Sign in to follow this  
Followers 0
av8or1

Deciding on a controller

69 posts in this topic

Just out of curiosity, what are you attempting to see with the photo-eye, and what do you plan to have the program do when it is blocked or clear?

Share this post


Link to post
Share on other sites
I have the sensor wired to operate dark (O.D.) so that the sensor will "close the switch" and send 24 volts to the PLC when the beam is broken. The program turns on a light. Or off, depending. Fell dead asleep last night after returning home from work. Long week and I spent several nights playing with this project until the wee-hours, so...will resume work today and update the thread with the findings of the experimentation. Thanks, Jerry

Share this post


Link to post
Share on other sites
Ok I played around with the sensors a bit this afternoon. And I don't have good news to report thus far. A bit of background: My sensors are photoelectric polarized retroflective, AB cat #42EF-P2MPB-F4. The reflective tape that I have is AB cat #92-99, so as it turns out I don't have the lowest reflectivity tape after all. This tape is advertised to be at 77% capacity of the "ideal" AB cat #92-39 reflective discs. Finally, I re-measured the floor-to-ceiling height and I remembered slightly incorrectly: it is actually 8' 3", not 8' 7", which should be to-the-better, albeit by only 4 inches. I digress. To conduct the testing, I created a reflector of my own in an attempt to replicate the size of AB #92-39 (3 in) by cutting the tape into strips, removing the backing tape and placing them on a piece of (flat) cardboard. I then connected one of the senors, taped it to the ceiling (temporary measure) and began playing around with the cardboard. At distances of about 5 feet or less, I could get good reception with acceptable margin. However at the longer distances I couldn't even establish reception. It was still daytime, so it was difficult to see where the laser beam was at the longer distances. Therefore I will try again tonight to see if I can make this work. It should be easier to spot and play around with the emitted laser beam at night. That said, I am not hopeful. I bought these particular model sensors at the suggestion of an AB representative who listened to my application plan and then made this recommendation. However the distance simply appears to be too great to affect a system with respectable reliability. As I mentioned, I am willing to give it one last try, but I suspect that I will have to make a run to the AB distributor on Monday during lunch and exchange these sensors for a different model. I think that the AB Series 9000 will be more suitable for this longer range application. They're a tad more expensive, but would seem to offer a degree of reliability that the RightSight sensors just cannot muster, at least not with the reflective tape, which is the only suitable reflector in this application. Will update again later tonight. Any feedback is appreciated. Thanks! Jerry

Share this post


Link to post
Share on other sites
Ok we have confirmation: these sensors just won't work in this application. The maximum range where I can establish both a recognized reflected light beam and have sufficient margin was at about 7 ft. Anything beyond that and I could barely get the sensor to find the reflected light, let alone a decent margin. And that is with a 5" x 5" reflective pad (there came a point where further enlargement didn't make sense and wouldn't work, but for the sake of completion I did it anyway). So this will set my project back several days, but on the positive side it has been a learning experience. I've discovered that I don't like sensors. Nah in all seriousness I knew at the outset that the achilles heal of this entire project would be the accuracy and reliability of the sensors. Without that, the remainder of this project is pretty much pointless. The sensors were and are the most vital component in the design. So it's back to the drawing board. On Monday I'll give my local AB distributor a call to see if I can stop by to return the RightSight sensors that I have and then pick up some of the Series 9000s. I'm in a suburb of Austin, the local distributor is The Reynolds Company and they just did an office move. Supposedly they'll be ready for business on Monday and I hope that proves to be the case. I'd like to finish things off and move on to some other projects that I have waiting in the wings. One of which is to work on this Ethernet adaptor stuff so that I can tweak the Micrologix program from the study if need be, as Bob suggested. And that will indeed be the case, as I just completed a re-wire of part of the panel tonight to include one more output (buzzer/flashing light). Naturally I need to add a couple of rungs to the program accordingly. So does anyone have any experience with the Series 9000s? Thanks!

Share this post


Link to post
Share on other sites
Two suggestions: 1. Perhaps a Photo-eye prox might work in your application. These can be found with a range out to approx. 9ft (Telco or Banner approx $140) and they only need a target such as a car or something to block them. 2. Some type of motion sensor might work for what you are trying to do, but they might be hard to find in 24 V. Another option would be to use a 110 VAC motion sensor that you can probably find at Lowes or Home Depot and wire that back to a relay to trigger an input on the PLC.

