Latest / The Refrigeration Mentor Podcast / Episode 343. Basics of CO2 Controls - Programming CAREL with Kevin Mullis (Part 3 of 4)

Transcript

Episode 343. Basics of CO2 Controls - Programming CAREL with Kevin Mullis (Part 3 of 4) · The Refrigeration Mentor Podcast

• 0:00
  Trevor: Welcome everybody to this four part series. We're in part three today, diving Into More Controls Programming. This is what I get asked all the time and, and my courses. People message me on Facebook or LinkedIn or Instagram. It's like, I wanna learn more controls and this is why we're doing it. My good friend Kevin Miller series, helping me out and sharing some of these knowledge 20 years in controls learning CO2 controls too, not just right, just diving into CO2 controls.
• 0:01
  I wanna thank all the people that have been reaching out since we started this saying, Hey, can you get me a copy? I don't want re wanna rewatch it. I've been learning a ton from this and this is what it's all about is sharing and growing. I'm super excited to dive in here 'cause we're gonna be diving into programming with grl and we're gonna see that we, we talked about RDM last week.
• 0:02
  You know, you might see, oh, that's another vault free relay set up that I've got there. Or, ah, I see we've got digital inputs, analog inputs, probes. Yeah, I recognize that from another controller. I've just gotta adapt it to what I'm doing here. So hopefully we'll see the similarities. We'll also see some things that are unique to Carell as we go.
• 0:03
  right. Let's get into it. So we're gonna base this on the NPX Pro. This is pretty standard case controller for Rell that we use on CO2. Reason being, it can drive electronic expansion valves. 'cause we use electronic expansion valves on CO2, right?
• 0:04
  Kevin: so we had pulse valves, as some of you guys will know them, AK vs. Is the, the brand name from Danfoss that everyone over here knows a pulse valve by, because it is the defacto pulse valve. They were around on HFCs for us in the late nineties, early two thousands, and they still around on CO2. You know, it's a, it's a, it's not a new piece of kit Stepper valves have been around a long time, but have really come to the forefront with CO2 for efficiency, as you say, in their controllability.
• 0:05
  Kevin: And it Yeah, absolutely. And it's, if you look at the board layout for that, if you imagine you haven't got the manual, I was just gonna say, get the manual everyone, everyone's got a smartphone.
• 0:06
  'cause you're like, you know, you might get lucky and two pins make a reading and you'll be like, okay, that sensor's reading, but it won't be the sensor you think it is. And some sensors will be not reading at all and you'll get a horrible mix of things or. Crazy things like at the end there. Just to jump to that, that's the communications plug.
• 0:07
  What the resistive rating of that relay is. You know, IE what amount of heaters or similarly connect to it, what the inductive rating is. If you wanna wire a compressor straight to it or a motor type load, it will tell you all sorts of things down this list. If you move down to digital inputs and probes, tell you what type of probes or what type of analog inputs it can take into that.
• 0:08
  It's gonna make your life a lot easier. And no excuse today, like, you know, no, I don't think so. And it's, it'll it, for the two minutes you spend doing it, it makes your life a lot easier because then even when you're running through parameters, you'll be like, ah, remember H zero does something, what does H zero do?
• 0:09
  Use that relay vault free. You can use it to switch a digital input on something, or you can use it for low voltage if you're using it in the control panel. Or you can use it for main's voltage. If you wanna switch like a compressor or some fans directly, they leave that up to you. The commons aren't linked.
• 0:10
  Trevor: Yeah, that's something important that a lot of people don't, don't know.
• 0:11
  Yeah. So if you've got like a hundred watts of of heater, it's gonna pull the same current Yep. Whether it's starting or been running. So you can get more reliable with that. But good practice, I would say, you know, how much is a, you know, a 10 amp contactor compared to buying a new controller? And how much hassle is it changing a controller versus changing a a relay.
• 0:12
  And if you look at the diagram for the probes on this unit, it'll show you there's a ground for S one, S two and S3 first three probes, and there's a ground for S four, S five and S six. And the reason for that is those probes can be set, those groups of probes can be set to different types and different functions.
