Internet Explorer is not a supported browser for TI.com. For the best experience, please use a different browser.
Video Player is loading.
Current Time 0:00
Duration 28:33
Loaded: 0.58%
Stream Type LIVE
Remaining Time 28:33
 
1x
  • Chapters
  • descriptions off, selected
  • en (Main), selected

So the products that we'll cover today are the transceiver interface portfolio from TI, which include devices for industrial communication and other standards, such as I2C. Here we see a few different major categories of these devices. We'll cover each of these portfolios individually. But at a high level, we're representing devices like RS-485, 422, and 232 transceivers, CAN and LIN transceivers, which are often used in automotive, but frequently used in industrial, as well, devices like I2C translators, buffers, muxes, and switches. And we'll also briefly talk about IO-Link transceivers.

The first major category that we'll talk about are the RS-485 products. We break the 485 portfolio into several major categories here, each one with individual products. So there's a set of standard transceivers, which would implement just the 485 physical layer function without a lot of extra features on. These are sort of nice standard general purpose devices. Many applications require a little bit broader protection. And so for those, we have a portfolio of ESD, EFT, or surge protected devices. And we'll talk about a few of these hero products listed a little bit later on in the call. But these integrate a level of transient immunity into the device, intended to be able to withstand relatively harsh environments, like you may have in some industrial applications.

Another category of devices would be the bus fault protected devices. These are ones that support a wide voltage range. These can be useful in applications where the 485 transceiver may be subject to things like short circuits on its interface lines and need to tolerate relatively high voltages without incurring any damage. We also have a category with sort of special feature devices. These are ones that do the standard 485 functionality, but may have certain special features on them in addition.

An example of that would be automatic polarity correction, which would allow for miswiring, where the A and B or Y and Z wires are unintentionally swapped. Devices with that correction feature would be able to correct for that automatically. There are also devices similarly that would have an invert function to do a similar function, but manually, devices with equalization, devices with support for lower logic level, and devices with higher output swing for long reach applications.

And then, we won't touch on this much today, but TI does also have a portfolio of RS-422 transceivers. These tend to be quad channel driver and receiver devices. So this shows at a high level some of our newer devices and how they differ from one another in terms of key features. One of the major differentiators among the 485 transceivers are the supply voltages used. So we see a set of devices here with a five as the second digit of the part number. These support a five volt VCC only.

The devices with a four in the second digit here support a range of voltages from three to 5.5 volts. And it's a common question whether or not these are interoperable. And yes, they are. It doesn't matter exactly what supply voltage is used, as long as the transceiver meets the differential signaling requirements of the RS-485 physical layer, which all of these do. So in applications that are five volt only, the left half of this portfolio would apply. But if an application needs to support 3.3, or if it needs to support a range, the right half might be a better fit. There is also some advantage to the wider range devices on the right side, even in five volt applications, since they tend to provide a little bit larger output swing, which can be useful for improving noise immunity in noisy environments or extending cable reach in cases where there's a large loss in the cable.

So within each of those portfolios, there's a range of different features or differentiators, major one being the data rate. The ones listed here are the maximum signaling rate that are supported. This is based on the slew rate of the output driver. Any of these devices could operate below this maximum limit, but it's generally best to choose a device that is not significantly faster than required for an application for signal integrity and EMI reasons. So that's why we keep a portfolio of devices at various speed grades to get full use of the slew rate limiting of the output drivers.

There's also various levels of ESD protection, as you can see here-- up to 18kv protection against IEC 61000-4 contact discharge. A lot of these devices also have EFT protection built in for the IEC 61000-4-4 standard. If you're not familiar with this, this is a test where large noise pulses are coupled capacitively onto a communication line. And the transceivers need to withstand this and, in many cases, continue to pass data correctly. So that's what we mean by this EFT feature.

Beyond this, there are just a few devices that implement, as well, IEC surge protection, which is an even higher energy stress compared to ESD. So the THVD1419 and the 1429 device listed here feature up to two and a half kilovolts of IEC surge. This is because they have integrated in a fairly robust transient voltage suppression diode. All these parts shown here are half duplex, meaning they send and receive data over a common pair of wires. We also have a similar family for full duplex.

I won't spend a lot of time talking over these, because these are based on the same technology in the half duplex family. But just know if applications require full duplex communication, the same benefits that these newer devices provide can be achieved in full duplex, as well. All of these devices for half duplex and full duplex are a pin for pin compatible with the standards, say, SN 75176 transceiver footprint. In many cases, smaller package options are available, as well.

