PA3GUO I don't know. I will test it more and share the results.
Today I tested it on-air. The amplifier gets hot, so I put it in the balcony. I also connected the output to spectrum analyzer (see the attachments)
I do not have a spectrum analyzer, but a quick and dirty set up this afternoon using a pluto SDR looked quite OK. My portsdown (plus small LNA) connected to the input gave 44dbm on the output for a 125ks stream. Power supplied 27,5v and 5A.
Need to do some proper measurements tomorrow.
beware : you get 36dBm from 10dBm input - means 26dB Gain, ok
-increasing input power to 20dBm you write output power is 43dBm thus only 23dB gain -
seems it is beyond P1dB and might be nonlinear !
Hello. I ran into an issue with this amp today.
For 2 days, I was running it and had many QSOs with strong signal reports and clean output modulation.
Today I did the same, but in the middle of a QSO, suddenly my power dropped a little bit and my voice became distorted!
I have checked everything and even changed my SDR to make sure the problem is not them.
It seems the PA is in an under-drive mode. If I set the TX gain on SDR Console to 100, it works and the output signal is strong, but it's distorted. If I reduce the gain, my signal becomes very weak!
I provide ~25dBm to its input.
Any ideas? Is it possible that the input stage of the PA is damaged? How can I best troubleshoot it?
Mehdi / DF2HF
I will check the capacitors today and will follow those posts.
I was initially concerned that I have damaged the first stage.
Is it safe to remove the faulty bias capacitors entirely and not replace them?
If you have acces to a spectrum analyzer you can build yourself a simple probe / sniffer
up to ghz
solder a 450 ohm smd or better 2 times 910ohms parallel at the end of a short piece of
semi rigid coax cable followed by 100nF cap to decouple dc :
this will represent a voltage devider with 50 input of yout spectrum analyzer with 10:1 ratio, thus
doing 20dB attenuation --
with this probe you can check input / output level of your amplifier (relative measurement ) check if the stages still have gain and reasonable dB and you can locate the stage easily
Made some pictures of one of my probes as described above to make things more clear -
Its by far the best tool i ever build for rf measurements in circuit - its a simple cheap and all passive 20dB probe
wondering why it has never been described so far !!!!
if you add a ground clamp/spring at the tip to get gnd connection you can do absolute power probing anywhere in your circuit - but it also works without ground spring -
resistors in this case .2 times 910 ohms parallel in 0402 ( 0603 also ok) and cap 0402 100nF
Thanks. I will do that as well.
In the initial tests, I couldn't find a bad capacitor.
I measured the voltages on Gain and Drain of every stage.
It seems there's a problem in either N100 or N110, because I don't have a voltage in the gate of BLF6G22L.
Going to test more and will update you if I manage to find the issue.
One more update: I cut BLF6G22L`s access to N100 and provided an external voltage directly to the gate.
This caused the idle current draw to increase from 200mA to 1.4A and also increased the output power again, so the problem exists in that section, but I think I have messed up something because the output is a little bit distorted and has an echo in the voice.
I faced another weird issue, but this time on another amplifier.
The idle current should be around 1A, but it's only 6mA !!
When I touch the gate of the transistor at the top of the picture, the current increases to 1A.
What can be the issue? I'm checking all the connections. The only thing I can think of is a loose connection to ground somewhere.
I really hope you are not using it like that - there is no ground for the transistors, no heatsink
My best guess at the bias circuit is in another thread here:- https://forum.amsat-dl.org/index.php?thread/2906-can-i-just-ask-about-a-pa-from-bisonelectronics/&postID=11857#post11857
Post #83 Perhaps that may help fault find.
Well you never know. There are some operators inexperienced with microwave amplifiers having troubles. Please can you attach a picture of the actual amplifier as I spent a fair bit of time trying to see if there was anything wrong I could see on the PCB. I didn't see anything but it's possibly not the same PCB.
Touching the gate of an LDMOS amplifier is usually a bad idea if it is done with anything conductive as it risks oscillations that will destroy the gate. If it starts taking a lot of current that's probably what's happening.
As you are seeing no bias current either there is no drain supply, insufficient gate bias or you have a destroyed gate and the transistor is dead. Hopefully one of the former. Is there bias voltage and if so what is it? You can measure this at the capacitor before the RF choke which in this case is just a thin PCB trace. Sometimes decoupling capacitors can fail short but I think that's unlikely. I would say it's just below the switch on threshold.
G0MJW Thanks for your time.
Well, I'm also a beginner to microwave. My experience in this area is limited to my QO100 setup (this is the 3rd amplifier I'm tinkering with, after the Chinese WiFi booster, and the other Doherty amp)
The voltage at drains seem to be normal (28V), but the gates are 0.4V and 0.5V, so you're right about the threshold and that's where my problem is.
I've attached some pictures. Please note that this is not my final setup and I'm not gonna use it like this. I just wanted to make sure that this device is not faulty, and only then I would spend time on a proper heatsink. So no RF input is given to it. Just idle measurements.
Also as you see in the pictures, the sources are not grounded, so before I do the test setup, I connect the source to ground manually.
These modifications are according to the manual provided by the seller (removing the capacitor in the input, doing 2 snowflakes, cutting some traces in the output path, and removing the final isolator)