An Overview of RS-232 and RS-485 Serial Communcations

Hensen Wong

MODERATOR:

Hello and welcome to today’s National Semiconductor online seminar. I’m Michelle Miller. I will be your host.

Before we begin, I’d like to go over the operation of your seminar interface. Slides will appear in the upper right section of your interface. If you’d like the slides to be larger, click the “Enlarge Slide” button. Slides will automatically advance. At the bottom of your interface is an interactive Web browser set to a Web page containing additional resources for this seminar. Questions may be submitted at any time and the presenter will address them as time permits.

To ask a question, click the “Ask a Question” button on the left which brings up an entry window. Enter your question and then click the “Submit” to send the question to the presenter. This seminar will be available in our archives shortly.

Finally, National Semiconductor owns and is responsible for all the content in this seminar. Today’s topic is “An Overview of RS-232 and RS-485 Serial Communications.” Today’s seminar will be given by applications engineer Hensen Wong of the Standard Analog Products Group.

Welcome, Hensen.

WONG:

Thank you, Michelle. Hi, everyone. I hope everybody’s having a good day at work today.

Here’s our agenda for today. First I want to discuss some of today’s top new serial interface standards. Second, we’re going to discuss about the benefits, disadvantages of RS-232 and RS-485 communications. Then we’re going to discuss key RS-232 and RS-485 applications. And lastly, we’re going to discuss about the different ESD standards and how to protect interface products against ESD.

Here’s a list of some of the popular serial interface standards. First at the top of the slide here, RS-232 is an asynchronous data transfer. It’s also a point-to-point communications. That means it’s only two devices on the Bus and the maximum cable length is about 50 to 100 feet. The maximum data rate with the fastest transceivers right now in the market are about 1 megabit. The next standard is USB; stands for Universal Serial Bus. It’s also asynchronous data transfer, but this standard can have quite a few more devices on Bus to handle up to 127 transceivers on a Bus and the cable length is about 15 feet maximum. If you’re using a USB 1.0 the fastest data rate can go is about 12 megabits, but the newer standard, which is USB 2.0 can go up to 480 megabits, and that’s quite significantly faster because with that rate of increases rivals the firewire, which I’ll talk about later.

The third standard we have is RS-422. It’s asynchronous data transfer, too, and in this particular standard what we have is called multi drop. One driver can be in a Bus with multiple receivers and the maximum receivers on the Bus is about ten. Cable length is significantly farther. It can transfer data up to 4,000 feet, which is good. Compared to metric it’s about 1.2 kilometers, and the data rate is about 10 megabits.

The next standard is RS-485, which is very similar to RS-422, except for some key advantages I’ll discuss later. And one of the key advantages here is that it can transmit up to 256 transceivers can sit on a Bus, and the data – the farthest distance is about 4,000 feet on cable and right now the fastest transceivers right now are running about 50 megabits.

And lastly is firewire, also known as 1394. This standard is a little bit different. It’s an isochronous data transfer. It’s different from the previous four standards. I’ll highlight the difference between isochronous with asynchronous later in my presentation. Firewire can have 63 devices on a Bus and openly transfer data up to about 15 feet of cable. If you look at the maximum data rate, 1394a can transfer 400 megabits or the b version can go up to 800 megabits right now.

Let’s go over RS-232. RS-232 is originally designed to have data communication between two devices. It’s not intended for multi protocol or multi point type applications. That’s reserved for the RS-422 and 485. Initially it’s intended for a short distance communications. So as I say, it’s only up to about 50 to 100 feet, and all that depends on data rates and type of cable you’re doing. RS-232 is also what you call unbalanced load. What this means is signal travels along in one cable as reference to ground. One of the key advantages of this is low cost. You know how you take one cable to send data through. But it also, there’s disadvantage to that, too. The disadvantage is less noise immunity because of that, because of the reference to ground. So if ground is shifting, you could have noises that are invalid data.

Next standard is USB. It’s a low-cost solution and it’s very popular now with the consumer in the market for PC peripherals. One of the key advantages is called hot pluggable. Devices can be connected and it’s all (inaudible) detected as all connected there. So there’s not much setup, so it’s good for consumer applications. They can support up to 127 USB devices on the Bus and like I talked about earlier, this 1.0 about 12 megabits and USB 2.0 is 480 megabits.

