output cursor location ?

Bill Martin wrote:

Hello,

 

Any way to access/output current cursor location on board editor at

run time, for example with a function key press? Interested in

generating motion control commands to an external device for

inspection of boards.

 

Thanks,

Bill

Not easily, but possible.

I used the following technique.

I wrote a ULP which was run using a function key that was assigned to that

ULP. The process was, position the cursor and hit the Function key and I

attained the cursor position.

The secret lies in (@) which is the current mouse  position See HELP /

Editor Commands / Command Syntax

But you cannot simply read the cursor position within a ULP. You have to use

it. I drew a line from the cursor position to a far distant point  on a

custom layer that had an obscure name.

The ULP exited and immediatly re-ran itself with a specific argument this

time to indicate re-entry. Satisfied it was a re-entry it looked for the

custom layer name and for a  line with one end at the known location. The

other end of the line was recovered as this was the cursor co-ords. You have

to write code to clean up the line on the custom layer each time you

initially enter the ULP

I hope a future subversion of V6  has cursor position obtainable from within

the ULP.

HTH

Warren

--

Viewed / responded via the newsgroup at

news.cadsoft.de

On 04/27/2012 12:37 AM, Warren Brayshaw wrote:

Bill Martin wrote:

>> Hello,

>>

>> Any way to access/output current cursor location on board editor at

>> run time, for example with a function key press? Interested in

>> generating motion control commands to an external device for

>> inspection of boards.

>>

>> Thanks,

>> Bill

 

Not easily, but possible.

 

I used the following technique.

I wrote a ULP which was run using a function key that was assigned to that

ULP. The process was, position the cursor and hit the Function key and I

attained the cursor position.

 

The secret lies in (@) which is the current mouse  position See HELP /

Editor Commands / Command Syntax

But you cannot simply read the cursor position within a ULP. You have to use

it. I drew a line from the cursor position to a far distant point  on a

custom layer that had an obscure name.

The ULP exited and immediatly re-ran itself with a specific argument this

time to indicate re-entry. Satisfied it was a re-entry it looked for the

custom layer name and for a  line with one end at the known location. The

other end of the line was recovered as this was the cursor co-ords. You have

to write code to clean up the line on the custom layer each time you

initially enter the ULP

 

I hope a future subversion of V6  has cursor position obtainable from within

the ULP.

 

HTH

Warren

 

>

Thanks for that pointer. I tried what seemed the obvious way to

circumvent the "not available in ULP" issue, with the following:

ASSIGN F12 'RUN chk_args.ulp (@)'

however, the (@) is not expanded and passed as an argv[] parameter as

would be desired. So it looks like your method may be the only way to do

this. Actually, until/if Eagle can pass coordinates to a ulp, I might do

better to look at using gerbv to emit coordinates, there I can alter the

source code if needed...

Regards,

Bill

Thanks for that pointer. I tried what seemed the obvious way to

circumvent the "not available in ULP" issue, with the following:

ASSIGN F12 'RUN chk_args.ulp (@)'

however, the (@) is not expanded and passed as an argv[] parameter as

would be desired. So it looks like your method may be the only way to do

this. Actually, until/if Eagle can pass coordinates to a ulp, I might do

better to look at using gerbv to emit coordinates, there I can alter the

source code if needed...

Regards,

Bill

Hi Bill

you don't get away to re-write gerbv that easily.

Have a play with the attached ULP. It may have value for you.

Place it on your ULP paths and read the description/guidance from Eagle

Control Panel

Regards

Warren

--

Web access to CadSoft support forums at www.eaglecentral.ca.  Where the CadSoft EAGLE community meets.

On 04/28/2012 05:03 PM, Warren Brayshaw wrote:

Thanks for that pointer. I tried what seemed the obvious way to

circumvent the "not available in ULP" issue, with the following:

ASSIGN F12 'RUN chk_args.ulp (@)'

however, the (@) is not expanded and passed as an argv[] parameter as

would be desired. So it looks like your method may be the only way to do

this. Actually, until/if Eagle can pass coordinates to a ulp, I might do

better to look at using gerbv to emit coordinates, there I can alter the

source code if needed...

 

Regards,

Bill

 

Hi Bill

you don't get away to re-write gerbv that easily.

 

Have a play with the attached ULP. It may have value for you.

