The following is a lightly edited transcript of a FPTC talk given by Chris Heberlein.
I don’t know how many people are familiar with Balluff. Balluff is a company that makes detection products, sensors. We also make measurement products, which we’re gonna talk about today, identification, RFID, bar code readers and connection, networking, cabling, things of that nature. My focus is on measurement and position sensors and I follow the hydraulics industry as well as a lot of applications and factory automation. Everything from steel making, anywhere you’re gonna find hydraulics that need a lot of feedback and position, sawmills, tire making, molding machines. That’s what I do for Balluff.
In cresting this talk, I was trying to figure out, “Okay so, what would be important to people in the hydraulics industry?” And kind of following up from Tony at Aggressive Hydraulics on cylinder positioning, my manager, Henry Menke, wrote a white paper, which you can see, called “Improving the reliability of hydraulic cylinder position sensors”. I thought it was very appropriate and try to give you some insight into some of the challenges that you face and how we work through some of these applications. Sometimes it’s a sensor, sometimes it’s rearranging something in the application, and we just want to share a little bit more with you.
We’re talking more specifically on magnetostrictive sensors and we’ll get into what that truly means and how those work. When we wrap this up, happy to take some questions, I’ve already talked to a few people about applications out by the booth. We’ll get started.
First, I want to get everybody on the same page is … What is sensor positioning 101, what kind of technology we’re really talking about, what kind of jargon that you hear, and I’ve already heard some already, kind of polled the audience, what they call these things. Then we’ll jump in to some of these harsh environments. On easy applications, you got a lot of options. It’s when you get into really challenging applications because of the environment that your options become more and more limited, or you have a lot more downtime and you’re always looking for another way to solve the problem. That gets to … You get to the point, “How can I increase my uptime?” We’ll touch on that as well.
All right, getting to the basics. When it comes to positioning, and Tony kind of talked a little bit earlier is, you really have two methodologies: you have end-of-stroke sensing and you have continuous sensing, kind of like the on/off light switch or the dimmer switch. When it comes to hydraulic cylinders, magnetostrictive technology, which I’ll define in a moment, gives you continuous position sensing. You know where you are all the time.
The sensor itself costs more, the information you’re getting is more. The end-of-stroke sensors are less expensive, but only tell you when you’re fully extended or fully retracted. I mentioned earlier some of the continuous sensing applications that I’ve seen and worked on, sawmills, steel mills, tire presses, things of that nature, end-of-stroke can be simplistic just like almost anything that you’d want to know that your cylinder just got fully retracted or fully extended. I did mention magnetostriction, that’s a technology, there are Hall effect sensors, there are LVDTs, there are all kinds of technologies that are for hydraulics, specifically hydraulic cylinders, but we’re gonna cover magnetostriction as far as what are these things called. So I call them magnetostrictive sensors, transducers, what … Anybody else?
Don’t all shout at once. All right, I’ll help you out. LVDTs, probes, wands, MDTs, cylinder feedbacks, scale, I’ve heard all these things. LVDT is actually a tran … it’s a linear variable displacement transducer or transformer. It really is a transformer. A lot of people will call magnetostrictive technology an LVDT and it’s just … They’re getting a linear feedback but that’s actually specific to a technology. You get a lot more information from a probe that goes inside a cylinder versus end-of-stroke. How it gets manufactured is, the effect is called gun drilling, so they … Let me see here … Check out my little toy. The cylinder head and shaft is completely drilled out so that we can put our high pressure stainless steel tube into the entire stroke of the sensor. On the piston-head, the cylinder manufacturer’s gonna mount a magnet and on the end cap it’s threaded or there’s bolts and you attach the transducer on the outside. That’s one method. Another method is you can actually encapsulate the whole thing on the interior back end of the cylinder.
Basic functionality is, as that magnet goes back and forth over that tube, our sensor is getting a position feedback and providing an output. In this case, we’re showcasing analog voltage. There are many, many other types of output interfaces that are available. Everything from the old fashion start/stop pulse with modulation, SSI, Ethernet based IO-link, CAN Bus, so there are a lot of different interfaces that are available and they all have specific characteristics to them. The speed of the output, the resolution, the accuracy, all those taking into consideration, which output interface you choose.
