500 Khz full bridge RF amplifier SMPS design and debugging
Hi. I am in the process of designing a Voltage controlled SMPS with VOUT varying from 20 - 200 V set based on user input. Maximum current rating is set at 4 Amps. This SMPS will be the power source of Full bridge RF amplifier, amplifying sine wave(500 Khz) to few thousand volts (Ex: VMax : 2000 V).
With the initial design and development of SMPS in place, the RF amplifier is able to work properly with SMPS VOUT values upto 70 V. Post that value, any increase in SMPS voltage, VOUT becomes unstable and varies randomly. Impedance of the RF amplifier load varies from 10 - 4000 ohms and maximum driving current limit is set at 500mA.
I would need some help in understanding the factors that dictates the proper operation of SMPS and RF Amplifier.
Re: 500 Khz full bridge RF amplifier SMPS design and debugging
yes i worked at a huge telcom place, and the SMPS guys failed to test the RFPA SMPS with an RFPA load, and when they eventually did, it wouldnt work properly, because the RFPA had a big capacitor at its input, and this made the smps unstable when it eventually got connected to it.
4 million pounds worth of 4G test data went down the drain.
But anyway, i woudl think an SMPS will not have a feedback loop fast enough to be able to see the 500khz.......it will just look like a constant load.
I imagine the RFPA will see the ceramic output caps of the smps as an LC circuit though.
Re: 500 Khz full bridge RF amplifier SMPS design and debugging
Hi.
@mtwieg : SMPS was tested with dummy load and it worked properly. RFPA drives a step up transformer which can be seen by the SMPS as an inductive load.
@treez : The load which RFPA feds varies from 10 to 5000 ohms, given the output voltage of RFPA is in thousands of voltages, loads expects very high currents [ex : V = 2000V, R = 10 ohm, I = V/R = 200 A]. But the current supply limit of the SMPS is 4 A. Do you see a possibility that RFPA trying to pull more current from SMPS than it is required. If RFPA looks like a constant load, how do we derive the average current drawn by it ?.
@Easy peasy: Yes. That could also give in to my problems. Do you have any suggestions to overcome it ?.
Hi.
@mtwieg : SMPS was tested with dummy load and it worked properly. RFPA drives a step up transformer which can be seen by the SMPS as an inductive load.
I highly doubt that the RFPA presents an inductive load, regardless of whether it has a transformer on the output.
As treez suggests, the 500kHz generated by the RFPA is probably far outside the control loop bandwidth of the SMPS, so it should look roughly like a resistive load with some shunt capacitance.
@treez : The load which RFPA feds varies from 10 to 5000 ohms, given the output voltage of RFPA is in thousands of voltages, loads expects very high currents [ex : V = 2000V, R = 10 ohm, I = V/R = 200 A]. But the current supply limit of the SMPS is 4 A. Do you see a possibility that RFPA trying to pull more current from SMPS than it is required.