100W Beginner

[ajsoukey]

Hello,

I just graduated and have started to get into RF/radio though my arduino hobby so I am a bit of a newbie when it comes to amplifiers. 

I have been having similar issues as spriya (Sathya) where my board is being supplied with 1W (0.2 V after 40 dB attenuator) from the 5W amplifier to my oscilloscope with a 50 Ohm feed through terminator. The preamplifying chip was supplied with a 20 MHz amplitude modulated signal and the wave packet had excellent shape after setting the bias to 68 mA. 

I got everything soldered on the 100W board and set the bias fully counter-clockwise and supplied with 50.0V from a CNC 48V power supply I had laying around. The resulting RF was very distorted on the output at approximately 1.5 Vpp (after attenuation) but improved in shape as I turned the potentiometer but after a minute, the signal suddenly dropped to 0W. I purchased a replacement MRF101 but I would like to ask some clarification questions before I destroy another transistor.

Please note:

-I have checked all of the resistances and capacitances of the components for correctness. 

-I checked to see if the signal was making it through the circuit by putting in a 1 mW signal and probing it throughout the circuit with no signs of unsoldered connections. 

-I have already gotten a 100x60 mm heat sink with brand new thermal paste and a fan to hopefully prevent overheating. 

The main confusion I have is where exactly is the bias voltage measured on the circuit and is it possible to roughly set it before soldering the transistor in as was done by Nick in 2021 (WA5BDU)? 

If I had a thermal issue, should D1 be a problem when trying to hook up another transistor to the board? 

By thermally coupled should I include thermal paste/heat transfer pad to the diode or just press it down for contact?

Are there any other suggestions of what I should check with my setup?

Best,

-AJ

[jcve...@gmail.com]

Hi AJ,

Yes the waveform directly out of the amplifier has a high harmonic content the must be filtered to produce a distotrion-free waveform.

Regards,

Jim 

[jcve...@gmail.com]

To answer your question specifically:

where exactly is the bias voltage measured on the circuit and is it possible to roughly set it before soldering the transistor in as was done by Nick in 2021 (WA5BDU)? 

You can measure the bias voltage at the gate of the MRF101A. The bias voltage is unimportant; the bias current is the relevant measurement. It should be 100 mA. Always set the bias voltage to the minimum value then monitor the drain current and set the voltage for a drain current of 100 mA +/- 10 mA.

If I had a thermal issue, should D1 be a problem when trying to hook up another transistor to the board? 

I'm not sure what you mean by 'should D1 be a problem'. You can test D1 by applying the bias voltage to the PCB and measure the voltage across it; should be ~0.65V

Also confused by 'hook up another transistor'. If you replace the MRF101A MOSFET you will need to set the bias voltage to minimum and repeat the bias setting procedure.

By thermally coupled should I include thermal paste/heat transfer pad to the diode or just press it down for contact?

I usually put a bit of heatsink compound on top of the MRF101A MOSFET so that it contacts D1 to help transfer the heat. 

Are there any other suggestions of what I should check with my setup?

It sounds like your amp is working as designed. If you are expecting an undistorted sine wave output, you need to filter the output.

Jim.

[ajsoukey]

Hello,

I forgot to specify that I was low pass filtering my signal after each amplification step so the shape of the pulse envelope isn't a problem of filtering but of the bias. 

I have replaced D1 with a new diode which measured fine but I had an extra just in case. I bought the MOSFET and replaced the 50 ohm resistor which measured in tolerance but looked damaged and once removed, split in half. I'm thinking my shotty spacer setup for the transistor had a lead accidentally make contact with the heatsink shorting something and killing the board as I pressed down to adjust the variable resistor. 

Before I test it again, where exactly am I placing the multimeter probes for measurement of the drain current? I assume it is at D and G pads on the top. I just wanted to check since the 5W had specific I_sense pins that could be connected and I really don't want to risk it. 

Is it possible/safe to turn on the power supply voltage for either/both of the 5W or 100W amplifiers without an input signal as long as the output is connected to a load? 

Have you thought about doing a single board that can cascade a 0.5-1 mW signal all the way up to 100 W on a single 50 x 100 mm PCB? I'd be interested in a stacking format for the arduino mega (2" x 4") running everything. 

Thank you for your response,

-AJ