What’s neat about the USB protocol is that it supplies power to peripherals. With RS-232 you have to provide your own power. USB can actually provide up to 2.5 watts of power to its peripherals. So a lot of things you can power up through the USB port. But that power is also dependent on the USB port itself. So you have to look at the USB port, the hub, see how much power it can supply the devices.

Next I’m going to talk about RS-485 standards. It’s a multi point, what we call multi point because multi transceivers can sit on Bus. But this is dependent on how many devices can sit on a Bus. It’s dependent on the key parameter, the receiver input impedance. What RS-485 defines is the term called unit load. One unit load means that you handle 32 transceivers on a Bus, and if you look on most the data sheets, the RS-485 device, one unit load device has a receiver input impedance of 12 kiliohms and they can handle 32 transceivers. When they talk about a quarter unit load, the receiver input impedance is four times, so basically it’s 48 kiliohms and because of that, you can handle 128 transceivers in a Bus.

And there’s another one they call, one-eighth unit load. You guess from that it’s eight times, right? So eight times 12 is 96 kiliohms on the receiver input and it can handle 256 transceivers on a Bus. This is important specs because if you look at it, the driver’s point of view is seeing all the transceivers in parallel to itself, so if you have a high impedance, you put in parallel, the impedance will drop. So if you have a high impedance, you can see that it’s necessary to have a high impedance if you put a lot of transceivers in a Bus.

Next we have an RS-422 standard. It’s very similar to RS-485, but there are key differences that differentiates the two. In most cases you can use 485 transceivers on a 422 standard, but you cannot do it the other way around. You can’t use a 422 in a 485 application. I’ll highlight why this is so. Because I know a lot of people get confused about this because a lot of times when you look at a lot of data sheets, they are RS-485 and 422 compatible, and yet we see some 422-specific transceivers, they only usually list 485 because there’s a reason they’re not compatible.

So one of the key differences is that a RS-422 is what we call multi-drop. It’s only one driver can be on a Bus and remember on RS-485, you have multiple drivers that can be on the same Bus. Another key difference is the driver output voltage limits. And RS-422 is only used for -0.25 V to +6 V, whereas RS-485, the range is from –7 V to +12 V. So just right away you know, the output swings are different. The output signal levels are different, too. On RS-422 it’s +/- 2 volts, which is RS-485 is +/- 1.5 volts. Another key difference is the driver load impedance is different. On a 422 is only 100 ohms, which is the RS-485 is 54 ohms. Another difference is the input voltage range for receiver. On a 422 is +/- 7 volts, which is the RS-485 is minus 7 to 12 volts. And you see, here’s another difference here that you should definitely look at. It’s the receiver input impedance on a 422 is only 4 kiliohms, whereas the RS-485 is at least 12 kiliohms to have a one unit load. So these are key differences why you can’t use a 422 transceiver in a 485 application.

Next I’m going to talk about some – one of the new standards is people who use Apple computers probably know as firewire, but in the PC world just about everybody refers to it as 1394. They’re both the same standard. It’s based on Apple’s developed technology and was adopted to industry about 1995 under name IEEE 1394. It’s a popular product in like DVDs, camcorders, portable disk drives, MP3 players that people play music like the Apple iPOD.  Also it’s popular in high-quality audio-video applications. One of the key advantages of firewire or 1394 is what we call isochronous data transfer. What it means is it’s guaranteed real time delivery of data.  However under other four standards it’s asynchronous. Data sent in packets let’s say is broke up in pieces and is reassembled at the other end. So it takes times to reassemble the packages in order to have data presented. So in video or audio applications, you can’t do that. So what 1394 does is real time. Data is transferred continuously. It’s almost like talk on the phone lines. It’s a good analogy is that there’s continuous feeding of data. There’s no dropage of data. So it’s real good for audio-video applications. You know, like when you’re watching a movie, you want the movie to stop for a few seconds to reload and continue on. So that’s why it’s a good standard for other applications, but a 1394 is specifically designed for audio-video applications. And what they call this is isochronous data transfer.

One of the other key advantages of 1394 has its own power to – it’s versatile. It can power up to 45 watts of power to its peripherals. With USB can only do 2.5 watts. Another advantage is that it’s a plug and play format. It is hot pluggable.  You don’t need to shut down PCs to add devices. You can just plug and the PC recognizes it and it will turn it on.

One of the key advantages it can also connect 63 devices on a Bus. This offers like peer-to-peer type technology, which you’re probably very familiar with. Multiple PCs can share the same say firewire camera so you can view images.