I’ve mentioned magnetostrictive technology a few times. The actual process is the magnet itself gets close to a ferrous conductor and it actually deforms that conductor. So here’s our magnet, it deforms the conductor. We actually send a pulse down that conductor because it’s an electrical pulse and the conductor is ferrous, it creates another magnetic field. Where those two fields collide, there is a collision and there’s a mechanical shockwave that travels down this, what we call the wave guide. And you can see with the little blue … It’ll do it again … The shockwave goes in two directions. On one direction, it actually is … We have a dampening area, so it doesn’t get reflected back. In the other area, it goes to a pickup coil. There is a stop watch, we send this pulse out, finds where the magnetic field is, the mechanical ripple comes back, we hit the stop watch again, so we know how much time it was and we put that out into an interface: analog, SSI, Ethernet, a lot of different variations.
If you’re interested in more of this, we’ve got some out on our table but also from our website … That’s incredibly annoying but we’ll just work through it. We call them Balluff basics, so if want to know truly of more about these technologies, we have them available for you.
All right, getting to harsh environments. Temperature extremes can really be challenging. We’ll start with extreme heat first. So when you get to extreme heat, there are some processes like steel making, some of the curing processes in the tire industry, you’re just living day to day with hot extreme temperatures and those extreme temperatures can destroy electronics or shorten their lifespan. I tell you, my son loves turtles, so I had to include the turtle reference here, but the three areas of heat transfer; conduction, convection and radiation … When you get to conduction is what are you directly connected to and that heat transfers through there. Convection, air is blowing overtop of a hot or cold service and carrying that to it and radiation comes from a heat source with infrared.
How do we combat some of these? Number one is, in the case of conduction, is maybe it’s not feasible, but can you relocate the sensor? Secondly is maybe you had a cylinder that had a rear clevis and that clevis is connected to something that’s connected something and it’s all metal. Is it possible somewhere along that line that you’re trying to break that conduction? Could you put a thermal barrier in between there somewhere? I think Milwaukee Cylinders, that’s gonna be talking tomorrow, they have a water-cooled hydraulic cylinder. Could you put an air cooling system or a water-cooled hydraulic [inaudible 00:14:31] temperature to reduce the temperature?
When you get to convection and radiation, “Can you install thermal guards, like on your car?” Underneath your car, there’s heat shields all over the place. You’re breaking the convection or radiant heat away from electronic device or something that you want to last longer.
Then we come to the sensor itself. Maybe, if you had shortened life on electronic device is, go look at your sensor. What is it rated for? A good sensor for linear position, 75° C, maybe 85° C. Maybe you want to work with your sensor supplier to see, “Hey, do you have specific engineered options that go even higher, maybe a 100° C?”
Also, is the cable. Not all cables are equal. Our standard cable for our product is PUR. There are other cables that are out there, PTFE. They’re rated up to 200° C. There’s also what we call the sacrificial cable strategy. If you are running … Instead of buying a transducer with a 20 meter cable going back to your control box. Then when that sensor dies, you got to rewire 20 meters of cable. That’s not a lot of fun. Maybe you want to go a transducer within the connector or with a pigtail. When that transducer dies because of high heat, you unplug it, you don’t have to rewire that whole cable. That’s a home run cable, goes to that short connection and off you go.
Still too hot? You put the sensor in a box water-cooled, air purge, you have a lot of different options there and that happens a lot in the steel industry.
Now, if you go in the other direction, so extreme cold, which we see a lot in oil and gas is … what are your options there? What happens when you’re in a very cold environment? Like -40 and -50 C? The low temperature breaks down the gas breaks down the O-ring, the cable, the sensor elements, they can all become brittle. The actual semi-conductors, they actually start functioning the way they need to function. Again, you could probably look at the sensor supplier and see, “What is it really rated for?” And look for something to be rated as low as you need to go, see if there’s a special engineered option from them to go even lower and then some manufacturers tell you to … Instead of sitting idle, that you would need to power the sensor all the time, so that it warms those electronics to keep them and back in that operating range.
