Questions about TCAN1051

Hi ttd,

"How is the number of maximum node?"

In CAN networks, the practical limit on number of nodes is often a result of the wiring topology used (length/number of stub nodes, parasitic capacitive loading, etc.). However, the theoretical maximum value in the absence of cabling concerns can be calculated based on the transceiver specifications. The TCAN1051-Q1 has a minimum differential input resistance of 30 kOhm and its output driver is specified to drive load impedances down to 50 Ohms. Therefore, up to 100 TCAN1051-Q1 devices can be placed on a bus that implements dual 120-Ohm termination while still operating to ISO 11898-2 requirements.

"What the meaning of the voltage of VBUS and VCM?"

These are similar terms, but VBUS refers to the voltage at either CANH or CANL while VCM refers to the average voltage between CANH and CANL.

"How is the current value(mA) to drive CAN bus line?"

Typically the differential output of a driver is a little more than 2 V and the load impedance of the CAN bus is 60 Ohms. Therefore, the typical current sourced by the driver is about 33 mA. The minimum/maximum values can be found similarly based on the output voltage specifications in the datasheet.

"Does TCAN1051 have hotswap function?"

Yes, the bus I/Os are high impedance until fully powered and do not produce glitches at power-up.

"When TCAN1051 is separated from CAN bus during power-on, is it possible to break transceiver?"

I would not expect this to damage the device. Damage should not occur as long as none of the absolute maximum ratings are exceeded.

"What the meaning of the spec of 'Ios(ss_DOM)' "

This is the current flowing into either CANH or CANL when the TXD input is low (dominant) and a fault condition is present that results in a short-circuit between these pins and a power supply voltage. The shorting voltage ranges are given in the "test conditions" column for the spec.

"Does it mean short between CANL and GND?"

This case is covered by the spec, although the actual current when CANL is shorted to ground would be close to 0 mA. (This is because CANL is a pull-down driver that would try to pull the signal to ground.) You would get higher currents when shorting CANH to ground or when shorting CANL to a larger positive voltage.

"What the meaning of the spec of 'Ios(ss_REC)' "

This is the same as Ios(ss_DOM) but measured in the recessive state (TXD input held high). The currents are much smaller in this state since it is not strongly driven by the CAN driver.

"I'd like to know the propagation delay from TXD to RXD(dominant)."

This is the tPROP(LOOP1) parameter.

"I'd like to know the propagation delay from TXD to RXD(recessive)."

This is the tPROP(LOOP2) parameter.

"About the question of 「7.2 ESD Raring」, does this spec shows without external parts like R and C ?"

For most specs, the value is with no components installed on the CAN bus and the IC stressed via an ESD strike applied via a short PCB trace. One exception is the SAE J2962-2 ESD testing; this was evaluated using a test module with an additional 100 pF capacitance from each CAN line to ground.

"Especially, does "±4000V of IEC61000-4-4" mean the test of supplying the voltage to pin of CANH and CANL?"

No, in this case the +/- 4 kV pulses were applied to a capacitive clamp (in accordance with the IEC 61000-4-4 standard for electrical fast transient testing) which then coupled the noise onto the wires used for CANH and CANL signals. This is a similar to test to RF noise immunity; you can read more about it in this blog: e2e.ti.com/.../what-is-an-eft.

Please let us know if you have any additional questions.

Regards,
Max

ttd,

Yes, your calculation is correct. Note, though, that to calculate the maximum possible current you must also use the maximum voltage. In this case that would be 3 V. A voltage of 3 V across a 50-Ohm load would correspond to a current of 60 mA.

We have tested this device against IEC 61000-4-6 requirements (JIS C 61000-4-6 is the Japanese standard based on this) and saw it pass without requiring additional components. You can find the results in this thread:

e2e.ti.com/.../1955364

The device was not designed to withstand JIS C 61000-4-5 surge pulses, though. It may be able to pass the 1 kV level without external components, but I would recommend evaluating this to know for sure. You may want to design a PCB to support the additional protection components so that they can be unpopulated if proven to not be needed.

For a reference on different protection components that can be used with TCAN1042-Q1 to give higher levels of IEC 61000-4-5 protection, you can review this reference design:

www.ti.com/.../TIDA-00629

Max

There are a couple of parameters that relate to this.  One is the UVVCC threshold, which allows for the device to behave in a predictable way (described in the datasheet as "protected mode") when not fully powered.  This is further described in Section 9.3.3 of the datasheet.  Another important feature is that the leakage current into the device is very low when in this protected mode, meaning that the unpowered transceiver will not load the CAN bus.  This is indicated by the Ilkg(OFF) specification and is described in further detail in Section 9.3.4.

Please let me know if this isn't clear or if you have other questions.

Regards,
Max

Hello ttd,

That is a good catch, and I believe it may be an error in our documentation. The test results on page 21 indicate a failure at 2 kV, and so the maximum passing level reported in the summary on page 13 should read 1 kV. Sorry for the confusion.

Only pins 1-3 of SM712 were used in this case so that the same PCB could accommodate other TVS diodes in a single (rather than dual) configuration if desired. Two are used per net on the schematic in order to give flexibility in placement (i.e., before or after the common mode choke), not so that two could be populated simultaneously.

Regards,
Max