Four Steps to End Encoder Problems

you okay so welcome back to those of you you who have signed on with us early and those of you who have just joined us if we want to take this moment to welcome you all to this latest in the design world webinar series you are here today for the webinar entitled four steps to end encoder problems and our sponsor today and our presenter will be need ik ab tron automation and a couple of things before we get started here this webinar will be at the design world online.com website afterwards and everybody who is registered for the webinar will also get a copy of it in their email inboxes as well keep in mind that after the presentation concludes we will have a short Q&A session so at any time during the webinar if you think of a question for our presenter feel free to type it into the question box there and we will do our best to get to all the questions or as many of the questions as we can towards the end and also one more item here if you are on Twitter and if you would like to tweet about this webinar the hashtag for the webinar is there on your screen hashtag DW webinar so once again my name is miles Buddha mere I’ll be the moderator for today’s webinar I’m a I’m the senior editor with design world magazine covering motion control and electronics and test and measurement as well and our presenter today will be Brian winter so let me tell you a little bit about Brian so Brian is the encoder product manager with need a cap tron automation brian is a experienced industrial controls engineer he has over 25 years of direct hands-on experience with encoders and motion control applications ranging all the way from ounce inch servos to up to 5,000 kilowatt generators he’s worked in marketing and application engineering roles for companies such as IBM General Electric molex and as we’ve mentioned already he’s currently with new deck F Tron automation corporation a little bit about his educational background Brian has a Bachelor of Science in electrical engineering from Rensselaer Polytechnic Institute and he’s also a senior member of the I Triple E as well so he brings plenty of knowledge with him here today so I think I will probably stop talking now and pass it on over to Brian so Brian anytime you’re ready feel free to take it away Thank You Myles I appreciate the introduction and the opportunity by design world to have have this webinar we are trying to work to get information out to customers and people in the marketplace about how to improve overall and co2 reliability and we think this is a great medium and a great opportunity for us to reach people from design world just one note we do know that there are competitors out in the world and whatever under US antitrust laws we do not want to communicate with our competitors in any way direct or indirect so if you are a competitor of our company or your manufacturer of encoders or other peripheral equipment please do excuse yourselves from this webinar and we do apologize that you were invited by accident so again with that disclaimer let’s proceed so we promised you four steps to end encoder problems there’s a whole range of issues here and so we’re just going to kind of step through them and at the end obviously we’re going to try to keep the power pointing to a reasonable amount and then we’re going to let you go live and ask your questions Myles is collecting those via your online support and you can also send those in so let’s let’s start so the first thing you’ve got to do when you look at encoders is recognize the kind of problem encoders are a funny duck there are funny breed of with equipment they’re electrical and they’re mechanical so very often when you try to diagnose them the first thing you’ve got to do is tell the difference between the kinds of failures that you’re seeing in your encoders unfortunately there are a lot of encoder failures out there in the world and so let’s try to differentiate them let’s look at electrical failures look mechanical failures and then what I like to call mysterious failures and customers like to not use the word like in that sentence because they hate mysterious failures and then replacement

issues so electrical failures I define pretty simply the encoder is still going to spin or move we’re talking primarily here about rotary encoders but applied to linear ones as well but in some way the outputs aren’t making your machine satisfied they’re not closing the loop you’re not having velocity control your machine isn’t going to the right position so when we try to break that down within electrical failures we’ve got a couple of basic things we’ve got the engine inside what we call the guts so you can certainly have the guts failing use in some way that they’re not making a good position or velocity signal or possibly your outputs are failing me the mind inside the encoder is fine but its ability to communicate with you is damaged and unfortunately really really commonly there’s nothing wrong with the encoder and the problem is somewhere else this is something I’ll repeat a couple times in this seminar but one of the things it’s really important understand is across encoder returns over 50 percent of the returns that are made to encoder manufacturers over 50 percent there’s nothing wrong with the encoder the manufacturer will test it they will find nothing wrong and then they ship it back to the customer so that it’s really important to keep that number in mind whenever you think you’ve got an encoder problem remember 50% of the time the problem is probably somewhere else and you might not want to ship that encoder back to the repair shop or the encoder manufacturer until you’ve diagnosed it a little bit more so here’s what quadrature encoder incremental pulses should look like so we’ve got phase a phase B and what are called the complements a and a not you see that this is kind of an idealized picture where the signals are separated by 90 degrees or half a pulse so it’s really easy for me to tell what direction the encoder is