Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

transmit/receive switch help

Status
Not open for further replies.

prototypingfool

Electrical
Nov 3, 2009
23
US
I prototyping a GPS LNA circuit that shares and antenna with a transmitter (1 watt). I'm trying to find a switch that will protect the LNA when the transmitter is on becuase the isolation is poor and i'm afraid the LNA will be blown.

If i had steady voltage i could simply use a switch and control voltage but don't have that luxury. The only voltage i can use is when the LNA is active (3.3V) for a set time. After the DC shuts off the transmit port goes active.

Is there a SPDT T/R switch available using only one source of voltage to do the switching? Something that doesnt require a control voltage? Thanks in advance!!

 
Replies continue below

Recommended for you

Best choice is to use the 3.3V and buffer it with a transistor to switch some device at the devices required voltage and current. Also when you transmit, the switch, rather than leaving the GPS input open (disconnected) needs to terminate it to a 50 Ohm load, or short it to ground so less leakage signal gets into it.

Now for the OTHER QUESTION! What kind of antenna do you have, and what kind of application where you would want to transmit into the same antenna you use for a GPS! Especially at 1 Watt! Most antennas are narrowband, and a GPS antenna is no exception. GPS needs a omnidirectional antenna and good noise figure which translates into a narrow band antenna, and usually a LNA located close to the antenna. I can't think of another band located close to GPS frequency that you would want to transmit into at that power.

The antennas I am aware of that have both GPS and something else, like cellular, or 2.4GHz ISM band, or 915/868 MHz ISM band, etc are actually two antennas in one plastic housing, and have two coax leads.
 
Comcokid, so simply place an RF-type transistor in line that will make the connection to the antenna, and when there is no voltage, the transistor (switch) will close??

Thx,

F
 
Proto
No, not exactly what I meant. You need an RF switch, and one capable of the RF power level you are using as well as a great deal of RF isolation between the ports. The RF switch may be a IC, relay, or module.

Most RF switches require 5 volts, 12 volts, 28 volts (typical commercial or mil) or +/-5 volts to operate. Since your LNA is controlled by 3.3 volts, a transistor or other buffer is going to be required to drive the switch.

Switches for RF need to be constant impedance - usually 50 Ohms.

Now, RF switches don't always have the best isolation. This is a problem when the switch is used to connect an antenna between a high power source and a lower power source - like the TX/RF of a transceiver. 1 watt is +30 dBm. A GPS signal at an antenna is less than -100 dBm - more like -130 dBm. The signal level have a difference of 130 to 150 dB. To the non RF person, these numbers may not mean much, but dB is a logrithmic unit of measure, so this is 13 to 15 orders of magnitude difference. A typical RF IC switch may provide 30 to 60 dB of isolation, so you may need an additional switch on the LNA port that connects or terminates the LNA into 50 Ohms when the transmitter is active (called an absorptive switch). Some electromechanical RF switches, or RF switch modules have these various features in a single unit.
 
Understand..thx, but trying to find a switch with one voltage line (and no control) is the problem. Appreciate the help.

f
 
A T/R switch only needs to provide sufficient isolation to prevent damage to the receiver input, plus some reasonable safety margin.

The T/R switch certainly doesn't need to keep the transmitter leakage into the receiver at the same level as typical received levels. That would (normally) be unnecessary.

One gotcha might be if the receiver's AGC (if it has one) time constant took too much time to recover compared to the transmitter's repeat cycle. The transmit leakage is likely to desense receiver (in addition to the time that the receiver is disconnected) and the overall timing needs to be considered.

Not to mention interrupting the GPS cold start acquisition process - adding receiver drop outs may make a long process even longer.

There are other techniques like using a limiter on the receiver input, but that bring its own issues.

A good question hasn't yet be answered. Why are you transmitting into a GPS antenna? Considering the (typically) RHCP, gain pattern, and frequency band of a GPS antenna - the application isn't immediately obvious.
 
VB,

I have a single puck antenna with two ports (prototype). Isolation between both pins are terrible (-5dB) as expected. I'd like to protect the LNA
when in transmit mode. They never operate simultaneously. Catch is i only have DC when LNA is on.

Some thoughts i have is: When LNA is off, and 30dBm is transmitted into antenna will poor isolation blow input to the LNA that is off? Also, if isolation is low, will the LNA actually be affected when receiving? So only though is a SPDT or DPDT where can rectify the TX RF (with a directional coupler) to DC for some control voltage during Transmit, and use the 3.3VDC to control the switch during receive. Dont know if anything is the right approach. Any thoughts?

See jpeg.
 
+30dBm - 5dB = +25dBm = 4 volts RMS = 5.6 volts peak. This may easily vary if the assumptions (like 50 ohms) are off.

That's likely to be sufficient to damage the LNA.

And even if it's not, there's still the issue that the LNA front end might (will probably) go non-linear and generate spurious products that would not meet legal requirements.

Now - why are you transmitting 1 watt roughly on GPS band using a RHCP with the main beam aimed directly up into space?

 
thanks, not transmitting to gps. ;) Was a hypothetical
again, thanks!. Any way to private message here?

f


 
If you have two antenna ports and the transmit signal is not in-band of the GPS, then a simple solution is to put a passive filter on the GPS port. GPS SAW filters would be one option, or it could be a simple lowpass or highpass filter.

What is the frequency(s) of the transmitter, and what is maximum no damage limit for the LNA?

Peter
 
Comcokid,

Where can I find info (diagrams, publications) on this advice of yours:

"Also when you transmit, the switch, rather than leaving the GPS input open (disconnected) needs to terminate it to a 50 Ohm load (...)."

Many thanks


 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Top