The gist of the article is about unapproved substitions, and I agree 100% with the author. An unapproved substituion is one reason for the KC Hyatt skybridge collapse we discussed several months ago.
The specifics of the article bring up a slightly different question. Should color be a differentiator? I know fully well the color standards for wiring, use it in my control panel designs, even agree that such a standard makes a ton of sense.
But what about color blindness? Somewhere between 5%-7% of the male population has some form of color blindness. For those, labeling and positioning are the only way to differentiate wires or controls.
@TJ McDermott: Being "colorblind" doesn't mean being unable to see colors; it just means that it's more difficult to differentiate between certain colors.
I have deuteranopia, which is one of the most common forms of colorblindness. While most people have three types of color receptors (red, green, and blue), my eyes only have receptors for red and blue. Since I don't have specialized receptors for intermediate-wavelength colors, they don't appear as distinct to me as they probably do to you. For example, the colors gold, light orange, and light green all look pretty similar to me.
This site has some good pictures that illustrate how colorblind people see things.
You're definitely right about labelling wires. Whenever I have to deal with color-coded wires (red and green; blue is not a problem for me), I usually have a non-colorblind person tell me what color the wires are. Then I attach small pieces of tape labelled "RED" or "GREEN."
I suppose that if someone wanted to make my life more difficult, they could either lie about the wire colors, or else switch my pieces of tape when I'm not looking. Thankfully, I've managed to get along well enough with my co-workers that it's never happened.
Interestingly enough, a small percentage of women are said to be tetrachromats, meaning that they have four types of color receptors instead of just three. That would make them better able to distinguish between colors than regular people.
Interesting topic regarding color differentiation. Regarding wires, you just can't be in industry very long without running into stuff that is not color-coded according to traditional colors (ex. red = Vcc; black = GND). Almost like whoever manufactured it was operating from an entirely different system but in reality was probably using wire that was most readily available to them. What is even worse is if they use a different gauge because they had the "right color" in that gauge - especially if they go the wrong direction for the current rating! I wholeheartedly agree that production should seek an ECN before arbitrarily making any changes, but it depends on the specific production facility and management as to whether or not that will be enforced..
Getting back to those folks who are colorblind, I can see how other methods of distinguishing wires would be helpful. Back in my school days there was a guy that was color blind so he couldn't look at a resistor to tell its value - he almost always had to ohm it out. And different shades of the same color also cause confusion among folks who are not colorblind...
Interesting discussion on color blindness and tetrachromatics. As neat as a tetrachromat might be, I don't think they would perceive more colors, just perceive the colors differently (since the different cones are somewhere between red and green responsive). Makes me wonder if anybody might be responsive to IR or UV light?
But back to the article: I've had similar issues with our production department (when we still had one). Occasionally we would get a change request from production, on product we had been making for years without issue. When we (Engineering) would inquire what had changed, the answer would be "nothing". They had been building the product as we wanted them to (but not to the drawings/BOMS). Eventually, someone would actually look at the drawing or BOM and halt production until the change request was processed. Of course, this always happened when an important schedule date was looming.
Actually we're all responsive to IR (and presumably UV). The response is just way, way down the curve. The eye (like the ear( is a logarithmic sensor with a range that spans many decades.
In about 1974 I was working in a darkroom testing IR LEDs and phototransistors, using a sniperscope from time to time when I needed to adjust something. After about 3 hours I I realized that I could see the IR emitters without aids. I was able to complete the testing with unassisted vision.
Indeed. I'm red-green colorblind and have to ask for help from time to time--or use really bright lights. Just this Sunday I was mending a cable that the puppy had chewed through in the car. She severed all seven wires, and shortened three of them.
The light blue was easy, as was the orange, and the black, and the black-with-white stripe. I had to ask for help to confirm that the remaining ones were a brown and two identical dark-grays.
