RF Power Detector IC Usage

[yanir]

Hi, I'm interested in using an RF power detector IC to measure a RF signal source, such as this one: http://www.linear.com/product/LT5538.

My question is if the RF signal I want to measure is branching off into this part on the way to another destination, how can I avoid creating a stub (thereby degrading the signal)?

Thanks

[yanir]

Thanks! This is perfect. I even found many options in surface mount packages.

[Neganur]

There was recently an email from Analog Devices regarding this topic. Let me see if I can find the link.

Edit: try this

http://app.h.analog.com/e/es?s=1548954099&e=54641&elq=f9833675f99e473192791a91efff20de&elqaid=3731&elqat=1

[VK5RC]

What frequency/ies and power levels are you dealing with?
Some of the microwave PAs just use a coupling line, a diode/resistor and capacitor, and "Bobs your Uncle" as Dave would say!

[yanir]

What frequency/ies and power levels are you dealing with?
Some of the microwave PAs just use a coupling line, a diode/resistor and capacitor, and "Bobs your Uncle" as Dave would say!

Frequency is 915Mhz, power level -115db to -60db in the main use case. I want to measure two antenna input sources and select the stronger signal. There are occasions when the transmitting device will be very close to the receiving device, so the power level will be much higher, > 0db.

BTW these RF sources are bidirectional, from what I've been reading this should be an issue with a directional coupler, correct?

[Neganur]

I'm not sure I understand what you're saying.

Antenna input source? You mean, the antenna receives the signal you want to measure? Or is the same antenna used for transmission as well? Perhaps a sketch of what your system looks like would help.

[yanir]

I'm not sure I understand what you're saying.

Antenna input source? You mean, the antenna receives the signal you want to measure? Or is the same antenna used for transmission as well? Perhaps a sketch of what your system looks like would help.

Yes antenna is connected to a transceiver. See sketch attached for what I'd like to do.

[KJDS]

You just need a comparator with some hysteresis to compare the outputs of the detectors and switch the switch. However there are a number of things to be aware of. When transmitting, you'll need something to lock the switch in one position, and also make sure that the detectors don't suffer any damage and that they recover from overload in time. Don't rely on more than 10 dB of directivity from the couplers.

You'll struggle to get the detectors anywhere near sensitive enough -114dBm is thermal noise in 1MHz, far better to use two receivers and either select the best or combine the outputs.

[yanir]

You just need a comparator with some hysteresis to compare the outputs of the detectors and switch the switch.

That's what I was thinking of using.

However there are a number of things to be aware of. When transmitting, you'll need something to lock the switch in one position, and also make sure that the detectors don't suffer any damage and that they recover from overload in time. Don't rely on more than 10 dB of directivity from the couplers.

So, this is what I wasn't sure about, are you saying that during Tx the coupler will pass power out on the coupled port? I guess I was thinking that these are 'directional' and one way. Am I wrong?

You'll struggle to get the detectors anywhere near sensitive enough -114dBm is thermal noise in 1MHz, far better to use two receivers and either select the best or combine the outputs.

Hmm, well -114dbm is the worst case, my threshold can be set much higher for antenna selection.

[KJDS]

Directional couplers are directional, but they are not perfect. With a perfect antenna and a very good coupler design you might get 30dB of directivity, but realistically, don't rely on anything better than 10dB

[AF6LJ]

I see another problem with this.
If you are actually looking at the antenna there is no selectivity, near by strong signals adjacent to the frequency you are using will cause you to have false readings.
It is much better to have two receivers and a "voting system". This is a tried and true technique to do what you are trying to do. The receivers provide quality of signal information and those two QOS signals are compared.

[KJDS]

Nearby signals needn't be adjacent, a few hundred MHz, like a TV transmitter will also do the job, as will L band radar.

[AF6LJ]

Nearby signals needn't be adjacent, a few hundred MHz, like a TV transmitter will also do the job, as will L band radar.

Exactly my point.
The OP needs to take a second look at how he is going about solving the problem.

[yanir]

Nearby signals needn't be adjacent, a few hundred MHz, like a TV transmitter will also do the job, as will L band radar.

Exactly my point.
The OP needs to take a second look at how he is going about solving the problem.

Fair enough, RF is not my main profession, but I tend to be doing more and more of it these days, which is fun. I'm using an off the shelf module so I'm trying to do things on the Antenna side as that is what I have control over.

Are you suggesting two complete modules (transceivers) or two receivers (this makes more sense, but I'm not designing the whole system).

[AF6LJ]

Nearby signals needn't be adjacent, a few hundred MHz, like a TV transmitter will also do the job, as will L band radar.

