OpenPnP hardware conversion pitfalls to watch out for?

[Wireb]

Recently our maker space became the second hand owner of a gen 2 shapeoko that was 90% assembled then moved and stored several times. All the parts are there and in good shape other than all the fasteners are loose and wiring is a mess. Vote at the last meeting was a PnP machine would get used way more than a small CNC mill. 

So now off trying to do some research on what is needed to convert it to OpenPnP. For the most part 0402 parts will be the smallest most of the folks (well mostly me) will use but generally everything will be 0602/0805 or larger. On the high end some leadless QFNs and similar (no BGAs at this time but may get the itch someday) 

Have a decent machine shop and stock of parts at home so fabricating up parts should not be a big deal and prefer it over just buying. (Funding is limited for the project so rather use what I have on hand). With that in mind what are the groups recommendations on the following:

One pick head or two? (Currently there is the lead screw Z but have some small ball slides in my collection I can swap out to make dual Zs.)

Minimum recommended Z travel distance? 

Camera recommendations? 

Through hole stepper vs spindle with external belt driven stepper?

Stepper size? (can easily find hollow NEMA8 and NEMA11s just don't want to be under powered with the vac line slip ring)

Z axis probe? (More or less how do you figure out when the tip is in contact with the part or should I just go with dead reckoning?)

Best bang for buck nozzle option? 

Anyone tried paste dispensing? ( I agree stencils are the way to go but most of the folks do one off (well 3 off from OSHpark) projects with SOIC and 0805 so not worth the cost of stencils) 

Any other hardware considerations I should look at before firming up the design of the new Z head?

Just trying to make sure I don't paint myself into a corner or drain all our project funds trying to the the Z built. 

Thanks for any help / pointers.   

 

[Michael Anton]

If most folks are just doing one offs, then I would recommend pasting and placing the parts by hand.  A PnP machine is really only useful when doing a number of something, since setting up the machine usually takes longer than stuffing a few boards by hand.

If you have a laser cutter, you could probably cut your own stencils pretty easily.  I use acetate overhead transparency film, though I cut it using a craft cutter, but a laser cutter would probably work just as well, if not better.

[Wireb]

That is true but some of them in the past have been quite large with a few 100 resistors and caps on them. Those jobs even one offs make some sense to automate. On the other end we have some folks making 100s of cards for sensor projects which will be a god send vs collecting up everyone around and passing out the tweezers :)

Know that paste dispensing is hit and miss but was one of the items requested to be investigated. Mostly just want to know if anyone has tried it and if so how did it go. Depending on that feed back and if the required mods would like to get at least what ever Z mounting support needed laid out now. 

[matt]

Dispensing is broken in develop branch atm, but i've got a pull request
that should get it back enabled - Jason going to take a look in next
couple of days. That said, i've not had the machine pasting yet - its
only the plan routines that i've fixed up and simulated. I don't know if
anyone has had the machine dispensing at any time with OpenPNP
controlling it yet?

I've also added Glue Dispense for double sided boards that need glue
dropping under the component. Its only really basic at the moment - adds
a splodge of glue at the component centroid - should probably have
patterns associated with it (various research papers about that tell you
the best glue pattern to use dependant on the package size), but thats a
job for another day.

> --
> You received this message because you are subscribed to the Google
> Groups "OpenPnP" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to openpnp+u...@googlegroups.com.
> To post to this group, send email to ope...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/openpnp/74778b0c-857c-4181-8a0e-b67413855ffc%40googlegroups.com
> [1].
> For more options, visit https://groups.google.com/d/optout [2].
>
>
> Links:
> ------
> [1]
> https://groups.google.com/d/msgid/openpnp/74778b0c-857c-4181-8a0e-b67413855ffc%40googlegroups.com?utm_medium=email&utm_source=footer
> [2] https://groups.google.com/d/optout

[Cri S]

Dual head is mostly for speed. Saving money, it's not needed.

You need to be sure the shapeoko don't have any mecanical issue, backlash, .... . If it have that, resolve it first, if possible.

You could mount webcam on Z axis, travel around with openpnp and checking error with fiducial centering, the bsh script i have recently posted.

If you have errors >= 2 pixels (0.08mm) it's not worth doing the conversion as the mechanic is not good enought.

