ST25RU3993 Intermittent Read

[Jeff C]

I made a new board based on the EVAL board (removing single ended xmit).

My issue is even after tuning the radio, my tag detection performance is below the eval board (same parts/layout as far as I can tell, minus 2:1 mux for switching tx paths).

Is there any way I can hook up an oscilloscope to see if there's anything strange/clipping with my tx/rx signals?

Here's some symptoms:

Using the same antenna, at roughly 5m, a tag is consistently detected by the EVAL board.

Changing nothing but the UHF RFID Board + retuning, I can still read the same tag, but on the Windows GUI the tag disappears now and then unless I "shake" the tag.

Using a handheld vulcan rfid power meter, I see that my output is consistently lower, and animatedly sags and recovers then sags again. While on the eval board, power is pegged at 0dB max constantly (assuming same physical spot/distance from same antenna).

On my board, the vulcan rfid power mapper might be pegged at max, while an rfid tag immediately below it can't be read.

A tag might be read with relatively good RSSI, then suddenly after being moved 1cm futher, completely become unreadable (rather than a little worse RSSI) (Linear Q/I 40/255, 255/30, 251/183)

I notice with higher RxGain, that sometimes performance drops significantly if output level is not lowered. Am I clipping the Rx path?

During tuning (slow mode), the eval board will slam my power meter to 0dB and beyond and rattle the needle.

My radio board will simply have the needle pegged to the max, but no rattling.

Based on the 3.0 firmware... My board has a slightly different GPIO ordering, but I've already remapped the C #defines to ensure operation.

Clearly everything works in general, but it's not as stable/reliable as the eval board.

I put together a second board which functions roughly the same (a little better, but nothing to note). I pushed the PA voltage to 5v and yes it helps, but same tag disappearing act unless tag gets shaken physically.

Where should I start?

I have an oscilloscope and spectrum analyzer but have no idea what to look for.

[Bart Herse]

Hello Jeff C,

I made a new board based on the EVAL board (removing single ended xmit).

--> I am assuming that you have removed the internal PA output - right?

Is there any way I can hook up an oscilloscope to see if there's anything strange/clipping with my tx/rx signals?

...

Using a handheld vulcan rfid power meter, I see that my output is consistently lower, and animatedly sags and recovers then sags again. While on the eval board, power is pegged at 0dB max constantly (assuming same physical spot/distance from same antenna).

On my board, the vulcan rfid power mapper might be pegged at max, while an rfid tag immediately below it can't be read.

--> I suggest to connect the SA to the antenna port to check the frequency stability and the output power.

A tag might be read with relatively good RSSI, then suddenly after being moved 1cm futher, completely become unreadable (rather than a little worse RSSI) (Linear Q/I 40/255, 255/30, 251/183)

--> Configure the register 0x10 to 0x80. Connect an oscilloscope probe to OAD2 and check if you can detect a Miller encoded tag response. It often helps to connect a second osci probe to OAD to see the PIE encoded reader commands. This way it becomes easier to see which point in the communication is currently shown on the osci screen. If you see the reader commands but no tag responses then the tag has not enough energy to turn ON.

During tuning (slow mode), the eval board will slam my power meter to 0dB and beyond and rattle the needle.

My radio board will simply have the needle pegged to the max, but no rattling.

--> once the tuning is completed, what reflected power level is reported. How does it compare to the EVAL reader?

I put together a second board which functions roughly the same (a little better, but nothing to note). I pushed the PA voltage to 5v and yes it helps, but same tag disappearing act unless tag gets shaken physically.

--> when shaking the tag the field strength distribution around the tag changes and therefore the tag might find at some point enough energy to turn ON. Are you using frequency hopping, or are you transmitting on a single frequency?

Cheers,

B

[Jeff C]

Hi Bart,

Sorry I didn't realize you responded. Let me give you some feedback.

Removed Internal PA? Yes traces removed entirely and left disconnected at pads of ST25RU3993

SA to antenna port - This is where my knowledge is weak. How can I safely attach my SA to the antenna port without frying the SA? An antennuator? A DC Block? Not sure what values of these widgets to buy. Do you happen to know?