Next standard is RS-232. As I talked about earlier, it’s a point-to-point type application and only connect two devices. You can’t connect multiple devices. And we talked about it’s single-ended on an unbalanced load. Data sent on one line as referenced to ground as you see on the photograph there. It’s mostly used in like PC-type applications. It’s basically the serial port on a PC. Next we will highlight some of the key parameter RS-232.  For example, the driver load output voltage is the limit is use about five volts. The driver open circuit voltage is about greater than 25 volts, and the driver can only supply drive current about 100 milliamps. So significantly less than other standards. But one of the standards to also highlight is the driver, what we call slew rate. It’s how fast it transitions from low to high is the slew rate. Basically, the maximum slew rate  is 30 volts per microsecond. The driver receiver output impedance is 300 ohms, and the driver input impedance is about 3 to 7 kiliohms and the receiver input limits is +/- 12V to +25V. So the set is really designed many years ago in the bipolar technology. In the CMOS center we don’t really swing that – those high swings, but it was originally designed like 40 years ago, which you can understand by looking at numbers, they have pretty high voltage  swings.

And last aspect to talk about is receiver threshold. The minimum is +/- 3 volts swing. It recognizes the high-low. Some of the key advantages of RS-232, like I said, it’s ubiquitous. It’s everywhere. You know, every PC has a serial port and it’s very simple to do the communication. It only takes three wires to do a two-way communication: basically a driver, a receive and the last wire is of course ground. And that’s also a very cost-effective solution. RS-232 transceivers are really cheap in the market and it’s everywhere. That’s why it’s used widely.

Next I’m going to talk about some of the typical applications for RS-232. The number one application is basically PCs and laptops. It’s also used in point of sale equipment, PDAs, GPSs, PC peripherals like printer, faxes, mouses, cell phones, and also network equipment like hubs, bridges, routers. The number one reason why RS-232 is very popular is because it’s everywhere. Consumers are slowly migrating to USB but RS-232 will be still be strong in commercial and industrial sector. So as the consumer market goes away from RS-232 to USB, there’s still a strong need for RS-232 in commercial, industrial sectors. The reason why is that for a lot of those sectors, they don’t need the fast data rate of USB and RS-232 is quite adequate and they can see equipment out there right now that a lot of people are not willing to change over yet. So in the long term I think commercial-industrial will be migrating to USB, but it’s going to take time – years. So for now there’s still a lot of need for RS-232 for commercial and industrial sector.

Here I’m going to talk about some of the pin assignments for PC serial port. This is based on a DB9 connector. The three necessary signals for two-way communication for RS-232 are basically pin 2, receiver data; pin 3, transmit data, and of course ground. The remaining are optional control signals intended for communication showing the ready status of the device. For example, ring indicator which is pin 9 forms the presence of a carrier signal on a phone line. The DTR is the Data Terminal Ready, and DSR  is the Data Set Ready, which provide information concerning the readiness of the communication channel for the RTS, which is called the Request To Send, and CTS, Clear To Send, signals provide information about whether a device is ready to receive a signal.

Here we have RS-232 handshaking. If you look at the top of the slide here, the first one basically no handshaking required, basically, the transmit is connected to receive, and transmit is connected and ground. So that’s basically simple RS-232 communication with no handshaking. The second one below there is the full handshaking with all the control signals connected in its proper order for full handshaking communication between two devices.

Some of the requirements for RS-232 in the PC market is for most of the serial port on the PC is a 3 x 5 transceiver. What I mean by that is it’s three drivers, five transceiver. And one of the requirements is usually at least 120 to about 250 kilobits per second data rate. For laptops usually it’s a 3V supply type device, three to five volts, requires a low quiescent current to maximize battery life, and desktops, on the other hand, basically looking for multi supply voltage transceiver, usually five volts for the receiver and +/- 12 volts for the driver.

The ESD requirements for the PC market usually are like 15kV ESD for the human body model and IEC standard. I’ll highlight later, what the ESD standards means.

Of course, the last requirement is low cost. You know, if you’ve gone to your electronic store recently, you look at, you get a PC for a couple a hundred dollars. So there’s reasons. There’s the cost for these parts are fairly cheap nowadays and that’s why the PCs you can get for a couple a hundred dollars.

Some of the RS-232 devices that National offers for PC market for desktops are DS14185 which is a bipolar multi-voltage transceiver. That means it’s five volts supply and a +/- 12 V supply. It’s a 3 x 5 transceiver and the maximum data rate is about 300 kbps.

