hey, thats the problem of the fucking country I live in. Nobody cares if you don't buy billions a year. (An then they moan there is not enough smart people interested in stuff... when almost nobody cares to support anything technical here)... But that would make a whole completely new different thread.

Now back to RF amps please. Rant over.

[CopperCone]

can you give an example of some wideband 10w chips?

[Yansi]

So, seems the matching networks are based on the S11 and S22 respectively.

To continue with my example  of 1627MHz amplifier with 2SC5773 (datasheet):
Thanks to the thorough colleagues from Hitachi, we have at least some  S parameter tables in the datasheet present.
So for example picking from the table on page 9 (Vce=3V, Ic=50 mA, Zo=50ohm), at 1600MHz we obtain S11 of MAG 0.482 ANG 141.

Now using the formula that z = (1+Gama) / (1-Gama), I can calculate the input impedance of the transistor to be zs = 0.387+j0.306

So, this should be at least a crude (but still better than nothing) estimate of an input impedance of the amp.  In absolute value, |zs| = 0.493.  The reference Z0 is 50ohm, so if I understand right, the amp has an input impedance of 0.493*50 = 24.7ohm.   Interesting. No idea if this is correct or how far from reality, but at least gives idea, what kind of matching network would be needed.

As a matching network for the amplifier input, it seems a simple LC lowpass network should be enough, with the series element connected against the input terminal and the shunt capacitor against the transistor. How to calculate these values still I have to figure out, so gimme a while...

EDIT: Well, I made it the other way round. The series element should be against the smaller impedance, while the shunt against the larger one
EDIT2: Oh... it may not work at all that way either. Doh!

[Yansi]

Last time I calculated the input impedance of the transistor to be 0.387+j0.306 in the normalized values or 19.35+j15.3 ohm denormalized.

Using the PDF attached, I tried to figure out the input matching, using chapter "Impedance Matching of Complex Terminations" on page 11.  Assuming the following model:



I choosed the method to absorb the inductance.  I calculated first the Q required to match the 50ohm into the 19.3ohm real impedance, leaving the jX=15.3 ohm to be absorbed in the low pass matching network.

Q = Qs = Qp = sqrt (Z0/Zs -1) = sqrt (50/19.3 - 1) = 1.26 

Then proceeded to calculate the matching components:

L = Qs*Zs / 2pi*f = 1.26*19.3 / 2pi*1.627E9 = 2.38nH
C = Qp / 2pi*f*Zo = 1.26 / 2pi*1.627E9*50 = 2.47pF

So the reactive part of the transistor gets absorbed, leaving  jX / 2pi*f - L = 15.3/2pi*1.627E9 - 2.38 = 0.88nH for the external component (meaning one would not rather add anything, huh). Just an external shunt capacitance of 2.2puff will be needed to match the input of the amplifier to 50ohms at 1.627GHz.

Phew...  Having no practical experience in this field at all, I can make no better than just putting a prototype on PCB and having it measured on a VNA (which is what I am going to do, after I design the output network also)

[rfeecs]

can you give an example of some wideband 10w chips?

Qorvo (TriQuint), Wolfspeed (Cree) and Analog Devices (Hittite) all have several.
Some examples:
http://www.qorvo.com/products/p/TGA2578
http://www.qorvo.com/products/p/TGA2963
http://www.wolfspeed.com/cmpa0060025d
http://www.wolfspeed.com/cmpa2560025d
http://www.analog.com/en/products/rf-microwave/rf-amplifiers/power-amplifiers/hmc7885.html

You will never be able to make a wide band high frequency power amplifier with discrete packaged transistors.  The package parasitics are just too large.  The options are hybrid chip and wire or MMIC.  For mmwave, about the only viable option is MMIC.

[G0HZU]

Your numbers for the input match look OK to me but for your first attempts at this stuff you might find it best to try to use this transistor at a lower design frequency.

Looking at the s parameters for 3V 50mA the most (unconditionally stable) gain you can expect to get at 1600MHz is less than 9dB. In reality you might get 8dB. But if you add just a whiff of emitter parasitic inductance this number quickly drops by several dB. So you could easily end up with an amp with less than 5dB gain if you want it to achieve unconditional stability at 1600MHz.

Or maybe work with a better transistor? The BFP182 in your app circuit looks to be more suitable assuming you can find one. Maybe 12.5dB (unconditionally stable) gain at 1600MHz? However, I would want something even faster than the BFP182 for a design at 1600MHz...

[Yansi]

Thank you for quick checking. Also doesn't the "emitter whiff inductance" make the amp more stable, at the expense of gain of course?

Yes, the BFP182 looks really better, have at least bunch of them here do I could test those too. However I lack any detailed datasheets for this one, namely the S param tables.

I see only two ways obtaining those figures: Either estimate them from the hybrid pi model (which is I think way over what I am capable of mathematically), or just simulate them, if I can find a SPICE model for the BFP182 (but I saw something right in the datasheet I think). Using LTspice .net command should not be as complicated I think. But never done that before.

By the way what other readily available cheap transistors for 1.6G would you recommend?  (haven't done any research yet, just if there any favorite ones)

[CD4007UB]

Yansi, are you sure that you haven't left out some capacitors from your circuit diagram?

The UHF pre-amp circuits that I've been studying all use a capacitor to ground the top end of the load resistor (39ohm in your diag) that is connected to the drain (in your case, collector). That point is then an RF ground (but not a DC ground), and the series inductor (e.g., L2) further decouples it from the DC supply (so, the inductor is not part of the drain/collector load).

Maybe your circuit operates differently, but I'm pointing this out in case you've inadavertently missed out the decoupling capacitors.

[G0HZU]

Thank you for quick checking. Also doesn't the "emitter whiff inductance" make the amp more stable, at the expense of gain of course?

Yes, emitter inductance can be used to improve the input match and the stability but you will need a faster transistor to allow you to exploit this I think...

My advice on this would be to get hold of a linear simulator to help you. You can just about get away with something like RFSIM99. Its free to download even though it is very old. At work I usually use the Genesys and SonnetEM simulators combined as this allows the influence of the PCB layout to be simulated quite well. The two simulators work together and the results are processed and displayed in Genesys. But I think you could get by with RFSIM99 for initial fumbling about with this stuff.

I would suggest you search for a few app notes for BJT designs for GPS (1575MHz) LNAs as a starting point. These will often give a realistic and sensible PCB layout, a parts list and some plots. Aim to reverse engineer what they did and see if you can replicate their results? Then maybe change the circuit to get a bit more power.

For example:
https://www.infineon.com/dgdl/AN155.pdf?fileId=db3a304319c6f18c0119ec027dee534f

There are quite a few others online. But note that these will be designed for low noise figure rather than signal handling etc. Successful RF amp design really does require you to draw up a set of requirements for what you need because some amps will be optimised for noise figure, some for signal handling and some for best match to 50R.

Also get some experience with RF (lumped) component engineering especially if you have access to a decent VNA to experiment with these components alongside the simulator.

The BFP420 is a decent device although you have to watch the low voltage limits. I've got a few here and they are cheap to buy at Farnell. But you will need to be wary of instability issues up to many GHz with this device. You might be better with something a bit tamer than the BFP420 and maybe you should try designing at a lower frequency before trying for 1600MHz. I've got discrete BJT design experience up at a few GHz but these days I either use PHEMT transistors or MMICs up at these frequencies.

I think a lot of the classic books covering this stuff are very dated. I think the best way to circulate/present this stuff (in 2017) is by video/DVD using a decent simulator and real test gear. Books are OK but they can't cram in enough words/pictures to compete with a real person talking in front of a simulator or a VNA. But the presenter has to have talent at 'presenting'!

Many years ago at my place of work we tried a video training course on RF design by Les Besser. Les Besser is a very experienced and clever RF designer. However, it was all presented in a very dull and dry manner (hour after hour) meaning it was hard to concentrate. It was on VHS videotape format so that shows you how old it was!