There are a lot of 485 transceivers in the industry and also from TI, but use of these newer ones for new designs in a lot of cases makes sense, just because 45 has been around for a long time. But these represent a newer generation of technology. So we're able to implement better performance, higher degrees of protection, and things like that in the new device compared to the older legacy portfolio that you may already be familiar with. But beyond that, these can be considered drop in replacements for many existing transceivers.

So I'll speak a little bit more specifically about a few of these devices. This shows a little bit more detail about the THVD24xx family. This specifically falls into that high voltage fault protection category of devices I mentioned earlier. It is able to withstand plus or minus 70 volts on the signaling lines without incurring any damage to the transceiver. So in the presence of a high voltage short, the communication, of course, won't be working, because you have a short circuit on the lines. But once the short is removed, there won't be permanent damage to the transceiver that would then require a resoldering or repair of a unit.

So for field installed units, or for things that are costly to replace, or are particularly susceptible to these sort of higher voltage conditions, a device like this is-- it's easier to use and can make the implementation a little bit simpler. It also saves, in many cases, some external components that are often used for this sort of protection, like relatively larger TVS diodes in conjunction with series fused devices like PTCs. This device is part of the family that supports the wide voltage range, so from 3.3 to five volts. And five volts, it has a larger amplitude, since the amplitude scales with the supply voltage, which means it's able to meet PROFIBUS requirements, as well. So for applications requiring a PROFIBUS transceiver, this device could be used.

This is another device that would fit under that category of surge protected. You can see in the block diagram below the integrated TVS diode. This is sized to be able to handle IEC surges up to two and a half kilovolts when coupling onto the balance line. So for things like outdoor applications or things that could be subject to a very large transient stresses rather than DC stresses, a device like this could be used and hopefully save some external components that are used for voltage clamping.

We have several resources available for RS-485. We're regularly putting out content in terms of application notes, blogs, videos, and reference designs. You can see an overview of them on this slide. And also feel free to browse ti.com/rs485, where you can find an overview of all the products, as well as an index of technical content. And of course, feel free to post on our engineer to engineer forum.

I can now briefly talk a little bit about RS-232 products. This is a also fairly broad category of devices broken into several different categories. But I think it may be simpler to look at it as sort of a portfolio view. Here for the 232 devices, there's a little bit further differentiation, because the channel configuration may vary from application to application. So here, you see different columns for different Tx and Rx channel counts, whether it be a 1/1, 2/2, 3/5, or whatever else is needed for the application. Typically, it would depend on whether the additional handshaking lines are being used for flow control in the application, or if just simple Tx and Rx functionality is needed to drive the transceiver selection here.

The vast majority of the devices mentioned here integrate a charge pump to be able to take a 3/3 or 5 volt voltage and generate the higher voltage bipolar supplies that are needed to drive the RS-232 signaling. And so they don't need external larger voltage supplies. And many of these devices are able to work at either 3/3 or 5 volts, as well. One device, in particular, is capable of operating at 1.8 volts, or even down to 1.65 volts. That's the TRS3122E device. This has kind of a unique charged tippler design, which is able to still generate RS-232 output voltage levels, even with the very small VCC level.

So for applications requiring 1.8 volt IO, and especially ones that wouldn't necessarily have an additional 3.3 or five volt rail available in the system, this can be a device that can help simplify the power implementation, not requiring any additional rails just for the transceiver portion. Beyond that, there's some other differentiators. Like the 485 portfolio, since these would communicate to the external world via ports on a system, they are subject to ESD stresses. We have several devices that have 8kv contact IEC ESD protection, as well as 15kv air discharge protection. You can see those in the second to last row on this page.

And in general, any 232 device that has an E suffix on it would have this level of ESD protection-- the level four IEC ESD protection. The E is for ESD. Similar to the 485, there are sort of standard pin outs accepted in the industry for these devices. So these tend to be pin for pin compatible with other similarly numbered devices that you may know of. And in many cases, small package options are available.

Some devices we saw on the previous slide feature various low power modes. This gives a quick summary of those. Many devices are able to support just a manual power down, but we also offer automatic power down features, as well, such as auto power down or auto power down plus. The difference between these-- auto power down senses when an invalid voltage level is present for a while, and then shuts off the device to save power. Auto power down plus takes that a little bit further and shuts off the device when it notices that the bus is idle or that there's no transitions present on it for some time. So these are both sort of automatic power saving mechanisms.

Like with 485, there's several tools available for RS-232. These can also be found on our website at ti.com/rs232. Here's an overview of the I2C portfolio. It encompasses several different types of devices. So there are level shifters, buffers, and hub devices. These are things that are intended to solve signal integrity challenges, either by just adapting voltage levels or by rebuffering or redriving signals in order to accommodate larger load capacitances on the bus, or perhaps more nodes. And in some cases, for hub devices, they could help with fan out function, as well.