Share this post


Link to post
Share on other sites
johnboy -> Thank you for the feedback! Not sure what #1 is, if to be open, but a max of 9 ft seems a bit concerning to me, based on this experience. That is because it's right at the edge of the range I need and I wonder how well it would work at such range. One of the original thoughts that I had was to use a motion sensor, but all of the ones I found had too wide of a sensing area. I need something that is fairly pin-point in terms of accuracy and don't want a wide beam by the time that beam reaches 8.5 - 9 ft. And so I dismissed motion sensors relatively quickly. Good thoughts though! UPDATE: I returned the 42EF RightSight sensors today at the Reynco office, no problem. Need to purchase a different sensor now or else this project is dead. As I mentioned, I have been considering the AB Series 9000s. I decided to take a closer look at things just for fun, when... After doing a little more research I discovered the RightSight 42EF-XXXXX-XX sensor's laser cousin, the RightSight LaserSight 42EF-XXXXX-XX. These boast a larger margin and longer sensing distance than LED based sensors. Specifically the one I am considering is the 42EF-P8KBC-F4, which is essentially the same sensor "package" that I originally purchased, just using a laser instead of an LED. The RightSight LaserSight claims a maximum sensing distance of 15 m/49.2 ft, with a maximum spot size of only 16 x 20 mm at maximum distance. Much more what I need in this application. This particular model will also plug-right-into the 4-pin DC micro QD cordsets that I have, and of course the wiring is identical. It does have the teaching function and appears to be D.O. only, which is what I had the system to be configured anyway, which will save re-wiring. The two are priced similarly, with the 42EF RS LS being a couple of dollars more expensive per unit than the Series 9000. So I plan to call and speak with AB during the next couple of days while I research it a bit further prior to making a decision. I'd like to avoid having another go-around with this if possible, though again, I have learned a significant amount, which is always good. Anyway, I'll stop now. Keep you posted. Thanks! Jerry

Share this post


Link to post
Share on other sites
Is your target that will be blocking the Photo-eye at floor level? If it's a few feet off of the floor, perhaps a prox or diffuse type eye might work and you won't need the reflective tape to bounce the signal back to the eye, or you can mount the eye on the wall to detect the target. Of course if the Laser eye is within budget, go with that since the beam will definitely be smaller.

Share this post


Link to post
Share on other sites
Hmmmmm .... well "no, not really" seems to be the answer I'm coming up with to your question. The target is not right-at floor level, no. What I am trying to sense is a vehicle entering the garage. So the height will depend on the vehicle in question, varing from just a foot or so (front end of sports car) to 6+ ft (top of SUV). If you know a fair amount about prox sensors, I'd like to learn more. Unfortunately all of the proxes that I researched (inductive) simply didn't have the range I needed, usually max-ing out at about 4 feet or so. I also looked into the diffuse option, but the AB representative told me that object color would play a factor in that scenario, as would general reflectivity of the object being sensed. Therefore I opted out of the diffuse stuff. I don't really like the reflective option because of the need to place a reflective material on the floor, but it seems (and that is the keyword 'seems') like the only really viable option in this application. And you're dead-on with the budget call, but heck, I blew that some time ago. Now it's either finish or else flush the money I've already spent and cannot recoup down the drain, which I am not inclined to do. Thank you for the continued feedback, I appreciate it! Jerry Edited by av8or1

Share this post


Link to post
Share on other sites
Jerry, Your AB Rep was right about the diffuse PE's, usually if you set them up on a light colored target, they have a hard time with a dark colored target and vice-versa! However, Telco has a prox (P/N SPP2903C $100.35, w/ cable P/N 80459A $20.27) that will see out approx 3m and is good at seeing light/dark targets. Take a look at that, and if you don't think it would work, then maybe a Telco (or I prefer Banner) ultra-sonic sensor would work, but then you're looking at $400-$500. Another option would be a see-thru eye which can have a range up to 200 ft or more. With this option though, you would have to set up a transmitter and receiver, probably on the walls. My primary occupation is industrial automation, but I do like to play around with home automation as well. If you tell me how you plan to block or trigger this eye (I'm assuming your car but you said that a motion sensor is too broad), maybe I can come up with another idea as well. Good luck, John

