Hello there,

Here's the situation:
I have a project which contains 16 RFID antennas placed in a 4x4. We read these antennas using a single sm130.
So we decided to use 5 CD4052BE. 4 to choose the line and 1 to choose the column.
The CD4052B device is a differential 4-channel multiplexer having two binary control inputs, A and B, and an inhibit input. The two binary input signals select 1 of 4 pairs of channels to be turned on and connect the analogue inputs to the outputs.

The idea was to use the CD4052BE to choose which antenna to read, then to move onto the next one to sweep the 16 antennas.
When I tested it,I tested it using 4 antennas and a single CD4052BE, I set the inhibit input to 0 (so that the antennas are not off) then A and B to a 0 or a 1. However, no matter which of the 4 antennas I tried to read, it would not select a single one, I could set the A and B to read the value from the antenna 1 and still be able to read the value from the antenna 4 if I placed the RFID tag on it. Making it impossible for me to know which antenna would send back the RFID tag ID.

I was powering the CD4052BE with Vdd=+8V and Vee=-7V.
Each antenna has ANT1 and ANT2 outputs and 3 GND. Each ANT1 and ANT2 from each antenna are connected to the X0/Y0, X1/Y1, X2/Y2, X3/Y3 and the result of the selected channel using A and B would be sent to the sm130 then to the microcontroller.

Unfortunately, I don't have schematics to show you, however, I can tell you that the value sent by the antennas was correct, the problem is that I can't tell which antenna I was reading from.

Is there something that I am missing on how to use the CD4052BE ?

Thank you for any help you could provide,

Here's a datasheet about the cd4052b.
https://www.ti.com/lit/ds/symlink/cd4052b.pdf

 

It would sure help to see a schematic, but let's wing it...
What frequency is your RFID system operating at? 13 MHz might be pushing things a bit (a lot).

Are you aware of the on resistance of the switches? The resistance might be a substantial problem particularly when transmitting.

It might be better (or possible) if you have one RFID transceiver per antenna and then multiplex or scan the outputs of the receivers.

 

What is the antenna characteristic impedance?

 

The tags we use are inside a mold. It's 13.56Mhz tags. 13.56 MHz Mifare PCB Antenna - 55x55mm. And a 13.56Mhz SM130.
Using 1 SM130 per antenna would be expensive, so, unfortunately, we can't do that.
I don't know much about the antenna sorry, we just used what we already had and couldn't find documentation about it.
About the resistance of the switches, I wasn't aware of it. How is it a problem and how could I fix it?

 

I hope somebody can come up with a more creative solution, but in the mean time you might want to consider RF relays.

https://www.digikey.co.th/products/en/relays/high-frequency-rf-relays/963?k=rf relay

 

Have you considered using PIN diodes for switching ? Here is some information on the subject.

Edit. I have just noticed that PIN diodes are not suitable below 100 Mhz.
Les.

 

Hello there,

Here's the situation:
I have a project which contains 16 RFID antennas placed in a 4x4. We read these antennas using a single sm130.
So we decided to use 5 CD4052BE. 4 to choose the line and 1 to choose the column.
The CD4052B device is a differential 4-channel multiplexer having two binary control inputs, A and B, and an inhibit input. The two binary input signals select 1 of 4 pairs of channels to be turned on and connect the analogue inputs to the outputs.

Actually, two groups of five i/o pins. Each address connects two i/o pins in each group at a time. XCOM and YCOM are the "common" pins that connect to each of the other pins in their respective groups.

The idea was to use the CD4052BE to choose which antenna to read, then to move onto the next one to sweep the 16 antennas.
When I tested it,I tested it using 4 antennas and a single CD4052BE, I set the inhibit input to 0 (so that the antennas are not off) then A and B to a 0 or a 1. However, no matter which of the 4 antennas I tried to read, it would not select a single one, I could set the A and B to read the value from the antenna 1 and still be able to read the value from the antenna 4 if I placed the RFID tag on it. Making it impossible for me to know which antenna would send back the RFID tag ID.

Its operation is clearly shown in the truth table on the datasheet.

I was powering the CD4052BE with Vdd=+8V and Vee=-7V.
Each antenna has ANT1 and ANT2 outputs and 3 GND. Each ANT1 and ANT2 from each antenna are connected to the X0/Y0, X1/Y1, X2/Y2, X3/Y3 and the result of the selected channel using A and B would be sent to the sm130 then to the microcontroller.

