haskal
/
kicad
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

current with latest freerouter's clearance support

This commit is contained in:
dickelbeck 2008-10-29 23:57:04 +00:00
parent f723c540d0
commit 8c76e2e644
1 changed files with 12 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
pcbnew/specctra_export.cpp
View File

@ -983,15 +983,12 @@ void SPECCTRA_DB::FromBOARD( BOARD* aBoard ) throw( IOError )
int curTrackWidth = aBoard->m_BoardSettings->m_CurrentTrackWidth; int curTrackWidth = aBoard->m_BoardSettings->m_CurrentTrackWidth;
int curTrackClear = aBoard->m_BoardSettings->m_TrackClearence; int curTrackClear = aBoard->m_BoardSettings->m_TrackClearence;
// The 0.5 is to give freerouter a little extra room, this is 0.5 mils. // Add .1 mil to the requested clearances as a safety margin.
// If we export without this, then on import freerouter violates our // There has been disagreement about interpretation of clearance in the past
// DRC checks with track to via spacing, although this could be a // between Kicad and Freerouter, so keep this safetyMargin until the
// result of > testing vs. >= testing in PCBNEW's DRC. // disagreement is resolved and stable. Recently Kicad started adding
// double safetyMargin = 0.5; // 0.1 mils to what it is given, so we can use zero for now.
const double safetyMargin = 0.0;
// recent freerouter code adds .1 to the requested clearances, and I want .5
// so subtract .1 so we actually get 0.5.
double safetyMargin = 0.4;
double clearance = scale(curTrackClear); double clearance = scale(curTrackClear);
@ -1004,13 +1001,13 @@ void SPECCTRA_DB::FromBOARD( BOARD* aBoard ) throw( IOError )
rules.push_back( rule ); rules.push_back( rule );
// On a high density board, a typical solder mask clearance will be 2-3 mils. // On a high density board, a typical solder mask clearance will be 2-3 mils.
// This exposes 2 to 3 mils of bare board around each pad. So we need at least // This exposes 2 to 3 mils of bare board around each pad, and would
// 2 mils *extra* clearance for tracks which would come near a pad on // leave only 1 to 2 mils of solder mask between the solder mask's boundary
// to the edge of any trace within "clearance" of the pad. So we need at least
// 2 mils *extra* clearance for traces which would come near a pad on
// a different net. So if the baseline trace to trace clearance was say 4 mils, then // a different net. So if the baseline trace to trace clearance was say 4 mils, then
// the SMD to track clearance should be at least 6 mils. Then, because // the SMD to trace clearance should be at least 6 mils. Also add our safetyMargin.
// freerouter seems to be off about a 1/2 mil on clearance tests, add sprintf( rule, "(clearance %.6g (type default_smd))", clearance + 2.0 + safetyMargin );
// another .5 mil for this test discrepancy as a safety margin.
sprintf( rule, "(clearance %.6g (type default_smd))", clearance + safetyMargin + 2.0 );
rules.push_back( rule ); rules.push_back( rule );
/* see: http://www.freerouting.net/usren/viewtopic.php?f=5&t=339#p474 /* see: http://www.freerouting.net/usren/viewtopic.php?f=5&t=339#p474