• 0:13
  And then when they come to do the probes that control the valve, like the suction probe or maybe the defrost probe, they'll do use PT 1000. They deem it may be more accurate, maybe the type of sensor they want to use. So it's very possible that you can have, with some manufacturers, two different types of probe connected to the same controller, which is perfectly fine for the controller.
• 0:14
  Carole can use four to 20 milliamp. We do quite commonly, the advantage to say we discussed previously as well, is that if you use a voltage trans user like this and you've got an island with 10 cases in, let's say, you can share that one transducer. You'll just take a two wire cable between your controllers and you'll take out the signal wire and you'll take out the ground and you'll join all of those together.
• 0:15
  The, the flip side of it, and again, something to watch for yeah, especially with Carell, is if that cabinet gets a false and you turn it off or it blows up, you've lost the transducer for all the cases. So you are reading will look quite sad. There is this,
• 0:16
  The last resort, if you like, is a setting in the controller called safety, evaporation, temperature, which is Trevor might remember when we last did Corre, was the one we struggled to find in the manual for ages. But yeah, it's, there's a setting called safety evaporation, temperature. If the controller loses its transducer input, it will set a fixed temperature into its evaporation condition.
• 0:17
  And the problem if you, if you leave it at minus 10, is that that is gonna make the controller think it's got a heck of a lot more super heat than it probably has. So it's gonna rack the valve open on that cabinet. If you've got it set for 5K, super eight, it's gonna let the suction go to minus five and you'll probably find your flooding back at that point.
• 0:18
  But it's knowing. What to look for. Look to think. Ah, right When I've looked at that controller, the trans users' wired to that one. You know the other, if you're lucky enough to have the displays on all the cases, you may see an S six fault come up on the other cases. But very common now for retailers not to put the display on the cases.
• 0:19
  And yeah, if you test between, if your trans users is wide to S six, then between 31 and 30, if it's wide to S seven, between 29 and 30. And you'll see the return voltage that the controller is getting from the transducer. And as you say on the, on the five volt supply rail test, it ideally, I mean by all means, test it with, it wired up first.
• 0:20
  So take the wires out, check the circ, the outputs with no wires connected, and then you'll know whether your fault sits in the controller or sits in your cabling. 'Cause it's, it's real easy, especially in retail, things chew on cables and like we, we know things live in stores that chew on cables and it's, it's just as likely to chew through it and short it as it's to chew through it and open it.
• 0:21
  When we've, when we've got blends, you know, 4 0 7 or 4, 4 8 or whatever running around that system, depending on the type of controller, when you put that number in, it's either gonna give you the bubble point or the dew point and we may have to apply an offset to that to get the cabinet to work. And as, as we've discussed before, Trevor, there are many interpretations of what the correct thing to do is there.
• 0:22
  They tell you what they want. Carell lets you decide for almost everything. The only thing it doesn't let you decide is relay three. So it tells you a little bit there about how we might set those relays up. It's very helpful. It tells you which parameter. Which relay as well. So if we get a parameter H one, we're gonna set what Relay four is doing, et cetera, et cetera.
• 0:23
  You can tell the relay it wants to be a light or a fan. It's things that people recognize doing and that's, that's just. To make it easier for everyone and to make it more easy for somebody to go, I wanna throw this controller on instead of using this one.
• 0:24
  So this is what this, to be honest, so thank you, all of you for, for showing up and I see a lot of people in there didn't put their, that been, know a few of these, Brian and few others. Lewis, my man. So welcome.
• 0:25
  And that gives this board a lot of flexibility. And the beauty of it is that nine times outta 10, if you go to a controller that's failed like this, it is not the valve driver board that's gonna have gone. What'll have happened is somebody will have gone the board wet or it'll took an electrical surge, et cetera, et cetera, or a relay's burnt out.
• 0:26
  So do it carefully and don't bend the pins over. I would imagine, and this is, this is the caveat I would imagine, if that controller was under warranty when you took it off your truck swapping the valve driver Bos, it's probably gonna avoid the warranty on that controller. But in, if you've got a, you know, let's say you've gone to a store or a warehouse that's got a freezer down and it's got a hundred thousand dollars of food in it.