Check for squareness errors. On 30x40cm , 2mm error is max error, you should measure 50cm.

If that is ok, the conversion is cheap, and beside resistors, setting up components is time consuming, assume 3 Minute/component. 

If you have 10 components, then total time is 40 minutes for setup only without having placed anything. Then pressing run, it needs (for 200 components) 15 minutes to be finished.

If there are 3 pcb, it's 85 minutes, and on placing it without pnp, it's what 50 minutes? 

[Cri S]

For dispenser, that is 250$ value (including compressor).

cameo stencil cutter cost 50 bucks and works faster/better, especially on 3 offs.

[Graeme Bridge]

you might be better looking at Johns assisted PNP

https://groups.google.com/forum/#!topic/openpnp/Emf2CoZMfyI

[Wireb]

Thanks for the link to the assisted that should take care of most of our "three offs" but we need a full auto option as well since we have a upcoming build that needs ~400 identical boards assembled. That is the main reason for the conversion. 

From my brief reading of that thread (at work will read more tonight) it sounds like mechanically a assisted and openPNP machine are the same (well camera locations are different as no uplooking and possible two downs) 

Initial checkout on the gantry everything looks good with no wear/slop in the bearings and very little backlash (once all the belts and motors were tightened up) Will measure the actual backlash and check squareness next time I am over there but overall the build looks very solid. 

So from the original list of questions:

*Answered* One pick head or two? (Two is faster and allow different tips without swapping)

*Partial* Anyone tried paste dispensing? (So far recommendation is to use a stencil cutter. Anyone happen to have a link to a tool chain to go from KiCad data to something the stencil cutter can handle? ) 

Minimum recommended Z travel distance? (just need a ballpark for when I am digging though my collection of ball slides)

Camera recommendations? 

Through hole stepper vs spindle with external belt driven stepper?

Stepper size? (can easily find hollow NEMA8 and NEMA11s just don't want to be under powered with the vac line slip ring)

Z axis probe? (More or less how do you figure out when the tip is in contact with the part or should I just go with dead reckoning?)

Best bang for buck nozzle option? 

[Jason von Nieda]

Hi Peter, answers inline:

One pick head or two? (Currently there is the lead screw Z but have some

Two is nearly as easy as one, almost doubles your speed and allows you to make less nozzle changes. There's several head designs for two that makes it pretty easy, including the other Peter's design that can be laser cut. I recommend two.

 

Minimum recommended Z travel distance? 

I've found 25mm is a good bare minimum. More is better, and there is no real value in going over about 40-50.

 

Camera recommendations? 

The ELP 720p camera is kind of the OpenPnP workhorse, and I recommend it.

 

Through hole stepper vs spindle with external belt driven stepper?

Hollow stepper is cheap and easy and most of the nozzle holders floating around are geared towards using it. External gets you better resolution. I recommend starting with a NEMA 8 hollow stepper.

 

Stepper size? (can easily find hollow NEMA8 and NEMA11s just don't want to be under powered with the vac line slip ring)

Both are fine. The NEMA 8 has enough power to overcome the friction from the slip fitting. 

 

Z axis probe? (More or less how do you figure out when the tip is in contact with the part or should I just go with dead reckoning?)

This is in progress in OpenPnP and there have not been many hardware designs for it yet. Perhaps you can come up with a good one and share? :)

 

Best bang for buck nozzle option? 

Most people are using Juki nozzles right now. The 50x series. Ray has recently done a run of nozzle holders for these and may have some left. 

 

Anyone tried paste dispensing? ( I agree stencils are the way to go but most of the folks do one off (well 3 off from OSHpark) projects with SOIC and 0805 so not worth the cost of stencils) 

I wrote the initial paste dispensing code and Matt has just revisited it. It's very new and very basic, at the moment. I have used the original code exactly 2 times and it worked, but the results were not impressive. I was using a syringe type pressure dispenser and it was hard to get exact drops. There are several people working on auger style dispensers.

 

Jason

[Mark Harris]

I have to admit, i don't get the fascination with "but its faster to do it by hand" thing... so what? Its a hell of a lot more painful by hand than going off and doing something productive in the time that machine plods along doing its thing.