[Yansi]

Yansi, are you sure that you haven't left out some capacitors from your circuit diagram?

The UHF pre-amp circuits that I've been studying all use a capacitor to ground the top end of the load resistor (39ohm in your diag) that is connected to the drain (in your case, collector). That point is then an RF ground (but not a DC ground), and the series inductor (e.g., L2) further decouples it from the DC supply (so, the inductor is not part of the drain/collector load).

Maybe your circuit operates differently, but I'm pointing this out in case you've inadavertently missed out the decoupling capacitors.

Very sure about that. I have already posted a photo of the circuit I reverse engineered, I may post a very detailed macro photo of that board if you want to investigate yourself. As was said, it is a kind of negative feedback from the collector.  More Ic --> more drop across that feed resistor --> leads to less bias current, therefore a decrease of Ic. (But never seen this before too.)

[Yansi]

I see at least four ways of obtaining those figures.  #1 on the list involves looking a little more carefully, with #2-#4 being very distant next-resorts..
https://www.infineon.com/cms/en/product/rf-and-wireless-control/rf-transistor/low-noise-si-transistor-up-to-25-ghz/BFP182W/productType.html?productType=db3a3044243b532e0124c960d6f062e6#ispnTab7

Something is REALLY wrong if you don't end up with S-parameters from the manufacturer for a device that is specifically marketed as an RF transistor.  Most likely also SPICE, ADS, MWO etc. packages as well.

Yes, the BFP182 looks really better, have at least bunch of them here do I could test those too. However I lack any detailed datasheets for this one, namely the S param tables.

I see only two ways obtaining those figures: Either estimate them from the hybrid pi model (which is I think way over what I am capable of mathematically), or just simulate them, if I can find a SPICE model for the BFP182 (but I saw something right in the datasheet I think). Using LTspice .net command should not be as complicated I think. But never done that before.

Well that does make life a LOT easier, wouldn't you say? 

Thanks, didn't notice them. 

[CD4007UB]

OK, Yansi, but I wasn't querying the DC biasing/feedback, but the RF grounding. All I can see is a 1uF electrolytic capacitor on your supply, which won't act as an RF ground. Unfortunately, I don't seem to be able to post a circuit to make the point clear. But, as a general rule, with UHF circuits it's important to watch the RF grounding, as poor RF grounds produce unwanted coupling that can produce instability.

[G0HZU]

It's hard to tell from the picture but the 1uF cap looks like an SMD tant cap. We can see that they haven't tried to connect this cap directly to the signal trace. So even though some 1uF tants can still present a fairly low impedance at 1GHz I think there may be another ceramic cap on the other side of the PCB or maybe the 5V rail is arranged as an inner 'power' layer in the multilayer PCB that will have its own self capacitance to the ground layer(s).

[Yansi]

There really is NONE other decoupling caps. I have already reverse engineered the whole PLL VCO part of the board.
Yes, the board is an 8 layer custom and exotic stackup, whole board being 3mm thick. Uses blind vias (which makes a pain in the as job to trace some signals!). On the other side of the board, theres the PLL, VCO and LDOs. Nothing is related to the other side. Yes the 1uF cap is tantalum. On the other side, there is only another 10uF tantalum after a LP2951 regulator. Nothing else powered from it, everything having its own local regulator, the whole board having I think 18 LDOs.

I have cut out the PLL VCO part of the board and already try to power it to have some fun out of it, and it works. The cut exposed the internal stackup. Unfortunately all layers on the side of the shielding rings do present only GND, so no idea if under the the UHF circuit may be a whole layer of V+ for the RF amps. It is possible.

[G0HZU]

Getting back to +13dBm 6GHz MMICs I had another rummage in my MMIC eval graveyard here at home and found an old milled test board for the Stanford/Sirenza/RFMD/Qorvo SBB-5089Z. I've used this part a lot over the years, but never at +13dBm at 6GHz.

My test board originally was designed for use from 20MHz to 3GHz but I changed the caps and bias choke to suit 6GHz operation. I retested it this morning and could get +13dBm at 6GHz without having to drive it very hard. It was beyond P1dB but still gave +13dBm at 6GHz.

Can you get these in your country? They are only $4 USD

https://www.digikey.com/product-detail/en/rfmd/SBB-5089Z/599-1007-1-ND/936981

These parts are Qorvo parts but they are quite old and date right back to the days of Sirenza at least.

http://www.qorvo.com/products/p/SBB5089Z

Obviously, to get good gain performance up at 6GHz you would need to make a PCB with a very good layout with lots of ground vias and use decent ceramic capacitors and a well behaved bias choke at 6GHz. Otherwise the performance will drop away well before 6GHz. You also need very good test gear and test accessories to prove the RF performance with reasonable certainty. I can show you the performance plots/data I got if it helps?

[G0HZU]

I had a tinker on the Genesys/Sonnet simulator to see if I could compete with the 5089Z at 6GHz by trying to make something with a cheap but fast BJT from my stash of transistors here at home. I found that if I used a fast SiGe HBT/BJT (BFP650F £0.50ea) the s parameter data suggested maybe just over 11dB gain was possible up at 6GHz. But the simulator predicted this would drop to less than 9dB with a real PCB layout with the various losses and parasitics. By changing to a SiGe part with a lower Pdiss spec I could get maybe 10.5dB gain according to the simulator. But I don't have that transistor here.

However, I can only achieve this with a fairly narrow bandwidth. Probably only about 1GHz (-1dB) BW at 6GHz? I think the P1dB would be >+15dBm and the input and output match would be OK but this amp would only be suitable for something like ISM or WiFi or maybe the 5.6GHz ham band. Certainly not a general replacement for a wideband MMIC but would something like this still be of interest?

The ATF-54143 PHEMT would work here too and would maybe give 10-11dB gain but these things seem to be very expensive today at about £4 each. I have quite a few of them here but I don't know if you can buy the ATF-54143 in your country. Can you buy SMD BJT transistors in the BFPxxx range?

[Yansi]

Hello! Sorry for not being able to respond quicker, have not much free time during the work days.

The reason why I mentioned the 6GHz MMIC would be for a whole separate thread. I head a few interesting ideas and design challenges for me, that I could try. The MMIC should serve as an LO buffer to drive a 13 level MMIC mixer, but in a rather large 1 octave range (approx 3 to 6GHz). But I need first to make some more babysteps in RF design, before I proceeding to such designs.

I have already looked a bit at the RFSIM99. It seems it will be a very powerful tool, thanks for the tip. Haven't played with it yet, but definitely will. I have also here a copy of Ansys Designer student version, which I obtained mainly to design and simulate microstrip filters. (already made a few of these physically, and measured them partially) Maybe the Ansoft Designer can also simulate using a "black box" two port component with loaded S parameters into it. That would become veeery handy too!

I see some ATF-54143 on Aliexpress for reasonable prices. (the BFP420 is there too)  Probably some NOS, I think one could probably trust those, maybe.  Some BFP devices are in stock at Mouser (Mouser is the very available large distributor here, Digikey unfortunately is not, they want a utter nonsense money for shipping. Farnell is here available easily too, but they want nonsense money for the components instead. Their pricing is mostly very noncompetitive).

For completeness and for those interested, am attaching a photo of my last microstrip bandpass etch test and a measurement of its S11.  Didn't measure anything else (apart from having it "noised" on a spectrum analyzer to see a sort of S21), as the VNA I have access to has only single port available for measurement. (Do not have the other connector converter for the crazy HP RF connectors :-/) The filter should have been 100MHz wide, however measures about 200 meg. Dunno why that happened, but the main variable is the mostly unknown material I made it on. I basically made an well educated guess of the permitivity (and haven't been very far off! The filter center is 1.46GHz, but should have been 1.5). Loss tangent of course unknown.

