If you read no further, here is the summary: Ignore drill back off errors.
Here is an explanation of what drill back off is, and why just about anyone can ignore warnings about it, today.
The reason we need not worry about this error today is that no contract manufacturer is going to alter the layout by backing off traces in such a manner. As the drilling is done by an NC (Numerically Controlled) robot drill, there is no need for a visible target. In fact, the drilling will typically be done on the board blank before any copper patterning even takes place.
Why this design rule check is still available in packages like DesignSpark is unclear to me, but fortunately, the check is optional, and can be turned off by unchecking a box. Most other layout software doesn’t even have the check at all.
(Visited 1,927 times, 1 visits today)
Thanks for that. I’m using Design Spark and there seems to be no way of routing without drill back off errors whether I’m routing manually or not. Now I don’t need to worry. Except of course I do. In which case I’ll try your formula out. Shame the DesignSpark help pages and fora don’t seem to have a clue about this.
Glad to help. I went through the same anguish on my design, and had to hunt a bit to figure out what was going on. I thought that I posted this tip to one of the DesignSpark forums (fora?), but my memory at this age isn’t what it used to be. I’m old enough to remember the era when a production engineer might actually do a drill back-off. (In fact, I remember when boards were laid out manually with physical tape – no gerbers involved.)
Okay so now I’ve had a good play in DesignSpark and finally cracked it. Your formula is wrong. The correct formula (at least for DS8.1.1) is t must be equal to or less than p – h. Works every single time. Why this should be makes no sense to me. Your equation made more sense but doesn’t work in DS. Now I need to work out erroneous cm-cm errors when components are miles apart! I’ll get there…eventually…
More investigations…apologies I had the wrong pdf output settings in DS. Your formula does work after all! So does mine as far as back off errors in DS are concerned, which was all I was thinking about from the title of your blog. Why these back off warnings matter in DS is if you’re outputting to a pdf to etch your own boards – the pad holes are obscured by the tracks so you can’t see where to drill precisely. Both your and my formula work to back off the tracks to leave the pad holes perfectly visible. What your formula does (and mine doesn’t) is that it takes care of drill breakout in addition to drill back off. Most excellent indeed.
Thanks for that. I’m using Design Spark and there seems to be no way of routing without drill back off errors whether I’m routing manually or not. Now I don’t need to worry. Except of course I do. In which case I’ll try your formula out. Shame the DesignSpark help pages and fora don’t seem to have a clue about this.
Glad to help. I went through the same anguish on my design, and had to hunt a bit to figure out what was going on. I thought that I posted this tip to one of the DesignSpark forums (fora?), but my memory at this age isn’t what it used to be. I’m old enough to remember the era when a production engineer might actually do a drill back-off. (In fact, I remember when boards were laid out manually with physical tape – no gerbers involved.)
Okay so now I’ve had a good play in DesignSpark and finally cracked it. Your formula is wrong. The correct formula (at least for DS8.1.1) is t must be equal to or less than p – h. Works every single time. Why this should be makes no sense to me. Your equation made more sense but doesn’t work in DS. Now I need to work out erroneous cm-cm errors when components are miles apart! I’ll get there…eventually…
More investigations…apologies I had the wrong pdf output settings in DS. Your formula does work after all! So does mine as far as back off errors in DS are concerned, which was all I was thinking about from the title of your blog. Why these back off warnings matter in DS is if you’re outputting to a pdf to etch your own boards – the pad holes are obscured by the tracks so you can’t see where to drill precisely. Both your and my formula work to back off the tracks to leave the pad holes perfectly visible. What your formula does (and mine doesn’t) is that it takes care of drill breakout in addition to drill back off. Most excellent indeed.