I'm about to start a 4hr hand assembly job - several of two different boards I need to have done for a client demo tomorrow. I also need to write code for the client demo tomorrow, and test a bunch of other stuff... so right now I'd much prefer to have a PNP taking 10hrs to assembly my boards I could do in 4hrs, so I could have those 4hrs to get other things done that can't be automated.

My neck has some damage, and leaning over placing parts (i only used 0603s on this board, knowing it was going to have to be hand assembled I used big parts) really does a number on my neck. My eyes are not so great either, and with 4hrs sleep the past 3 nights from working on this project I really could do without hand assembly in my day right now.

So yes, I completely understand the wish to automate a tedious job of even 1 offs.

For stencils, I bought some stencils for this from http://OSHStencils.com, stainless, same day laser cut - very cool. Typically PCBCart/SEEED will do a stainless stencil for $15-20 for me. OSH pricing is very good, and the quality is much nicer than what we've received from SEEED. We also had the stencils the next day, we did a 2 day turn on the 4 layer boards for this project with Amitron down in the USA due to the timeline, so couldnt get stencils with the boards. These boards have several 0.4mm and 0.5mm pitch parts which i'm told don't come out well with the craft cutters.

If you're doing mostly 0805 and SOIC then the $50 craft cutters are fine, but for 0.5mm pitch TSSOP or QFN/DFN you'll need something laser or chem etched.

--

You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/4bd40515-ef0f-4c89-a2df-8855a177d6fc%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Jason von Nieda]

I agree completely. I've been doing this for 5 years now, and I've heard the "it's faster to do it by hand" argument literally hundreds of times. It's never helped. Everyone has their own reasons for wanting to build a machine. The purpose of this group is to encourage people to do that, not discourage them.

Jason

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/CAJMBTyQZC2QG9i1ToMk1i%2BLN60zPJaUV4VCvmjDyR7jUaV%3DL%2Bg%40mail.gmail.com.

[matt]

I've got an auger dispenser (and a project where i'm going to need it
coming up fast... hence sleepless nights fixing machine and contributing
OpenPNP code...) so when i'll report back in a about three weeks on how
well it works :D *fingers crossed*

> --
> You received this message because you are subscribed to the Google
> Groups "OpenPnP" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to openpnp+u...@googlegroups.com.
> To post to this group, send email to ope...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/openpnp/CA%2BQw0jxc5v2s141sDF6moWN-q6P2a6FhLkfXtiR8iqVnkO1z8w%40mail.gmail.com

> [1].
> For more options, visit https://groups.google.com/d/optout [2].
>
>
> Links:
> ------
> [1]

> https://groups.google.com/d/msgid/openpnp/CA%2BQw0jxc5v2s141sDF6moWN-q6P2a6FhLkfXtiR8iqVnkO1z8w%40mail.gmail.com?utm_medium=email&utm_source=footer
> [2] https://groups.google.com/d/optout

[Wireb]

Matt, Do you have some mechanical prints or basic dimensions of the screw paste dispenser? Currently just looking to reserve space and add mounting holes on the machine for a future upgrade.

Jason on the ELP 720p I see it comes in USB and IP flavors with and without IR (all all for about the same price) any advantages to one vs the other flavors? Also any ideas where 1.8deg rotational resolution would come up short? (guessing large / fine pitch parts but trying to size if we would ever run into it). 

Jason / Mark I totally agree and am in a similar boat. If I have a good tool chain and can just let it plod along vs hand assemble I would use it for one offs as well.Hoping I can set it up to easily swap over from openPNP to assistPNP as the second option would be used by the bulk of the folks. Those of us with large batches of cards to do or are out of time and don't mind the extra programming can use openPNP. 

So from the original list of questions:

*Answered* One pick head or two? (Two is faster and allow different tips without swapping. But cost delta of 2 heads is low vs 1 so may be worth it.)

*Answered* Minimum recommended Z travel distance? (between 25mm to 50mm recommended)
*Answered* Stepper size? (NEMA8 have plenty of torque to overcome vac slip ring so just a cost / size tradeoff on the two)

*Answered* Best bang for buck nozzle option? (Juki 50x series recommended)

*Partial* Camera recommendations? (ELP 720p looks like it comes in USB or IP with / without IR)
*Partial* Anyone tried paste dispensing? (So far recommendation is to use a stencil cutter. Anyone happen to have a link to a tool chain to go from KiCad data to something the stencil cutter can handle? Matt is working on the screw dispenser)
*Partial* Through hole stepper vs spindle with external belt driven stepper? (most tip holders are for direct mount. external mount gets higher resolution. at what package size does 1.8deg resolution become an issue?)
*Partial* Z axis probe? (Currently a work in progress in openPNP so no set winner yet.)