The next we have is a DS14C335. This is a CMOS technology based device. It has an internal charge pump. That means you can use a single supply voltage supply to the chip, so the charge pump actually supplies internal +/- 12 V for driver internally. What it needs is some what we call charge pump capacitors to create  the +/- 12 V and the size of the capacitor is about 0.47 uF.

Next we have the DS14C335. The DS14C535 is also a CMOS device. This is a 5V device.   It has a charge pump and the charge pump capacitor of 0.1 uF and a data rate of about 200 kbps.

Next I’d like to talk about the RS-232 in what we call POS markets, point of sale. You guys probably use it one time or another without realizing it’s actually RS-232 device connected to it. But they’re usually connected on the bar code readers, receipt printers, the bar code label scanners. Usually connected to an RS-232 transceiver to the cash register.

One of the new devices that we have to offer in the POS market is our newest LMS202E. We’re excited about this device because this device has ESD protection that can handle up to 15kV. The human body as well as the EN-61000 standard, which is the same as IEC. You know, when the European union joined, they changed their standards. What used to be IEC now it’s called 61000 4-2. It’s basically a second source to Maxim’s MAX202 and again, what our target market for this device is the POS arena and it’s used mostly in the bar code scanner. The data rate for this one is 230 kbps. And this – a few things to also know about this device is that it has an internal charge pump, so you only need the five volt supply to power this device.

Next I’d like to talk about RS-232 and the networking market. Some of the key specs for the networking market is some of the transceivers they use 1 x 1 or 2 x 2, 3 x 2, 3 x 5, or 6 x 10. You say, what’s a 6 x 10? Basically what it is is actually two serial ports on the same chip. Also, some of the key specs is that it should at least 120 kbps and most of them are using your single supply with 3 volt/5 volt part or a multi voltage 5 in a +/- 12 volt, and a lot of the network market requires a 15 ESD rating on their part.

Next we have a list of basically some of the RS-232 parts we offer in the network market. You know, it’s 3 x 5s that we talked about in our previous one on the PC market, and we have the 5 x 3s, which are typically used as the peripherals to the PC, which is a complement to the 3 x 5. Usually used in modem, which is our DS14196 is ours. And if you take a look at those, 4 x 4, 4 x 5, and the 2 x 2s we’ll talk about.

Next I like to talk about RS-232 and PDAs and GPSs market. I know a lot of the new PDAs are migrating to the USB for serial port, but we still have some Legacies, PDAs are still using RS-232. And typically they’re using like 120kbps data rate and they’re mostly using single 3V supply voltage and a shutdown feature. They want to save battery life, so a lot of times they have shutdown features when they’re not using the serial port. Also they want to use the one that integrated ESD protection like 15kV ESD protection.

Next we have RS-232 in the cell phone market. Most of these are 2 x 2 or 3 x 2 transceivers, and one of the key things is that most of the time the chip is not actually built into the phone.  Actually, most of the transceivers are built into the cable. With the chip actually in the cable, just plug the cable to the phone which plugs into your PC.

Next we have the RS-485 standard. As I said, we talked about it’s a multi point, handles up to 32 units with a one-unit load equivalent and a fault that we called a balanced line. It’s differential signal. The signal is actually sent over to two cables, A and B.  So the difference between the two cables is actually the signal itself. And the fastest data rate right now out are about 50 megabits per second.

Well, some of the key advantages of RS-485 over 232 is 485 is a single supply voltage, whereas RS-232 you require a 5 V and +/- 12 V. If you don’t have a charge pump, then you need to actually supply external +/- 12 V, but if you have RS-232 transceiver with just single supply, you need external capacitors to generate the +/- 12 V. What RS-485 does, it just needs 5 V because it just swings over to 5 V at the output. One of the key advantages also is that it has ability to do multi drop. As I said earlier, we have 256 transceivers on a Bus, which RS-232 can only have two. And RS-485 can drive data a long distance. It can drive things 4,000 feet, which RS-232 is about 100 feet maximum. And the 485 is also very fast. You’re talking about at least 10 megabits for the slow ones. The fastest RS-485 transceivers out there right now are running about 50 megabits. And RS-232 right now, the fastest transceivers right now are 1 megabit. One other key advantage that most people using 485 is, the better noise immunity. Because of the differential signal, any noise is subjected externally to the cable is basically subjected to both cables. So when you take the difference of the two, noise is actually canceled out. So it’s better noise immunity. So that’s one of the key advantages why people are using 485. And I know a lot of people who are in the consumer arena might not be familiar with the RS-485 because most of this is used in the industrial and commercial applications. I’ll go into some of the applications where it’s being used later.