turning because in this case B is rising before a and then a follows B you see the knob signals are there those are provided to allow you to have differential input which allows you to ignore a lot more signal noise and error so this is an idealized signal but no one signal ever looks like that it looks kind of like this this is a shot straight off a customer thing complete with a nice whack to the front of the oscilloscope screen but this is still looking pretty good here this is a pretty idealized in incremental encoder signal again I see that a is rising and in this case 10 B is rising so you’ve got a encoder rotating getting good pulses I can see a couple of nice characteristics here I’ve got good square waves they’re separated by half away for 90 degrees and you can see there’s not a lot of noise or other strange things going on so this is a pretty pretty good situation for the customer now this is again a real-world customer scope signal and you can clearly see there’s something very badly going wrong here now you don’t have a square wave at all so in this example this is an example of what happens when an encoder output is loaded too heavily what I mean by that is it’s trying to make a square wave but it doesn’t have enough current enough capability the transistors are too small the line driver is too small on this encoder to make good square waves so in this example the guts of the encoder are working fine but the output just can’t do what you’re asking it to do you’re going to see this kind of problem very typically with long cables long cable runs often make square waves into things that look like rounded edges so this is this is a typical thing and what’s the problem here well the problem is that you’re not seeing distinct ones and zeros that the drive or control can turn into position or velocity information so this example of a real-world problem on outputs now what if you don’t have an oscilloscope how are you been a debug and encoder or maybe you’re not an expert with oscilloscopes most people aren’t I work around encoders you know all day long and a customer gives me a different or new scope and I struggle with it so there’s definitely an issue there in terms of being able to troubleshoot an encoder if you don’t have an oscilloscope so we’re going to talk a little bit further in the seminar about self diagnosing encoders and I consider those really important because most customers either don’t have a scope handy or their technician doesn’t know how to operate the scope or it isn’t easy if you look in this configuration how are you going to get in the middle

of that wiring to get your signals and monitor what the encoder is doing so a pretty difficult thing so I’m going to really emphasize self-diagnosing encoders is a very very important part of helping you eliminate the problem so now let’s move on from electrical failures to mechanical failures and unfortunately these are all too common as well most encoders have bearings so you can hear a grinding sound from inside of the bearings are malfunctioning sometimes the bearings fail entirely and the shaft freezes or the hollow shaft won’t turn with the main shaft or there’s another problem where actually the encoder will still mechanically turn but you don’t have any output at all electrically the signals still look like the encoders not turning so here’s an example of a nice big rugged encoder bearing and just to give you guys a point of reference this encoder bearing goes into one of our severe Duty encoders and it’s about a 50mm across barring it’s a very rugged bearing that provides excellent support there’s actually one on the other side of what you’re looking at there is a magnetic rotor system so this is an example of a good healthy bearing you can see from the side that the races are smooth there’s nothing strange going on it’s all looks new and uniform like a king from the factory now here’s a comparison of and to encoder assemblies from hollow shaft encoders and you can see clearly on the right hand side that that’s a good unit there’s no significant disruption but now on the left side we’ve lost the shield to the bearing and you can actually see disruption that the balls aren’t even in place the shielding is not working correctly and you can see also if you look more carefully kind of a blackening to the shaft so these bearings of malfunction completely in this type of encoder and obviously it’s not going to work so this an example of a mechanical failure identified another in customer sent up this encoder to troubleshoot once we took it open it was pretty easy to troubleshoot because the little glass disc literally fell out and all these little bits came with it this is a pretty common construction technique for optical encoders they use in a glass disk inside and so it is very susceptible to vibration and shock for instance need a cap Tron never uses glass disks in our encoders because we consider them very vulnerable to this kind of failure but there’s a more subtle failure here too if you look at the encoder housing on the left-hand side of the picture of the upper left corner you notice that the inside of the aluminum housing is just corroded white that’s a really bad sign because what that means is that encoder has been receiving contamination it’s getting saltier it’s getting some content corrosive thing inside the bad news for us is in the encoder business if you’ve got an optical encoder and and other things are getting inside water dirt or dust it’s going to cause a malfunction of the encoder that’s because the sensor is trying to shine a light through the disk and and view that light and if you’ve got water or dust going on in an optical encoder that’s bad so this is really kind of a dual failure one the disk broke but I’ll bet you probably the reason the disk broke is because somebody hammered on the encoder and the reason the hammer about it is because it was intermittent before that time our least favorite category mysterious failures so the encoder causes some sort of machine problem but there isn’t an obvious mechanically electrical problem so one of the things we’ve talked about here is optical encoders and you have a light shining through a disk and on the other side of the sensor that comes from a factory all sealed up with a certain amount of factory air nice dry air maybe 70 degrees or 20 C and then what happens is you make the encoder hot using the machine or maybe it’s out in the sunshine and the Sun shines on it the air expands it pushes past the seals now maybe it’s nighttime or the machine stops running and it’s cold you form a vacuum inside that encoder and it starts to breathe it starts to inhale all manufacturers rape their seals as a certain IP rating IP 65 IP 69 and whatnot but the bad news is that’s the first day it shipped not a year later or six months or six years later so as the

