search

oresat / oresat-c3-hardware

C3 (Command, Communication, and Control) our Command and Data Handling (C&DH) card.
Other
16 stars 7 forks source link

5th order LPF between AX5043 SE output and PA needed #22

Closed glennl closed 3 years ago

glennl commented 4 years ago

The requirements document (block diagram) shows a 5th order LPF between the AX5043 SE output, and the input of the PA. This has not been added.

Glenn and Vigely should confirm it is really needed.

glennl commented 4 years ago

After discussion with Vigely, it's determined it is needed. There are significant magnitude harmonic products coming off of the AX5043, and the PA design will likely amplify them. The filter choice is the one Vigely used in his thesis work. and can be found at: oresat-c3-rf/eagle/lpf-5th-order.sch

NB: Since the signal is still considered "small signal" at this point before amplification, small 0402 lumped components can still be used.

There was also discussion about adding another 5th order LPF after the PA, but we decided to leave this out because, (1) the PA will not be running anywhere near its 1 dB compression point (low in its linear range), and (2) the component sizes will need to be bigger, and the board is already getting a bit full. After testing this C3 Rev-4 version, we will need to decide if the is a good decision.

andrewgreenberg commented 3 years ago

Glenn, II need this designed. Can you do that?

glennl commented 3 years ago

I'm working on this right now. Should be small 0402 components, and there is reasonable space available on the board for this.

This should currently be included on the board by copying the existing LPF layout on the L-Band Rx (page 3/7) verbatim; whats between the L-Band AX5043 and the MAC24+ mixer, and placed between the 70cm SE output and the PA. I will get the actual values for these lumped components. I'll run it past Vigely to confirm the values.

NB: The signal flow direction is the opposite WRT the AX5043 however. Signal is out of the AX5043, and into the PA.

glennl commented 3 years ago

Andrew, Thanks for placing the circuit. We can close this issue once I determine the lumped component values.

glennl commented 3 years ago

The very same LPF is used for the UHF Tx between the AX5043 output and the PA input (436 MHz), and the L-Band's IF between the mixer output and the AX5043 input (452-462 MHz).

Checking the component values with a 5th order LC Chebychev LPF calculator, the values produce a filter of the following specs: Cutoff freq: 560 MHz Zo: 50 ohm Passband ripple: 1 dB

This has been confirmed with Vigely