SparkFun Forums 

Where electronics enthusiasts find answers.

Questions about the BatchPCB service

Moderator: robacarp

By One~Zero
#7957
I've been trying to submit a board, but it keeps coming up with DRC errors
(2 on top, 190 on bottom) for the 8 mil trace width/clearance minimums. I've submitted
boards without any problems before, so I'm a little confused as to why this keeps
popping up with these errors. The bottom errors seem to be a result of a large negative
layer on the bottom (board size is approx. 3in by 3in). My DRC check in Eagle 4.13
checks the board with minimums of 10 mil, and produces no errors when run.
By Caffeine
#7982
One~Zero wrote:I've been trying to submit a board, but it keeps coming up with DRC errors
(2 on top, 190 on bottom) for the 8 mil trace width/clearance minimums. I've submitted
boards without any problems before, so I'm a little confused as to why this keeps
popping up with these errors. The bottom errors seem to be a result of a large negative
layer on the bottom (board size is approx. 3in by 3in). My DRC check in Eagle 4.13
checks the board with minimums of 10 mil, and produces no errors when run.
Rebuild your ground pour with 12mil spacing. See if that helps
By Philba
#7990
Assuming you are using a polygon for this, go to the change tool, select isolate, select 12 mil (or .012 if in inches, .3048 if metric) and then click on the polygon boundary. It might take some messing around to hit it if the polygon is clipped by the board dimensions. hit ratsness to see if it took.
By One~Zero
#8000
OK, I played around with the board some more, but I keep getting the same fault indication after submitting the board for DRC. I've tried isolate at 12,13,14,16...and even changed my routing grid up to 13 mils.
So, I decided to submit an older design that I had sent in back in September (before the new site was up), and now this one fails with 200+ errors on the bottom. I had no problems with submitting the older board before...matter of fact, I have 2 of them sitting here on my workbench.
So my question would be: Why is a board that I submitted before, that had passed, now failing under the new system?
By Caffeine
#8002
One~Zero wrote:Why is a board that I submitted before, that had passed, now failing under the new system?
Sparky used to fix problems like that instead of bothering the customer...
By One~Zero
#8004
Don't get me wrong, I appreciate everything they've done and everything they are doing now. But, if my DRC is passing my board at 13mil spacing, how could I possibly be failing on 8mil tolerance with BatchPCB?
By Philba
#8037
have you looked at the failure locations with viewmate or similar app?

also, have you tried isolating the problem? maybe it's not your ground pour at all. I know text on the copper layers will make batchpcb barf but then eagle drc will flag it, too.
By djr7710
#8039
If you run it through FreeDFM (www.freedfm.com) it'll show you images of where you fail their DRC, which should allow you to isolate the problem.
Dave
By One~Zero
#8042
djr7710 wrote:If you run it through FreeDFM (www.freedfm.com) it'll show you images of where you fail their DRC, which should allow you to isolate the problem.
Dave
When your design fails, they send you images of where it fails. No matter what I change my isolate to, it gives a large amount of errors on the ground pour. I'll try FreeDFM and see if it shows anything different, but it's getting a little frustrating not being able to submit a design that passed under the old system, but not under the new.
By One~Zero
#8043
Philba wrote:have you looked at the failure locations with viewmate or similar app?

also, have you tried isolating the problem? maybe it's not your ground pour at all. I know text on the copper layers will make batchpcb barf but then eagle drc will flag it, too.
No errors appear with eagle DRC, which is set, at least for my boards, to have a lot higher tolerances than BatchPCB. There is no text close to where the errors are appearing either. I'll try removing the ground pour totally and see if it still reports errors.

Much thanks to everyone for your suggestions by the way :D
User avatar
By sparky
#8091
Ground pours are a major culprit. I've seen a few go by with GPs that passed. Many fail. I think it has to do with how the trace width and placement are being interpreted differently between programs. When all else fails get rid of the ground plane if possible.

-Nathan
By One~Zero
#8093
sparky wrote: When all else fails get rid of the ground plane if possible.
-Nathan
:( I was afraid you might say that. I in fact just resubmitted it with the ground plane removed, but it came back this time as saying that the check couldn't be completed and would be handled by a human within 24 hours. I did run it through FREEDFM, and it showed no errors. Thanks.
User avatar
By sparky
#8098
I don't see your submission for the past 4 hours - in which I've corrected a few DRC bot problems.

Really. Resubmit - the bot can take it now.

-Nathan
By One~Zero
#8100
sparky wrote:I don't see your submission for the past 4 hours - in which I've corrected a few DRC bot problems.

Really. Resubmit - the bot can take it now.

-Nathan
I resubmitted earlier, about 1:00 p.m. EST. I'll have to try again in the morning. Thanks again.