• 0:27
  Trevor: before, Kevin. If you build a good relationship with that manufacturer and the salespeople, you know.
• 0:28
  And you can perhaps keep those expansion boards on your truck as well, so that you're not swapping an old part on them. You have got capability to do either straight off the bat with what you've got. Just worth noting on the stepper driver. Boards, when you actually look at it, it's quite easily identifiable.
• 0:29
  PWM board doesn't need it because the PWM valve automatically shuts on the stepper board. It's got a battery. So as soon as you power that controller off the battery, that this big capacitor, as I say, the capacitor will store enough charge that it'll be able to drive the valve shut. So that you, you are not leaving that valve open it and your case uncontrolled.
• 0:30
  Kevin: Yeah, yeah, yeah. Well, and you can tell they're all talking the same language when you've got a super cap and ultra cap and you know, all, all that good stuff. So you can recognize what that part is.
• 0:31
  So if, as you see on that, the lead points towards where the V is stamped onto the controller, if you were to flip that lead 180 degrees and it was pointing in the direction of the E, when you tell the controller to be shut, the valve will be a hundred percent open and vice versa. Now, you might think that is really easy to work out, that's been done, but what tends to happen is that you power the controller up, it starts driving, and once it's in the mid position, it's really difficult to spot that it's happened because if you are commanding it 50%, it doesn't really matter whether you are, whether you're the right way around or not.
• 0:32
  You know, if it's, if it, if it's set for 32 Fahrenheit, turn it up to 60 so that the valve shuts. If it doesn't shut and it goes wide open, chances are you've got the valve wide backwards. So it's really important to remember with valves that you have gotta get the sequence.
• 0:33
  Well, what's going on here? Yeah. Why have I got different colors? And that's when it starts. Mistakes start happening. And cable lengths actually another good one. Yeah, it's mentioned here the step of signal needs accuracy. You, this, this valve has four 80 steps from open to shut on a corral valve.
• 0:34
  So there are specified cable lengths for the different gauges of wire. We tend to always work in truth to, for myself, for no more than about 30 to 35 meters. Corral is saying with thick wire, look there, you can get up to 50 meters. Your chances of getting error increase with length the chance of getting interference in that cable.
• 0:35
  When you come to close the valve, the controller will usually put more steps than it thinks is necessary because it knows it's gonna have lost a bit of position. So if your valve's got four 80 steps, if it's going to full closed, it'll often drive to 500 or 520 steps because it's, it's absolutely fine for it to touch the seat on the valve and just push it a little bit because at some point you need to, the valve needs to know that the calibration's right, so it gives it a little bit extra in a safe place, which is shut to make sure that it hasn't lost steps.
• 0:36
  So be aware with stepper that what the controller thinks it's doing is not always what the valve thinks it's doing. It's there's, there's no feedback. There's no mechanical way that it will know.
• 0:37
  This comes from people's knowledge of solenoid valves and pulse valves. If you pull the coil off, that doesn't change the valve position, it'll just go shut. It's only got two positions open and shut. If you pull the cable or the coil of a stepper, as you say. It'll still keep pulse in it. It doesn't know it's not moving the valve.
• 0:38
  It, it can be the case, especially with the smaller valves, that the mechanism can get jammed up by a tiny bit of debris. Yeah, and you'll, you'll pull the filter off and go, that valve looks perfect, but you'll try and drive it. You'll see it just go, it can't move because something's stuck in the screw and you won't know that with the valve fitted.
• 0:39
  Trevor: sorry. Yeah, before we dive in there, I wanna go back to cable lens. I've seen, I've been at sites where there are 250 feet, you know, almost a hundred meters long, but you have to do a proper calculation. And like Kevin said, there can be a lot of interference.
• 0:40
  And I was, I was working on a, the contract, the technicians were working on a warehouse and I was fitting them with ex valves, and it was like, yeah, 200 feet. And then I had to convert for meters to the feet. I'm like, well, we're way over. And so you have to go in to do resistance calculation to make sure that it's gonna be spiting out, that exact resistance at that time.