We also have IO expanders, which don't directly modify the I2C signaling, but they would have an I2C port that would allow for configuration of a bank of GPIOs. So in an applications where a microprocessor may be GPIO limited and it would need an additional bank of four, eight, 16 or 24 IOs, an IO expander could be used. And with just a two wire I2C connection, you'd be able to get this bank of IOs. And then, there's also a family of switches and muxes. These are useful for communicating from one bus to many, whether it be two, four, or eight, or more than that if you want to use multiple devices in parallel.

These can be useful in applications where you may have high capacitance that you need to segment by only enabling certain paths at certain times, or it could be used in applications where there are many slave devices that all share a common address, and to avoid address conflicts you need to shut down the path to certain slaves at certain times. So to see kind of a portfolio overview of the level shifters, buffers, and hubs, you see this view with various sort of differentiators spiked out. The main things that you might want to look at on something like this are what supply voltage ranges are supported on both sides and also the type of output that's used-- whether it uses a static voltage offset, which is common in many buffers, or whether it's something like a dynamic offset.

Those each have their trade offs with respect to how robust the direction control may be versus whether or not they can easily meet the requirements of the other target devices, whether they be master, or slave, I2C devices on the bus. For GPIO expanders, mostly depends on the number of GPIOs needed, which you can see there's a range of different counts supported, from four to 24-- as mentioned before, in various packages sizes that would obviously scale with the IO count.

All of these, in terms of the I2C performance, support kind of a standard I2C interface up to 400 kilohertz, or in a few applications up to 1,000 kilohertz. And you may also consider things like the address used in terms of how many devices could be shared on a bus, as well as whether a reset or interrupt output is needed. Interrupt outputs, in general, are fairly useful so that you can sense, in the case of an input, whenever an edge transition is detected. That way the microprocessor could then go poll the input port register of the IO expander and find out what has changed.

And this shows the I2C switch and multiplexer portfolio. These would be differentiated primarily by the number of channels supported. It should be noted, as well, that these generally can support a level translating function as well, just by using different pull up voltages on different sides of the switch devices from the main bus to each sort of component slave bus. Level translation is supported. And similarly, these have things like interrupt outputs and reset inputs as needed.

So I'll speak a little bit more about new devices in development. The TCA9511A is a bus buffer that is hot swappable. So for applications where you may have, say, live insertion from a back plane card into an active back plane and you want to avoid any sort of data corruption, a device like this could be used. The way it achieves that is by implementing a pre-charged circuit that brings the sort of connector side SCL, SCA, SDA lines up to a voltage before the connection is made. And that way you don't see a dip on the bus when this card is inserted that might look like a glitch to the other active devices on it.

This device does also buffer the signals. So if you have a large capacitive load on either side of the device, it's able to re-drive the signals through that and features rise time accelerators as well, which are a feature that help to drive that additional capacitance without necessarily requiring a very low external pull up resistance, which would require the other target devices to sync larger DC currents when pulling the line low. Larger external resistances can be used to lower current, and the load to high edges can be driven by the rise time accelerators.

Similar device in this family would be the TCA4307. It's also a hot swappable buffer. This device features stuck plus recovery as well, though. And what that means is that it senses when there may be a condition on the I2C bus that holds it in a stuck state. And it's able to automatically recover the devices that are stuck by generating clock pulses on the SCL lines to advance each slave device through its I2C state machine to the point where it releases the bus. So this is nice, because it doesn't require extra software intervention or special firmware concerns. It's a simple hardware solution that's sort of transparent to the higher layers.

As with the other products, you can, of course, find more information on ti.com under ti.com/i2c. You can see a few popular application notes and trainings here. I'll mention, as well, on the ti.com/i2c page, we do have a new design tool that's launched we call I2C Designer. This is a tool that allows you to input the specs of your I2C master and slave devices-- a number of different ones. Things like the input high and low thresholds, the output levels. And it will recommend an 12C bus topology made up of switches or buffer devices that would ensure compatibility between all the inputs that were provided.

So you can find that on to ti.com/I2C. I'll talk a little bit about that the CAN devices. Similar to RS-485, these are devices that have existed for a long period of time. TI has a long history in these. But in many cases, the newer devices feature a lot of performance improvements over the older ones. And so for new devices, it may make sense to consider newer devices in place of the pin for pin legacy equivalents that may be in use today.

We have the portfolio broken out here in two major categories. These are transceivers that are powered via 3.3 volts, or powered via five volts. A lot of people aren't familiar with 3.3 volt CAN and that it exists, but it is useful in many applications where a five volt round may be generated only for use of the CAN transceiver. So these 3.3 volt devices are able to generate a large enough differential signal to meet the CAN standard requirements. And so they are interoperable with five volt transceivers, but without requiring the five volt rail.