[G0HZU]

Your BPF looks tidy. I don't see any scalpel marks or soldering iron burns on it yet though One thing of note is that the (1.5dB?) return loss in the stopband makes it look like the material is very lossy or maybe you measured it with some cable inline after you calibrated the VNA?

RFSIM99 is going to struggle a bit up at 1600MHz because it's really just aimed at lumped element design. So it isn't easy to model microstrip unless you import it as a 2 port model or if you use the basic transmission line model. But at a stretch you can use something like Sonnet Lite to model and export short lengths of microstrip as a two port model for use in RFSIM99 as a two port data file.

RFSIM99 would have been fine for designing and simulating an amplifier at 145MHz. There are several classic ways to (reliably) make a broadband 50R amplifier with a BJT at 145MHz but I guess you are more interested in higher frequencies? But if you have a decent BJT, a 145MHz amplifier with quite good 50R matching and good stability and flat gain is quite easy to make. You can almost do it all with a pocket calculator with no need for s parameter data but having a simulator and a 2 port model of the transistor makes it easy to simulate and predict the performance.

At work I have access to the latest SW from Microwave Office and Agilent Genesys and Sonnet but I still like to use the old version of Eagleware from 2004 for quick and dirty design work here at home.

When making one off designs for amps or oscillators I like to use as few critical components as possible so for the simulation of that BFP650F amplifier the only critical parts were the BFP650F and two (Kemet HiQ CBR 0603) ceramic caps. But these were still low cost parts. To keep the design simple I used radial stubs for the bias feeds and simulated it in Genesys and Sonnet combined together as in the image below. This does mean it is quite big. The whole PCB is about 25mm x 25mm in size. I did use a decent PCB material but this wouldn't be that critical for this design. But the PCB does need to be quite thin to minimise the inductance of the through vias.

As you can see it only has about a 1GHz -1dB bandwidth at best so no good as a replacement for a decent MMIC. But it was fun trying to get a simple PCB layout that preserved the gain as much as possible up at 6GHz.

[KE5FX]

There are quite a few MMICs that operate up at medium or high power levels at 6GHz. I had a look here at home in my fairly modest stash of MMICs and found an old eval board for the Hititte HMC788A and the Macom MAAM28000-A1. The MACOM part is old and quite expensive but there are MMICs from various manufacturers that cover a wide range of power levels up to 6GHz.  eg there are wideband MMICs that can go to >+40dBm at 6GHz.

MAAM-011206 looks pretty good, P1dB about +18 dBm at 6 GHz, usable to 15 GHz, about $11/1 at DigiKey.  Quite a bit cheaper than the HMC788A.

[Yansi]

The VNA was uncalibrated in the first place. The dudes operating are well... how to say it. Corporate employees.  And the nice dualport 3GHz VNA is used for singleport measurement of 13.56MHz loop antennas for NFC. So if there is something, that does not have to be done, they do not do it, they do not have the tools for it. They said something like the machine has a metrology certificate, but only up to 30MHz.  So currently am searching for anyone else in the "neighborhood" who has a DC-3G dualport VNA, that can be calibrated at least using the short-open-50-through.
There was also a 20cm SMA-SMA cable between the VNA and the filter, so a bit of loss might be introduced by that and the PCB should be a variant of FR4, but FR4 should not have very high loss at 1.5GHz either.

I have the Ansoft Designer here to simulate higher frequencies. I didn't experiment with that part of the software yet, so I do not know its limitations, but from what I know about it so far, it should work. I have also the Sonnet Lite here, but I must admit I hate that. Its craptacular GUI drives me nuts and the memory limit prevents me to simulate anything more complicated than a line with few stubs. 

Yes, you guess right, my interest is at higher frequencies, but nothing especially high. The current goal I set for myself was to make at least somewhat usable receiver for the 3.4GHz amateur radio band.  I already made a few experiments, designs based on very old books that were designed in a way they might work only by a random chance. So quickly reverted back to basics, to learn a lot first and make the things more properly with more modern components.

But currently I still need to get a good understanding of the impedance matching, and make a few experiments with that to confirm the theory works. At least finish the experiment with the 2SC5773 at 1627MHz and maybe make some more for different frequencies with different transistors. Currently need to get some free time to try design the collector matching circuit.

[G0HZU]

I couldn't resist actually building the 6GHz BJT amplifier so I milled the PCB and there are only a handful of parts on the board as I bias it manually with a second supply. So It was quick and easy to make.

See below for a quick measurement with my VNA and an image of the PCB. The response is shifted down slightly to about 5.6GHz and I think this is partly because I rounded up the capacitor values from the ideal. I haven't bothered to change the caps as it's probably more use to me on the 5.7GHz band anyway Plus I didn't want to inflict any rework scars on the PCB and spoil its photoshoot. However, I didn't want to waste any of my ultra fine PCB rivets on this board so the via holes are all just wired/soldered. So it looks a bit clunky.

But the plots below show that even up at 6GHz the s parameter models for these BFP transistors seem to be quite good and it shows how powerful Genesys and Sonnet are when combined. Like you I don't like the user interface of Sonnet but when combined with Genesys all the user inputs are done in Genesys and Genesys launches and controls Sonnet remotely and exports the PCB to Sonnet and then re imports the simulation data. It then uses this data to produce the graphs for S21 etc.

It's the first time I've used the cheap Kemet caps. I would normally use ATC 600S caps here but they are expensive. I don't know how accurate the Kemet caps are up at 6GHz and I think they are the reason it is slightly off frequency.

But it still looks good. I got about 9.4dB gain at the peak. Slightly more than the simulation but that's partly due to the extra solder around the device and this reduces the emitter inductance a bit I think.
I also tested it for P1dB and got about +15dBm P1dB on a decent lab power meter and the OIP3 was about +28dBm. But I could have biased it harder to improve this I think. I can't measure the noise figure very reliably here at home. Maybe one day I'll measure the NF at work but that won't be for some time.

Just happen to have a picture of SBB5089

Thanks. Looks like there's a lot going on inside that device!

MAAM-011206 looks pretty good, P1dB about +18 dBm at 6 GHz, usable to 15 GHz, about $11/1 at DigiKey.  Quite a bit cheaper than the HMC788A.

Every year the MMICs just get better and better... Nearly 30 years ago when I first started work on RF converter design the WJ A25-1 amplifier was considered to be very special. But it was always a bit weedy in terms of OIP3 and P1dB and it only worked to about 1500MHz and the noise figure was just under 4dB. But it was also expensive! There are amplifiers available now that would have been beyond fantasy back in those days

[Yansi]

Fakin hell! The firefox crashed when I was writing a lengthy response. 

Look what I have found! Less than $3 on Mouser: TRF37A75: http://www.ti.com/lit/ds/symlink/trf37a75.pdf
+12dBm OP1dB at 6GHz.  Not as bad.

So I need to make some progress with the discrete transistor amplifier experiment. I need to design the collector circuit, however am unsure how to proceed with that. I guess using an RF choke is a preferred way. But how to pick a value? What impedance should it present at the design frequency? For example to achieve 1kohm, I'd need about 100nH at 1.627GHz. That seems rather lot.  Still the impedance will have to be matched to 50ohm, so maybe a lower impedance of the choke make be sufficient. I guess I have to combine the S22 of the transistor with the impedance of the collector circuit being in parallel and then the resulting impence will be used in the impedance matching design process. Is this the correct approach?

Also, some times I see also a small resistor (tens of ohms) in series with the collector choke. What is the purpose of this resistor? Could not find a definitive answer for that. (Even the amplifier I have reverse engineered has that).

I still use common whatever SMD 0603 capacitors I have available, the small values being generic NP0/C0G.  I think these are still good enough until couple GHz, aren't they?

Thanks,
Yansi

[G0HZU]

The simplest starting point when choosing the collector components is to decide what output power you want and what DC operating point you want for the transistor. The stuff I've written below is just a crude/ballpark analysis and it may contains nuts/typos as I'm in a bit of a rush. I've got a few things to do tonight but hopefully all the stuff below is OK...

