Wire-Wrap Prototyping

[David]

So I was debugging a wire-wrapped prototype at work the other day, the first time I had really had hands-on experience with it. My first thoughts were that it was an absolute nightmare to debug, however after several days I have now come around to the idea. In fact, I don't understand why more people don't wire-wrap prototypes! The advantages seem very obvious; no soldering, easy dissasembling, complex 'routing' without restrictions of strip-board/breadboards etc, cheap, fast... I may have to look into purchasing a wire-wrap tool and giving it a proper go myself.

 

[JohnS_AZ]

Just a little historical perspective ...

The board pictured is from a 1980's-ish Acromatic airborne telemetry data decommutator. This board is -not- a prototype or a one-off, it's a full production product. There are a number of reasons for producing wire wrapped boards, but the one I found most interesting was that they were more physically reliable than PCBs in high vibration environments.

Also, there used to be (maybe there still are) fully automated wire wrap machines that could churn out wrapped boards like this.

[Jon Chandler]

The Nicolet 440A Mini Ubiquitous Spectrum Analyzer was one of my favorites.  This is late '70s technology and it's completely made of wirewrap boards.  My attempted troubleshooting on it was a failure.  No matter how carefully I traced a wire and compared it to the schematic, one end of the wire was a steady 5 volts, and the other end had a square wave.  Time for the professionals.

[GeoffS]

I used to work on Univac 1100 mainframes back in the 1970's.
Thousands of wire-wrap connections in the main cabinet made tracing a wire almost impossible.
Fortunately, they were very reliable and most faults were failed logic cards.

[Mechatrommer]

...one end of the wire was a steady 5 volts, and the other end had a square wave.  Time for the professionals.

need my pro advice? well.... u missed something in the "middle" no matter what. i repeat, u missed something in the "middle". thats a nonsense in term of physics!

[Jon Chandler]

Yeah, it's obvious that a solid wire can't have a signal at one end and a steady level at the other.  But doing my best  to trace the wire, even with a well-done schematic I couldn't prove otherwise.

Look at the first picture of this thread and you should be able to see how difficult it is to trace a wire through the mess.

[djsb]

This is where a HP 547A current tracer would come in handy

http://www.alliancetesteq.com/node/17204

It could be used with a logic pulser to inject a signal and trace it through a wire (it detects the magnetic field around the wire).

[sonicj]

i have a similar device for tracing wires behind walls. its called the fox & hound. works great! heres some notes on building one. http://www.sentex.net/~mec1995/circ/foxhound.html

[logictom]

I did wire wrapping once at uni and I won't be doing it again. It took so long to do it, obviously something that would decrease in time the more you did it, but it was so difficult to keep track of what was connected to what, if you wrapped a wire wrong at the beginning you had to lift all the wires above it. What a pain the arse, give me a breadboard and some wire any day of the week

[Neilm]

The good thing about wire wrapping is it teaches you disipline and meathodology - mess it up and you have to undo half your board.

Neil