Usually, the supply voltage needs to be high (VDD=15v, VSS=GND VEE=-15) for these device to operate at max frequency.

Unfortunately, I don't have schematics to show you, however, I can tell you that the value sent by the antennas was correct, the problem is that I can't tell which antenna I was reading from.

Is there something that I am missing on how to use the CD4052BE ?

I don't know how your setting the address inputs but unused inputs should be at ground or pulled down to grroud.

You can do a simple DC test by setting the address and checking continuity between the pins.

eT

 

I hope somebody can come up with a more creative solution, but in the mean time you might want to consider RF relays.

https://www.digikey.co.th/products/en/relays/high-frequency-rf-relays/963?k=rf relay

I'll look in that direction, but I'm not sure how to work with RF relays so I'll have to search it.

Have you considered using PIN diodes for switching ? Here is some information on the subject.

Edit. I have just noticed that PIN diodes are not suitable below 100 Mhz.
Les.

I have my microcontroller working at a clock speed of 120Mhz (because of some LEDs that needed it). Is that what you mean?

I don't know how your setting the address inputs but unused inputs should be at ground or pulled down to grroud.

You can do a simple DC test by setting the address and checking continuity between the pins.

Yes, I understood the truth table. A and B allows to control with Xi/Yi sends back it's value to Xcommon/Ycommon.
At first, I only tried to control A and B but soon noticed that if I didn't set the inhibit input it would work correctly. So I connected it like A and B to the microcontroller, and then with the microcontroller I just did output_low(PIN_NAME); or output_high(PIN_NAME); to either send a 0 or 1 signal.

As soon as I can (in 2 weeks) I'll try changing my supply voltage values. So, are you saying the supply voltage being too low could make the CD4052BE
switch on it's own?

 

Do not power 4052 from +15 -15 as someone suggested. You will destroy the part.
1. Are the 4052 control lines connected directly to the micro?
If not how are they connected?
2. What is the voltage on the micro?

 

Do not power 4052 from +15 -15 as someone suggested. You will destroy the part.
1. Are the 4052 control lines connected directly to the micro?
If not how are they connected?
2. What is the voltage on the micro?

The uC and the 4052 don't share the same supply voltage. The only lines directly connected to the micro are A, B, Inhibit input. All then the Xi/Yi are connected to the antennas and the Xcommon/Ycommon is connected to the SM130 which is itself connected to the uC in the I2C pins.

I can't tell you what supply voltage is on the uC itself right now (don't remember) because I use a test circuit board (which is on +8V).

 

The uC and the 4052 don't share the same supply voltage. The only lines directly connected to the micro are A, B, Inhibit input. All then the Xi/Yi are connected to the antennas and the Xcommon/Ycommon is connected to the SM130 which is itself connected to the uC in the I2C pins.

I can't tell you what supply voltage is on the uC itself right now (don't remember) because I use a test circuit board (which is on +8V).

 

4052 vdd = same voltage as micro

4052 vss = micro ground

4052 vee = -5 volts

 

Yes, I understood the truth table. A and B allows to control with Xi/Yi sends back it's value to Xcommon/Ycommon.
At first, I only tried to control A and B but soon noticed that if I didn't set the inhibit input it would work correctly. So I connected it like A and B to the microcontroller, and then with the microcontroller I just did output_low(PIN_NAME); or output_high(PIN_NAME); to either send a 0 or 1 signal.

The device is a analog switch, so whatever signal is present at XCOM appears at Xn, and YCOM at Yn, at the same time.

The INH pin needs to be pulled down to ground. This disables the INH pin and doesn't allow it to "float".
Connecting +V to the pin inhibit's the chip. Digital inputs should always be connected to either +V or ground. If your microcontroller has "weak" pull downs for the pins connecting to the A/B pins on the CD5052 they should be enabled. Otherwise, pull each of the A and B pins down to ground with a 10k resistor.

As soon as I can (in 2 weeks) I'll try changing my supply voltage values. So, are you saying the supply voltage being too low could make the CD4052BE
switch on it's own?

I'm saying if you allow pins A, B, or INH to "float", it could cause random circuit behavior (like switching on its own).
The higher supply voltage will allow the chip to operate at maximum frequency. You can see this in the datasheet.

You can see in the simulation below that both X and Y pins are active at the same time but independent I/O circuits. X is passing a DC signal, while Y is passing a sine wave signal.

eT