If something like the BFP420F is used it will typically be run at 3V Vce so there isn't much in the way of voltage swing available at the collector. So if you want to produce +13dBm (20mW) with fairly low distortion, you would probably want to operate the transistor at 3V and about 30mA bias current and present it with a collector load of 100 ohms or maybe a bit less. This is the load the transistor sees look outwards into the L match. Not the same as looking back inwards obviously.

Here is why 100R is a reasonable target for +13dBm power...

Power = (Vpk*Vpk)/(2*Rload)

With 3Vce at the BFP420F you will get maybe 2Vpk (4Vpkpk) of voltage swing before severe distortion and limiting kicks in so in your case with a 100R load seen by the transistor...

Power = (2*2)/(100*2) = 0.02W = +13dBm. Bingo.

If you present it with too high a load then you need lots of voltage to get +13dBm and the lack of (available) voltage swing from the BFP420F will limit your power (before distortion). So you won't achieve +13dBm power.

If you present it with too low a load resistance then the 30mA bias current will limit the (distortion free) power to a level below +13dBm.

A reasonable target would be Vce 3V, Ic 30mA and a load of about 100R if you want +13dBm. So a typical L match from 100R to 50R at 1620MHz would have 10nH as the collector choke and a series 2pF cap to your 50 ohm output. But these values are only approximate and you need to play with your simulator and model for your transistor to optimise it all.

Putting a 10R resistor in series with the collector choke will help with stability and also supply isolation/filtering and can also help with matching and it won't cost much in terms of loss. It should be OK to use cheapo ceramic 0603 COG/NPO caps here but there will be some variability between manufacturers I guess. Worth a try and they should be fine at 1620MHz as long as the overall inductance within the structure is low.

If I was designing the amp I'd do things a bit differently than this simple/classic approach but it should work OK.

[Yansi]

Well thanks a lot for that explanation, I completely missed the voltage x current point 

So for a 100 ohm collector load impedance, I need 10nH.  I will then combine that with the S22 of the transistor and calculate the match. Let me try that, I am looking forward to what I can come up with. Only paper, pencil and my trusty old calculator. Just need to exercise my math skills a bit.  Then I'll try to work out the Ansoft, if I can simulate that maybe including layout.

I will try to finish the design experiment with the 2SC5773, as I have these available right now. Even though the C5773 is not very good at 1.6GHz.  (The BFP420 and others will be on the way.)

[Yansi]

Well now I am a bit confused about designing the output match. You write:

. So a typical L match from 100R to 50R at 1620MHz would have 10nH as the collector choke and a series 2pF cap to your 50 ohm output.

But where did the S22 of the transistor go into that? If I  use 10nH collector choke plus 10ohm in series, combined with S22 of (0.217  -174°, that is 32.2-j1.55ohm)  results in 29.3 + j7.69 ohm, that has to be matched to 50ohm.   Or did I miss something?

[G0HZU]

In reality, the 10nH in the simple L match will need to be tweaked (down?) because the transistor will have some self capacitance. This self capacitance will make a 10nH choke look bigger than it really is at 1620MHz.

Also, the transistor obviously doesn't have perfect isolation from input to output so tweaks to the input match will upset the output match and vice versa. A bit like jelly wrestling. That's why it's important to use a decent simulator that can look at this stuff in real time. So changes at the input can be optimised for the optimal output circuit.

I'll see if I can find time tomorrow to set up a harmonic balance analysis.This is a frequency domain analyser that can display non linear behaviour (in a steady state). It can show real time (small signal) matching and can also show steady state waveforms in the time domain.

Note that your 2SC5773 isn't a good choice to play with here. I've got to go now... I've lots to do tonight... sorry...

[Yansi]

I already got it, that the C5773 is crap at 1.6GHz. But still I'd like to try design the amp with it, and then have a compare with BFP182 I have available also.

Currently sitting in front of the RFSIM99 and toying with it. I will definitely love this thing! The creators definitely took a piss to make a very good UI for their software. It took me just minutes to learn how to use that thing. Quite spectacular compared to the today's complicated anti-intuitive software full of shiny colors and animations...

I put in S params of the BFP182, calculated input match manually and simulated. Well, I wasn't very far off (still the collector circuit was just 50R terminated, need to design the collector circuit somehow).

When adding the 10nH+10R collector load, the automatch in RFSIM99 will generate two component matching network. Same as I thought it should have been used. 
I also toyed a bit with the "emitter whiff" ( ) and been watching its effect on the gain. The theoretical maximum with the BFP182 being 12dB under ideal conditions, I think one could expect like 8dB from a real design. (looking forward to measure a real prototype, huh!)

Will need to find the touchstone files for the C5773, or to make my own based on the coarse datasheet tables. Just to be able to compare the performance of the two transistors.

//EDIT: I have made the SnP files for the C5773 myself, based on the tables in the datasheet. Haven't found anything more detailed from the manufacturers. Then tried to simulate this transistor at ideal conditions: 8dB of gain. At maybe more real conditions, about 5dB to be expected it seems.
Attaching the files if anyone interested.

[G0HZU]

I put together a quick harmonic balance (combined with linear) simulation using the BFP420F and see below for the graphs etc. The gain was good at about 15dB and it also shows that the Vswing of 2Vpk at the collector is as expected at +13dBm (shows little distortion from a sine wave with harmonics at about -20dBc in the spectrum plot) and the input and output match is quite good. The 2pF cap is there at the output but the 10nH inductor at the collector is down to about 6nH because of the capacitance inside the transistor and also the series inductance in the decoupling cap and also inductance/loading effects in the transistor. So it has shrunk from 10nH to about 6nH.

I've used a non linear model of the BFP420F here and this generally isn't as accurate as a decent s2p model when it comes to small signal matching. But when I dropped in an s2p model at 3V 30mA the results for the match were very close. But I think to get the best results for a 'real' circuit the effects of the PCB layout will need to be modelled using Sonnet and the various matching components adjusted accordingly.

It looks like you are getting to grips with RFSIM99. It's a shame that the developer stopped working on that software project because it had great potential.

[Yansi]

Yes, the RFSIM99 is a very friendly app.

So finally, I cobbled together two experimental amps, one with 2SC5773, the other with BFP182. Neither of them worked as expected, but I did not expect the performance of both to be so piss poor.

The input shunt capacitor for matching may work in theory (in the simulator and on paper), but not in practice.  I could get like ~2dB of gain at 1627MHz from the 2SC5773. After desoldering the 2pF input shunt, gain rose to about 4dB. Well, that at least a bit does what the theory thinks it should do.
However the big disappointment was the BFP182. Couldn't get more like 2dB of gain, even without the input shunt cap.

So something has been done very, very wrong. I checked biases of both amps an they seem to be correct (apart from the slight variation due to hFE dispersion), not far away from the design goal. I am suspicious about two things: 1) Very bad layout, 2) Very bad inductors.  I used some not so known type of inductors I got in a set some years ago and they might have very crap performance at these frequencies. Combined with the bad PCB layout, it is a recipe for a disaster. What do you think?

[G0HZU]

The layout doesn't look great and the circuit seems odd because you have put a resistor and a cap in the emitter of the BPF182. However, it might be possible to salvage some performance with the BPF182 version...

Can you make it look like the modified image below? This now has the emitters grounded directly. You have to be careful how you bias it now via the separate Vbias line, start with a volt or two at Vbias and bring it up until the collector current goes up to maybe 20mA.

You need to drill and fit ground vias where there are red dots and use a separate Vbias supply. Also, run Vcc at 5V to start with. Because you can run this device at a higher voltage, the 2pF cap at the output can be smaller and this will reduce the bandwidth but the gain should improve.
Also note that some components have been removed or moved elsewhere. Also a lot of component values have changed. Also the collector choke is now just a bit of very thin wire bent up in the air like a hairpin. You could make this wire from a single strand taken from some 702 hookup wire. This way it will be easy to bend and tune without ripping pads off the PCB. Push it away from vertical towards the PCB to reduce the inductance. This way this inductor can be tuned in circuit. Or you could make a tiny solenoid inductor of about 7nH here instead.

