Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Remove I2C Chip from production module

Status
Not open for further replies.

soumen21

Member level 5
Joined
May 7, 2011
Messages
81
Helped
3
Reputation
6
Reaction score
3
Trophy points
1,288
Activity points
1,801
In my design, I have a chip working with an I2C interface. The board is in production. Now I want to remove the I2C interfaced chip as the intended function of the chip is not required anymore.
I2C is working at 400KHz. If I remove the chip, the I2C_CLK will be unterminated, which means I2C_CLK trace can act as an antenna. I understand, that for the lower part of the RF range, noise is expected to travel along lines rather than radiate from the equipment. Does that mean I can avoid radiated emission test if I remove unmount the chip? What other test is recommended so that I can safely remove the chip from the board without respinning the board?

Regards
Soumen
 

Why wouldn’t the trace act as an antenna WITH the chip installed?
Why don’t you just disable the I2C CLOCK?
I2C clock is open-collector; if you remove the terminating resistor, there’s no current flow
 

I2C lines are not terminated in the interface, both master and slave drivers show high impedance in idle state. They are terminated with pull-up resistors, the effect at RF frequencies is however small.
 

Why wouldn’t the trace act as an antenna WITH the chip installed?
Why don’t you just disable the I2C CLOCK?
I2C clock is open-collector; if you remove the terminating resistor, there’s no current flow
--- Updated ---

Same I2C_CLK is used for another device as well. So I can't disable the clock.
 

Hi,
Now I want to remove the I2C interfaced chip
"Remove" means
* desolder them from boards, or
* just not assemble them on new boards

the I2C_CLK will be unterminated,
As FvM wrote "Terminated" is the wrong expression.
Do you mean the I2C typical "pull up resistor" is inside this chip? This is quite unusual.

While I2C (according specification) should be open_drain (open_collector) with external pull up ...I've seen many implementations where SCL is driven push pull. This prevents SCL from floating, but it also means that it's not able to run
* multi master mode
* clock stretching mode

Does that mean I can avoid radiated emission
"Radiated" emission by the SCL trace on your board?
Before you talked about "receiving" HF, now you talk about "transmitting" HF. Is this what you meant?

400kHz is not considered HF in the meaning of EMI/EMC. The edges usually are not that sharp.
On a PCB with solid GND plane (what should be essential when talking about EMI/EMC) I see no general hardware problem at all when one I2C slave chip is not assembled.

Klaus
 

Same I2C_CLK is used for another device as well. So I can't disable the clock.

I2C lines are not terminated in the interface, both master and slave drivers show high impedance in idle state. They are terminated with pull-up resistors, the effect at RF frequencies is however small.
Same I2C clock goes to one more device.
Hi,

"Remove" means
* desolder them from boards, or
* just not assemble them on new boards


As FvM wrote "Terminated" is the wrong expression.
Do you mean the I2C typical "pull up resistor" is inside this chip? This is quite unusual.

While I2C (according specification) should be open_drain (open_collector) with external pull up ...I've seen many implementations where SCL is driven push pull. This prevents SCL from floating, but it also means that it's not able to run
* multi master mode
* clock stretching mode


"Radiated" emission by the SCL trace on your board?
Before you talked about "receiving" HF, now you talk about "transmitting" HF. Is this what you meant?

400kHz is not considered HF in the meaning of EMI/EMC. The edges usually are not that sharp.
On a PCB with solid GND plane (what should be essential when talking about EMI/EMC) I see no general hardware problem at all when one I2C slave chip is not assembled.

Klaus

Remove me not mounting on new boards.

Below is the schematic. SCLK currently goes to three devices and I don't want to mount U103 in new production. modules.

1649578663553.png
 

Hi,

so the pull up is external (R135) and has nothing to do with U103

I see no problem at all..
I see no "termination" and thus no termination problem
I see no problem with the use of other devices
I see no problem regarding HF, EMI/EMC and radiated emmision test

The only thing I see now is the software problem. Will it work properly when U103 does not respond?

Klaus
 
Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top