Rf Power Amplifier

[ElCap]

hey
im designing a rfid system for a project at 915MHz
im getting 5.5dbm out of the transmitter and need approximately 35dbm out of the antenna to compensate for the free space loss and to reach the tag 2m away.

Therefore i basically need a 30db amplifier. the only thing is i cant find a 30db amp that is on an ic. there are 30db amps out there but they are physically too large for my use.

I have been told that i could put 2 amps in series in order to get my 30db gain. eg. 10db amp feeding a 20db amp.
the only thing is the output of the 10db amp has to be lower than the max input power of the 20db amp.

I have done some research into this but cant seem to find a good amp pair to use in my design.

Any help would be appreciated.
Rob

 

[VEBill]

When selecting amplifiers for the transmit path, don't just look at dB of gain. You need to switch your thinking to absolute power levels (for example: picowatts versus gigawatts). Gain is a consideration, but real world operates with absolute power in and absolute power out.

There's a big difference in the amplifier depending on the absolute power level. Two 30dB amplifiers: one an IC, the other fills a room.

You'll also have to consider licensing. I don't think that you're allowed to transmit X watts in the 915 MHz band.

 

[benta]

30 dBm is 1 W.
I don't think that you'll get away with that at 915 MHz legally.

Benta.

 

[Comcokid]

Benta has a good point. You can operate up to 30 dBm in the 915 MHz ISM band only under certain conditions. Any you have to be careful because the harmonics of the 915 MHz ISM band fall in some forbidden bands - your amplifer will have to be real clean to keep the harmonics acceptable.

A good ISM radio can get 1600 meters on only 15 dBm into a PCB trace antennas. I'm suprised that your RFID requires so much power for only 2 meters.

 

[ElCap]

well the transmitter puts out 5.5dBm (-24.5dB) and the our tag sensitivity is .5mW (-3dBm) and we have a path loss of approx 35dB. so we approx need about 30dB gain to reach the tag.

i see what ur saying bout the total output, ve1bll, i sorta wondered why that was on the datasheet lol!

as for the ism radio being very long range, that depends on the receiver sensitivity. i read that some have bout -110dBm. our tag has -3dBm.

we might need to look at a tag with a lower sensitivity. as i cannot find any amps that will give us a large enough output power.

are there any other ways to get around this problem?

 

[ElCap]

also we are using passive tags so it doesnt have a power source, so therefore needs power from the reader to work.

 

[VEBill]

'passive RFID tags' - a phrase that would trigger my alarm bells (except for ultra-short range, proximity tags).

 

[Comcokid]

I knew RFID tags were less sensitive - especially if passive. But -3dbm detection level! Yikes!

I think you would do better to have a acoustic piezo parasitic power generator and a fog horn as a excitation source to power your passive RFID. Or maybe one of those audio systems that they put in a crapped-out sedan that rattles from the subwoofer at the stoplight. But I jest.

Seriously, you need to make or buy a power amp.

To design with discrete bipolar or MOS devices almost always requires the proper knowledge and tools (a RF CAD simulation program and VNA with s-parameters). Or, if you're like me - working in a company that doesn't believe in capital equipment - you have to resort to easier-to-use RF MMICs, simple Excel spreadsheets, and trial-and-error with the careful prayer and crossed-fingers when you do go to a test lab for FCC testing. Of course, the target production costs of your device doesn't allow you to buy an off-the-shelf amp that meets your needs or to use a outside design consultant. Have I read between the lines of your original post correctly?

Two MMICs will allow you to get your 30 dBm power level. Finding a "pre-driver" device that has the right P1d and IP3 might be a problem, but remember, you can always add a attenuator pad, which also compensates for some inter-stage impedance matching sins.

Another thing to consider, is that the FCC allows 30 dBm in your transmitter is omnidirectional at the antenna (as I recall). Having a perfectly omni antenna is unlikely. So you will need to plan for maybe only -27 dBm at the last stage.

 

[VEBill]

If these passive RFID tags (the tags themselves) are off-the-shelf, then the OEM should provide all the end-to-end system information you require.

If you're inventing new passive RFID tags yourself, then you might have to back-up a bit. It's not an easy problem (tricky bit* = passive).

[* except for ultra-short range, proximity tags]

 

[ElCap]

oem - original equipment manufacturer?
well the company were buying the tags from is gonna give us a demo of their system. They sposedly get an EIRP of 36dBm, 4W, which is the power limit in aus for 915mhz. so ill find out what type of amp their using cos i cant seem to find a amp chip and we prob wont have enough time to design and build one either. hmmmm would it be hard to design one??

 

[bogeyman]

If the receiving tag is always in one place try using a parabolic dish antenna at the Tx to focus the power onto the tag, you will be able to use a lower Tx power to get the thing to work. Your existing antenna and one of those there satalite dishes may well do the trick.
PS Please don't put out harmonics in our band.
Check your RF output with a spectrum analiser to be sure that you are not generating out of band interference.

 

[europus2002]

Hi Elcap,
Is there a margin for employing directional antenna. Try to add gain upto 10 dB at the antenna so that u can swipe off one amp stage....Capacitively load the antenna if the antenna dimension gets bigger in proportion to yr wavelength....This will save u from employing one more stage of amplifier and hence the Noise figure...How does this sound.