Homebrew PCBs

Ok, this is the story so far. I asked for and got great help from the group about a PCB CAD program, Diptrace. It is a lot less of a P.I.T.A. 
than Eagle. But now I need more help. When I tell DT to make a board, using the as best as I understand DRC rules of my board house,Olimex. I get DRC errors all around the chip. No place else. It seems too me to be because this chip is so freaking small that 8mil spacing is too big to allow the DRC to permit no errors. I reason this because if I set the minimum trace spacing to 1mil (too small for any board house I know), I get no DRC errors. So, this is my question. Am I doing something wrong? Or is there a board house that can make a PCB with the resolution required for a MAX668eub? This chip is driving me nuts! I have at least 3 projects that require this chip so I am really motivated to get this problem solved. In theory, I can use a chip from Linear Technology, the LT3780, but that chip has a much greater external parts count and since it is not in the DT part library, if I let them make a library part for it , there is no guarantee it will not have the same problems I have now. Help!!!

--- On Fri, 26/11/10, First L <adastra_2001@...> wrote:

From: First L <adastra_2001@...>
Subject: [Homebrew_PCBs] MAX 668eub
To: Homebrew_PCBs@yahoogroups.com
Date: Friday, 26 November, 2010, 3:50







 



  


    
      
      
      Ok, this is the story so far. I asked for and got great help from the group about a PCB CAD program, Diptrace. It is a lot less of a P.I.T.A. 

than Eagle. But now I need more help. When I tell DT to make a board, using the as best as I understand DRC rules of my board house,Olimex. I get DRC errors all around the chip. No place else. It seems too me to be because this chip is so freaking small that 8mil spacing is too big to allow the DRC to permit no errors. I reason this because if I set the minimum trace spacing to 1mil (too small for any board house I know), I get no DRC errors. So, this is my question. Am I doing something wrong? Or is there a board house that can make a PCB with the resolution required for a MAX668eub? This chip is driving me nuts! I have at least 3 projects that require this chip so I am really motivated to get this problem solved. In theory, I can use a chip from Linear Technology, the LT3780, but that chip has a much greater external parts count and since it is not in the DT part library, if I let them make a library part for it , there is no guarantee it will not have
 the same problems I have now. Help!!!

As Douglas Adams once said...Don't Panic.click on the verification tab, and select the design rules tab.Adjust the pad to pad spacing, pad to track spacing etc, and re-verify.Export as gerber or DFX and send to your favorite board for there verification, stating that they should not proceed with a product (board) unless the design rules comply.Hope this answers your problem,chester




















    
     

    
    


 



  






      

[Non-text portions of this message have been removed]

Hello again Adastra 2001,
Right, the problem is the MAX668EUB pad footprint in the DIPTrace is incorrect !!!
I have created a folder called chester in the files section.
This currently contains a PDF from Maxim of the footprint for the uMAX10 package and a .lib file called Maxim, just copy it into your own library, open and look for the uMAX10 footprint.
Best to copy the MAX668EUB into your own library, edit library, and attach the new pattern to the part.
In the first post I sent, I did not check out the part itself as I was busy with something else.
Having a bit of spare time on my hands, I loaded the DIPTrace part from the standard library.
Converted to PCB, and tried connecting a few 0.2mm tracks. I then proceeded to verify, with the same results that you experienced.
Back to the drawing board, went to Maxim's home page for the data sheet and footprint D/Sheet.
Bingo ! The pads of the library component are too WIDE and as such will only pass verification with settings of 0.1mm.
Over the years I have come to rely less and less on Standard component libraries and as such I now create my own components. This gives me the satisfaction that any any errors are mine and mine alone.
Still you now know the why and wherefore of the error, and now have the means to proceed with your original project. Have fun !!!! chester
P.S. DIPTrace staff are currently re-compiling the standard libraries and will release them with the forthcoming update soon.



      

[Non-text portions of this message have been removed]

I knew I wasn't crazy. I will try this and report back Thank you 



      

[Non-text portions of this message have been removed]