OTOH it's very easy to work on my Whirlpool drier which labels each end of each wire with a distimctive identifier and has matching identifiers on the schematic helpfully placed on the back of the machine. These folks are a model for friendly design.
The color does matter, in a way. If substitute wire is used in a system, it may not work. In this case, relying on someone to choose correctly, is chancy. I have another story coming out where someone installed 4km of wire that "looked the same" as what I suggested. There was a difference between the speaker wire they installed, and the shielded larger gauge wire that it needed.
However, taking into account for someone's color blindness is something I did not consider. If that is the case, perhaps a second person should be consulted. I assume that color blindness is something a field tech or line worker should admit? I will be sure to ask about this issue in the future.
Cabe, absolutely, the standards for wire color coding should be followed. They make sense! But during retrofits and upgrades, I expect to find contrary or insanity for color coded wires. Opening a panel often reveals a rat's nest instead of neat, orderly wire ducting.
Color coding can be a pain in the neck too, when it's a large-count multiconductor cable. Is that wire red with a white trace or white with a red trace?
Adding a second person to the process means adding cost.
Asking an employee to own up to a weakness that could cause them to be discriminated against is simply not going to happen.
Cabe Atwell wrote: "I assume that color blindness is something a field tech or line worker should admit? I will be sure to ask about this issue in the future."
Surprisingly, normal color vision might not even be a requirement for the job. In 1963, as a new high school graduate, I attempted to enlist in the Navy, in a program that included a degree in Nuclear Engineering and service as an officer on a nuclear submarine. At the pre-induction physical my color-blindness was discovered. I was ineligible for the neuclear program and asked the recruiter what other electronics options were available. He responded that the only one was Telephone Installer.
Given that it was common to wire buildings and ships with 25-pair and 50-pair telephone cables, that was surprising to me then, as it still is.
(I turned down the program and went to college instead.)
Having built, instlled and serviced equiptment I have found that it takes very little time to do it right the first time, and it makes for much more work to take a shortcut. Not having the right color wire generally means that someone chose to cut back on an order to save a few dollars and someone in the field wants to get teh job done so they use what is at hand rather than what was specced. It also seems that the panel that was wired neatly is generally wired correctly and seldom needs to be accessed. Conversly, the panel than was wired poorly is the panel that will be accessed again and again. It would seem that the guy who cares to do it neat also cares to do it correct.
Having the correct wire colors also saves time when it comes to repairs. Instead of having to look at a bundle of 100 wires, you have a bundle with 20 wht/blu, 25 blu, 15 red, 15 orange,a few white and the rest are green. If they are all the same color it is abotu certain that he wire you want will be the last one you trace out.
I work at a government R&D laboratory and as such we build numerous prototypes very few of which actually are converted into actual products (we do a lot of speciallty measurement systems). One prototype that was actually turned into a product had a similar problem when it went to production.
The instrument is a very large autonomous air sampler first designed and built back in the early 90's. We built several prototypes and succesfully demonstrated the system. The government decided they wanted to build several of these instruments to place around the globe. A defence contractor won the bid to produce said instrument. We turned over our designs and a prototype to the contractor and were basically held out of any interaction with the contractor.
Once the contractor finished the production prototype we had a look at the system. It performed as well as our system and they actually fixed some features but as a whole they did not change much wiring or components. One thing they did do was change all the wiring to white wire. Nearly every wire in the system was white. One component in particular is a 3U rack mount electronics box.
This electronics box has 115VAC, various DC power supplies and many signal wires. There are over one hundred wires in the box and all of it is point to point wiring. When we designed and built our prototype we used standard colors for all the wires inside this box. It made troubleshooting, adding and subtracting functions much easier during the prototype stage. Comparing the contractors version and our version the only difference was that all the wire was white. All the components were the same and positioned in the same location in the box.
We recently redesigned this box for the contractor. The main reason for the redesign? WIRING PROBLEMS! They said that they subcontract the production of the box. When they get the box back they always have wiring problems, ALWAYS. When the system is deployed and put into operation they have additional problems when they troubleshoot and correct problems.