fields are rotated they wear so now that vacuum inside the encoder becomes too strong and it draws in moist or dusty air it’s the encoder that’s an example of something that’s going to cause a problem because you’re going to condensation the worst problem is during the day it’s going to boil off and your encoder is going to work fine so here’s an example of a standard decoder is it working or is it not it’s really hard to tell from the outside you need an oscilloscope you need something else to help you try to troubleshoot and debug here’s been one of my favorite slides that came a picture came in from a customer and you see on the these are two relatively you know mechanically interchangeable encoders and the customers sent us this slide and we were trying to figure out okay which one of these encoders is working and which one is not and the real key was to read the customer’s note the type that doesn’t work and the type that works so ETA was able to identify in this case that there was a malfunction in this so-called interchangeable unit as an example what was happening in this application this is a motor with a brake and the brake has a very powerful magnetic field so the app Tron encoder is designed to withstand that powerful magnetic field in this example the competitors unit was not it had a problem where it would lose the marker pulse the once per revolution pulse and what would happen is the customers machine would become lost and that they would issue a fault so a pretty subtle problem in this case but it and it took some sleuthing on our part to figure out what was going on in the end what we actually proved with the customer’s help is when he actually clamped the shed the shaft and a vise and turned the brake on he could watch the pulses from the other guy’s encoder go away so we were able to actually help the customer or say okay this is why the problem is occurring now here’s an example of a self-diagnostic encoder this is so much more helpful because now the encoder has got a green LED that’s shining up and it’s giving you information it’s telling you that not only does it have power but the self diagnosing digital signal processor onboard is telling you I’m making good square wave my pulses are 90 degrees apart all this other information and the nice thing is it’s presenting it to you in a simple go/no-go fashion because you don’t want to have some scrolling LCD display on this thing well the quadrature is okay but the phasing is not so good or you know people don’t know that kind of information about encoders you want a simple red-green indication so this type of encoder provides that type of feedback information directly by using self-diagnosing technology used to be that was pretty expensive technology nowadays that can be included in a whole range of models so for example Nita capped Ron has several Oklahoma series of encoders that fit different motors and applications all of which that include self diagnosing led they also have a remote alarm contact that’s pretty important because now not only does the machine have red green but the operator may not be standing there you can see this machine this is behind a paper machine there isn’t a guy standing there so now we can bring an alarm contact out to the remote drive or PLC cabinet and it’s quite effective because if you if you don’t have an encoder fault you’ll know it if you do have an encoder issue it’ll alert you so much simpler to diagnose and much quicker to repair and furthermore these are preventative Diagnostics they look ahead they ask is the signal getting bad and that happens before the drive actually has a trip or the machine has a problem so here’s an example of what a conventional absolute encoder looks like and you can see there’s just incredible amount of gear work inside it’s basically a little watch very intricate so for simplification need a cap Tron has attacked this horrible system and for our magnetic absolute encoders we don’t use gears of any kind but you can see this isn’t something you want to take open and try to troubleshoot yourself it’s just really mechanically pretty fragile so after on doesn’t make need a catch on doesn’t make encoders like this but even if you have some some from another competitor it’s probably not something you’re gonna be able to