*New* Down looking camera mounting should it be on the Z (moving) or should it be fixed to the gantry (fixed distance from bed)?

[Cri S]

I have a program that convert gerber to dispense points for gauge22 needle.
I have t100us value of 11.58 , time per volume.
In reality it use gerb2tiff/bmp/png and then with opencv checks the
result image and
generate the output file and output image doing some calcs and the
program itself
have some options too. I made the software a long time ago.

Example of output file:
DOT 2202.000 4265.833 12.604 28.65 8125 701
DOT 2238.000 4265.833 12.604 28.65 8125 701
DOT 3553.333 3945.833 11.232 25.53 5750 496
DOT 126.667 3932.500 11.232 25.53 5750 496
DOT 1099.378 4026.076 77.733 176.69 1906000 164594
DOT 1845.000 3868.133 12.847 29.20 8604 743
DOT 1845.000 3903.533 12.847 29.20 8604 743
DOT 3553.333 3850.833 11.232 25.53 5750 496
DOT 126.667 3839.167 11.232 25.53 5750 496
DOT 2357.346 3893.229 55.696 126.60 701083 60542
DOT 2108.516 3892.500 60.136 136.69 882472 76206
DOT 1098.333 3770.929 81.175 184.51 2170528 187437
DOT 1452.000 3515.833 12.604 28.65 8125 701
DOT 1488.000 3515.833 12.604 28.65 8125 701
DOT 705.333 3515.833 12.604 28.65 8125 701
DOT 741.333 3515.833 12.604 28.65 8125 701
DOT 2203.667 3512.500 12.604 28.65 8125 701
DOT 2239.667 3512.500 12.604 28.65 8125 701
DOT 3553.333 3200.833 11.232 25.53 5750 496
DOT 126.667 3189.167 11.232 25.53 5750 496
DOT 3553.333 3105.833 11.232 25.53 5750 496
DOT 126.667 3094.167 11.232 25.53 5750 496
DOT 1608.412 3144.288 55.695 126.60 701056 60540
DOT 1360.325 3142.923 60.244 136.94 887250 76619
DOT 860.679 3143.229 55.696 126.60 701083 60542
DOT 611.992 3143.743 60.244 136.94 887250 76619
DOT 2848.876 3513.333 81.135 184.42 2167306 187159
DOT 2359.012 3140.352 55.696 126.60 701083 60542
DOT 2110.325 3140.410 60.244 136.94 887250 76619
DOT 3104.855 3511.796 77.723 176.67 1905250 164529
DOT 2203.667 2765.833 12.604 28.65 8125 701
DOT 2239.667 2765.833 12.604 28.65 8125 701
DOT 1457.000 2765.833 12.604 28.65 8125 701
DOT 1493.000 2765.833 12.604 28.65 8125 701
DOT 705.333 2765.833 12.604 28.65 8125 701
DOT 741.333 2765.833 12.604 28.65 8125 701
DOT 2948.067 2765.000 13.046 29.65 9010 778
DOT 2985.267 2765.000 13.046 29.65 9010 778
DOT 3555.000 2450.833 11.232 25.53 5750 496
DOT 125.000 2435.833 11.232 25.53 5750 496
DOT 3555.000 2357.500 11.232 25.53 5750 496
DOT 125.000 2340.833 11.232 25.53 5750 496
