Philips CSM

Subscribe to Thread

Hi, I'm searching for a software that optimizating the fe... - May 10, 2008 by andras

I have done noting about optimization, except manualy choosi... - May 16, 2008 by jmelson

Jon, I tried to compile your code, but had to make 2 change... - May 17, 2008 by sarason

OK, here's a sample component file : 1 0.1uF 0.... - May 18, 2008 by jmelson

andras

#54702

Philips CSM | 10 May, 2008

Hi,

I'm searching for a software that optimizating the feeder setup on CSM machines and may generates the board and feeder files. Could anybody help with such software?

Regards: Andras

reply »

jmelson

#54777

Philips CSM | 16 May, 2008

I have done noting about optimization, except manualy choosing the feeder location of the most common part or two.

I do have a program that takes Protel 99SE pick and place files and converts them to the placement file. I haven't done a program for the feeder file, I just use the machine to set that up, although I can see the reasons for having a program that does it RIGHT the first time!

I have put my c program online at http://pico-systems.com/codes/cvtmount.c It is very basic, but you give it a file with a list of component types and their feeder locations and pick-up angles in the feeder tape, and the Protel P&P file, and it produces two files for top and bottom side mounting. These have only the M lines for the mount file, I just use a text editor to edit in the standard location of the board origin, and hand edit the board fiducial coordinates. it automatically flips the coordinate system for the bottom side. I use the main stop and front rail as the origin for bottom side, and calculate the offset from that for the lower left corner for front side.

Of course, if you don't use Protel, your P&P input file format will be different, so you will have to adjust the code for that.

Jon

reply »

sarason

#54784

Philips CSM | 17 May, 2008

Jon, I tried to compile your code, but had to make 2 changes for happy compilation under Borland C++ Builder. Specifically: #include void outNum(double RealNum, long SigFrac)

Now the question, could you post your component_file? Or was it generated by another piece of code?

regards sarason

reply »

jmelson

#54788

Philips CSM | 18 May, 2008

OK, here's a sample component file :

1 0.1uF 0.00 1 24 1.00K 0.00 1 22 10.0K 0.00 1 3 1000pF 0.00 1 3 .001uF 0.00 1 25 1.0uF 0.00 1 2 100pF 0.00 1 30 10M 0.00 1 26 22uF 0.00 1 31 2pF 0.00 1 23 100 0.00 1 21 BAV199INCT 90.00 1 17 MA2J11100 0.00 1 38 AD8132ARM 0.00 2 30 CHIPFIX 0.00 3 28 MMBT2222AFSCT 90.00 1 39 AD706AR 0.00 2 5 4.7 0.00 1 4 39.2K 0.00 1 19 5.1K 0.00 1 34 .018uF 0.00 1 25 MMBTH10 90.0 1 43 CHIP16 0.00 3 27 33uH 0.00 1 26 100uH 0.00 1 40 LTC1660CGN 0.00 2 37 MAX4614EUD 0.00 2 18 2.67K 0.00 1 32 4.75K 0.00 1 7 SN74AHC245PWR 0.00 2 20 10uH 0.00 1 35 2.2K 0.00 1

The first column is the part number, it corresponds to the component # in the CSM's component file. The 2nd column is an exact match for the component types called out in the P&P file. I have to edit that file if I have any spaces in the component type. the next column is if the P&P file assumes a different orientation than the part comes off the tape. Most of the components come off the tape aligned the same way the P&P file expects, so I enter 0.00 as the rotation. That MMBT2222 transistor isn't, so I had to put in 90.0 to compensate. The last column is the head to use.

Jon

reply »