I can't guarantee this will help but you should see some more gain as long as you don't mind taking a scalpel to your PCB and making the changes shown.

Note that the bias method here is very basic and not suitable for a final design as the bias point will be very temperature sensitive and also will vary according the device beta. But you can use this crude method during development as long as you swap over to active bias for the final design.

Thinning down the trace where shown will make this section look like a series inductor. Trim it down if it seems to help with S11. But do this bit last.

[Yansi]

Oh well. Rather than bastardizing the board, I can make a new one. I base the layout of the BFP182 board on the piece of gear I have reverse engineered. The amp there is working slightly below 1000MHz, so I guessed the layout will be at least a bit good for 1600MHz too. But maybe not. Attaching a detailed photo of the original amp.

So are you telling me, that to get at least some decent performance at 1600MHz, one can not use emitter degeneration (resistor + bypass cap) at all?   If that's the truth, then okay, I get it. But that brings another question: How to bias that little bastard. Having to manually tweak two power supplies may be good enough for prototype but certainly not for a final product.

How to bias a common emitter stage with its emitter directly on GND?  I can see a few ways to do it:

1) Resister in between base and collector (as used commonly in AF amp) , but as there is zero to non DC resistance in the collector load, this is not gonna cut the mustard.

2) Resistor from base to supply voltage. Will, work, seen that done even in the appnote AN155 from Infineon, but I think this is very "meh", will drift like crazy with temperature, as the hFE changes with temperature and also variations between different transistor's hFE makes this very unfriendly for production of more pieces. Every amp will need to be tweaked separately.

3) The last resort I can think of is to close a feedback loop for the bias. I think of a PNP transistor on a collector current shunt, controlling the base current. This should make the bias current very stable (up to the point of the Vbe tempco of the PNP). But will this work? Isn't that overkill? Of how is it done in real designs?

I will need to try simulate what your change to my circuit does first and to understand it. Now I have to go have a lunch. I'll continue afterwards...

[Yansi]

So I tried to replicate your suggestion in the RFSIM99. It seems it does something, but the return loss is not very good, about -10dB.  (Note: Lowering the output cap to 1.2 puff makes it slightly better to -15dB S22, but S11 still -10dB).  Lowering the emitter whiff also makes it better. How did you come up with the input matching circuit?

[T3sl4co1l]

Well, the large supply of vias sure stands out to me.

The lack of any connection between connector housings and top side ground also looks rather suspicious, but maybe that's more of a noise problem than a gain or oscillation problem.

Emitter bias: it's fine, as long as you have enough bypass caps and vias to keep the ground return length sufficiently short.

Otherwise, you end up with a significant amount of reactance there, which is degenerating the transistor.  Or often, making it oscillate!

If you can't make the emitter stub length short enough, then you have no choice but to ground it to plane, directly (red dots in marked-up picture above), and use a collector resistor for bias feedback (as discussed earlier).

BTW, regarding port impedances: when 1/s12 << s21, the amplifier has high isolation, and the port impedances are very nearly s11 and s22, respectively.  When s12 is relatively large (which is often the case in common emitter amps and such), the feedback is significant, and this affects the port impedances as well as the stability.  A wideband amp with low dB's gain shouldn't have too much problem, but this is significant in tuned (narrow band) amplifiers, where you're trying to push as much gain as possible (indeed, you may then be trying to push the maximum stable gain, another parameter to be aware of!).

I would hope RFSim is taking this into account, so you should be fine there.  Just to add this as FYI!

Note that inductors on the order of 7nH are just a few mm of thin trace, where "thin" is Zo some ratio above the system impedance (i.e., a reasonable inductor is a length of 100 to 150 ohm trace, in a 50 ohm system).  When specified in these terms -- as transmission line lengths -- you should get even more accurate answers from RFSim.  Hence why you might want to scrape at some traces, or beef them up as the case might be.

Indeed, you can express the emitter impedance, and everything else, that way: as transmission lines.  I don't know if you can include all those components around the transistor, in RFSim, to create a model that is well representative of the physical circuit, but if you can, it should provide more insight into how your circuit differs from expectations.

Tim

[Yansi]

With transmission lines, it definitely looks more interesting. I had to change the values a lot.

I know (can calculate using tools (SaturnPCB toolkit) and verify with a little help from formulas here) that a line of 10mil width (Which I am able to etch at home with very repeatable results) has ~105ohm Zo on my substrate, which means about 6.1nH per 10mm.

I could use PCB inductors (transmission lines) into the design as well, that is a very good idea! As mentioned above, putting transmission lines into the circuit I had to make a lot of changes in the simulator, to make it fit back where I want. (1627MHz and a decent match).

Here's the result. I think it looks decent (but still may be not corresponding so well with reality).  Match is at -20dB,  S21 being reported 9dB, which would be great if it would work that way in reality.

[G0HZU]

How did you come up with the input matching circuit?

In a hurry

See the very basic sim below.

I spent more time doctoring the PCB image than playing with the sim because the sim is very crude and doesn't model the PCB layout. So I slapped in some standard cap values and I guessed the inductance and the amount of trace thinning required. But these will need to be tweaked to suit the PCB layout.

Note that the transmission lines in the sim below are actually FR4 (0.031" PCB thickness) microstrip models rather than just transmission lines. In an earlier post I did suggest you either use the raw transmission line model in RFSIM99 or better still would be to use Sonnet Lite to model the microstrip. You can probably do the whole thing in Sonnet Lite if you can live with the clunky user interface. ... but I don't know the limitations of Sonnet Lite. Will it allow s2p imports etc?

[G0HZU]

But that brings another question: How to bias that little bastard?

The last resort I can think of is to close a feedback loop for the bias. I think of a PNP transistor on a collector current shunt, controlling the base current. This should make the bias current very stable (up to the point of the Vbe tempco of the PNP). But will this work? Isn't that overkill? Of how is it done in real designs?

Have a look at the applications section in the datasheet for the classic ATF-54143 PHEMT. Although this is a FET device the active bias circuit can be adapted to suit your requirement. It's worth reading this section in the datasheet because it gives a worked example on active biasing.

[G0HZU]

I base the layout of the BFP182 board on the piece of gear I have reverse engineered. The amp there is working slightly below 1000MHz, so I guessed the layout will be at least a bit good for 1600MHz too. But maybe not. Attaching a detailed photo of the original amp.

So are you telling me, that to get at least some decent performance at 1600MHz, one can not use emitter degeneration (resistor + bypass cap) at all?

I looked at your commercial board a few days ago and I think the key design drivers for those amplifiers are to get modest gain at 1GHz with fairly low P1dB, but to also achieve decent reverse isolation and good stability across a wide frequency range.

I don't think you can adopt the same emitter 0603 R and C in your amplifier at 1620MHz using a BFP182 if you want to achieve a decent gain. eg 10dB or more. But you might be able to use 0402 parts here and a very tight layout with the ground vias fitted very snugly. However, the primary reason I said it was odd that you had used them in your own PCB was because they weren't there in your early RFSIM99 simulations at 1620MHz. You just had an inductor there.... although it did seem to have a lot of inductance.

[Yansi]

But that brings another question: How to bias that little bastard?

The last resort I can think of is to close a feedback loop for the bias. I think of a PNP transistor on a collector current shunt, controlling the base current. This should make the bias current very stable (up to the point of the Vbe tempco of the PNP). But will this work? Isn't that overkill? Of how is it done in real designs?

Have a look at the applications section in the datasheet for the classic ATF-54143 PHEMT. Although this is a FET device the active bias circuit can be adapted to suit your requirement. It's worth reading this section in the datasheet because it gives a worked example on active biasing.