take apart and diagnose so replacement issues that’s very common you see a lot of option codes out there a lot of OM specifics a lot of lead time so you’re going to get a lot of replacement issues that happen with encoders that you know it’s broken but now you need one and you’re trying to find it there’s just an awful lot of possible combinations out there okay so I’ve given you all the nasty problems out there so how are we going to try to solve these problems you’ve identified the problem so the first thing is you get rid of what I like to call claptrap which is mechanical stuff outside the encoder so here’s an example of an application and you can see you’ve got encoders hung off the shaft with a coupling you’ve got another one that’s belt driven and then there’s an additional coupling outboard there’s just a lot of mechanical stuff going on here this type of system can often be replaced by a no baring encoder that mounts directly on the motor so you can eliminate a lot of these additional items that can cause failures the encoder itself might or might not have failed but maybe the belt broke or the coupling broke or you’ve got shaft run-out or wobble there’s a lot of mechanical problems here so try to get rid of mechanical problems by getting rid of what I said claptrap here’s another example a belt drive off axis you’ve got additional bearings you’ve got additional problems with reliability and then you’ve got the small encoder hung off the bottom of the system here’s an example of a system that’s got a lot of couplings in it now you’ve got an encoder in the middle and you’ve got a mechanical over speech on the back that’s a pretty complicated system for instance instead Nita gab Tron offers integrated over speed switches so if you need a mechanical over speed switch for safety we can mount it directly and as an integral part of the encoder and get rid of those couplings in this application we could have also offered a hollow shaft encoder that would have directly hung on the main shaft so we could have gotten rid of a number of mechanical points of failure we talked a lot about these internal gears and multiple sensors that are going on as I talked about before need a cap Tron our magnetic absolute encoders don’t have ears we use an electronic counting system which does not require a battery or any other backup system and it’s a very much simplified device we think mechanically simplifying is really important for encoders so 7 over 3 I advise you to do is you know get rugged what do I mean by heavy-duty this is a big problem between encoder manufacturers because people will call something heavy-duty more rugged so they show you this picture and they show it on your screen say this is a heavy-duty encoder is this a heavy-duty encoder well looking at this picture it looks pretty nice machined out of metal looks looks nice and rugged but if I show you this picture you suddenly see whoa this encoders a fraction of the size of a heavy-duty encoder this encoder might have a 50-pound bearing or a 20 pound bearing and the encoder on the right has a 2,000 pound bearing so there’s enormous difference between what manufacturers call heavy-duty encoders so definitely you want to look for heavy duty models but also make sure when the guy sells you heavy duty that it really is more rugged than the other models that you’re currently using comparing across manufacturers can be very difficult press the manufacturer for bearing ratings sizes what’s the shaft size try to figure out what characteristics he has that makes them more here’s a here’s a good example the bottom set of bearings is from a common HS 35 a encoder the top set of bearings are from our X our X P and X our 45 model so you can see these these are going to fit the same application this encoder can fit the same application but you just get a feeling for the enormous difference in varying sizes here’s here’s one that I love this is a really nicely sealed up encoder cam beautiful except somebody doesn’t bother to put the extra pins in the connector so now I’ve got a leakage point in my encoder where the connector can leak and to save money the OEM didn’t include the little o-ring that you can include for this type of connector so this is really a guaranteed failure point that you’re going to see on this encoder so count your pins make sure somebody put pins in all the little holes because otherwise that’s going to get into your optical encoder and again I’ll emphasize our absolute encoder for instance have no internal gears and it also uses a magnetic sensor system with magnetics I can ignore dust or water all sorts of things because magnetics works through these materials as opposed to trying to

shine a light through a disc and it can be blocked by dust or dirt or water here’s another example of a sealed magnetic encoder sensor so this is an incredibly durable item it’s cast aluminum and then the electronics are all power potted in a solid brick of plastic so you can give this thing a big sturdy whack and it’s not going to hurt it you can dip it water it’s not going to have a problem this is really going to provide a much higher level of reliability than a typical optical encoder for instances of magnetic heavy-duty model here’s another innovation that you can do you can eliminate all those belts and couplings and hollow shafts and so on by using a no bearing or modular encoder this technique is really only possible using a magnetic encoder you have a rotor in the center that spins on the motor shaft with the motor shaft and then that sensor that you saw a census that disk from the side so there’s no contact there’s no bearings there’s no seals it’s just a tremendously reliable system we see typical lifetime of encoders like this ranging in the 15 to 20 year time span because really there’s no reason for them to wear out and step number four I think this is perhaps most important is to get diagnostic and that is you need to make sure that your encoder is telling you when there is a problem remember the number I told you that’s 50% of all the encoders returned they returned but there really wasn’t a problem so you need to have Diagnostics in your coder so you can right away eliminate that 50% of the issues to say oh it’s not the encoder it’s something else or equally important a good diagnostic encoder can tell you it’s not a problem with the signal before there’s a problem on your machine it’s really important that it’s predictive that it’s looking ahead at the quality of the signal trying to decide am I going to make good signals or am I headed toward a problem so again this one’s really easy to tell I’ve got these LEDs on it and like I said this is not just a power light this encoder is trying to tell us hey I’ve got good quality signals it’s meted measuring quadrature phase separation it’s measuring jitter it’s measuring a lot of characteristics of the signal and