• 0:41
  Stepper valves are a different ball game on that. You've gotta, you've gotta account for that in the cable length and the thickness of the cable. So, yeah, just, you can always speak to your sales rep or the, the technical team at whichever controller manufacturer. And if you say, I've got 400 feet of 80 gauge wire on this, they may say okay.
• 0:42
  You can get a 12 watt coil, you can get an 18 watt coil, but depending on what mechanism that board's going use for pulsing that valve, the track coils on pulse valves generate a lot of inrush current. The, the current they consume when you first pulse it, when the magnetic field builds is much higher than the running current.
• 0:43
  So if you've got a huge pulse valve that wants a 25 watt coil or a 30 watt coil, it's not gonna do it for you. The, the other rating on there that isn't usually a problem with our applications, there's also a minimum drive on it, and that's gonna be what enough to switch the transistor. You need enough current through it.
• 0:44
  Trevor: Why is that that important? 'cause I've seen people before say, oh, well I got a transformer here. I can put these five controllers on this one transformer that's 50 va, but you need 25 VA for each controller. And then you got some valves that are working when you first start up. Oh, everything looks like it was working, but when it starts to run, you don't get enough juice to push the valves over to.
• 0:45
  So the beauty of using Volt three is that the controller gets a nice clean power supply from, let's say an isolation transformer, and you can use the dirty supply that's ev all the other mechanicals are connected to, or your contact us and everything else to be switched through your relays. And that way, firstly, you don't overload the supply that's feeding the controller, but also you keep that supply clean, it doesn't disrupt the electronics.
• 0:46
  So it's really important to to account for that. Because they do take quite a bit of power because you're lifting quite a heavy magnet at the end of the day. Yeah. I wanted to talk a little bit about when we actually get into setup, and I apologize for the maths on a Thursday evening or Thursday daytime.
• 0:47
  Ah, it doesn't happen always. So years back, somebody came up with the idea of what we've always called CPT, calculated product temperature, which is taking into account some of the supplier, some of the returner. Now, as Carell, nicely tell you there, one of the advantages of doing this when you control purely on supplier, let's say minus three, let's say we've got a meat case and we've set minus three degrees c.
• 0:48
  CPT, if you use that formula, allows you to account for that because it drops the it drops the point at which the control will operates. So if you, if you've got an air on that's come down, a return air that's come down at night, then it will satisfy easier. The, the control will stop at, at a higher air off because it's a ratio of the two temperatures.
• 0:49
  It's basically midway between the two numbers. If it's 50 50, if you are taking an air on of one temperature and an air off of another, and your ratio is 50 50, your sh your CPT, your virtual prob I'll call it, is right in the middle. But the same formula works with any ratio you like. You can say, I want you to pay more attention.
• 0:50
  It's not as complicated as it looks. All it tells you is you've gotta take the percentage of one, you use the one value using the percentage of the other, and divide, divide it by a hundred and it will give you the value. So
• 0:51
  Why do I worry about superheat when I'm maintaining a proper temperature? And same, same with this, like now I got night. I got day. So I can run my system differently in the different stages of that to save energy. And we're gonna see a massive shift. I, I see it coming back, a lot of floating suction, floating head low or increased su lower superheat set points at certain times to really fine tune a system.
• 0:52
  they can't simulate that easily. They can create a draft, they can force the doors if it's a door cabinet to open and shut, but it's really difficult for them to simulate the real conditions in a store.
• 0:53
  Kevin: on Yeah, leaning on the door. Yeah, yeah, exactly. Exactly. Yeah. And you know, it's, they'll do, and, or they'll pick up a box out of the freezer, turn it over, have a look at what's in it, check the price, the doors open all the time. And so you tend towards more the return out.
• 0:54
  Yeah. And we'll probably go over a few minutes on the couple next slides. I don't mind if you've got time, Kevin, to, to Yeah, of course. Think this, this is important here. 'cause I get in all my supermarket training programs in my CO2 training programs. Like whoever, I don't understand PI PNID. Should I be changing it?