Some of the typical applications of RS-485 applications is point of sale equipment. You probably have used this many times at department stores without realizing it. You go into a Wal-Mart recently, those little price checkers that are kind of spread out throughout the store, you just bring your merchandise up there and just check the prices. That’s actually using a 485 transceiver in the box. When you scan the bar code, the information actually gets sent on a 485 transceiver through a long cable to a PC somewhere within the store. So this way not all the price checker needs to be uploaded with information. It basically goes to a remote PC that has all the pricing information and it sends the data back to you. And this way it’s easy for – one reason why they do this is that it makes it easy for the stores to update prices. They can easily update one PC, which will take care of all prices to the store. Plus a lot of times the data rates are so slow on these bar code readers because they’re only sending a few kilobits. So you’ll need to really have a high-end PC at each terminal to actually store prices. They just send it to remote PC and basically send it back.

Another of the typical applications is using factory automation. And another application is in network hubs, bridges, and routers, and another application where you use RS-485 is called in the heating, ventilation, air-conditioning systems. These are typically in most of the office buildings. Something that’s real popular in Asia right now is power meters. RS-485s being used in all the power meters in China.

First, I’m going to talk about RS-485 applications in PLC or in the factory automation. You’re looking at this diagram here. You have on the bottom of the diagram. There’s four PLCs monitoring a manufacturing process. PLC monitors and collects data and sends it via RS-485 by (lines in red) to a master station, which is right above there, which collects the data and stores it. There’s an Ethernet network after that that connects various PCs to it. So this way it allows factory workers to log in to view, to monitor, to process and to also make any necessary changes to the process during the manufacturing. So we see, this way you can move the factory worker away from the floor and actually control the process away from the machines itself.

Next I’m going to highlight some of the devices that we have introduced in this area for the factory automation. One of the devices is the LMS75ALS176 and another device is LMS75LBC176. Both of these are RS-485 transceivers. The second source is TI devices. One of the key matters that we do offer a little bit over TI’s is that we have a little higher ESD rating. We have a 2kV human body model, whereas the TI’s 1.2 kV we actually tested ourselves. On TI’s, it doesn’t even specify the ESD rating, we actually tested the devices. They pass about 1.2kV human body model. And the data rate’s about 35 megabits. As I said, this primary target most of the PLC arena.

Next we have an LMS1485, which is also a RS-485 transceiver. It’s a second source to ADI’s ADM1485. This is a much better ESD rating. It has an 8 kV human body model versus ADI’s 2kV.

Next I’d like to talk about the RS-485 application in heating, ventilation, air-conditioning unit. RS-485 is used in basically connecting the temp sensor or thermostat in the room to the master controller. If you’re sitting in the office right now, and you look to see your thermostat, there’s usually RS-485 transceivers connecting that from there to basically the control unit in the building. What it does is that’s after sensing the temperature in the room and yet when you’re making adjustment to the thermostat, it actually sends the data to the controller, which could be controlling that, the temperature, turning on and off the AC to heat or cool the room up. So that’s how RS-485’s being used.

One of the main reasons why they use this standard is because RS-485 is able to send data a long distance, you know, like I said earlier. RS-485 can send data 4,000 feet on cable. So in your typical office building, you’re probably running hundreds of feet if not thousands of feet just from one end of the building to the next. Also, its robustness.  It is a differential signal.  So remember, differential, any noise subjected to the cable is canceled out. So you can have a reliable, low-cost solution to a heating, ventilation control unit.

Next I’ll talk about profibus. Profibus is basically a standard that’s developed in Europe. It’s pretty popular in factory automation. It’s really originally started by Siemens. It’s well adopted in Europe. I think the U.S. is slowly grabbing onto the standard, but it’s taking a lot of time. One of the key advantages is it’s an open standard.

Look at the key specification on a Profibus. Basically, Profibus is an offshoot of RS-485. We will look at the standard

The key difference is they look at the driver output voltage.  For RS-485 is 1.2V.  For Profibus be compatible to 2.1V. Today there’s only a few RS-485 transceivers that are profibus compatible. to improve the drive capability, The driver ouput is the 2.1 V driver output instead of the 1.2 V output of RS-485. The rest of the standard is pretty much the same standards. Basically it’s just a derivative of RS-485.