Exactly my point.
The OP needs to take a second look at how he is going about solving the problem.

Fair enough, RF is not my main profession, but I tend to be doing more and more of it these days, which is fun. I'm using an off the shelf module so I'm trying to do things on the Antenna side as that is what I have control over.

Are you suggesting two complete modules (transceivers) or two receivers (this makes more sense, but I'm not designing the whole system).

In a voting receiver system two or more receivers are used.
Their outputs are fed to a steering network and each receiver provides a signal quality voltage.
In the old days this was AGC voltage in the case of an AM receiver, or the last limiter stage in the case of an FM receiver.


Now if your receiver is switching between two antennas at the same location you can adapt this principal for use on a system with two directional antennas and one receiver, but you still need that quality of signal voltage.

[VK5RC]

Is the 'nearby' transmitter the other antenna on 915MHz or are there other RF sources nearby in frequency or distance?
If nearby (but not exact) in frequency, cavity filters can be helpful but bulky.

[AF6LJ]

Is the 'nearby' transmitter the other antenna on 915MHz or are there other RF sources nearby in frequency or distance?
If nearby (but not exact) in frequency, cavity filters can be helpful but bulky.

That would work, there is a rub however...
Here in the US There is a lot of Part 15 items on 900 MHZ Part 15 is the regulations that govern all that license free consumer stuff. The 900 MHZ band is rather populated here in the US.

[yanir]

Is the 'nearby' transmitter the other antenna on 915MHz or are there other RF sources nearby in frequency or distance?
If nearby (but not exact) in frequency, cavity filters can be helpful but bulky.

Both Antennas are fed to the module (through a switch), and are both intended to be used for Rx and Tx based on the best voted signal strength.

I'm starting to think that AF6LJ is right and that I need to rethink the architecture of this design.

[AF6LJ]

It is worth rethinking.
I don't know if the modules are your design or if they are off the shelf products.
Ether way; if you can get access to the receiver may find a voltage that is perceptional to the incoming signal strength. Once you have that you are most of the way there, some logic to select the best antenna, a little logic to keep the last selected antenna during transmit, and look for the best antenna when the signal drops below a set value.

Assuming that you are receiving a signal from a moving source, you are going to want to switch between antennas when the system is idle so you don't miss a transmission.

[VK5RC]

If the Transceiver Module is not easily accessed esp the separate Tx and Rx paths I am thinking along the same lines as AF6LJ; the 'logic' sniffing which Antenna has the 'best' strength, then setting the antenna select relay and perhaps another relay to switch out the Rx signal strength sensing unit during Tx.

A 'public' frequency increases the likelihood of interfering signals and might cause some false switching 'onto' a false but stronger than yours signal. Some modest signal encryption, perhaps using spread spectrum and phase shift keying may help e.g. Joe Taylors WSJT ?MAP65 type algorithm. I am not sure if your transceiver modules may do something similar to this.

Do you know how wide the Tx/Rx band pass is?
What is your data rate?
i.e How close to the Shannon S/N limits are you?
Just a few semi random thoughts.

[yanir]

If the Transceiver Module is not easily accessed esp the separate Tx and Rx paths I am thinking along the same lines as AF6LJ; the 'logic' sniffing which Antenna has the 'best' strength, then setting the antenna select relay and perhaps another relay to switch out the Rx signal strength sensing unit during Tx.

Yes, this seems to be the only way to do this with my module.

A 'public' frequency increases the likelihood of interfering signals and might cause some false switching 'onto' a false but stronger than yours signal. Some modest signal encryption, perhaps using spread spectrum and phase shift keying may help e.g. Joe Taylors WSJT ?MAP65 type algorithm. I am not sure if your transceiver modules may do something similar to this.

The module does support encryption.

Do you know how wide the Tx/Rx band pass is?

No

What is your data rate?

115.2 Kb

i.e How close to the Shannon S/N limits are you?

Not sure, but I think it's generally good.

Sorry my knowledge of RF system design at this point is more dangerous than proficient. I'm solid at implementing modules, full chip solutions, and antenna integration, but getting deeper than that is a learning curve. One that I'm thoroughly enjoying BTW. I think I'll spend some more time researching and work on side projects. I'd love to do a from scratch transceiver. Anyone have a good project suggestion that would teach the fundamentals of RF system design?

[AF6LJ]

As far as good RF projects go; This is a hard question to answer, because only you know what your needs are. I can suggest look for amateur radio related projects because there is just about everything under the sun in terms of projects to look at, analyze, and use as a jumping off point to expand your own knowledge. I learned quite a bit by reverse engineering gear but that was 40 years ago and at least two technology generations ago.
The Smith Chart will become your best friend, if it isn't already