assuring you hey those are good quality signals remote diagnostics this is equally important so the local deli is good but also get remote diagnostics on a absolute encoder you typically get kind of a smart communication bus option you can get a word or a piece of data off your message that will tell your PLC if your encoder is working well and if it’s happy for an incremental encoder you often just get a simple relay context go/no-go they’ll no fail red green and then again it’s very useful information to bring back to your drive or machine because you can save the customer a lot of troubleshooting time and you can also predict if there’s going to be a failure so in summary these are my four basic steps to hugely reducing encoder products you’re going to recognize the problem we talked about the different types of problems that you’re going to recognize you want to simplify your mechanical mounting book we’re – an N and mounting today we want to recognize the mechanicals just make them fundamentally more durable and then you want to get Diagnostics combining all of these four steps it’s really going to greatly reduce your encoder failures it will decrease your time to repair and it will increase your machine uptime so that’s our basic message but I’m sure you’ve got various questions and areas of interest to you so with that I’d like to turn it back over to Myles for questions ok and well I’m gonna take some time here – thank you Brian for that that very interesting and thorough and detailed presentation I I learned a couple things to hear as well and that’s um I might have some questions – we if we get to those so anyway I do want to open it up now – to the Q&A session here so if you do have any questions and there have been some questions coming in so we’ll see how many of those we can we can get to so um let’s see here maybe we’ll maybe we’ll start off here okay so we have one question here let’s start off this is a the question dealing with cable length I think this is probably when you were talking about the excess loading

problems and maybe some of the some of the line drivers as well and the question is long cables meaning in terms of what like in terms of tens of feet or hundreds of feet or I guess what is there some kind of cable length that’s that that would be considered long or I think that’s that’s the general question so maybe what what length is considered long or is that kind of subjective based on application or great that’s a really good question and again it depends vendor to vendor historically if we talk about incremental encoders people put a relatively weak line driver on it a see most based technology for those of you that are from way back in the day and anything over a hundred feet was considered long you would often end up with signal problems anything over a hundred feet then a lot of the incremental industry went to a more standard line driver which was not see most base often called a seventy two seventy two that was kind of like the default and now you’ve got a range of performance you could run five volt a thousand feet but twenty four volts you could only run about two hundred to three hundred feet before you got problems so what Nita capped Ron did is we’ve taken heavy duty line drivers and coupled them with additional protection circuits so that we can actually drive 24 volts way more than 1500 feet so it depends on the vendor that you’re talking about in terms of what length is acceptable there are people that will consider a hundred feet long most people in the middle will consider 250 feet long Anita gapped Ron we’re looking at distances of you know 1500 feet or more are considered long one people advice I want to give you is a lot of people try to solve this problem with a repeater device we know about these and they are out there the problem I’m concerned about is those tend to add to the complexity uncertainty you’ve got to wire those two you’ve got a pirate power those devices so apt Ron doesn’t typically recommend repeaters for extending length okay maybe a maybe a bit of a follow up with that then what what would you suggest for for kind of a solution to that problem is it just is it just as simple as maybe just getting a shorter cable run or maybe you know maybe changing up the actual the the power capacity of the line drivers themselves or is there some other type of solution that you might recommend there’s a couple of solutions to your miles and that’s a good you put it in a subtle way and that’s good if you if you think about the problem in terms of you’re trying to drive a signal down a cable a good analogy is a kid wiggling a jump rope up and down not actually twirling it but just wiggling it you’re kind of wiggle make a kind of a wave down that rope if you think about that rope as if the longer it gets the harder that little kids arm has to work to make that wiggle so often you’re not in control of how long your cable distances are your machine is your machine but what you can do is use a higher quality cable that is one with lower capacitance lower capacitance cable will provide your abilities basically to wiggle a lighter rope if you want to make the analogy there it’s it’s not like the big anchor rope that you might see on a ship you’re wiggling a small light rope and it’s easier to control there is a trade-off there because often the high-low fastens cable can have high voltage drop so you want to be be aware so for instance need a cap tron so a special cable that has both no voltage drop and low capacitance so that’s one combination you can do is to improve your cabling then the other thing you can do is upgrade your line driver nita gastrin and a number of other vendors offer a range of line drivers and you’ll be able to find a little table in the book that shows you the distances they rate for the voltages so a low power line driver might give up the ghost like I talked about at 200 feet but a high power one can do the same job the same app location and might even be available in the same model and would drive more than a thousand feet so there’s a couple ways to solve the problem okay good that’s all that’s that’s all good good details turn to another question now mounting so question here is its as for mounting our chef couplers required or recommended or something else perhaps that’s that’s a good question as an encoder manufacturer we really don’t like couplings couplings are a necessity of evil in a lot of