Next I’m showing you how the RS-485 is used in the profibus application in the factory. You see right here as you look at the diagram, you have a – there’s RS-485 profibus which connects various sensors on the Bus which connects to the PLC. Much like the factory automation diagram I talked about earlier. You must use RS-485 Profibus compatible transceivers. These are being used mainly in the factories in Germany.  I think about 50 percent of the factories use the profibus type standard. In U.S. is learning, we’re only about 11 percent or so. So what I see, it’s still growing but it’s extend out for a few years and it’s widely adopted in Europe, but U.S. is slowly grasping on to this standard.

Here’s something you’re probably not familiar with. This is an RS-485 in the power meter applications. You see this is very popular in China. How this works is, the power meters you typically have in house, you monitor how much electricity usage. What they do is they connect the RS-485 transceivers to it, which can send data through the RS-485 network to a PC in the power central office, which could obtain data on your usage of electricity. You don’t need a meter person to come to your house and read the meters. So this is very popular in China I think because of the vast regions they have to cover, and they can’t afford everybody to go out there and read the meters for millions and millions of people. So they use RS-485 network to basically do the power reading. So this is also used in the gas metering business, too. One of the key devices we have for the power market is the LMS1487. This is a second source to Maxim’s 1487 and one of the key specs for the use of 1487, it can handle up to 128 transceivers on a Bus. So they aren’t concerned about the data rates because they’re only driven by kilobits of data, just reading the meter on it. The power meter, don’t need to read every millisecond so you really don’t need that high data rate transceiver. What they are looking for, the key is actually how many transceivers can they hook up on the Bus. So, 128 means you connect 128 homes to that particular meter. This is a key application in power meter markets.

Next I’m going to talk about ESD. I know many of us think ESD it’s almost like black magic sometimes. But you look at the photograph. When you walk across the carpet and you touch a device and you zap it and you see what happens. On the bottom right-hand corner of the device you can see actually flash aluminum. That’s what it means is right there is that ESD charges actually gone through the I/O pin and it’s heated up the aluminum trace on the die and actually melted it.  When they melted it, of course, you will shorten the life together and the device will fail. Another thing that often happens is that if it doesn’t flash the metal, can actually blow a hole through the oxide, and short traces. So, you see, it’s harmless to us. When you walk across the carpet and touch the doorknob and it’s – you get a little shock. Oh, but to IC, it’s almost death to them because they’re subjecting to high stress. If they don’t fail, they might be critically injured. They might be just work to surviving for a few more months, but eventually it will die.

On the next slide, I’m going to talk about some of the charge voltages generated by doing some simple things. Like, walking across a vinyl floor, you say, oh, how many volts can you generate through that? But let’s say in California, walking across a simple vinyl floor you can generate up to 12kV. Walking across the carpet and  touch the doorknob, you can discharging 35kV at the doorknob. Of course, there’s not much, if you’re subjected to that much current for long, voltage with that current, you’d probably die, but you’re talking about milliseconds of discharge. So this way, you just get a quick shock and it’s over. So you look down a list. You can generate thousand and thousands of volts just by doing some simple tasks.

Next I’m going to talk about ESD testing methods. I know there are a lot out there and there’s a lot of confusion out there. What’s difference to human body model to machine model, charge device and the IEC standard.  IEC 1000 4-2 also known as ENC61000 4-2. Next we are going to highlight what these methods are and what do they mean. Human body model is simpliest.  It’s originally defined as the mill standard 883C, but it basically simulates a human discharging on IC. So like I said, this is the model of you walking across the carpet and touching the doorknob.  In the same way you walk across and touching the IC.

The next model we have is called machine model. This standard was originally developed in Japan. It’s popular in Japan and the automotive industry. What this says is it simulates the charge of a large object discharging on an IC. So let’s say for instance you have an automated pick and place machine placing IC’s onto board for assembly. If your machine is not properly grounded and your machine’s actually building up charges through movement.  So if it’s charged up and it goes, pick up the IC, it will discharge on the IC and actually damage it. So that’s what the machine model simulates.