DOT 2358.910 2394.289 55.647 126.49 699250 60384
DOT 1612.243 2394.289 55.647 126.49 699250 60384
DOT 860.577 2394.289 55.647 126.49 699250 60384
DOT 2110.325 2392.923 60.244 136.94 887250 76619
DOT 1364.356 2392.921 60.326 137.12 890889 76933
DOT 611.992 2392.923 60.244 136.94 887250 76619
DOT 3105.679 2392.019 55.696 126.60 701083 60542
DOT 2856.992 2391.257 60.244 136.94 887250 76619
DOT 2952.000 2017.500 12.604 28.65 8125 701
DOT 2988.000 2017.500 12.604 28.65 8125 701
DOT 2203.667 2017.500 12.604 28.65 8125 701
DOT 2239.667 2017.500 12.604 28.65 8125 701
DOT 705.333 2017.500 12.604 28.65 8125 701
DOT 741.333 2017.500 12.604 28.65 8125 701
DOT 1453.667 2015.833 12.604 28.65 8125 701
DOT 1489.667 2015.833 12.604 28.65 8125 701
DOT 3553.333 1699.167 11.232 25.53 5750 496
DOT 126.667 1694.167 11.232 25.53 5750 496
DOT 3553.333 1604.167 11.232 25.53 5750 496
DOT 126.667 1599.167 11.232 25.53 5750 496
DOT 3108.516 1645.352 55.744 126.71 702889 60698
DOT 2860.374 1645.416 60.152 136.73 883167 76266
DOT 2359.012 1644.895 55.696 126.60 701083 60542
DOT 2110.183 1644.167 60.136 136.69 882472 76206
DOT 860.246 1645.121 55.774 126.78 704028 60797
DOT 611.992 1645.410 60.244 136.94 887250 76619
DOT 1609.012 1643.229 55.696 126.60 701083 60542
DOT 1360.183 1642.500 60.136 136.69 882472 76206
DOT 2952.000 1269.167 12.604 28.65 8125 701
DOT 2988.000 1269.167 12.604 28.65 8125 701
DOT 2203.667 1269.167 12.604 28.65 8125 701
DOT 2239.667 1269.167 12.604 28.65 8125 701
DOT 1455.333 1269.167 12.604 28.65 8125 701
DOT 1491.333 1269.167 12.604 28.65 8125 701
DOT 703.667 1267.500 12.604 28.65 8125 701
DOT 739.667 1267.500 12.604 28.65 8125 701
DOT 3553.333 945.833 11.232 25.53 5750 496
DOT 126.667 945.833 11.232 25.53 5750 496
DOT 3553.333 850.833 11.232 25.53 5750 496
DOT 126.667 850.833 11.232 25.53 5750 496
DOT 3107.243 897.622 55.647 126.49 699250 60384
DOT 2358.910 897.622 55.647 126.49 699250 60384
DOT 1611.745 897.621 55.695 126.60 701056 60540
DOT 2858.658 896.257 60.244 136.94 887250 76619
DOT 2110.325 896.257 60.244 136.94 887250 76619
DOT 1362.689 896.255 60.326 137.12 890889 76933
DOT 860.182 895.352 55.744 126.71 702889 60698
DOT 610.751 895.000 60.258 136.97 887861 76672
DOT 3553.333 242.500 11.232 25.53 5750 496
DOT 126.667 229.167 11.232 25.53 5750 496
DOT 2981.667 214.167 9.287 21.11 3250 280
DOT 3553.333 147.500 11.232 25.53 5750 496
DOT 126.667 135.833 11.232 25.53 5750 496