• 0:55
  So we're back to getting to temperature, switching the thermostat off, letting the case temperature rise, switching the thermostat on. But that may, that every time you switch a case on and off, it's a little bit of disturbance in the rack pressure. So we want stability for energy and one of the ways we can achieve that is we can tell the controller, you get near to set point.
• 0:56
  So the change that the rack sees is much smaller. Now. One case going on and off in 50 or 60 attached to a rack isn't gonna disturb the rack a lot. If you've got 60 cases all doing this at the same time, switching on and off, you might have 10 or 15 cases that are coming on and going off. If you can stop them from going through that process, it does smooth the rack performance out.
• 0:57
  Now I'm gonna stop adjusting the super heat. And as I say, if it moves, starts moving back the other way, the controller's gonna say, I'll let the super heat come down a little bit again. And instead of doing that up and down a lot, you'll just move up and down a little bit. There is a downside in some cases and literally on some cases, and that is if you, when you back the super heat up and you make the super heat big, what you are effectively doing is using less and less of the evaporator to absorb heat.
• 0:58
  So you have to apply some caution to where you use this. If you've got challenging environments or damp environments, you've gotta be careful with using this. We used to see this a lot with blends that had a high glide because the temperature at the back of the coil is really cold anyway compared to the rest of the coil, and you would build a heck of a lot of frost at the back of the coil.
• 0:59
  CO2 isn't as bad for it because the temperature's stable across the coil. But use, try the facility out. If you've got problem with your, your stability in the store, try the facility out. But just maybe try a, a relatively small band. So you might find that 50 or 60% of the time it stops the case satisfying on temperature and sometimes it will.
• 1:00
  Kevin: know, so, yeah. And, and Danfoss does this automatically to a degree, it, with Danfoss we've talked about before, you have a super peak minimum and a super peak maximum.
• 1:01
  Kevin: So, PID is a way that we get electronic systems to respond to changes in load changes in conditions. And all it is, is a mathematical formula. You feed it some numbers and you tell it how to react to what it sees coming in. It will have a value that it's trying to achieve and it will look at the value that it's at and it will be like, how do I get there?
• 1:02
  But it breaks down relatively simple. So your P, which is your proportion, is all that means is how far am I away from my set point? Am I a little way away? Am I a long way away? And it sets the speed or the amount, not necessarily the speed, the amount the valve responds or the motor responds or whatever the object is, to being a distance away from the set point.
• 1:03
  So what I'm gonna do is I'm gonna start adding more and more response because I've been away from the set point, even though it's only a little bit, I've been away for a long time. And so as you get closer, your proportional gets less, but your integral gets more and they, they both work together to try and get you to the exact set point.
• 1:04
  So if your, your set point is here and your temperature is here, you might go boom past it. And what happens with, with proportion, you tend to oscillate. You'll go either side of it. So what derivative looks at is it will say, I'm moving really fast, so I need to calm down a little bit. And it will watch the rate of change and slow your rate of change down.
• 1:05
  And it's a really good way of demonstrating how PID works in practice for us. Most case manufacturer has controller manufacturers. Have PID settings that will work in 95% of circumstances. Yeah. If you are fiddling with the PID, you've got an unusual set of circumstances going on with a case. There will often be times when we talked about response time on RDM, all response time on rack controllers is doing an RDM is adjusting the PID, but it's doing it in a way that RDM are managing internally.
• 1:06
  So the characteristics of the evaporator, even if they haven't done that, the case controller manufacturer. Puts in some settings that are gonna be, you know, goldilock settings. They're not too fast, they're not too slow. So don't get to PID as your first thing if your case isn't stable. Because the chances are if your suction pressure's relatively stable, your probes are all working, your fans are running and you're not iced up.
• 1:07
  Trevor: need. Yeah, so now that's a big thing. So I've had people in some of our course, the CO2 course, and we're going through the case controller and if the, if the P and i and D are changed, I ask 'em like, and it's still not working, just put it back to manufacturer settings. Start, start from scratch.
• 1:08
  And you've always gotta remember if you're gonna a service call on a case controller. The chances are that case controller was working one day ago, two days ago, three days ago, and it was working just fine with the settings it had. So what's different about the day you've rocked up? Nobody's probably changed the settings unless your colleagues' been there the day before and it was working before you got there.