The next model we have is called a charged device model. This is actually opposite. This device is charged – the IC is charged up already and when you place it onto the board, it discharges. The charge is already on IC. The charge is looking for path to ground. And when you place it onto the board, it discharges. So this happened because let’s say improper handling device. Let’s say the device is in these rails and they’re not ESD protected rails that device is holding, they slide back and forth and they build up charge. There’s no way for it to discharge so that actually by shaking device in those tubes, you’re actually charging them up. What happens when you pick them up and place them on the board, it discharges and damaged the device.

The last standard I’m going to talk about is same as IEC1000 4-2. This is basically a European standard. The European Union required that ESD immunity testing for all electronic products. And conditions for the ENC marking before shipping to the members in the country, which is European Union, you have to meet this EN61000 standard. Originally wasn’t really an IC standard. It’s basically a system requirement. It means when you ship an end product, it’s supposed to meet the standard. But a lot of the OEMs have mandated IC manufacturers to actually meet the standard. So what their belief is that if you get the IC to meet the standard, the hope is with that they can have a higher probability of passing the system standard.

Next, a little more detail on human body model. If you look there, this is basically how it works. The model is basically discharging 100 pF capacitor through a 1k ohm resistor. If you look at the amps versus the time curve, the rise time of this is basically 10 nanoseconds. So you’re basically from zero to the maximum in about 10 nanoseconds. And the peak current is about 1.33 A, about 2 kV and inabout 10 nanoseconds.

Next one, we’ll look at the machine model. Machine model is very similar to this, but there’s a slight change in the model. If you look at it, it’s still discharging 100 picofarads cap, but the resistor is zero. So because of the no resistor, you know the current will be much higher. So if you look at the machine model, the peak current is about 7 amps versus the 1.33 amps human body model. So this a much more severe test because you’re talking about the current, which is much higher than the human body model. And you also look at the curve below there. I have it over different inductances. The inductance  includes not only the bondwires in a chip, but also the traces that the chip’s connected to also plays a major factor in what the chip actually sees. With a 0.5 uH, you get close to about 7 A of current. You look at that little curve below that, which is inductance of 2.5 uH, you notice that the current has dropped significantly. It’s only maybe about 4 amps or so. So there is this quite big drop in current based on the inductance alone.

Next I’m going to talk about the charged device model. Look at this. This is – if you look at the curve here, the rise time of this is much faster. Time of this is only a .5 nanosecond. From the previous two models we have like 10 nanoseconds to get up to the peak. But this one goes in about half a nanosecond. So a lot of times, the ESD structure may not even turn on on time. You see, a lot of times we have this ESD protection and why are we still having failures? But sometimes the ESD event is so fast, your ESD protection doesn’t even have time to even turn on to protect the device. So by the time you detect it, the ESD event has already gone past. So that’s important way the charged device model is 0.5 nanoseconds is a real important test. And again, the peak current is 6A.

Next, I’m going to talk about the EN 61000 standard. Actually there’s actually two requirements within the ENC 61000 standard. (contact and air-gap discharge). I’ll talk a little more about it. You’re basically going through the same thing with basically 100 percent pF capacitor to a 330 ohm resistor, but the difference between the two is that on contact, the contact is basically if you use ESD gun, you actually place the ESD gun contact point on the particular pin and they zap it. The really advantage of that is that you get more of a repeatable test. Because it’s not based on your environment.  If you have an air-gap between the ESD gun and your chip, depending on what type of relative humidity and temperature,the results may vary. The contact test is really more consistent. You can pretty much get a consistent time after time, it’s repeatable-type test. The disadvantage of that is that it’s not really realistic. Most times when you touch a device, it would actually discharge before you actually touch it. So that’s not really realistic test, but it’s basically a more repeatable test.

The next test they have is called air-gap test. Air-gap discharge mimics more closely to where it happens real world.  When the charged object gets close to a device that’s not charged, it will arc across through air and actually zap the device. You actually have a small air-gap between the ESD gun and the IEC pin and you zap it. So to meet the IC standard, air-gap, you have to be at 15kV, the contact at 8kV. Look at the curve there, the rise time is quite fast. You are talking about 0.7 to 1 nanosecond rise time to the peak.  That  it quite fast.  It’s a very stringent test. You look at the peak current, you’re talking about 37A of current. You’re just talking about only, you know, less than nanosecond. That’s why you don’t get killed when you get zapped with that. You just get a good shock.

