TCAN1044A-Q1: TCAN1044A-Q1 RX Error issue

Hi Kygo,

Can you provide a scope shot of the CANL and CANH wave forms when the glitch occurs? My guess is that the glitch occurs on the dominant to reccesive edge of the bit on the CAN bus. This is most likely due to ringing on the CAN network which occurs when you are communicating at this fast speed with un ideal loading conditions.

If my guess is correct then I would suggest upgrading to CAN SIC or lower the speed of your CAN network. We can discuss this more when you take those scope shots of the CAN bus.

Best,

Chris

Hi Chris

Please confirm that the attachment meets the requirements.

In addition, we have found that this RX Error will disappear after a period of time (about 2~4 minutes).

Product application, the CAN speed  cannot be reduced

CANFD RX Error measurement_20221103.pptx

Kygo,

Which one of those signals is supposed to be the CAN bus? I am pretty confident this is a problem due to ringing on the CAN bus. Have you tried switching the TCAN1044A for a TCAN1462? The TCAN1462 is SIC capable and will help when you are communicating at those speeds.

Best,

Chris

Hi Chris

The provided signal is the waveform of CAN_H, compare the difference between OK/NG.

Is TCAN1044A not recommended for high speed transmission?

So TCAN1462 was developed for signal improvement and replaced TCAN1044A?

Are TCAN1462 and TCAN1044A PIN to PIN? What is the price difference between the two?

In addition, we have used NXP TJA1042, and RX Error did not occur

Kygo,

Were you using the TJA1042 under the same speed, bus loading, and network configuration as the TCAN1044?

The TCAN1044A can be used for high speed communication, however, it is the physics of transmission cables that limits how fast a TCAN1044A can communicate. 

With CAN SIC what we do is actively driver the dominant to recessive edge so that we can reduce ringing on the bus. This allows us to communicate much more reliably at faster speeds. On your scope shot you show two signals:

I assume the top is TXD and the bottom is RXD correct? Can you probe the CANH pin and the RX pin during this same transmissions? I want to see what the CAN bus looks like.

Best,

Chris

Hi Chris

We are replacing TCAN1044 with TJA1042, other applications are the same.  

the waveforms of the scope are all CAN H, top is the waveform of OK, and bottom is the waveform of Error.

the main is to compare the difference between the two waveforms.

Kygo,

The point in the CAN frame that has a higher amplitude is showing multiple nodes transmitting at the same time. Since this is outside of the arbitration phase, what's likely happening here is that multiple transceivers are transmitting an error frame. The fact that a stuff error is getting reported indicates that there is likely a timing issue here. What is the sample point set to, and is it possible to give a bit more margin either earlier or later in the bit?

Regards,

Eric Hackett