I post here the resulting Gcode from this file:

(press time: 293.541 sec)
(volume: 0.55666 ml)
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 72.6186 X 22.7584 Y 1.5302 Z 3.47828 D 0.0145394 ml 7.6619 sec )
G0X72.6186Y22.7584
G0Z1.5302
G4P0.0001
M7
G4P7.6619
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 75.7428 X 5.4356 Y 0.23589 Z 0.536194 D 5.3258e-05 ml 0.028 sec )
G0X75.7428Y5.4356
G0Z0.23589
G4P0.0001
M7
G4P0.028
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 78.9178 X 22.8092 Y 1.41343 Z 3.21285 D 0.0114587 ml 6.0384 sec )
G0X78.9178Y22.8092
G0Z1.41343
G4P0.0001
M7
G4P6.0384
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 90.2462 X 3.7592 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X90.2462Y3.7592
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 90.2462 X 6.1468 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X90.2462Y6.1468
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 90.2462 X 21.6154 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X90.2462Y21.6154
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 90.2462 X 24.0284 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X90.2462Y24.0284
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 55.9816 X 51.2572 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X55.9816Y51.2572
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 90.2462 X 40.7416 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X90.2462Y40.7416
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 53.594 X 41.7576 Y 1.52745 Z 3.47193 D 0.0144611 ml 7.6206 sec )
G0X53.594Y41.7576
G0Z1.52745
G4P0.0001
M7
G4P7.6206
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 90.2462 X 43.1546 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X90.2462Y43.1546
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 40.8686 X 41.7322 Y 1.41468 Z 3.21564 D 0.0114887 ml 6.0542 sec )
G0X40.8686Y41.7322
G0Z1.41468
G4P0.0001
M7
G4P6.0542
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 90.297 X 59.8932 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X90.297Y59.8932
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 37.846 X 51.2064 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X37.846Y51.2064
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 37.8714 X 32.2326 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X37.8714Y32.2326
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 36.9316 X 51.2064 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X36.9316Y51.2064
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 36.957 X 32.2326 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X36.957Y32.2326
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 15.5448 X 60.7822 Y 1.5302 Z 3.47828 D 0.0145394 ml 7.6619 sec )
G0X15.5448Y60.7822
G0Z1.5302
G4P0.0001
M7
G4P7.6619
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 34.544 X 41.7068 Y 1.52745 Z 3.47193 D 0.0144611 ml 7.6206 sec )
G0X34.544Y41.7068
G0Z1.52745
G4P0.0001
M7
G4P7.6206
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 15.5448 X 79.8576 Y 1.5302 Z 3.47828 D 0.0145394 ml 7.6619 sec )
G0X15.5448Y79.8576
G0Z1.5302
G4P0.0001
M7
G4P7.6619
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 21.844 X 41.783 Y 1.41666 Z 3.22021 D 0.011537 ml 6.0797 sec )
G0X21.844Y41.783
G0Z1.41666
G4P0.0001
M7
G4P6.0797
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 17.907 X 70.2564 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X17.907Y70.2564
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 18.796 X 32.2072 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X18.796Y32.2072
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 17.907 X 89.3064 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X17.907Y89.3064
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 18.8214 X 70.2564 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X18.8214Y70.2564
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 18.8214 X 89.3064 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X18.8214Y89.3064
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 3.2258 X 3.4544 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X3.2258Y3.4544
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 21.8694 X 60.8076 Y 1.41343 Z 3.21285 D 0.0114587 ml 6.0384 sec )
G0X21.8694Y60.8076
G0Z1.41343
G4P0.0001
M7
G4P6.0384
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 3.2258 X 5.8166 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X3.2258Y5.8166
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 56.896 X 70.2564 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X56.896Y70.2564
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 21.8694 X 79.8322 Y 1.41468 Z 3.21564 D 0.0114887 ml 6.0542 sec )
G0X21.8694Y79.8322
G0Z1.41468
G4P0.0001
M7
G4P6.0542
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 3.2258 X 21.6154 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X3.2258Y21.6154
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 56.896 X 89.2048 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X56.896Y89.2048