Thanks for the tip, will read it. I have only skimmed through and do not understand why the PNP has the current monitoring shunt in its emitter, instead of base. Guessing so the shunt can be designed with any voltage, not Vbe only.

[G0HZU]

It does explain how it works in the text. Best thing to do is study the worked example and then simulate the system in spice over temperature etc.

One classic/amusing aspect of RF design is that if you gave a junior engineer or a student a task to design an RF amplifier using a fast BJT they would often produce an oscillator up at maybe 4GHz. They would then spend considerable time trying to tame it.

But if you asked another student/engineer to actually design /create an oscillator up at 4GHz they probably wouldn't know how to do it and if they tried they would typically fail to get it to oscillate.

[cdev]

I would like to share my experience making cheap LNAs with commodity off the shelf parts for use with my SDR collection (for receive only)

This was and remains a easy cheap way to learn and make ones self a useful RF gain stage for almost nothing. MMICs simplify the process so much its really a phenomenally easy thing to do to use them. The various lessons I learned were invaluable in jump starting my electronics hobby.

Making my own PCBs I found that making the classic two sided design was asking for trouble even if I made great effort to add lots and lots of vias.

The first ones I made tried to imitate sent away boards but none of them worked very well.

I realized then that having a copper ground plane near to both sides of the trace on the top made them much more prone to problems so I ditched most of the top ground plane. key to success was putting a bunch of drilled through vias literally right underneath the gain device. Also I angled the device by 45 degrees to make the trace a bit shorter and straighter, so the input on one corner and the output on the other were closer to the coaxial connectors.

All of my LNAs ended up being quite tiny.

It was pretty easy and a lot of fun and even though they were small and simple, in the process of doing this I learned essential lessons in RF design fast. My first attempts were really bad as I was trying to use copper and perfboard instead of etching a PCB. They were also much too large for what I was doing. The first ones only worked up to around 500 MHz. The devices I ended up with were not good enough to keep, I redid them. I figured out the following which seems to be a good approach for people who want to make their own boards.

If I confined the signal path to a direct - short - straight trace of appropriate width on the top of the board and ran power to it on the top also, and did not put a ground plane on the sides of the top except where I had to, as part of the bias tee/power -keeping it short and using lots of bypass caps, it worked well.

I used edge mount SMAs and used a ground plane on both sides under them, at both ends but did not have any ground at all along at least one side, and on the other the grounds were used for anchoring bypass caps exclusively and the area was minimal - also I used lots of vias under the MMICs.  I made one attempt at a two stage LNA but it diidnt work well, however I did have success connecting two LNAs together when I used separate power supplies for each of them. (and not the same low voltage post-regulated power) These were very simple tiny boards. Typical size 25 mm long, 17 mm wide. The MMIC handles all the biasing and all I had to do was supply it with clean DC between 3-6 volts, either via a bypass cap through the case or via a bias tee using the coaxial cable, and block DC.

This approach is ridiculously easy and saves a lot of money. I ended up with good LNAs optimized for different uses-otherwise I would have had to spend several times what I paid for all the RF ICs for just one LNA.

The LNAs I made all have extremely decent gain and very low noise. One thing I found is you have to clean them really well, remove all the flux - if you do that and don't connect the LNA right to the antenna, use at least a short pigtail (10 cm will do) on both sides, which I think smooths out the impedance a bit, (putting a ferrite bead there around the coax to reduce common mode signals is good too) they almost always work quite well. Gain was likely consistent with manufacturers claims which are roughly 15 db gain from 25 MHz-4 GHz -

Seems to me gain continued both upward and downward without any radical changes so these devices likely would test out to be functional on pro test equipment.

I have used this same device for an optimized broadcast FM LNA using 75 ohm antenna and connectors and there it works extremely well too. Grounding the device well and simplicity seem to be the key to getting the best out of them.

[T3sl4co1l]

Difference being, one is stable and the other isn't.  Before I tamed this thing,



(uses BFR92AWs), it was squealing somewhere up over 2GHz.  Changed with proximity, so, it was also a rather overly sensitive Theremin.  I mean really, my Theremin (proper) only moves 10kHz when I touch it, but a whole GHz?

(After fixing it up, by the way, it's a solid 20dB gain, apparently pretty flat from a few MHz to over 700MHz.  And has suspiciously low noise of 1.7 nV/rtHz.)

Tim

[Yansi]

It does explain how it works in the text. Best thing to do is study the worked example and then simulate the system in spice over temperature etc.

One classic/amusing aspect of RF design is that if you gave a junior engineer or a student a task to design an RF amplifier using a fast BJT they would often produce an oscillator up at maybe 4GHz. They would then spend considerable time trying to tame it.

But if you asked another student/engineer to actually design /create an oscillator up at 4GHz they probably wouldn't know how to do it and if they tried they would typically fail to get it to oscillate.

I understand, how it works. But still need to read it to understand why it was done this way.

Yeah, some years ago, I attempted to make a narrowband VCO for 1627MHz. Yes, the same frequency of interest. You probably can guess, it didn't oscillate, but when it did, it wasn't 1627. It was all over the place and the tuning was severly nonlinear and non-monotonic - which is a very bad thing for a PLL to work with.  Maybe I will try that VCO again, but with a difference of some years more experience in circuit design.

I started designing another transistor experiment. Now with "microstrip inductors". Let's see, what it will do.  Not finished yet, but I will make the biasing as simple as a resistor from +5V. I think it is acceptable for a one off experiment, to tweak the Rb value for a proper bias current. (Otherwise I would make the biasing more reliable.) Do you have any suggestions about the layout, what I should change, before I proceed to etch it?



T3sl4co1l: What does that board do (or what should it do)? The layout looks quite "space generous" so to say for RF.

cdev: Fine, I get it. But MMICs are not the only solution to a RF amplifier.  For example, you are not going to build a HF receiver out of MMICs. Therefore I am attacking discrete circuit design.  And for a narrowband fixed frequency amplifier, an adequate discrete transistor is I think suitable solution.  Also, most MMICs suck at delivering power, the OP1dB being mostly well below +15dBm at MW frequencies, while a single transistor (when designed properly into the circuit - which is the reason why I am learning this) can deliver +20dBm from a single SOT143 device. For fraction of the price of a MMIC. 

//EDIT: Changed layout to a completed one.

[G0HZU]

Difference being, one is stable and the other isn't.  Before I tamed this thing,



(uses BFR92AWs), it was squealing somewhere up over 2GHz.  Changed with proximity, so, it was also a rather overly sensitive Theremin.  I mean really, my Theremin (proper) only moves 10kHz when I touch it, but a whole GHz?

(After fixing it up, by the way, it's a solid 20dB gain, apparently pretty flat from a few MHz to over 700MHz.  And has suspiciously low noise of 1.7 nV/rtHz.)

Tim

I only skimmed over that PCB but that first stage looks like a common base amplifier. The poor decoupling on the base (via a few mm of microstrip) is going to cause that first stage to lose any unconditional stability in a big way in the upper UHF region if that first device is a BFR92. Common base amps using fast BJTs are notorious for instability and it looks like there are more of them there in cascode amps.

The placement of the RF decoupling caps does seem odd as does the length of some of the hot traces.

You would never get that amplifier PCB through a critical design review where I work I can tell you why it would raise alarm bells but maybe you are happy with it if you have tamed it enough already.

[G0HZU]

Do you have any suggestions about the layout, what I should change, before I proceed to etch it?

If you gang the supplies together you will have to 'select on test' the bias resistor going to the base if you want to set the bias current. You might have to swap the resistor several times to get this right so be careful not to damage/lift the copper pads for this resistor with the iron.

Otherwise your latest PCB doesn't look to be easily tweakable for the printed inductors. Maybe add some extra artwork to make them easier to trim? Otherwise you could end up having to cut them out and use sticky copper tape in their place to make alternatives.

