Magneting data recording on steel

Prince

Member
Join Date
Jun 2002
Posts
284
I am vigilant to know if any of you know a way to write data magnetically on the steel. The prob is a pipe mill plant in which they want to track the manufactured pipe in different testing station. about 10 bits is enough.
I checked and saw a device called a wire recorder with which they used to write sound tracks. I wnder if some body is old enough to remember the way it was working.
 
Wire recorders worked basically like any magnetic tape recorder... actively modulate a coil on the head with your signal, and it magnetizes the bit of material under it. Then read the flux changes by passing it passively back over the same or a different coil.

I seriously doubt it will work for piping though, too much volume, varying ferro-magnetic properties...

Why not just slap a barcode on it?
 
Recording wire is made of a very high quality material with a very consistant composition. Your best bet to id your pipe is with a bar code or RFID tag. RFID would be more expensive, but more durable and easer to pick up, but you can print the bar code an a standard printer or maybey direcly on the pipe. The other option would be to paint the pipe with colored stripes and use a machine vision gizmo to read/ID them.
 
1 - why not barcode?

The condition for pipes movement is not a healthy proccess, pipes roll over rotate and crash to each other many times during the proccess. a normal barcode will not be the thing cause If I use it in one side the operator must move and search for the tag and then try to use the barcode reader that is not the automatic that can persuade my boss to choose. Do you know a barcode printer that can write in circular shape due to be read from any side?

2 - Why not RFID?

Anyway I had no experience with it but as far as I checked the technology, the main enemy for RFID is the steel which we have a lot in pipe making industry. and also big DC and Ac motors working in the Area. As far as I checked, I found no sensor which can read RFID from the other side of the pipes and the problem with dangerouse environment which easily can destroy any RFID tag is still existing...
 
I wouldn't be too quick to discount several earlier posts that mention etched bar codes. Use Google to search "Bumpy Bar Codes" or "Direct Part Marking" to read about several companies that develop these systems.

Some names include Sensis and Mecco, for starters...

Rick
 
Etched bars are real good for my application but the problem is that the Pipe rotates during the transfer proccess so an operator is needed for each Identification station to find the place of the marking.
 
Prince said:
Etched bars are real good for my application but the problem is that the Pipe rotates during the transfer proccess so an operator is needed for each Identification station to find the place of the marking.
Would etching the inside of the pipe work?
 
etching inside of the pipe is not that easy. and it won't be easy to read. and will still need operators for being read if the pipe rotates around itself...
 
Pricne,

I think no matter what you do you'll have to rotate the pipe at either the marking station or the ID station. Whatever you marking is (etched bar code, or color code etc..), you will need to either mark it around the entire circumference, or ID it by passing the sensor around the entire circumference.

Where ever it is, the "Rotation Station" would be a series of ball rollers place in a cradeling configuration. A rotating pneumatic tire would then come in contact with the pipe to rotate it for marking or I.D.ing it.

Personally I'd lean twards painting colored stripes around the parimiter of the pipe. Most of the paint would likely survive the process, and it could be ID either by a machine or visually.

However you didnt mention how many codes you need. Do you want to ID just size and schedule (pipe type), or give each pipe a specific serial number?

Color coding painted stripes for size and scheduale would be fairly simple. Basic colors like White, Red, Green, Orange, blue, grey, would be easy to pick up or see. If you had six possable colors in six bars/stripes, that would provide 46,656 different codes. More bars or colors would increase the possable codes dramaticly. (colors^bars = possable codes). The only drawback with painting wold be that it would require more maintenance to clean the spray heads, and restock the paint.

What I'm courious about is: what is the spicific purpose of the ID?? I think that would make a difference in how you'd mark the pipe.
 
Pricne,
as has been said an operator does not have to rotate the stock..a wheel could do the same..

Sorry just re read your post and you are worried about the bumping and bashing defaceing the bar code..hmmm
perhapos a bit more info?? are the pipes all the same size?..etc etc
 
I am working in a project for a factory producing pipes for Oil and Gas purposes in this way any pipe needs to be attached to a data base the system must read the related data from different stations in automatic mode and must record this data in database for office use.

Regarding the rotation I think if you consider that I have a new pipe every 20 seconds and I have about 23 different testing stations you will agree that it would be hard and expensive to make fast rotating stations for each station in that speed. I just want to identify each pipe for day, it woud be a number between 0-2000. The pipes for the day are the same size but whole in whole the system can make pipes between 8-20 inches...

Regarding the painting colors in circular mode I will get very happy if you lead me to a company which produce such device. can you give me more clue about a sensor? most of the sensors I know for the color are a bit expensive and I did not understand wether I must be the one who is doing the code for the color identification or there is a pre-made color identification methode and relative sensor for this purpose?
 
Last edited:
I dont know who would make a complete system to paint the color codes, however the concept isnt too diffucult.

You would have one marking station that would rotate the pipe. You need to track up to 2000 pipes so you will require maybe 5 colors and 5 stripes = 3000+ individual codes.

I have more thoughts on this, but right now I have to deal with another issue, so I'll be back later.

In the mean time answer this: Do your pipes stay in order after they enter the production/process line? Or do they get mixed up? If they dont get out of order from when they entered the process you might not need to mark them but keep track of them via an FIFO stack or shift register. It's just a matter of counting the pipes as they go by each station. However if the pipes branch to different lines then you back to actually marking them.

Later..
 

Similar Topics

Hello I have a s7-1200 and I would like to read the tags present in this controller with my controllogix controller. The two controllers don't use...
Replies
5
Views
142
Hi folks, I'm not as accustom with Siemens & WinCC, however I've been asked to increase the amount of data an existing application is logging...
Replies
2
Views
67
Has anyone migrated Proficy Historian data to a new server? I followed the guide to move all the data over, but when I run the utility, it stops...
Replies
0
Views
50
I cannot add SLC500 analog input tag (I: 8.3) to EZSeries Touch Panel Editor (V 5.3). I used all the listed tag datatype but it all says "Invalid...
Replies
10
Views
258
Hi, Wy we log data in PLC using DLG instruction, when we have option to log data in HMI
Replies
1
Views
80
Back
Top Bottom