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 34.544 X 79.8322 Y 1.5302 Z 3.47828 D 0.0145394 ml 7.6619 sec )
G0X34.544Y79.8322
G0Z1.5302
G4P0.0001
M7
G4P7.6619
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 3.2258 X 24.0284 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X3.2258Y24.0284
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 59.9186 X 60.8076 Y 1.41343 Z 3.21285 D 0.0114587 ml 6.0384 sec )
G0X59.9186Y60.8076
G0Z1.41343
G4P0.0001
M7
G4P6.0384
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 34.6456 X 60.7822 Y 1.53228 Z 3.48285 D 0.0145991 ml 7.6933 sec )
G0X34.6456Y60.7822
G0Z1.53228
G4P0.0001
M7
G4P7.6933
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 15.5194 X 22.733 Y 1.53055 Z 3.47904 D 0.0145494 ml 7.6672 sec )
G0X15.5194Y22.733
G0Z1.53055
G4P0.0001
M7
G4P7.6672
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 59.9186 X 79.756 Y 1.41468 Z 3.21564 D 0.0114887 ml 6.0542 sec )
G0X59.9186Y79.756
G0Z1.41468
G4P0.0001
M7
G4P6.0542
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 21.844 X 22.733 Y 1.4159 Z 3.21843 D 0.0115183 ml 6.0698 sec )
G0X21.844Y22.733
G0Z1.4159
G4P0.0001
M7
G4P6.0698
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 72.3646 X 89.2302 Y 2.06083 Z 4.68427 D 0.0355158 ml 18.7159 sec )
G0X72.3646Y89.2302
G0Z2.06083
G4P0.0001
M7
G4P18.7159
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 34.6202 X 22.7584 Y 1.53228 Z 3.48285 D 0.0145991 ml 7.6933 sec )
G0X34.6202Y22.7584
G0Z1.53228
G4P0.0001
M7
G4P7.6933
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 72.5678 X 60.7314 Y 1.5302 Z 3.47828 D 0.0145394 ml 7.6619 sec )
G0X72.5678Y60.7314
G0Z1.5302
G4P0.0001
M7
G4P7.6619
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 40.9448 X 22.8092 Y 1.41465 Z 3.21564 D 0.0114882 ml 6.054 sec )
G0X40.9448Y22.8092
G0Z1.41465
G4P0.0001
M7
G4P6.054
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 74.8792 X 70.231 Y 0.331368 Z 0.75311 D 0.000147647 ml 0.0778 sec )
G0X74.8792Y70.231
G0Z0.331368
G4P0.0001
M7
G4P0.0778
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 53.594 X 22.7584 Y 1.5302 Z 3.47828 D 0.0145394 ml 7.6619 sec )
G0X53.594Y22.7584
G0Z1.5302
G4P0.0001
M7
G4P7.6619
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 46.863 X 99.1616 Y 0.326314 Z 0.74168 D 0.000140994 ml 0.0743 sec )
G0X46.863Y99.1616
G0Z0.326314
G4P0.0001
M7
G4P0.0743
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 75.819 X 70.231 Y 0.331368 Z 0.75311 D 0.000147647 ml 0.0778 sec )
G0X75.819Y70.231
G0Z0.331368
G4P0.0001
M7
G4P0.0778
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 59.9186 X 22.8092 Y 1.41343 Z 3.21285 D 0.0114587 ml 6.0384 sec )
G0X59.9186Y22.8092
G0Z1.41343
G4P0.0001
M7
G4P6.0384
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 27.8892 X 95.7834 Y 2.06184 Z 4.68655 D 0.0355686 ml 18.7437 sec )
G0X27.8892Y95.7834
G0Z2.06184
G4P0.0001
M7
G4P18.7437
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 78.867 X 89.2048 Y 1.97416 Z 4.48742 D 0.0312215 ml 16.4529 sec )
G0X78.867Y89.2048
G0Z1.97416
G4P0.0001
M7
G4P16.4529
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 27.9146 X 102.26 Y 1.97442 Z 4.48793 D 0.0312337 ml 16.4594 sec )
G0X27.9146Y102.26
G0Z1.97442
G4P0.0001
M7
G4P16.4594
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 78.8924 X 60.7568 Y 1.41468 Z 3.21564 D 0.0114887 ml 6.0542 sec )
G0X78.8924Y60.7568
G0Z1.41468
G4P0.0001
M7
G4P6.0542
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 3.2258 X 97.5106 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X3.2258Y97.5106
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 3.2258 X 99.8728 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X3.2258Y99.8728
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 78.9686 X 41.783 Y 1.4159 Z 3.21843 D 0.0115183 ml 6.0698 sec )
G0X78.9686Y41.783
G0Z1.4159
G4P0.0001
M7
G4P6.0698
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 75.8952 X 32.2326 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X75.8952Y32.2326
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 75.8952 X 51.2572 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X75.8952Y51.2572
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 74.9808 X 32.2326 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X74.9808Y32.2326
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 74.9808 X 51.2572 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X74.9808Y51.2572
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 72.644 X 41.783 Y 1.52786 Z 3.47294 D 0.0144725 ml 7.6266 sec )
G0X72.644Y41.783
G0Z1.52786
G4P0.0001
M7
G4P7.6266
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 59.9186 X 41.783 Y 1.41468 Z 3.21564 D 0.0114887 ml 6.0542 sec )
G0X59.9186Y41.783
G0Z1.41468
G4P0.0001
M7
G4P6.0542
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 18.8214 X 51.2572 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X18.8214Y51.2572
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 56.896 X 32.2326 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X56.896Y32.2326
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 17.8816 X 32.2072 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X17.8816Y32.2072