[T3sl4co1l]

T3sl4co1l: What does that board do (or what should it do)? The layout looks quite "space generous" so to say for RF.

Sorry, didn't make that very clear -- 20dB wideband amp.  Idea was to use relatively many stages, at relatively low gain, to maximize bandwidth, without cranking up the bias (BFR92s aren't very powerful) or resorting to MMICs or what have you.

Wideband, mainly because I have more interest in time domain and EMC, though it does a fine job as a preamp for my spectrum analyzer, with a 1m bowtie antenna, to see much of what's floating around the airwaves here.

//EDIT: Changed layout to a completed one.

I like the emitter vias.   Don't like the suicide bias, but if you're okay with that, then that's fine I guess.  You should consider being able to hack in a trimpot, though.  Drift isn't something you can deal with by soldering in different chip resistors. 

I only skimmed over your PCB but that first stage looks like a common base amplifier. The poor decoupling on the base (via a few mm of microstrip) is going to cause that first stage to lose any unconditional stability in a big way in the upper UHF region if that first device is a BFR92. Common base amps using fast BJTs are notorious for instability and it looks like you have more of them there in cascode amps.

The placement of the RF decoupling caps does seem odd as does the length of some of the hot traces.

Yeah, not sure what the layout guy was thinking when he did that board...

Looking at one here that I "fixed", it's studded with extra resistors, and I think 100pF bypass caps, in strategic areas.  Several cases of scraped-off soldermask for desperately-needed ground pads near important connections (like the input's common base).

You would never get that amplifier PCB through a critical design review where I work I can tell you why it would raise alarm bells but maybe you are happy with it if you have tamed it enough already.

I'm still a bit unsure about a couple of the variants I've made (one seems to have dubious peaks around 500MHz, like it's nearly oscillating, or that I'm seeing conversion products against an unseen GHz tone; but anomalous response is not seen in the noise-source test), but they certainly cleaned up from what you see there.  Well, functionally clean.  The boards don't look so good... 

Tim

[G0HZU]

Are you sure you get a flat bandwidth up to 700MHz out of that module?

It scares the bejeezus out of me. Have you ever looked into it with a VNA? I know you have cured the instability but I'd expect to see spikes of negative resistance up above 2GHz from what I see in the version on screen. Even if you grounded the base really well the BFR92 itself has parasitics and the output signal path looks quite long before it reaches the next BJT stage. There isn't much ESR here and I'm not sure what the input Z of the next stage is at 2GHz but there's enough trace length to do a decent trip around a smith chart at UHF so I'd expect to see negative resistance at the board input 'somewhere' in the upper UHF region and maybe beyond this. But I can only look and guess. I think other engineers at my place of work would also be scared of that module in a review

I can suggest some 'alternative MMIC' circuits using BJTs if you want broadband 50R gain up into UHF with high reverse isolation. The difficulty is achieving unconditional stability as well as low noise figure and decent signal handling.
I haven't (properly) played with stuff like this for many years because a MMIC would almost always be a better/smaller/lower risk alternative.

The NRE costs of discrete design are a real killer where I work. 15-25+ years ago I could design discrete VCOs and also lots of variants of discrete small signal amplifiers because there were few alternatives. But the NRE costs of putting a VCO or exotic amp through formal testing are too prohibitive these days unless you are making stuff in huge volume on very low margins.

But even in those days I don't think a wideband (UHF) MMIC alternative would have been tolerated in a design review because of the risk of instability up at many, many GHz. i.e. at frequencies beyond the manufacturer's s parameter tables. There's nothing like putting a lid over a marginal amplifier design and then cooling it to -40degC to make it wake it up into oscillation

[G0HZU]

Some time ago I analysed a simple BC547B test board on a VNA to get s parameter data for the transistor and the VNA data (post processed on a PC) suggested this transistor could (in theory) oscillate up at 600MHz+ if the conditions were right. The faster RF transistors can hoot at frequencies way beyond what we are looking at here.

[CD4007UB]

(After fixing it up, by the way, it's a solid 20dB gain, apparently pretty flat from a few MHz to over 700MHz.  And has suspiciously low noise of 1.7 nV/rtHz.)

Tim

Just to pick up briefly on the point about noise, 1.7nV/rtHz is indeed a very small voltage by everyday standards. However, the noise from a 50ohm resistor at 290K (the standard reference temperature) is only about 0.9nV/rtHz. So, the noise figure (NF) for your amplifier is about 4.6dB (a noise temperature of about 550K).

LNAs usually have a NF <~1dB (a noise temperature of about 75K). For example, in our radiotelescope, we're looking at 1420MHz hydrogen emission from the galaxy, which has a noise temperature of ~100K. The receiver needs to have a comparable (or lower) noise temperature. That translates to a NF of about 1.3dB, and the LNA at the front end aims for a lower NF (which may not quite be realized in practice).

BJTs are not generally used in LNAs because of their shot noise (due to random fluctuations in base and collector current). In contrast, the main RF noise source in FETs is Johnson noise from the drain-source channel resistance. The high transconductance (gm) of pHEMTs means they have a low resistance (~1/gm), and hence low noise, combined with a high gain. They are, therefore, the device of choice for low-noise applications, including measuring the cosmic microwave background from the Big Bang (which has a temperature of about 3K), as described in https://arxiv.org/abs/1310.3088 .

It can be tricky to convert between the different ways of measuring noise. Some useful online calculators are at http://www.daycounter.com/Calculators/Thermal-Noise-Calculator.phtml and http://www.rfcafe.com/references/calculators/noise-figure-temperature-calculator.htm .

[cdev]

Lots lots more vias. move two vias around each ground leg closer to legs. Need many more of them. Right under ground pins.

Need vias all along sides, Go with big vias along trace and smaller vias behind it. (so as to avoid resonance) The closer to your device, the more via density.

Also if your software allows your doing it, maybe angle the chip - I dont know if the design - which has a pretty high via density in the red PCB below would even be enough but it might be. If it wasnt you could drill a few more holes so you could have even more right under the legs.
Best if your pins could go right onto or next to vias on both sides. The little chip would straddle a break in the signal path, the trace would be cut. Has to be a certain width to reduce the capacitance. Ideally there both side ground areas get connected. If you cant do anything like that consider leaving the side ground planes out and doing something like the mini-circuits suggested layout below.

(The pattern is for a PSA4-5043+ MMIC which, if you look around you seem now to be able to buy for as little as $0.75 sometimes.)

[Yansi]

Winner winner chicken dinner! Maybe... likely accidentaly... need to have that bastard characterized using better equipment.  (I.e. proper two port VNA, which I do not have access to, yet.)

I was able to get ~9dB of gain from it!

The suicide bias wasn that bad at all. I made an guesstimate about hFE - but close. So from the real circuit measurement, I calculated back the real hFE (was exactly the typical value from the datasheet, go figure!), then calculated the correct base resistor.  Almost bang on, running it at 19mA. (didn't have 21k, used 22k instead).

I should have made a bigger island with more vias on the upper emitter leg. (I hate hate hate the manual wire-through via soldering!) In a proper PCB manufacture, I wouldn't probably hesitate much to via the shit out of the board. But still, it seems to be (likely accidental) success. Even the simulation predicted 9dB of gain.

What I do not know, is the OP1dB of the amplifier.  I do generate the 1627.5MHz  carrier from some very old radio-amateur build. It is very unstable, drifting (amplitude wise, frequency is held precisely by an OCXO ) and I cannot vary the output amplitude finely. I can only stack 4dB pads which I salvaged a handful from old equipment.

What I did found, is the output of the 1627MHz LO is not even close to 50ohms. (which one could guess, based on how the LO output is done).  By connecting the amplifier directly to the LO output, I cannot get the  9dB gain.  If I insert at least one of the 4dB pads before the amplifier, the LO match to 50ohm gets probably better, and then I am able to get 9dB of gain. 