Moving away from temperature and the environment. Physical impact. If you are in the sawmill industry and some others, there’s a lot of stuff lying around and banging on the cylinder position sensor. A hydraulic cylinder is very robust and that’s not the part that breaks. It’s the sensor. You can see here, this one has a metal shroud on it and this one looks like they’ve replaced it at least once or twice and it’s missing its shroud. So what are approaches if you have rocks flying up on a mobile vehicle, or people climbing on top of equipment, which I hear quite frequently? Is guard, embed and upgrade.
So on the guard, this is a picture I took at a sawmill out west. You can see they put a large end cap over the back end of the cylinder and it’s just protecting the transducer that’s inside it. That’s all its job is, to protect it from flying objects.
Another one is to embed. Is take the transducer itself, instead of being outside, hanging outside, it’s completely embedded on the inside. So the cylinder wall itself is protecting it and the only thing that’s sticking out is the electrical connection.
Another option is upgrade. This is a cylinder that, instead of a threaded design, it’s got six bolts. It’s very low profile, it’s very robust stainless steel, so it is almost its own protective guard. Very robust sensor.
Shock. I would say that shock and vibration is probably my number one challenge that I … number one application problem that I work with customers on. Shock and vibration. And there’s not a lot that you can do about it. Shock is inherent to a lot of hydraulic applications and that’s why hydraulics are chosen. Is that there being away from servomotor or electric motor, is that they can really do a really tough job and they’re very robust, but you got to get the feedback and sometimes the feedback device could be the weak link on survivability. If you do have a customer, and I actually spoke to somebody this morning, sometimes they just live with it. They live with sensors that last six weeks and they change them out. They keep trying different things, so I’m trying to give you some ammunition if you are running into some real problematic applications, what you could possibly try.
One is look at the hydraulic system, the motion … If it’s on a motion controller. Look for robust sensor specification. Work with the sensor supplier to see if there’s some things that they don’t publish in their catalog, some really neat there, variants.
Then also, the connector. I think, about three months ago, there’s a lawnmower manufacturer out in Oklahoma or Kentucky, and man, we were getting returns, another return and another return under warranty and then we were like, “What’s going on here?” They got an air cylinder that they gun drilled and Bam! The violent shock that this thing was experiencing was breaking our products and it was … The shock was outside of our specification. What did they do? They worked with … They went further, instead of just taking another product off the shelf, plugging again, waiting a week or two after it breaks, and then going and get another one and another one, they started to look into the control system. Could they slow it down with the motion profile? A soft start and a soft stop on the extending retract. That’s an option for you.
Most sensors in the field are rated at least 100 G, for shock. Look for some that maybe go beyond that, like 150 G and some other variants. This is an application that I worked on in the northwest, at a cedar fence manufacturer, cedar fence board manufacturer. Their hydraulic cylinder sensor was lasting about between two to four weeks and they just replaced it. It just went over and over again, year after year. We got an opportunity to give them a sample to try. We used a special option, secret decoder ring says SA42. We inject a potting material into the electronics housing. It keeps the PCBs together during a violent shock. It’s been running for three years now, and they’ve changed out the rest of the sensors on the machine.
Last thing with shock is, go from an integrated connector to a pigtail. I mean it seems very simplistic, but you’re actually adding mass. This is a typical M16 connector. So you’re adding more mass on here, which is gonna just get more shock on it. Also, is that connector is gonna get cracks in it and that’s the point of flex and so, ingress protection all these things get violated there. What do you want to do? You do 200 or 300 millimeter small length of cable and then bring your connector here. You get all that flex from your cable. That’s just one less area of concern. It doesn’t solve all the problems, but I had a steel mill that went from six weeks to five months and they were very happy. There are still plenty violent applications in hydraulics, and to give customers more uptime, that’s what we want to do and that’s what we want to share some of these best practices with you.
This is one of my favorite. I would say outside of shock. Shock and vibration is my number one. My number two is moisture ingress. These are my number one headache and then my number two headache. So this is a rolling mill in a steel plant, and under this blue tarp here, is a transducer. They’ve got a zip tie, a tie wrap holding that little blue bag in place to add an extra layer of protection. Water can get into a lot of places. We’ve gone through a lot of headaches with customers and been a partner to them. They’re opening up the housing and water’s pouring out, and they’re like, “Where is it coming from? Is the gasket failing? What’s going on?” We have some recommendations for you.