• 1:09
  But then they'll put it back into operation and they'll be like, why the heck doesn't this case work now? And it's because somebody's tried to get around a mechanical fault with controller settings. A lot of retailers especially ask for their controllers to be locked down. At this point, I'm sure we're all familiar with that.
• 1:10
  Kevin: yeah. What, look at when it was working, go back on the graph to when it was working. See what's different about it. Unless somebody's altered a setting, virtually never be solved by setting change. If you've, the exception would be when we're talking about CPT before, if you go into a store and they say, the meat's gone brown.
• 1:11
  This is the last slide, just a little bit on defrost. The left hand side is just a really useful diagram that Carell gives you for what all your outputs are doing at a given point in a, in a defrost cycle, for example. So it will tell you what the defrost relay's doing, what the fan relay's doing, what the EV's doing, every point in a cycle.
• 1:12
  Problem with that is that people defrost a lot and they have a long recovery time, long defrost time. They don't set a termination temperature. They let it run to time. You think about your numbers for that. If you've got six defrost a day in a cabinet, 45 minutes each, and then you've got a 30 minute recovery delay.
• 1:13
  So just think about the total. It starts off really great because it'll come outta commissioning. Let's say the manufacturer wants 16 frost. It'll come out with six 30 minute defrosts. The case allows at once because of an external factor. Somebody will look at 45 minutes. And then something else will happen.
• 1:14
  So slight negatives. That's that coil. And then now all of a sudden you need to get that frost off there. Now you're, you're gonna run it longer. And then same thing I've seen terminations at 72 degree Fahrenheit, like almost 20 something like what your product's going bad every time if it gets up to that high, you know?
• 1:15
  So you would see that shelf temperature, probe alarm, regardless of whether it was in defrost or not. Not by all means, by no means does every manufacturer do this, but it's quite a nice way of ensuring that you're not cooking the cabinet. But as Trevor said, you know, the primary aim of that cabinet is to keep the food cold.
• 1:16
  People don't buy it. They leave it there because it does not look good. And same with the wilting of lettuce or thing. Go talk with them and build a relationship with them and they'll tell you, this hasn't been working very good for the last few months. I've been, my products only lasting here for two days and it used to be four days or three days.
• 1:17
  You can't shift that cold air off the coil. It's not keeping the product cold. You're gonna hurt what the evaporator iss doing. Just don't ignore the obvious and go straight to eye eye stuff and must need more defrosting. Even, you know, passing solenoid valve, you're never gonna solve a passing solenoid valve by putting more defrosts on.
• 1:18
  So, so it was causing it because it was a different time, different length. And that was that and that, that, and I bet I was like, I bet you if you go check the logs two months ago was the same thing. He was like, yeah, when I checked it was like two months ago. Somebody was there and washed a case, and two months before that somebody came wash the case.
• 1:19
  It's not as nice as adjusting the parameter, but eventually somebody's gonna have to do that work. The, the very last point I just wanted to make on this is staggering the defrosts across time on your rack. The other problem with having six or eight defrosts a day is that you might have 40 cases on a rack that have all gotta go on defrost in a four hour period.
• 1:20
  Trevor: Yeah, a hundred percent. And you, this is like a major and a hot gas defrost setup, ? Because you just don't have, you don't have the heat, you know, if you go and try to put 60, 60% so it's like one, one big case and then you gotta stagger, like you gotta walkin box, big walkin boxes, glass doors.
• 1:21
  Trevor: I love this. Kevin, we got a couple questions in there that we'll try to Yeah. Hit up pretty quick. Yeah, sure. Would you recommend installing an additional relay for each digital input and output in order to provide protection for the controllers?
• 1:22
  Anything that pulls a significant amount of current, it's worth putting a relay on. A good example these days is LED lights, LED light drivers have a whacking great startup current, huge inrush current, and that is really bad for blowing breakers, but it's also really bad for relays. So things like LED lighting Absolutely.
• 1:23
  What will happen is theus though, and you'll generally be fine. So it depends on your situation.