Heck.. how a proper RF generator would be handy!  Unfortunately, this is very out of my skill to build and very well out of what I could afford to buy, for this kind of frequencies. 

Photo, schematic below.


//EDIT: I have added a bunch of BAP64Q  PIN diode quads to shopping cart at Mouser, so probably I could use that to vary (and maybe even stabilize the amplitute of the LO source?)  By making another piece of this amplifier, I could probably get enough power to probably overdrive the amp and make Pin/Pout measurement by fine adjusting the PIN attenuator. ??

[cdev]

That bugs me a lot too! 

Working on getting a better setup where I can drill nice neat measured holes quickly.



Quote from: Yansi on Today at 13:27:40

I should have made a bigger island with more vias on the upper emitter leg.

(I hate hate hate the manual wire-through via soldering!) In a proper PCB manufacture, I wouldn't probably hesitate much to via the shit out of the board. But still, it seems to be (likely accidental) success. Even the simulation predicted 9dB of gain.

[Yansi]

I have no problem drilling more holes. Making some chips or dust is not a problem either, as I have a small machine shop at home, where I do all the electro-mechanicals a small metal working.  I hate soldering the wire jumpers through the board, as it is a very slow and boring process.

Now that a first working amplifier is done, I am not quitting yet, not at all!

For example I completely skipped the stability analysis here. Need to have a look back at the Sonnet Lite and the Ansoft Designer to see, if I will be able to simulate the whole amp including the PCB. And a lot of other things.

At least I am currently a bit more positively "tuned" after this. 

I will take S11 and S22 measurements of the amplifier tomorrow or on Tuesday (using the single port available only, non-calibrated VNA)


I have currently another goal for me to try and test.  Meanwhile I continue to study and experiment with transistor RF amps, for some projects I'd like to build, it would become handy to be able to multiply frequency. For example I have a 814 MHz LO I'd like to multiply by 2 to get the same 1628MHz I have now, and even multiply even that by two to get 3256 MHz. I should probably create a separate thread for RF multipliers, hopefully I don't get accused for spamming the forum with n00b questions too much!

//EDIT: Typos.

[T3sl4co1l]

I have no problem drilling more holes. Making some chips or dust is not a problem either, as I have a small machine shop at home, where I do all the electro-mechanicals a small metal working.  I hate soldering the wire jumpers through the board, as it is a very slow and boring process.

I have some PCB drills that are just the right size for some, whatever it is, 18AWG or so, bare wire I've got laying around.  A little finagling wedges the wire in the hole, then I cut it flat, and pound on the wire a bit, riveting it in place.  Solid riveted vias, what could be better?   (Glommed over with solder, of course.  I don't trust such a crude rivet to hold.)

Tim

[T3sl4co1l]

Just to pick up briefly on the point about noise, 1.7nV/rtHz is indeed a very small voltage by everyday standards. However, the noise from a 50ohm resistor at 290K (the standard reference temperature) is only about 0.9nV/rtHz. So, the noise figure (NF) for your amplifier is about 4.6dB (a noise temperature of about 550K).

LNAs usually have a NF <~1dB (a noise temperature of about 75K).

Yeah, not low as LNAs go, but it's quieter than the other amp that I made specifically for low noise purposes that's two or three times worse...

Tim

[cdev]

When I thread several through at once and then solder them all, one side and then the other, and then chop the tangle of wire off, with sidecutters, I waste a bit more wire but its faster.

[Yansi]

Well, I have currently the smallest drill bit 0.5mm, but using a rather thinner wire, so working multiple wires at once is not possible. it just falls out.

[cdev]

There has to be a better way and we have to find it. Has anybody ever used the little grommets they sell?

[Yansi]

Me personally not, but my friend uses it. It sucks. They are too large. I can make the wire-vias to take very small surface area on the pcb.

[cdev]

I know the inductance of larger diameter plated hollow vias is much lower than smaller ones but I have no idea what changes when the via is solid metal. Another option is a slot, maybe cut with a Dremel, a very thin line, with copper in it. Might be thin enough to solder both sides well in one motion.

I wonder how hard it is to electroplate your own vias? Using graphite and a low voltage power supply.

[Yansi]

I've seen people doing that on YouTube. But my goal is not mastering electrotechnology, but mastering circuit design.  For quick and dirty prototypes or  retro designs from THT components only, I etch the PCBs at home, as the photo-etch-process is quite quick and precise with very repeatable results.  The small PCB for the RF amp takes like about an hour up to hour and a half, from cutting a bare PCB to drilling holes in the finished board.

Currently I am making another piece of the board, that I will use to measure the other one's OP1dB (and the whole Pin/Pout characteristic).  Still I have to wait for the PIN diode attenuator to arrive, or find some PIN quad at home.  I have somewhere a large TV (or satellite?) distribution amplifier, and if my memory serves me right, it was absolutely full of PIN diodes. Maybe I could  bodge an attenuator from those? Mmm. Interesting idea.   

[Yansi]

Found the detailed photos of the TV distribution amp.  There is a fuckton of these diodes.  Marking seems to be AG or A6.

A6 might be BAS16 : https://www.infineon.com/dgdl/Infineon-BAS16SERIES-DS-v01_01-en.pdf?fileId=db3a30431400ef6801141b93811b03ff
But that is a switching diode. Useless for attenuators.

But I suspect the marking to be AG more likely.

EDIT:  AG may be ZMV835A, but varicap diode? I don't think so.  They must be either the switching diodes (but what for in the UHF distribution amp?) or PIN diodes.

EDIT2: Well.. They most probably are the BAS16. Even switching diodes can be used to switch RF (??), but with a rather low off isolation. Hence probably why there are four in series on the photo below!  The configuration seems to be a SP4T switch.

[G0HZU]

Glad it seems to be working a bit better now...  I'm not sure how you are measuring the power level but obviously, you have to be wary of mismatch uncertainty as this can give a fairly wide uncertainty window wrt your gain measurements.

It's a while since I did any formal VCO design but I could probably design a fairly decent narrowband VCO for you at 1620MHz with good phase noise and stability. But it would need soma fairly decent components. But if you want something a bit more average I could make something a bit uglier and cheaper?

[Yansi]

Well, what decent components exactly?  I can take an off the shelf Minicircuits JTOS can packaged VCO, slap that on board and use that. Have some of those here too.
But sure knowing how to design a VCO from discretes would be more interesting for me. If you could for example share some details how to do it (or why and how you did it), you'd be gold! Many thanks!

(Note: I'd probably vote for the cheaper and uglier, to make it from what I have available, rather then buying expensive new stuff).

I can only imagine building a Colpitts oscillator (I have built a bunch of those, but only at VHF), but am not too sure about how these classic topologies stand at these frequencies. Maybe using some quarter wave resonator may be good, but at 1627M, the resonators are pretty large and space hungry.

I have also a bunch of LMX2326 or TSA5055 PLL chips I could use to lock the VCO (either the MiniCircuits can or discrete would be more interesting) and produce a nice 1627 MHz LO.

To measure the power, I use my trusty old Advantest R3131 spectrum  analyzer.  Unfortunately, still don't have any RF generator in my lab, it is on the list to get one. But I can borrow a crusty (not trusty) old Russian one for 1627MHz if necessary. It has a cavity resonator inside and is mechanically tuned.

[Yansi]

So I dug through my document archives and found Infineon AN061: W-CDMA 2.3 GHz VCO using BFR360F and BBY58-02V, which seems to be the closest to what I am looking for.

Also found another interesting book on RF design: Randall W. Rhea, Oscillator Design and Computer Simulation .
And already started reading the book from Chris Bowick, RF Circuit Design. It is the very basics mostly (at least the beginning part of the book, but still good to remind that.

Attaching the VCO schematic from the AN061.

[T3sl4co1l]

Are you sure you get a flat bandwidth up to 700MHz out of that module?

Reply moved to here: https://www.eevblog.com/forum/projects/wideband-amp/msg1299433/#msg1299433

Tim