Mitigation, again, back to the heat transfer problem is, “Can you move the sensor?” It’s not an ideal thing, maybe it’s not even possible, but what could you do to move things. Maybe putting a splash guard. Number two, a more robust sensor and again, number three, let’s look at the cable again.
Can you get out of that wet area? It’s not likely but its splash guards, if we go back here, we got one big splash guard up here, there’s a yellow one, looks like this green one here is another one. They’re trying a little bit but maybe you can add some more, just extra guarding. Then you can always look at the sensor. What kind of rating, what ingress protection ratings do you have?
IP 68 is complete submersion. And 69K? Anyone? High pressure spray. If you ever see an IP69K test, it’s pretty neat. It’s actually at multiple angles of very high pressure, it’s like 1,200 to 1,400 psi and the product has to rotate. It comes from … It’s actually not an IEC spec, it comes with German DIN mobile vehicle specification. It’s been adopted in industrial applications, and it’s been working its way through the IEC, to become part of the IP rating system.
You see over here on the right, this is an IP69K sensor that’s been welded shut. It’s got a great ingress protection rating.
Evaluate the cabling. I have had a lot of issues with cabling, making sure the insulation had a proper drip loop, nicks and cuts in the cable. The other thing is people were like, “Yay, I got a IP69K sensor.” And then they have water ingress protection ratings … or problems. And then you say, “Okay, so you got a IP69K sensor. What cable did you use?” Then they go … They pull one off the shelf, it’s IP65, it’s IP67. If you’re using a really high sensor rating, IP rating, make sure that your cable is also rated for the same thing. Look at your jacket, the cable jacket. Can you put a tube overtop of the … To prevent nicks and cuts or falling debris on the cable. It’s kind of a cable within a cable.
Corrosion. This came out of one of the plants that I used to call on … in North Carolina, in a steel mill. When you get chemicals, liquids, gases in the air, and you got air purge, and you got some moisture in the air, those chemicals bond to the water and they’re using air purge to cool down the area because it’s in a steel mill. All that stuff has got some acid chemicals and it’s eating the sensor alive. This particular mill put a silicone [inaudible 00:29:46] dip on top of this to try to add more ingress protection for the corrosive nature. They got us involved and … when you get to corrosive environments, you really want to go from aluminum housings to a stainless steel housing. If you got really corrosive things in nature, 303 and 304 stainless, those are great for almost all the applications. If you have salt water or salt spray, that’s where 316 comes in play. It fights chlorine and chloride ions a little bit better than 303 or 304.
You’ve gotten all this way, you’ve done everything you can to protect your application but you still end up with a violent process. One of the gentleman I talked to earlier was in sawmill. And primary breakdown when the logs come in, those things ram into place, the endogger picks it up and moves it down, gets cut. Those get rolled out and there’s a lot of hydraulics, a lot of positioning and just very high shock, shock and vibration. They’re doing everything they can to keep uptime, but they just, it just becomes the point where they’ve lasted as long as they can, and they just need to keep running. If you need to keep running in a critical application, here’s three options for you.
One is, we do have customers that will run a dummy cylinder next to the main cylinder and they put a transducer in it, or they’ll just run a second transducer to get a redundant feedback.
Two is, use a redundant sensor. This is extremely popular in power generation. Some power generation applications, they cannot stop. They got to keep going. Downtime is just unacceptable. This one, you can do redundancy or three outputs, a third level of redundancy in the product.
Lastly is, okay, now you’re down, “How fast can a get back running?” You could do a rapid repair unit, where the flange is still threaded into the cylinder. Keeps all the hydraulic fluid inside. You can take a couple screws out, replace the wave guide in the electronic housing and go back in and you’re up and running again. Just kind of … In summary, hydraulics are used in lots of applications that are in a lot of harsh environments, and we can try to mitigate them, look at the applications as carefully as we can, make sure you match that harsh condition to the right sensor, and then have a strategy. If you have a critical application, maybe you want a redundant sensor or rapid replacement unit.
Filed Under: Sensors