So you can see a couple of different main categories of devices. There are these older SN65HVD23x devices, as well as these newer TCAN33X devices. In most cases, the TCAN33X devices will have some advantages in terms of internal ESD protection. These also come in very small package sizes. We have SOT packaging available on these. And they're also able to support the higher data rates required for CAN-FD. Some applications that would require very wide voltage range support, though, may still want to consider the HVD233, 234, or 235 devices, which feature minus to plus 36 volt tolerance on the CAN bus pins.

We also have a broad portfolio of five volt powered CAN devices. Now even though these are powered by a five volt PCC, in many cases 3.3 volt logic can be supported. Look for devices that had a VIO or a VRXD supply rail. And those devices will have the level shifting integrated into them. So for these, I would call attention to the right side-- the TCAN family of devices-- the 1042 or the 1051. These are newer devices from TI that are designed to support a very wide bus fault voltage range-- plus or minus 70 volts-- as well as a high level of integrated ESD tolerance at 15kv.

These also support up to two or five megabits to support CAN-FD requirements, and are fully compliant with that latest ISO CAN physical layer in terms of the FD timing requirements. This again talks about the TCAN33x family of devices. I covered most of these features earlier, so I won't spend a lot of time on this slide. Similarly for the 1042 and 1051 device. These are the latest and greatest devices in the five volt range, but they do support 3.3 volt logic, as well.

The difference between these two part numbers are just the low power mode supported. So you can see that on pin eight, 1042 supports a standby mode, in which both the transceiver-- or both the transmitter and receiver circuits are powered down, and a low power bus monitor is monitoring the bus for activity that can then be signaled to the microcontroller via RXD pin. The TCAN1051 is a little bit simpler in implementing just a silent mode, which simply disables the transmitter while keeping the receiver fully active. So it can be thought of as a listen only mode.

ti.com/can is a great resource for not only the products and parametric search to find the right products, but also several application notes, as well as things like reference designs, training videos, and other collateral. This gives an index of some of our most popular CAN application notes, covering some basic topics in CAN usage, as well as a few device specific things. You can reference these links, as well as ti.com/can.

And the last category of devices I'll talk about, and fairly briefly, are the IO link transceivers. IO link is maybe not as pervasive a standard as things like 45, 232, and CAN. So it you may not be fully familiar with it. It's intended for use in digital sensor applications where the digital output sensors may have an output stage designed to drive things like a push pull output at a high voltage-- say zero to 24 volts-- or have something like a PNP or NPN output stage that would pull high or pull low to those same rails.

But in many cases, that's a uni-directional interface. And so TI does have devices that would drive that high voltage output, whether it be a push pull, NPN, or PNP. But the IO link standard allows for that same connection and same sort of physical layer signaling to be used as a bi-directional signaling path that would communicate not just the sensor data from the sensor to a master device like a PLC, but also data in the other direction-- things like, say, configuration data to a sensor, calibration, things like that. So IO link has a physical layer, defines a mechanism for changing the direction of data communication over this line and toggling between sort of the simple IO pull high, pull low sort of mode to IO link communication, which is more of a data format in defined frames, rather than just a binary output.

So TI does have devices targeting more the sensor side to be able to output those voltages and also receive the binary signaling from the master device. These same devices could be used in master applications, as well, with some external circuitry. You can find some reference designs showing that on ti.com. These devices could also be used outside of IO link standard applications, though, for any sort of industrial communications where high voltage single ended signaling is needed. Which can be common, whether it be something like an open drain push, pull or open source type output, where maybe very high speed communication or differential communication is not needed, but something that's fairly robust that can send a high or low voltage is useful.

In those cases, the TIOL111 or TIOS101 devices could be used, whether it be an IO link application or not. And the benefits of these newer devices are in terms of thermal performance, the-- even at higher output loads-- so these are in the few hundred milliamp ranges. They don't have a very large voltage drop, so the power dissipation within the transceiver is limited. And they also have, in a very small package-- two and a half by three millimeters-- a high level of integrated EMC protection in terms of the voltage ranges supported, as well as the transient voltage protection.

So in many cases, this allows for things like transient voltage suppression diodes to not be used in the system or for applications with very high sort of surge immunity requirements, maybe smaller TVSs with larger clamping voltages could be used without exceeding the rating of the transceiver. So similar to the other categories of devices, you can look at ti.com/iolink for a variety of products, as well as application notes and training videos. And always feel free to post on the E2E forum at e23.ti.com. Our team monitors this on a daily basis and responds back same day to all requests.