applications but if you can get rid of the coupling do it and what I mean by that is look at a different encoder construction method and see if you can get rid of the little shafted encoder that’s coupled to the load because couplings are fundamentally a point of breakage they are you know they’re set screwed to the shaft and they can have problems so what we recommend is hey let’s look can you look at an old bearing encoder can you mount it directly on the frame can you go to a hollow shaft encoder that twirls with the shaft and now you eliminate the coupling we try to eliminate couplings wherever possible now if you’re in if you’re stuck with a coupling one thing is don’t try to go too cheap on the coupling you want a rotary multiple disk coupling you don’t want to use a piece of garden hose or other things because a lot of those have pretty nasty behavior so the the output of the control system won’t be accurate so we don’t make couplings but we do recommend high quality a couple flexible disk units so that eliminate the coupling wherever possible but if you’re stuck with it buy a good one okay sounds good all right well let’s keep keep moving on with the questions here so here is a kind of a small C it’s a application type of questions I believe but the question is what type of encoder do you recommend for high speeds up to 10,000 rpm 8-bit resolution would be plenty okay saying there are thoughts there if somebody is talking about 8-bit resolution that tells me they’re probably talking about an absolute encoder so the as opposed to an incremental encoder and a 10,000 rpm what you’re going to have to watch out for is an encoder that doesn’t have much drag or on the seals because you’re going to spin it and it’s going to get real hot so seals drag a lot so there’s going to be an interesting compromise how dirty is your environment versus how tough a seal you need that won’t get too hot one solution that I might propose is taking a look at a magnetic absolute encoder because we can tolerate more stuff getting inside the encoder we could use a lighter seal with less drag and run it up to 10,000 rpm certainly if you know if you’re if your application is just horrific ly dirty you’re going to have to worry about the bearings of the encoder but as long as you’re not too awful a magnetic absolute might be the way to go in that application okay all right there’s a question that kind of harkens back to one part of your presentation talking about we were talking about the self Diagnostics portion one question is what I guess what kinds of look the things can it can it tell you I know you were kind of saying that maybe the primary message sending you is that it’s working that signals are clean that that kind of thing is that and that would presumably be just with that green light flashing more or less right are there other kind of things that other pieces of information or data that a diagnostic type of setup can can can tell you or is it really just that hey I’m working fine and if the light changes color that means hey I’m not working fine or maybe you should check you know X Y or Z or something like that very good there really are two layers here of available diagnostic systems the most common one unfortunately in most encoders is no Diagnostics they’re a mysterious little black cam that you can’t tell they’re working or not but the big improvement above that is the simple go/no-go diagnostics and the reason these are so helpful is because they are going to tell you if the encoder is producing a high quality signal in our case I’ll just numerate the the specific things that’s telling you number one I was telling your power is correct you’ve got it the polarity of voltage are are correct for the system the second thing is telling you is that the digital signal processor and all associated equipment are passing continuous testing so therefore the circuit is working in every way the next thing is telling you is for quadrature it for incremental signals it’s actually producing good quadrature remember those pulses I showed you earlier they’re 90 degrees apart they’re the same width you’re getting consistent pulses you’re getting a and B you’re getting all those things is what that system is troubleshooting that’s what it’s self diagnosing then if we go one step further to an absolute encoder now because we have the luxury of sending you a digital message to your point Myles I can tell you much more about what’s going right with the encoder or more if there is a problem I

can be more specific about the encoder and say for instance the once per revolution apart the multi turn part of the encoder is working but the within round one revolution is inaccurate and so I can get much more specific with my error messages and send you those things so the one extreme is the mysterious black box you can’t tell why it’s working in the middle is the stop diagnosing system and the third one is the high level of diagnostic messages okay all right good sounds good maybe it’s just a very quick follow-up on to that but we know that the world is becoming more more wireless is there any you know any way that these encoders can have or can send their diagnostic data back wirelessly somehow or is it still pretty much a kind of a wired world when we when we’re talking about some of these applications and factories and some of these dirty places where these things have to have to that’s a very good question today as we talked about the majority of encoders don’t have any Diagnostics at all so that’s a shortcoming that I hope will be overcome in the next couple of years