Share this post


Link to post
Share on other sites
Hi John, Well I appreciate the feedback from an IA professional. There's been a considerable amount of valuable guidance on this forum, so thanks to you and all who have contributed, I appreciate it! I will look into that telco prox. The laser sensors that I am considering are a bit cheaper at ~$90 each and I already have the cabling in-place, but if I end up needing to buy the micro-cube reflective squares for them, then the price difference would be a wash. The transmitter receiver would be a great choice, but not practical in this application, as the sensing - according to my design at least - needs to be done in the vertical plane, which would mean that I would need to string out the receiver along the garage floor, which means exposing cabling to the pressures associated with being treaded upon by humans, vehicles and lawnmowers, not to mention those pesky pets who like to "play" with those types of things. So I need something that emits/receives a beam or that can just sense objects without the need for such emission/reception. All options that I investigated for the latter design choice just didn't seem viable in the "real world" and that is how I arrived at the selection of the AB 42EF sensors. The basic plan here is to have sensors mounted in/on the garage ceiling. As the vehicle enters the garage, these sensors detect the vehicle's entry progression WRT vehicle position within the garage. Motion sensors emit too broad of a range for this to be a viable option, IMHO. I need to know exactly where the car is to within a greater degree of accuracy than a motion sensor can provide. This is another reason that the AB 42EF sensors seemed to be a better fit. Ok so you're probably wondering why in the heck I care to such a detailed degree. Well I had wanted to keep the wraps on things a bit until I had the project finished, but in the interest of implementing the best design ... this is a parking aid system. A bit of background seems appropriate. My mother had complained of difficulty in parking her Jeep in my garage when she came to visit. Furthermore she expressed a general concern/worry for not wanting to hit anything. So I mentioned the usual stuff, tennis ball on a string (which I didn't like), stop sign stand, etc. She seemed luke warm to those suggestions and so one day I made the mistake of saying "well then how about I just hang a traffic light in there for you, would THAT help?" To which she replied "oh you could do that? That would be wonderful!" .... oh boy ... I hadn't intended for that suggestion to be taken seriously. But too late now. And so the story went. After some investiagtion I learned about PLCs and I decided that I would prefer to use one in this implementation rather than to solve the problem "electrically" with relays and such. I learned ladder logix software in about 30 minutes and had my program designed, running and tested in about an hour. I've augmented it since then, but didn't need to make any fixes. Ladder Logix is kinda cool in a way, but the programming was rather easy really; it's been dealing with the hardware issues that has been the difficult aspect of this endeavor. Ok short version ... These sensors need to detect the vehicle entering the garage in reference to an exact point on the floor of that garage. Of particular importance is the sensor that will illuminate the red right, meaning to stop. This must be fairly precise, as it will govern the vehicle's ultimate position in the garage and should be exact enough such that she can close door behind her without worry that it will strike the back of her beloved Jeep. So in my estimation motion sensors just won't hack it. Also, I have developed a way to make the sensor adjustable so that different vehicles can be parked in the same spot, which furthers the need for precision and that complicates the notion of using diffusion type sensors due to the varying color of vehicles that could park there. Anyway. Sorry for the novel. That is what I am up against. The project has gone way over budget, but it's for my mother and if it gives her some ease of mind when she visits, then hey, all to the good. And I've learned something in the process, so ... Ok I'll stop now. Thank you for reading. Jerry Edited by av8or1

Share this post


Link to post
Share on other sites
One thing you might consider is a light curtain type device that could be mounted at the door. Maybe one that's about 2' tall. It would need to be mounted at a height to "see" the rear-most part of all the vehicles (bumper probably). When it's clear, turn on the red light and allow the door to close. Unfortunately, this is likely going to be pretty expensive. I put together a quick "system" of a 16" transmitter-receiver pair with cables for about $660 at AutomationDirect.com. Another alternative may be to use some sort of ultrasonic distance measurement sensor. It may have to be aimed at something flat to work, though. If most of the vehicles are about the same size, you could mount the sensor on the back wall of the garage and point it at the front license plate. This may be complex, though, since it will stop the cars so the front is at the same spot every time instead of the back. A longer car than expected could cause problems and a shorter than normal car would have a lot of extra space at the garage door, which may or may not be useful for you. Check these out: http://www.automatio...nsors?learnsite Some of these can reach out to about 8'. Note: I've never used any of these particular devices, so I can't speak to their ease of use or reliability. I have, however, had generally good results from Automation Direct products. Edited by Joe E.

Share this post


Link to post
Share on other sites
Hi Jerry, Well I would definitely say that your situation is unique, but should be do-able none-the-less! Maybe a couple of low-tech approaches would work: 1. Depending on how close the Jeep is to the wall, you might could mount a limit switch on a piece of uni-strut with a homemade wand that would activate by the tire. This would be adjustable on the strut for various vechicles. 2. A vacuum switch on a piece of hose so that when the tires ran over it, it would activate the switch. (Similar to the full-service gas stations that would ring a bell when a car drove over the hose.) 3. I believe that there are pressure mats that can be purchased that activate a switch when a forklift or vechicle drives on them, I'm just not sure about the cost. In the meantime, I'll keep working on other solutions. I'll let you know if I come up with any!