Thank you for the note regarding OAD/OAD2 for Tx/Rx from the Tag. I'll try it. I'm not familiar but I think you make it sound reasonable enough to check.

During tuning (slow mode), the eval board will slam my power meter to 0dB and beyond and rattle the needle.

My radio board will simply have the needle pegged to the max, but no rattling.

--> once the tuning is completed, what reflected power level is reported. How does it compare to the EVAL reader?

On this, even after tuning, my board is weaker than the EVAL reader. My board is weaker during tuning. It is also weaker overall in airspace after tuning. Not sure exactly why but that's where I'm at so far.

[Bart Herse]

Dear Jeff C,

SA to antenna port - This is where my knowledge is weak. How can I safely attach my SA to the antenna port without frying the SA? An attenuator? A DC Block? Not sure what values of these widgets to buy. Do you happen to know?

--> it is recommended to use a DC block to protect the SA input. An attenuator is typically needed as well as most SA have a max. input power of 30 dBm.

-->!!Check the max. input power of your device!!

-->I would recommend to use a 10 dB attenuation in series with the DC block.

-->Check the power rating of you attenuator as well - it should accommodate the dissipated power.

On this, even after tuning, my board is weaker than the EVAL reader. My board is weaker during tuning. It is also weaker overall in airspace after tuning. Not sure exactly why but that's where I'm at so far.

--> What is the reflected power on your reader after the carrier cancellation tuning is completed?

Cheers,

B

[Jeff C]

Hi Bart,

You are right about the 30 dBm max on my SA. For trivia's sake I see:

Maximum average power 30 dBm, 3 minutes, fc ≥10 MHz, att > 20 dBm, preamp off

Maximum damage level 33 dBm, fc ≥ 10 MHz, att > 20 dBm, preamp off

I see that 30dBm is 1 Watt, so a 10 dBm attenuator would bring us down to about 20dBm.

The TQP9107 is stated to have a 35.5dB gain. Yet the ST25RU3993 Eval is rated at 29dBm.

I read that the ST25RU3993 transmits at 0dBm in differential, yet with a 35.5dB gain, the ST25RU3993-EVAL only transmits 29dBm. Where did the other 6.5dBm go? I was thinking since 0dBm is 1mW, then it would allow me to use 35.5dB directly against 0dBm allowing for 35.5db gain. Perhaps it's because the Eval kit doesn't drive the Amplifier at 5v?

Anyways, I'm asking since I wanted to drive the TQP9107 at the maximum 5V rating and wanted to understand how many dBm I needed to attenuate.

Let me gather the reflected power graph/numbers from the eval software after CC Tuning. I misunderstood your previous request as to read what my handheld power meter showed, but you're asking for reflected power, so I'll need to run to my lab to figure that out.

Thank you Bart!

[Bart Herse]

Dear Jeff C,

"The TQP9107 is stated to have a 35.5dB gain. Yet the ST25RU3993 Eval is rated at 29dBm."

--> Please do not mix-up output power with gain.

--> Gain is a relative figure hence the unit is stated in dB while output power is referenced to an absolute entity, 1mW that is, hence dBm.

--> For PAs, if you review the datasheet, look for the output power and/or the 1 dB compression point.

--> These figures tell you at which point the output starts to saturate (where the linear region ends).

--> Any increase in the input power for the PA does no more translate to the same increase in PA output power.

--> In other words if you would decrease the the input power for the TQP9107 then you would probably see the 35.5 dB of gain.

--> You are correct in terms for the supply voltage that the ST25RU3993-EVAL reader provides to the PA - is is below 5V.

--> The ST25RU3993-HPEV reader on the other hand does provide 5V for the PA therefore the output power is higher.

--> Note that the ST25RU3993-EVAL reader is USB powered while the ST25RU3993-HPEV reader requires an additional DC power supply for the PA.

I hope this explanation was helpful to you.

Cheers,

B