beginner needs to know if it will work

[Lancie1]

Unlike some of the garment materials which have different shades and textures I am using fabric and thread that is all dyed to match each other. Having not used the vision systems I don't know if this will be a problem.

Yes, it would be. There are color-detector photosensors that can see small color changes, but with color-matched thread, it would be difficult to detect a missed stitch.

What would be easy to detect is a small metallic fiber embedded inside the thread. Then, it might be possible to use a proximity sensor to detect the presence/absence of a stitch. If you could get a thread manufacturer interested in this concept, it might be worth testing out. I would think there would be a huge demand world-wide (if it works).

Carrying it one step farther, with a metallic fiber in the center of the thread, and with machines that use two thread spools, you could even wire up a thread-break detector circuit between the two spools, using a low-voltage resistance detector that connects automatically when the spool is inserted onto a spindle with a commutator contact built in. The contacts wire back to the resistance detector, which measures the current loop formed from Spool 1 to the machine stitch (where the threads will be "shorted" by the stitching), back to Spool 2. Any missed stitch should cause a sudden increase in resistance in the circuit, and a thread break would cause an easily-detected open circuit.

(If you get a patent on the above thread-break detector, I am reserving 1% of the future profits for thinking up the idea!- Lancie1)

 

[bce123]

metallic fiber in the center of the thread

dont think this one would fly through the airport metal detectors LOL

 

[Lancie1]

dont think this one would fly through the airport metal detectors LOL

Maybe not, but those detectors have to be set to ignore a lot of things - metal screws in broken bones; and metal buttons, zippers, underwire, and metal fasteners of all types on pants, bras, and coats. Think about this: We have about 5 liters of blood, and that blood has 2 to 3 grams of iron in it. That does not count all the other metal contaminants in our bodies, such as lead, mercury, aluminum, and so on.

Once I went to an aluminum plant where they used large DC-powered smelters to melt the aluminum. I spent several hours close to these smelters. The current was so strong that one of the guys I was working with threw his adjustable wrench up in the air about 10 feet high and 5 feet from the side of a smelter, and before it hit the floor, it flew sideways 5 feet and clamped to the side of the smelter. He had to use a pry bar to get it off. On the way home, I had to go through 3 airports, and I set off the metal detectors in every one, and got pulled out of the line and searched. After removing all metal from pockets and clothes, I still set the things off. I thought they would never let me go. Finally I decided the iron in my blood had been magnetized by the aluminum smelter currents.

 

[milldrone]

it is a 1" disk is rotating well under 1500 rpm. However once I have the system designed there are going to be many different applications for it. Some applications may go as high as 2000-2500 RPM.

The one thing that has me worried is the change in speed is only for a brief moment, say a .25 seconds or less. I am using this to see if a sewing machine skips a stitch.

The lack of a true HS counter in the CLICK makes this a little more difficult. My proposal is to use a 1024 ct absolute encoder and a prox to detect the end of a stitch cycle. Presuming you go with the 1 inch diameter wheel the approximate counts for 1/16" thread travel would equal 20 encoder counts. This type of encoder uses Gray code to translate encoder position. If the PLC misses a few counts it does not matter because the encoder shows the absolute position. There is a gotcha however. During a encoder rollover, the difference must be recalculated to allow for the rollover. This is not too difficult however, we just look for a number that is not in the anticipated range, and use permissives or conditions to enable the corrective math.

Also the direction the encoder is traveling also matters. I have made a presumption that the position number will decrease as the thread travels forward. If the direction is the other way, then the math to compute the difference must be reworked.

My delayed response is due partially to the lack of a Gray code translator for the CLICK. Some forum members from the Automation Direct users forum were gracious in providing a link to a "roll your own" version. Thanks go to Opzed and Bernie Carlton. Here is the site where the code for rolling your own was found. http://www.plcdev.com/using_ladder_logic_for_gray_code_conversion

The code below does not include any action on the results of the stitch length variation. You would also need a way to set a value to compare against. I'm suggesting a button to record a "good" value while operating or HMI to set this value. Lastly the code has not been tested, as I do not have a encoder or CLICK at my desk to test it on.

 

[Lancie1]

Sounds good. One thing to mention is that on many sewing machines, a breaking thread at the needle can cause a "jerk" in the thread back to the spool. Often this causes a snarl in the thread. In this case, it might even cause the encoder wheel to reverse direction. Maybe the logic could look for both a stopped encoder or a reversed encoder.

 

[Tim Ganz]

Wire Color

I am adding several roof fans to our building and I have wired each one to it's own MCC bucket with a soft start but we also have to tie these roof fans into the building fire systems so that when the fire alarm is tripped or when there is a fire it will turn off all the vent fans.

I plan to run wiring from the fire panel to each bucket and put a relay in each bucket to break the start/run circuit.

This will be 120 volt ac wiring that wil still be on when the bucket disconnect is off and I am seeing UL standards that say this should be yellow wire and NFPA standards that say orange wire? Which should I use? I found this info doing searches for old threads but I don't know how up to date or accurate it is now?

 

[Lancie1]

Tim,

I don't know about UL standards, but the National Electrical Code has required that any foreign or second-source power supply in a electrical enclosure (that is not disconnected by the enclosure main breaker) required the posting of a warning label. You might get around that requirement if your main disconnect has an auxiliary contact that could be used to disconnect the second-source.

EDIT: I see now that this thread needs to be referenced back to your other identical question on this thread:

http://www.plctalk.net/qanda/showthread.php?t=74052