Share this post


Link to post
Share on other sites
For what its worth, I would look at Ultrasonic sensors. Units made by Banner Engineering have the capability of reaching out around 25 feet. I have used these sensors successfully in applications where I am trying to detect large lift trucks.

Share this post


Link to post
Share on other sites
All- Wow, thank you for the feedback, I appreciate all of your ideas. They are all quite sharp in their approach! With specific regard to the notion of employing ultrasonic sensors, I investigated that option during the project's early days but didn't feel comfortable with some of the literature I read regarding the "beam width" associated with that type of sensor. There were claims of it being narrow throughout its range, but I couldn't see how it couldn't be interfered with. And so I was concerned about getting false positives, though I could have been off-base with such thinking. I read some more tonight about it and pretty much arrived at the same conclusion, though I learned more than the first time out. Also, the only ultrasonic sensor that had the range and narrow beam width was 48VDC, which would mean some re-tooling of the current system to make that work. Naturally, I could have missed others that are 24VDC and there were AC options but that too would require re-tooling of the already-existing configuration. I'm not opposed to that, but if I could use what I've already risked life and limb to run in the never-reaches of my attic space, I'd prefer to go that route if possible. So I called AB again today to get more information regarding the RightSight LaserSight sensors. My key question involved the reflector: what type of reflector would I need in this application (3 - 10 ft) to make that sensor work? The basic answer was that if I didn't go with the recommended reflector (AB #92-118) then using the 8.5" x 11" reflective tape (AB #92-104) would work with a laser sensor and would likely yield acceptable margin in that range rather easily. That said, the reflective tape that I currently have (AB #92-99) will not work with a laser sensor at all, so it would be a throw away. At only ~$9, no big deal, but what I didn't and don't want to get into is to have these little reflectors that are like miniature hockey pucks that are raised off of the garage floor. I could just see them getting crushed by the vehicle at some point, then I'd need to clean up the mess and buy a new reflector. And depending on how often that phenomenon occurred, that could get pricey. Reflective tape on the other hand will ahere to the floor with almost no raised section, so if it is run-over, no problem, just clean it off if you need to and keep going. The AB engineer also confirmed the answer to my question I had when considering #92-104: "Could I cut this sheet-of-paper sized reflective tape into smaller pieces?" It seemed like an obvious 'yes', but the issue that often rears its ugly head to bite you during implementation and testing involves the one question that you didn't ask, so I decided to ask anyway. That said, I plan to give it another day or two of research before making a decision. If I had to decide right now, I'd go with the AB option. So we'll see. Again, thanks so much for the input! Jerry

Share this post


Link to post
Share on other sites
Bob, Thought about you last night when I drug my PC, monitor, keyboard, mouse and power cord up into the attic so that I could add the buzzer logic and change one of the outputs to a different channel. Because of cord lengths I had to do some interesting contortionism with both me and the components in order to make the modifications too. It was one big mess of stuff running this way and that. Interesting. And then take everything back down again, of course. Having something as you describe above sure would have been nice. And FYI ... still debating/researching the sensor thing. Thanks, Jerry

Share this post


Link to post
Share on other sites
UPDATE: Ok after a good long discussion with a couple of AB reps regarding sensors (prox/laser/ultrasonic/etc) I have decided to go with the LaserSight RightSight sensors and use the 92-104 reflective tape. None of the local AB distributors (Dallas, Houston, Austin) have either in stock, so they'll have to come direct from the factory. This means it will require 3 or 4 days to arrive. Feedback appreciated. Thanks! Jerry

Share this post


Link to post
Share on other sites
I've been thinking about this a little. Where are you mounting the sensor? At the entrance to the garage or further in where you want the front of the vehicle to end up? At the entrance end would keep you from closing the door on the vehicle, but may allow a longer vehicle to go too far in and bump something. At the inner end will make sure you don't bump anything, but may leave a longer vehicle sticking out for the door to hit. If the arrangement of your garage door allows, you could do both but you would need 2 sensors. Control the "stop" light with the inner one and disable the garage door with the outer one (or maybe flash the "stop" light). Interesting problem...

Share this post


Link to post
Share on other sites
Hi Joe, Well there are several possible solutions to the problem really. In my design, I am using an actual traffic light that was once in service somewhere in the USA. It's the old-school type that illuminates with simple 60W light bulbs running on 120VAC (as opposed to LEDs). Specifically, my system will involve 6 sensors in total, as it will provide both longitudinal and lateral guidance. I will have 5 lights to provide such guidance, and 5 of the 6 sensors will match 1-to-1 with those lights. The standard red, yellow and green will handle the longitudinal axis. Two additional lights in the form of red arrows will do the lateral stuff. The sixth sensor exists solely to monitor the state of the garage door (open/not-open). To address your question of vehicle length, my system keys on the red light sensor (or in your words the sensor at the "inner end"). That is, the system can manually compensate for varying vehicle lengths via the red light sensor's mount (adjustable). In this way, you can move the sensor closer towards the inner wall for a longer vehicle or towards the garage door for a shorter vehicle. Make no mistake about that however, I am not attempting any form of automated length adjustment. Rather, I am mounting the sensors via standard-issue track lighting such that all one would need to do is to pull the vehicle in question into the garage, make note of the ideal parking location, then simply get a ladder and manually move the red light sensor by hand such that the laser beam is broken right as the vehicle encounters the desired stopping location. Or slightly before that actually, factoring in a driver reaction time of some sort. Finally, this same means of adjustment will be applied to the stray-left and stray-right sensors, while the green light, yellow light and garage door sensors will remain in a fixed position. A couple of pictures of the assembled traffic light are attached for reference. I completed the final wiring of the light last night. Will do a rudamentary test tonight and maybe post a video of that test. Just waiting on the sensors now really. Hope to get them by week's end and finish this thing off by the weekend. Thanks, Jerry Edited by av8or1

Share this post


Link to post
Share on other sites
This is a cool project! Thanks for sharing it with the board.

Share this post


Link to post
Share on other sites
Thanks Joe! The traffic light passed regression testing last night, so all I need now are the sensors. Hopefully by Thursday or Friday. 'Would like to be finished by the weekend. We'll see. I made a progress video too, but had a mysterious difficulty in uploading it today. Have to go back and look at that again. My camera is getting quite old now, so it's just as well. I might need to re-shoot after getting something new. Will post when I have something viewable. Anyway.

Share this post


Link to post
Share on other sites
I just picked up a Nikon Coolpix L120 {14 MP with 21X Optical zoom} for under $250 for work. Pretty sweet rig.

Share this post


Link to post
Share on other sites
That does sound like a nice setup. What is the storage media used? (mine is 8mm tape - if you can believe that! Early 90's vintage stuff) Here is a condensed version of the progress report, not sure it's worth posting here, but I thought y'all might be interested to see that I am at least making progress. Edited by av8or1

Share this post


Link to post
Share on other sites
Secure Digital Card - my 8GB card holds 2000+ pics or 30 minutes of 720P video

Share this post


Link to post
Share on other sites
Very Cool! Great job and Thanks for sharing.

Share this post


Link to post
Share on other sites
bob -> Great video rig! I'll have to look into that in a few weeks when I go to find a replacement. gbradley -> Thanks! Glad it is received rather favorably here. I was a bit concerned at first regarding posting about the project at all, thinking that the experts here wouldn't be interested in a home project like this. But I am glad that I did. The advice received in this forum has been tremendous. UPDATE: Well the hopes of completing the project this weekend fell by the wayside on Friday. I returned home from work and became excited because I noticed a brown box outside my front door. Went to pick it up, all prepared to test and implement the sensor "front end" only to find that the box contained the reflective tape (AB 92-104) but no sensors. Wha-? The local distributor is quite good at getting you next day delivery when you purchase from them, but for whatever reason neither the tape nor the sensors (AB 42EF-P8KBC-F4) were available in the Austin, Dallas or Houston offices (Reynolds Company). So they had to be shipped directly from the AB factory in IL. Not sure at this point if it was an oversight or if the two component types are being shipped separately. I'll call the local distributor tomorrow and see if they can track it down for me. Anyway, maybe by early this week I can finish up. FYI. And one sidebar question if I may. In the heat of the typical TEXAS summer it gets rather warm in my attic. I mean really warm. So I came up with the idea of installing some PC fans in the enclosure to help keep the components cool. However in reviewing the literature it almost seems like it would be unnecessary, as the maximum temps for the controller and the power supply exceed the max temp that I would suspect can be reached in the attic. Still, I decided to go ahead and implement something since I - admittedly and embarassingly - didn't exactly know how PC fans worked. So I researched it, learned a fair amount and developed a design that would work somewhat well. Not ideal, but ok. So the question for you PLC experts is simply this: in your estimation would supplemental cooling like this be necessary in this design (enclosure in an attic)? Have any of your designs included some form of a cooling system? Thanks again, Jerry Edited by av8or1

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0