THVD1406: THVD1406 output waveform with 120 ohm terminal resistor

Gary Mao said:

I would like to understand why the output voltage levels of the THVD1406 differ depending on whether a 120-ohm termination resistor is present or not.

      

So this device disables it's output driver after the t_device_auto-direction time occurs when the Dinput is high longer than that time. This is how the auto-direction feature is achieved. So in the cases where there is no termination resistors, the external pull up and pull down resistors will passively generate the VoD so the bus signal looks cleaner (wider margin). You want to have the termination resistors present when the bus is long or the communication is fast to prevent reflections. You're bus looks very slow so termination resistor may not be needed if this is the fastest and furthest distance for your application.

Now the second waveform on the right when the termination resistor is populated, when our device disable's its driver the VoD is generated by the external pull up and pull down resistors. And since the resistor values are much larger than the PFET and NFETs of the drivers, the VoD is much lower. When the DIN signal is a low, the device will actively drive the line but when the DIN is high, it will only drive the differential outputs for several microseconds then turn off and let the external resistors generate the VoD which is why you see the pulses on the A-B line go high. If you zoom into it, you will see its high for the t_device_auto-direction time.

Gary Mao said:

I have measured RS485 transceivers from other brands, and when a 120-ohm termination resistor is applied, both the high and low output voltage levels are significantly offset from the zero level, which seems ideal. How is this achieved? What is the main reason for the voltage level differences compared to the situation described in Question 1?

The other RS485 devices are actively driving the VOD (A-B) when DIN is high. Our device disables the driver after a set time to allow any other device on the bus to drive to avoid contention. Main reason is due to other RS485 devices driving the high logic and our device disabling. If you used THVD1400, you would see the same behavior as the other transceivers. This device (THVD1406) has the auto direction feature.

Gary Mao said:

If this difference is due to the choice of transceiver model, what features or specifications should be considered when selecting a transceiver?

Yes, this device specifically has a feature called auto-direction that disables the driver when DIN is high for t_device_auto-direction time. 

Theres several kinds of features RS485 devices can have. If you use a device like this, you ideally want the other transceivers on the bus to have something called internal fail safe protection which pushes the Vth+ negative to allow for more noise margin when the driver is disabled. Otherwise you may need to make the external fail safe resistors a lower value to get a larger VoD when the device has disabled it's driver. 

Gary Mao said:

In cases where the difference is indeed related to the transceiver model, how should a product designer decide between a transceiver that clamps the high-level voltage close to zero and one that maintains both high and low levels at a defined distance from zero?

Depends on noise margin. I recommend finding transceivers with fail safe protection included (Vth+ minimum is a negative value). If you expect the application to be used in an electrical noisy environment then I would recommend using lower resistance values on the Pull ups and pull downs on the A/B pins. Generally I recommend customers target atleast 200mV when the drivers are disabled because +/-200mV is the RS485 threshold values that all transceivers should be compliant to. The external pull up/pull down resistors should not exceed 375 ohms equivalent resistance per RS485 standard. 

The idle VoD can be estimated by eq:

Rpullup_eq+Rterm_eq+Rpulldown_eq = R_total

Vcc/R_total = I_bias

VoD_idle = Rterm_eq x I_bias

https://www.ti.com/lit/an/slyt324/slyt324.pdf?ts=1754344231731&ref_url=https%253A%252F%252Fwww.google.com%252F

This document walks through this in more detail with better defined equations but the one I provided gives a good enough estimate.

-Bobby

Hi, Bobby:

Thank you very much for your professional response — it has cleared up all of my concerns.

I am currently applying for samples of isolated RS485 transceivers without automatic direction control, specifically the ISO3086DWR and ISO3088DWR, for my project. I will proceed with testing and validation accordingly.