Using the correct color coding is very important, I've worked in TV stations which have hundreds of coax cables running all over the place, all of them one color, generally black. When one is troublshooting a problem, particularly when you've gone off the air, trying to trace a given cable through this snake pit is very frustrating. Any type of labeling seems to be the last thing anyone thought of during installation.
On another point, I needed some custom prototype power transformers and ordered them from a recommended source. The transformers arrived, all were of impecable quality, beautifully finished, except for two things, the color coding of the wires were not standard and there was no documentation. The primary did turn out to be black wires, the high voltage secondary was blue/grey/yellow, the filament windings, one was red, another green and third was purple/orange. Not exactly standard coding. There was no markings to tell which transformer was which except by measuring the voltage outputs. A high quality product totally messed up by using incorrect color coding and a lack of documentation. Needless to say, despite the quality, this transformer house is out of business now. What a shame, a couple of shortcuts by somebody, who knows who, killed the company.
I did send a letter to them asking about the color coding and documentation, I never did receive a reply.
For the record, the person installing the control panel & wiring was not colorblind. Saying all the wires are grey might have been misleading. As I said, it was just a lazy approach to handling the job at their end.
There are countless laws that regulate color of wiring. Imagine if our homes were wired up with whatever was laying around, or "looked like it would work." I have been in a few homes where the owner took it upon themselves to handle the power distribution. Luckily, they never had an inspection from the city walk through or they most assuredly would have been fined and required to hire an electrician to fix it all. Take the time, do it right is always the lesson.
There's no excuse for anyone deviating from a design without proper approval. This is why having a well-functioning process in place for submitting, documenting, and approving or rejecting deviation requests is so important.
That being said, I also don't have a lot of patience with self-righteous engineers who refuse to consider perfectly reasonable deviation requests because "manufacturing should just follow the drawing." These individuals need to get off of their high horse, analyze the consequences of accepting the deviation, and make an engineering decision, rather than a knee-jerk reaction.
Life is rarely neat and orderly; in real-world manufacturing, all kinds of things come up that can't all be predicted, and some of them may require deviating from or changing a design. A good engineer is able to analyze these things objectively as they arise, and to make sound recommendations.
While the production team in this example was obviously in the wrong and engineering was obviously in the right, the tone of the article seems to suggest that engineering is always right and production is always wrong. A little more humility may be in order.
One reason why production personnel sometimes make changes without asking engineering is that whenever they approach engineering, they are met with a condescending and dismissive attitude. So they just stop asking. This obviously isn't a correct response, but it's an understandable one.
At the end of the day, engineering and manufacturing (and quality, and purchasing, and accounting, and marketing...) all need to work together so that the company can make money and everyone's paychecks will clear.
In my first job that put me in the position of giving instructions to the production technicians, one of my first directives was "If something does not look right to you, please tell me". Not only did that assure that problems were reported as soon as they were realized, it also set the tone for good communications between engineering and production. Some things that looked like errors were just lack of clarity in the drawings, while there were some actual mistakes, although not many. Publicly thanking a tech for finding a problem did a whole lot for moral and greatly improved the cooperation that I got.
There was one interesting case of two detailers who assumed that I had forgotten to put the letter "K" after many of the resistor values in a design sketch that I gave them to detail. Since the first units built from the sketch worked well, more units were built using the finished drawing prior to my being able to check it. The technicians came to me complaining that the design could no longer be made to work at all, and asking what I had changed. It was not until I rescued my sketch that the problem became obvious. The techs got a laugh out of it, and I explained to the detailers in minute detail about not altering my designs. Once again I gave the directive, " if it does not look right, ASK ME". That seems to be the best way to get things right the first time.
