Author Topic: Selecting class of amplifier for a multicarrier FM broadcast  (Read 811 times)

0 Members and 1 Guest are viewing this topic.

Offline vu2tve

  • Contributor
  • Posts: 29
  • Country: in
Selecting class of amplifier for a multicarrier FM broadcast
« on: January 25, 2022, 09:14:03 am »
I would like some signal processing advice for a project. I'm working on an SDR based setup that produces multiple FM carriers in the FM broadcast band, and am wondering the class of operation of the RF PA.

Now, a straight answer for this would be of-course class-A + low pass filter, but I am wondering if I can get away with class-AB and a tight bandpass filter.

I did a couple of simulations in gnuradio to see the nature of harmonics being produced in class-ab, and while of course the class-a fft looks a lot cleaner, could I get away with class-ab + tight band pass filter.

If anyone could offer practical advice, it would be very helpful. Class-A would mean loss in operating efficiency, but if the distortions in AB would be too much, then I'd have to go with class-A (or consider AB-push-pull)

73 // VU2TVE

 

Online gf

  • Frequent Contributor
  • **
  • Posts: 725
  • Country: de
Re: Selecting class of amplifier for a multicarrier FM broadcast
« Reply #1 on: January 25, 2022, 09:46:29 am »
In your AB simulation, you distort each signal individually and then add them up - I guess this is not what you really intend.
Don't you want to add the N signals first, and then send them together to the (distorting) amplifier?
 
The following users thanked this post: vu2tve

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 19776
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Selecting class of amplifier for a multicarrier FM broadcast
« Reply #2 on: January 25, 2022, 10:02:00 am »
AB is fine if it's push-pull.

With distortion, I don't see that an effective result will be had for multiple carriers at a time.  If a diplexing filter can be had, this works with multiple class C amplifiers feeding in their respective channels; making such a filter, good enough to avoid crosstalk, is not trivial.  And, I guess, not what your synth is expecting (single pre-mixed output?).

Tim
« Last Edit: January 25, 2022, 10:04:08 am by T3sl4co1l »
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 
The following users thanked this post: vu2tve

Offline vu2tve

  • Contributor
  • Posts: 29
  • Country: in
Re: Selecting class of amplifier for a multicarrier FM broadcast
« Reply #3 on: January 25, 2022, 10:05:28 am »
You're right. I modified the flowgraph and here are the screenshots..

Also realized that it is really class-B, not AB

There's a number of harmonics, but perhaps with a BPF it could work.


I guess I could also test with actual signals and FM modulate them and demodulate and see what they sound like :)
 

Offline vu2tve

  • Contributor
  • Posts: 29
  • Country: in
Re: Selecting class of amplifier for a multicarrier FM broadcast
« Reply #4 on: January 25, 2022, 10:08:18 am »
Yeah. I want to avoid the class-C/diplexing filter. Single SDR output with multiple carriers to PA stage seems cleaner approach.
 

Offline vu2tve

  • Contributor
  • Posts: 29
  • Country: in
Re: Selecting class of amplifier for a multicarrier FM broadcast
« Reply #5 on: January 29, 2022, 02:49:30 pm »
Hi,

I made a calculator to help out with some numbers as a google sheet at this link. This is publicly editable, but I will change that after a few days have passed.

https://docs.google.com/spreadsheets/d/1H0mCZAM0MhHkbXDTtomCzDjg0lkKSL3_eBUJIuL1NSY/edit?usp=sharing

I would really appreciate some feedback and whether I made any mistakes. The sequence of calculations is as follows:

Desired output power per carrier (Editable) = 2W
Number of carriers (Editable) = 5
Characteristic impedance = 50 Ohm
Output rms voltage swing per carrier = Square_root(2W x 50Ohm) = 10V
Total output voltage swing = 10V x 5 (carriers) = 50V
Output power required of class-A amplifier; Pout = (VxV)/R = 50x50/50 = 50W
Max. achievable efficiency of class-A amplifier = 50%
DC power consumed by class-A amplifier ; P_tot = 50W / 50% = 100W = Power that will be dissipated by the heatsink
Drain supply voltage (Editable) = 50V
Rough estimate of Quiescent drain current = DC Power / Supply voltage = 100W / 50V = 2A
Load line resistance R_L = (VDD*VDD)/(2*Pout) = 25 Ohm
Max allowable Junction temperature (from Datasheet) = 175C
Case to junction thermal resistance (from Datasheet) = 1.1 C/W
Max allowable case temperature; T_case = 175 - (P_tot*1.1) = 175 - 110 = 65C
Ambient temperature (Editable) T_ambient = 30C
Target thermal resistance of Heat-Sink = (T_case - T_ambient) / P_tot = (65C - 30C)/110W = 0.4 C/W

Now 0.4C/W is probably difficult to realize, but these are the numbers for calculations' sake.

Does this look correct
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf