Routing fine pitch SMD

Eldon Hiebert wrote:

"John Giddy" <jgiddy@bigpond.net.au> wrote in message

news:gbu8m9$c5c$1@cheetah.cadsoft.de...

Eldon Hiebert wrote:

Hello all,

 

The router seems to have difficulty routing small pitch devices. My

MAX3000A

is TQFP-100 @ 0.5mm pitch. With a grid of 5mil it leaves many of the pins

unrouted, eventhough there is lots of routing space on the inner layers.

 

I have alreayd optimized placement for the airwires so that I get as few

crossings as possible and the shortest nets (tradeoff here), but there

must

be more I can do to this to ease the routing task.

 

Q: What are the specific clearances that I need to set in the router to

help

route these parts? (ie. increasing pad to via keeps the router from

placing

vias too close to the part and crowding neighboring pads).

 

Q: Does it make sense to manually route from the pads, out away from the

parts to provide more clearance for vias and wires?

 

Am I asking too much from the router? I wouldn't think so.... as there is

tons of free space. A small pic of what I have so far included.

 

TIA,

Eldon.

 

 

 

Your routing grid and the package pitch do not match, so the auto router

won't work.

You would need to change the grid to a metric value of 0.5 mm or a smaller

but compatible value to allow the package pins to coincide with the grid.

John G.

 

That's not the problem actually. The pitch is 0.5mm and my routing grid is

set to 5 mil which is about 3 times smaller than the fine pitch parts. The

router still seems to choke the area around the fine pitch stuff in such a

way that it is even impossible to manually route after the auto-router runs.

 

Using select with the auto router helps, but still does not complete the

entire part, and again, the crowding makes it hard to manually route

afterwards.

 

I guess I am just expecting too much from the autorouter, and need to stage

my route in phases. I really wish I could select routing air wires by class.

 

Thanks,

Eldon.

 

 

The grid may be "about" 3 times smaller than the pitch, but it is not

exactly 3 times smaller.  The origins of the pins are not on the

intersections of the routing grid.

Mixing metric and imperial measurements is asking for trouble in my opinion.

John G.

Eldon Hiebert wrote:

"John Giddy" <jgiddy@bigpond.net.au> wrote in message

news:gbu8m9$c5c$1@cheetah.cadsoft.de...

Eldon Hiebert wrote:

Hello all,

 

The router seems to have difficulty routing small pitch devices. My

MAX3000A

is TQFP-100 @ 0.5mm pitch. With a grid of 5mil it leaves many of the pins

unrouted, eventhough there is lots of routing space on the inner layers.

 

I have alreayd optimized placement for the airwires so that I get as few

crossings as possible and the shortest nets (tradeoff here), but there

must

be more I can do to this to ease the routing task.

 

Q: What are the specific clearances that I need to set in the router to

help

route these parts? (ie. increasing pad to via keeps the router from

placing

vias too close to the part and crowding neighboring pads).

 

Q: Does it make sense to manually route from the pads, out away from the

parts to provide more clearance for vias and wires?

 

Am I asking too much from the router? I wouldn't think so.... as there is

tons of free space. A small pic of what I have so far included.

 

TIA,

Eldon.

 

 

 

Your routing grid and the package pitch do not match, so the auto router

won't work.

You would need to change the grid to a metric value of 0.5 mm or a smaller

but compatible value to allow the package pins to coincide with the grid.

John G.

 

That's not the problem actually. The pitch is 0.5mm and my routing grid is

set to 5 mil which is about 3 times smaller than the fine pitch parts. The

router still seems to choke the area around the fine pitch stuff in such a

way that it is even impossible to manually route after the auto-router runs.

 

Using select with the auto router helps, but still does not complete the

entire part, and again, the crowding makes it hard to manually route

afterwards.

 

I guess I am just expecting too much from the autorouter, and need to stage

my route in phases. I really wish I could select routing air wires by class.

 

Thanks,

Eldon.

 

 

The grid may be "about" 3 times smaller than the pitch, but it is not

exactly 3 times smaller.  The origins of the pins are not on the

intersections of the routing grid.

Mixing metric and imperial measurements is asking for trouble in my opinion.

John G.