Okay, next I’m going to talk about ESD protection on interface products. Here’s a schematic of an ESD protection on RS-232. You will notice there are Transient Voltage Suppressors, called TVS devices, placed on the cable. You will notice we placed both on the driver and the receiver pin because usually when you have an ESD events that’s coming from the outside world, it would be mostly subjected to I/O pins. The rest of the pins on ICs are well protected within the casing of the device. So just by this example alone you know why ESD is such a concern for interface products. Most other IC’s are well encased in some kind of casing which protects them from ESD events. But for the RS-232 and 485 or any interface products, it’s the bus pins that goes to the outside. By you touching it, maybe grabbing onto the pin you could be discharging onto the serial port.  This will cause voltage transients on the bus. This is the reason why we need ESD protection on devices. So if you don’t have the ESD protection onboard on the IC, most will have to put some external ESD protection on the I/O pins. So like RS-232 pins, not every pin is actually 15 kV. Basically where we’re only putting the ESD protection is only on the I/O pins(Bus pins that goes out to the connector). The other control pins will not be ESD protected.

Next we have a RS-485 application. You see that on this diagram. We have TVSs connected on the differential wire there. And that’s how we protect devices from ESD events.  The TVS’s will actually turn on and basically take the charges to ground in protecting the I/O from being blown out.

A couple of things I want to help you guys out with future designs and helpful. We have an RS-485 Helpful Hints page. These are some applications notes to help you design RS-485 applications. So there’s lots of good information there to help you design hardy RS-485 transceivers to withstand the “Real World”. We have ESD helpful hints. I just went through the Web and I found some good sites on ESD protection. I know by my experience and   visiting customers, I noticed that a lot of customers are not addressing ESD events that well. The guys are just thinking ESD properly. But ESD is a very important design criteria for  interface product. It may not so for other type products. Other devices are not subjected to ESD as much as for the interface products.  It’s much more severe environment because it has I/O pins are exposed to the  outside world.  Any ESD event can easily zap and kill the device. Companies like Protek Devices, Semtech, California Micro, they have some good applications how to protect interface products and I hope you guys go through it.

So that’s all I have today. I appreciate the time for staying with me. So I’d like to turn this over to Michelle.

MODERATOR:

Thank you, Hensen. Before we continue our question and answer portion, I’d like to invite our viewers to please fill out and submit the survey form which should be appearing on your screen shortly. Your answers to this survey will help us in the development of new products as well as future seminars.

Now if you have any additional questions on this topic, please enter them in the question form and submit. We will address as many as time permits. Questions may be paraphrased.

WONG:

Thank you, Michelle.

Here’s a question, good question. Is RS-232 protocol an obsolete topic?

That question is no because RS-232 you know has gone away from the consumer realm. It’s still very active in the industrial and commercial realm like we talked about earlier, and cash registers. So the answer is no.

Here’s another question. Are there any  3 or 5 volt RS-2 transceivers?

Yes, there is. We look back on the slides here, we have – we do have single supply voltage transceivers which is still in charge on like the DS14C335, DS14C535. There’s a DS14C238.  Most of the CMOS devices have the built-in charge pump. 

Next question: How much ESD protection is needed for a connector exposed to human contact?

Well, I would say 15kV.

Here’s next question: How much does TVS’s device capacitance slow down interface?

That’s a very good question. TVSs, they have many different type ones, so for like – for high-speed, something like RS-485, you’re looking for something with very low capacitance, something like a 2 pFs. They do have various TVSs for different applications. If you actually go into my ESD Helpful Hints slide there, you look at those Web sites, there will actually be type of application and the correct TVS for the application. They’ll recommend different type of devices for different interfaces. So if you’re equipped for high speed, looking for something for low, low capacitors, if you’re looking for something like RS-232, very slow data rate, capacitance doesn’t play  a major factor, you could probably use devices with large capacitance.

Here’s the next question. For RS-485 and 422, what are connectors used for these steps? For example, what are the physical connectors?

For RS-485 and 422 are electrical specifications. They don’t define the physical connections. So it’s up to you to actually define what you want the type of connector, but I think a lot of times a lot of guys use like an RJ-45 or something like that type of connectors.

So, I’m open for questions. We’re running out of questions here. You guys, please send more questions if you guys have any more. We have maybe one or two minutes left.

MODERATOR:

I guess that concludes our question and answer portion. Thank you so much, Hensen. Thank you, everyone, for joining us for the seminar, “An Overview of RS-232 and RS-485 Serial Communications” brought to you by National Semiconductor and Yahoo! Broadcast.

(End of Presentation.)