is manufacturers catch on that this idea of let’s get diagnostics in there because it saves the customers so much time now we’re still in the first phase where a lot of customers don’t have wireless capability within their manufacturing facilities and so that’s the that’s where we’re at today as most of the Diagnostics today are still wired it is possible to carry some of these pieces of information wirelessly over a wireless Ethernet link or what have you but a lot of our customers today don’t have the infrastructure for it if they did it’s it’s relatively easy to bring the information out of the encoder and get it to them some people have experimented with cell systems and and seeing if they can add cell systems or some other information there and that is something that you can bring the encoder Diagnostics to is one of those cell based monitoring systems okay all right so that was a bit of a detail question maybe it’ll kind of broaden it out to a more kind of a wider question or a general question but you did a pretty good job they’re talking about you know how how some of these encoders can can fail or one of the common ways I guess but is there you know in your in your in in what you’ve seen I guess is there some one big common cause of failure or is there one kind of big thing that people should be watching out for or or or not I guess yeah at the risk of being repetitive I’ll say the number one cause of failure is no failure and that is 50% or more of those encoders that you think have failed haven’t and that’s because in some serious black box and you don’t you can’t tell that it’s not working or whether it is so that’s the number one thing that I think that everyone on this call should be really aware of is when you think you’ve got an encoder problem you might but you also might have another issues so helping having Diagnostics on the encoder is going to save you an enormous amount of time uh-huh so there’s a basically like you said earlier there’s there’s a 50% chance that it’s probably not that but it’s something else basically absolutely the other problems range all over the map and there probably isn’t nearly as much of a you know one big glaring thing that people can look out for I would say the one big glaring thing is it’s not broken oh okay so uh there’s a question about replacements so for instance how would you how would you replace an encoder with a replacement that perhaps might be bigger or or even smaller in some cases that’s an excellent question because I know e/m is going to build a machine with an encoder on it and they picked an encoder often OMS have to make cost-benefit trade-offs so they’re going to pick an encoder that’s going to last a year or to meet their warranty requirements but might not be the most rugged at the end-user application like we always say you know OEMs are very worried about the warranty length and for the end user they want the encoder to last for just a huge difference in expectations so what it what Nina Kapton does we make a whole set of models that are physically compatible with OEM mounts and fitting so you can take that small 58 millimeter flange and a 10 millimeter shaft you could mount in these types of applications where a light-duty encoder was originally installed and now you can put something with those thousand-pound bearings in

place with the magnetic sensors it’s still going to mechanically fit and maybe the housing might be bigger but usually the mating parts the important parts will fit the same or the hollow shaft will fit over the same shaft and now it’s the end your customer he’s going to see a much much more reliable machine he’s putting a fundamentally more reliable encoder that was designed to last somewhere between seven and 20 years instead of you know one two five okay okay and maybe the kind of work off of that topic I want to kind of circle back to something else you mentioned earlier when you were talking about the the concept of heavy duty I think and I think you made some really good points there too about just you know something can be labeled that but what exactly does that mean and I guess maybe the follow-up question that would be what what kinds of factors would you be looking for or should you be looking for so I think you mentioned you know mentioned looking at bearings obviously looking at shaft size maybe you know would it be other things like temperature ratings or IP ratings even or you know shock those those those kinds of things or something else very good question um I’ll say that there are a set of specifications that really dictate how rugged the encoder is unfortunately for the end user for the OEM trying to pick the product as I talked about there can be a number of things that are measured only at the time of manufacture so it can be very difficult so I’ll start with one that’s my favorite that you don’t want to use as a measure of durability don’t use IP rating as a measure of durability when a manufacturer is trying to tell you with IP rating is how well sealed up is the cam as we talked about with rotating seals you can end up with wearing seals so very quickly that IP rating degrade so I don’t recommend using IP rating as a yardstick for how tough your encoder is it’s it’s a yardstick of how hard the only the magnets are trying to make is optical encoder seals up a better a better measure of this is bearing size baring size absolutely dictates bearing wear and a bigger bearing will last longer in all these applications and that’s true across every vendor if you get an encoder with bigger bearings it’s going to last longer okay another another indicator of durability for me we make more magnetic and optical encoders so I don’t I’m not slanting the table one way or the other magnetic encoders are fundamentally more durable because other contamination doesn’t interfere with them water dirt dust and so on so we make magnetic and optical but for the heavy-duty applications we always recommend the magnetic technology those are the three things that I would