I find the stitching interesting. As both stages do work with the same frequency the stitching for rf could be the same. So the stitching effort at the second stage is made to improve thermal conductivity.
Congratulations! You found one amplifier module from Ali that seems to work. I got two different which died after a few seconds. The first had a RF mosfet in the output which was killed by insufficient cooling and no over-voltage protection ( the seller gave a supply voltage range which extended the limits of the mosfet). The next one had a non existing biasing of the last stage. But the first stage (the same chip as yours) of both modules is still alive and can be used for further experiments.
That is a good observation with the flat compression. Most amplifiers that I measure had flat compression characteristics, no matter if the gain characteristics are not flat. That have interesting 🤔 implication for intermodulation measurements for setting the output power per tone. You position that to fix power, at list 10dB (I usually set that 16dB) below 1dB compression to be in linear region. So the compression characteristics perfectly align with power requirements for intermodulation measurements ...in most cases.
It looks like the SMT parts were machine soldered, but perhaps under-soldered, with too little paste solder, and everything else was poorly hand-soldered, with the flux not removed being a common indication of sloppy workmanship. I'd remove the flux before I used it. It can sometimes lead to feedback oscillation and can often hide a poor solder joint. It will also increase the noise level on small-signal circuits. I recall as a kid working at a TV radio shop trying to find the hum problem in a new All-American 5 tube radio. I could find no defective parts even though I worked on it for half a day. The boss came back from doing home TV repairs, heard my report on what I'd done, took 1 look at it, then cleaned off the flux and it was fixed. Two minutes.
Can you work out which stage is compressing ? Can you refine the design and make it more betterer? You could have a whole series on buying cheap parts from AliX and refining the designs for cheap
3:16 KB2X3 -> RF Power MOSFET (Type Name: KB, Lot No: 2X3), Mitsubishi RD01MUS2B, 527MHz, Pout 1.6W, Gp 15dB, Vdd 7.2V
I find the stitching interesting. As both stages do work with the same frequency the stitching for rf could be the same. So the stitching effort at the second stage is made to improve thermal conductivity.
Yep. As soon as you get very far onto the ground plane away from the conductor it's only for thermal conduction.
Congratulations! You found one amplifier module from Ali that seems to work. I got two different which died after a few seconds. The first had a RF mosfet in the output which was killed by insufficient cooling and no over-voltage protection ( the seller gave a supply voltage range which extended the limits of the mosfet). The next one had a non existing biasing of the last stage. But the first stage (the same chip as yours) of both modules is still alive and can be used for further experiments.
That is a good observation with the flat compression. Most amplifiers that I measure had flat compression characteristics, no matter if the gain characteristics are not flat. That have interesting 🤔 implication for intermodulation measurements for setting the output power per tone. You position that to fix power, at list 10dB (I usually set that 16dB) below 1dB compression to be in linear region. So the compression characteristics perfectly align with power requirements for intermodulation measurements ...in most cases.
As a radio ham, I see all sorts of uses for that little guy...
I work for another rfic company and my coworkers all pronounce Qorvo the same way you did, if you were curious at all.
I like Qorvo Gold.
It looks like the SMT parts were machine soldered, but perhaps under-soldered, with too little paste solder, and everything else was poorly hand-soldered, with the flux not removed being a common indication of sloppy workmanship. I'd remove the flux before I used it. It can sometimes lead to feedback oscillation and can often hide a poor solder joint. It will also increase the noise level on small-signal circuits. I recall as a kid working at a TV radio shop trying to find the hum problem in a new All-American 5 tube radio. I could find no defective parts even though I worked on it for half a day. The boss came back from doing home TV repairs, heard my report on what I'd done, took 1 look at it, then cleaned off the flux and it was fixed. Two minutes.
I've got a very similar amplifier from AliExpress that does 2w at 5.8GHz. Also underwhelming, but it was fun to play with and cost less than 20 bucks.
I was explaining that P1dB to students this very week.
Can you work out which stage is compressing ?
Can you refine the design and make it more betterer?
You could have a whole series on buying cheap parts from AliX and refining the designs for cheap
the second device could be a RF mosfet
It seems PCB not dissipate heat adequately all around the second stage: poor welding, not welded the chip cooling fin?
Do you have an example of a really well designed board? Does that require more circuitry, or just better parts?
I'm assuming that the constant frequency power sweep is varying the voltage while maintaining a constant 50 Ohms?
here's how: th-cam.com/video/yGKWBpgN8PU/w-d-xo.htmlsi=tlhxJuKJgAClaP2K
These two green leaded components are inductors? My first thought was - leaded resistors. The assembly is horrendous.
Yes, they are inductors to block the RF from getting back into the power supply.
your channel is amazing for me like curiousmarc youtube channel and some other channel
Qorvo was born from the merger of two you're more likely familiar with: RFMD and Triquint (RIP).
Sweep 8uW to 80uW. Succinct-er.👍