Place it on your ULP paths and read the description/guidance from Eagle

Control Panel

 

Regards

Warren

 

Hi Warren,

Thank you for this, I will try it out. Had a cursory (no pun intended!)

look at your code...my gut tells me that this is the hard way, but has

the advantage that it is contained in the "Eagle universe", which is not

a bad thing. Having done embedded code for a LOT longer than I have been

attempting simple board designs, I may not be convinced that changing

Gerbv is so difficult...but there is another possibility: Eagle uses

dynamically linked libraries, so it may be possible to insert an access

hook that can be called from a .ulp program to get the current cursor

position from the X library code that actually moves the cursor upon

command from the Eagle program. That sounds like fun to look into...

Best Regards,

Bill

The ULP I attached a couple of days ago was a little hurried and had a bug.

It could not resolve a cursor point of (0 0).

The attached ULP has fixed that problem.

Bill, when you get a moment would you please describe the inspection device

you will be driving.

Regards

Warren

--

Web access to CadSoft support forums at www.eaglecentral.ca.  Where the CadSoft EAGLE community meets.

On 04/30/2012 11:40 PM, Warren Brayshaw wrote:

The ULP I attached a couple of days ago was a little hurried and had a bug.

It could not resolve a cursor point of (0 0).

The attached ULP has fixed that problem.

 

Bill, when you get a moment would you please describe the inspection device

you will be driving.

 

Regards

Warren

Thank you for the update...I had not stepped on the bug yet!

My idea is basically to construct an X-Y gantry with a USB microscope

mounted on the moving stage, coordinates obtained from the Eagle cursor

location would be sent to a controller, such that the microscope goes to

that location above the physical pcb. It's all in "thinking about it"

stage at the moment, I'm toying with the idea of also mounting an IR

laser on the gantry stage to see if I can do some no-touch soldering of

smt parts. I'm finding that my biggest problem with hand assembly of

prototypes is stopping the flea-like behavior of the tiny parts. touch

them & they jump somewhere else! Of course, it all could be moved

manually, no doubt for less money spent. Feedback on practicality of

this is welcome, I'm not hung up thinking it's the best idea ever.

Regards,

Bill

On 04/30/2012 11:40 PM, Warren Brayshaw wrote:

The ULP I attached a couple of days ago was a little hurried and had a bug.

It could not resolve a cursor point of (0 0).

The attached ULP has fixed that problem.

 

Bill, when you get a moment would you please describe the inspection device

you will be driving.

 

Regards

Warren

Thank you for the update...I had not stepped on the bug yet!

My idea is basically to construct an X-Y gantry with a USB microscope

mounted on the moving stage, coordinates obtained from the Eagle cursor

location would be sent to a controller, such that the microscope goes to

that location above the physical pcb. It's all in "thinking about it"

stage at the moment, I'm toying with the idea of also mounting an IR

laser on the gantry stage to see if I can do some no-touch soldering of

smt parts. I'm finding that my biggest problem with hand assembly of

prototypes is stopping the flea-like behavior of the tiny parts. touch

them & they jump somewhere else! Of course, it all could be moved

manually, no doubt for less money spent. Feedback on practicality of

this is welcome, I'm not hung up thinking it's the best idea ever.

Regards,

Bill

On 05/01/2012 06:43 PM, Bill Martin wrote:

My idea is basically to construct an X-Y gantry with a USB microscope

mounted on the moving stage,

...

I'm toying with the idea of also mounting an IR

laser on the gantry stage to see if I can do some no-touch soldering of

smt parts.

This is interesting... You'd have ramp the current up such that you

don't induce a thermal shock on the part.  Do you have an IR laser in

mind powerful enough to reach the melting point of solder?

    - Chuck

On 05/02/2012 04:59 AM, Chuck Huber wrote:

On 05/01/2012 06:43 PM, Bill Martin wrote:

>> My idea is basically to construct an X-Y gantry with a USB microscope

>> mounted on the moving stage,

...

>> I'm toying with the idea of also mounting an IR

>> laser on the gantry stage to see if I can do some no-touch soldering of

>> smt parts.

 

This is interesting... You'd have ramp the current up such that you

don't induce a thermal shock on the part.  Do you have an IR laser in

mind powerful enough to reach the melting point of solder?

 

     - Chuck

Not sure how much power is needed for such a small area. The IR rework

stations I have seen use hundreds of watts over a large area. I have

seen a 1W blue laser burn holes in plastic panels, but that is likely

different from a shiny metal surface. Reflection of energy could make

the whole idea go "flop". It's really not something I know anything much

about, perhaps that's why it is interesting. There are other obvious

limitations as well, for example it is not going to solder down too many

BGA parts...and inner layers of multi-layer pcbs can't really be seen

with the scope either. So, it only fits the case of assembling the kind

of pcbs that one might try in a non-commercial setting.

-bill