really point out then after that you start getting into more subtle things like gee does the guy put you know screws do the sight of his housing or the edges or what do we SEALs look like it’s a much more tricky thing but my advice to you number one is don’t use IP rating number two do use bearing size and number three do look at magnetics versus opticals okay good good advice okay so here’s a here’s another question about troubleshooting you were you were kind of looking at that and showing us some of those some of those snapshots basically of the auditory signal so here’s a question about troubleshooting the question is how can I troubleshoot the encoder output with the simple fluke meter let’s say not a scope that’s a really good question and the answer unfortunately is you can’t do very much with a with a typical fluke meter without a scope here’s why if you look at a typical encoder let’s say an incremental one it’s supposed to have a thousand 24 pulses per revolution so that means every one of those pulses represents about one third of one degree so if you’re trying to use your meter to say monitor the outputs and say is it a one or is it a zero is it 24 volts or zero number one any kind of small mechanical vibration and you can’t tell what it is because if that moves the shaft a third of a degree you’re going to see a different value the other thing is if you try to use a meter to measure the continuous voltage of the unit while it’s rotating you get some average value because meters think they’re listening to a CD not pulses so the meter might give you something meaningless like four point five volt output when the encoder is really making between 0 and 15 volts back and forth is pulses so the bad news for you with when you only have a flute meter is you can really only do the very

basics which is do I have power coming to the encoder to the right connections that’s you can try measuring other things like is a the opposite of a not but again if you’ve got any mechanical vibration it’s almost impossible to tell that because you’re the shaft can be moving a third of a degree pretty tough and that’s the bad thing boys that’s a bad thing about trying to use a flute meter to diagnose the encoder it’s really hard to do yeah yeah I mean the kind of the the very nature of those output signals kind of lends themselves to you know to having a having a scope and having to having to have to do it that way and pretty much you know no other way really is what it kind of sounds like but how many customers have a scope and how many customers know how to use it that’s the real the real difficulty is it’s not an easement to use and people don’t walk around with one in their toolbox exactly exactly but just about everyone has a you know every every tech has a has a simple DMM right so actly so that’s kind of the intent of putting Diagnostics on the encoder is to try to reduce that need for the oscilloscope because that is a it’s a very tough thing to walk out there with a fluke meter and try to see fewer encoders work okay all right there’s another question a question is is it is it possible to use magnetic encoders in radioactive environments ah very good question this is an interesting one this the the person contacting asking that question may be familiar that optical encoders have a interesting problem when under radiation their their glass disk tend to turn black so if exposed to radiation a lot of times these disks that are used to either make let the light through or not become completely black so they don’t let the light through it all so you’re going to so magnetic encoders can be used in applications with some radiation eventually you’ll end up with too much radiation and you can kill the circuits because they’re not these are not milspec red hardened equipment so it’s not something you want to plop on a satellite and fire into space without really not a good amount of radiation shielding with an advice that more might be more suitable for ultra high radiation environments is a resolver that there is there you’ve just got windings and to mechanical coils a resolver is really the traditional solution for ultra-high radiation but I can tell you magnetic encoders will withstand more radiation than an optical and still function okay good advice all right well and we’re getting closer to the three o’clock hour here so we’ve we tried to uh try to answer as many questions as we could here and so maybe we should kind of start to start to wrap things up here but I do want to remind you everyone I think can see on the screen there you have contact information both for for me and for Brian as well so if there’s anything any other question that you think of that you haven’t you didn’t get a chance to chime in with or you think of something after feel free to contact Brian I’m sure he will be able to answer your questions so I’d like to thank everyone for joining us here today and just as a reminder this webinar will be up at design world online.com and as well as it will be sent to everybody who has registered for the webinar and their email INBOX as well you can still tweet about this webinar on twitter using the hashtag DW webinar a couple of ways to connect with design world there we’re on all the social media outlets we’re on facebook and twitter and linkedin and google+ youtube as well and we can also keep this going as well if you’d like on the engineering exchange comm so once again Brian I’d like to thank you for your time today and for your expertise here and sharing it with our with our listeners and readers here on wish everybody a good rest of the day and rest of the week Myles I’d also just like to close by me decapitron does offer 24 by 7 technical support 365 days a year so we do want to make sure will answer your questions over the web by phone or whatever in a timely fashion so if you have any urgent or emergency issues we do offer that service available thank you so okay good enough thanks Brian okay thank you everyone and have a good day you you