Post Go back to editing

Control of external GaAs switches with ADAR1000

Category: Hardware
Product Number: ADAR1000

I'm considering to the use an external GaAs switch along with the ADAR1000 in order to have an external LNA and a PA. GaAs switches require two complementary voltages to control the signal path (PA to ANT or ANT to LNA). In my case, to set the TX path I need to set VA=0.5V and VB = -5V and to switch the RX path I need to reverse the voltages, i.e. VA=-5V and VB = 0.5V.

After carefully reading the ADAR1000 datasheet (page 38), I found that it provides two pins for driving an external GaAs switch:

TR_SW_POS = {0, 3.3} V

TR_SW_NEG = {-5, 0} V

TR_SW_POS and TR_SW_NEG are mutually exclusive (datasheet, page 42) so I can't have two wires, one with Vhigh and the other with Vlow, at the same time:

Can the ADAR1000 provide two simultaneous voltages as I mentioned above to drive a GaAs switch? If so, do you have an application example for the ADAR1000 with an external GaAs switch?



Trying to make the explanation a bit clearer...
[edited by: andresmmera at 12:52 PM (GMT -4) on 29 Jun 2023]
Parents
  • The ADAR100 TR_SW pins were not intended to drive GaAs switches, but rather drive silicon switches like the ADRF5024 or ADRF5026.  These switches only need a single control voltage.  The examples given would use the TR_SW_POS pin.

    Would a slicon switch have high enough linearity and compression to meet you application needs?  The switches listed as examples are operational through 44 GHz.

  • Yes, ADRF5026 is very well suited to our application, but we prefer to use a GaAs device we already have.

    We only have one polarization. As a workaround, I think I can get the other voltage signal we need from TR_POL [-5, 0] V. Then, I would just have to change the POL bit with SW_DRV_TR_STATE. Is there a problem with this?

    There is a slight degradation on the switch performance for lowering the drive voltage from 0.5V to 0V, but it's not so critical for our application.

Reply
  • Yes, ADRF5026 is very well suited to our application, but we prefer to use a GaAs device we already have.

    We only have one polarization. As a workaround, I think I can get the other voltage signal we need from TR_POL [-5, 0] V. Then, I would just have to change the POL bit with SW_DRV_TR_STATE. Is there a problem with this?

    There is a slight degradation on the switch performance for lowering the drive voltage from 0.5V to 0V, but it's not so critical for our application.

Children