Wire colors are an interesting challenge. For most of the auto companies, which were our major customers, there were not very many colors used. Red is for AC, blue is for DC, green is for frame ground, and black is for AC over 120 volts. Wire numbers at both ends of the wire are the way to go. And all terminals to be numbered the same as the wires that terminate in them. Of course the terminals must be in order, which does make tracing much simpler. Even color coded wires must have numbers, since not all cables use the same colors.
It's not just wiring that needs proper identification. Piping for water and process materials need the same identification. Imagine our surprise when water began pouring from a newly installed electrical panel. Before the conduit was completed and any wires had been pulled, the plumbers the floor above had mistaken the 1" conduit for a water line they were installing.
As a site engineer I learned to double check all tie-ins, and trace them back to their source. Once when tracing my project's supply lines, I was confused by a nearby fire sprinkler supply line I used as a reference. It was a freshly painted red, but seemed to change position when it ran through a wall. When the maintenance supervisor and I tracked his painter down, we got the answer. Before he started painting in the new room he'd made a point of checking again which pipe was red. He held the brush in that hand, then walked around the wall and began painting the pipe on that side red too... he just forgot he as facing the other direction.
You're right about asking the factory floor to tell you if it doesn't look right. I invite everyone to tell me their impressions on a new design. If a screw is hard to access, if something is difficult to service, or even, "gee, we used to have a wire that was this color, can we change it?" Of course you can't design by committee, and that's a terrible trap to fall into, but you can design a product that's built for production, service and the customer. There's a hidden agenda, too. I found that Production, Purchasing and Service would issue Engineering Change Notifications to alter product attributes to their preference. What's worse is that the ECNs would be signed off by Production Engineering and Development Engineering had no way of knowing what was changed. Sure enough, substandard parts would be substituted by Purchasing, and Production would stream-line a process that destroyed product integrity.
If you can get the other departments input during the prototype stage you can tell them why a component or process is specified.
@tekochip: That sounds like a bad ECR approval process, if design engineering had no input (or if production engineering could substitute for design engineering -- along the lines of, "If dad says no, go ask mom.")
I have found, through painful experience, the importance of having a well-thought-out process for approval of ECRs and deviation requests. At one place I worked, approval to use non-conforming parts only required the signature of one engineer -- any engineer -- and there was no documentation of the non-conformance other than "OK TO USE PER [ENGINEER'S NAME HERE]."
One engineer (who had a master's degree from MIT, and was not a dumb guy by any means) would sign these requests without even reading them.
It comes down to laziness, going home early, taking the path of least resistance. Even if you pay people more, they will only care for a little bit. Hence, future raises.
Scaring employees into working hard is how Apple and some other companies keep their engineers working diligently. Otherwise, more will try to find a place to nap on the job. I should know, found some nappers under their desks.
I have two issues with RVs. 1) using Romex a solid conductor wire in a vibration enviroment. It should be stranded. I believe it is criminal and hope some day the manufacturers that do this get sued out of business. The other is the change or lack there of across AC and DC systems. In AC green and bare are grounds. Black and colors are hots. White is neutral. Except for travelers. On my camper I put red tape on the + side. Both terminals are a mixture of black and white wires.
Absolutely right about the use of solid wire in RV s. I was asked to repair a problem of missing power in an AIRSTREAM trailer, and the very small outlet boxes were wired with solid wire, white and black. The problem was found to be a broken jumper strip on an outlet, making the repair was a real challenge because of the short solid wires. But adding a jumper between the two hot scrfews solved the problem. I am no longer impressed with the quality. Plush does NOT equal quality, nor does having lots of cute features.
Robots that walk have come a long way from simple barebones walking machines or pairs of legs without an upper body and head. Much of the research these days focuses on making more humanoid robots. But they are not all created equal.
The IEEE Computer Society has named the top 10 trends for 2014. You can expect the convergence of cloud computing and mobile devices, advances in health care data and devices, as well as privacy issues in social media to make the headlines. And 3D printing came out of nowhere to make a big splash.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.