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 56.896 X 51.2572 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X56.896Y51.2572
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 17.907 X 51.2572 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X17.907Y51.2572
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 55.9816 X 32.2326 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X55.9816Y32.2326
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 15.5448 X 41.783 Y 1.5302 Z 3.47828 D 0.0145394 ml 7.6619 sec )
G0X15.5448Y41.783
G0Z1.5302
G4P0.0001
M7
G4P7.6619
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 3.175 X 59.4614 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X3.175Y59.4614
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 3.2258 X 40.6146 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X3.2258Y40.6146
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 3.2258 X 43.0276 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X3.2258Y43.0276
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 36.8808 X 89.3064 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X36.8808Y89.3064
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 3.175 X 61.8744 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X3.175Y61.8744
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 37.0078 X 70.2564 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X37.0078Y70.2564
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 3.2258 X 78.5876 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X3.2258Y78.5876
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 37.7952 X 89.3064 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X37.7952Y89.3064
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 3.2258 X 81.0006 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X3.2258Y81.0006
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 37.9222 X 70.2564 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X37.9222Y70.2564
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 40.8432 X 79.8576 Y 1.41465 Z 3.21564 D 0.0114882 ml 6.054 sec )
G0X40.8432Y79.8576
G0Z1.41465
G4P0.0001
M7
G4P6.054
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 40.9448 X 60.8076 Y 1.41343 Z 3.21285 D 0.0114587 ml 6.0384 sec )
G0X40.9448Y60.8076
G0Z1.41343
G4P0.0001
M7
G4P6.0384
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 53.594 X 60.7822 Y 1.5302 Z 3.47828 D 0.0145394 ml 7.6619 sec )
G0X53.594Y60.7822
G0Z1.5302
G4P0.0001
M7
G4P7.6619
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 90.2462 X 78.8924 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X90.2462Y78.8924
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 53.594 X 79.756 Y 1.5302 Z 3.47828 D 0.0145394 ml 7.6619 sec )
G0X53.594Y79.756
G0Z1.5302
G4P0.0001
M7
G4P7.6619
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 90.2462 X 81.3054 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X90.2462Y81.3054
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 55.9816 X 70.2564 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X55.9816Y70.2564
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 90.297 X 62.2554 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X90.297Y62.2554
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 55.9816 X 89.2048 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X55.9816Y89.2048
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 90.2462 X 97.8154 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X90.2462Y97.8154
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 90.2462 X 100.228 Y 0.285293 Z 0.648462 D 9.42256e-05 ml 0.0496 sec )
G0X90.2462Y100.228
G0Z0.285293
G4P0.0001
M7
G4P0.0496
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 59.8678 X 98.8822 Y 1.41468 Z 3.21564 D 0.0114887 ml 6.0542 sec )
G0X59.8678Y98.8822
G0Z1.41468
G4P0.0001
M7
G4P6.0542
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 56.8452 X 108.356 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X56.8452Y108.356
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 55.9308 X 108.356 Y 0.320142 Z 0.72771 D 0.000133145 ml 0.0701 sec )
G0X55.9308Y108.356
G0Z0.320142
G4P0.0001
M7
G4P0.0701
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 53.5686 X 98.8568 Y 1.52745 Z 3.47193 D 0.0144611 ml 7.6206 sec )
G0X53.5686Y98.8568
G0Z1.52745
G4P0.0001
M7
G4P7.6206
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(DOT mm 46.863 X 98.2472 Y 0.326314 Z 0.74168 D 0.000140994 ml 0.0743 sec )
G0X46.863Y98.2472
G0Z0.326314
G4P0.0001
M7
G4P0.0743
M9
G4P0.002
G1Z7.06184F2600
G0X0Y0
M9
G4P0.002
G1Z7.06184F
(total time: 590 sec)


The cut code is:

check(char t, double w, double h, double x, double y, double r, double a, d
ouble l ) { double v,z,d, t100us;
// t == Rect, Square, Xcv (convex), Triangle, Circular, Unknow
if(u_mil(w)<50||u_mil(h)<50) v=7.5; else v=10; // stencil thickness
v*=u_mil(u_mil(a));
z=pow(v*24/(CV_PI*31),1./3.); // gauge22 = 16mil inner diameter
t100us=v/time_per_volume; // 11.58; // time constant
t100us+=0.3; t100us=(int)t100us;
d=v/z; // area of dot
d=sqrt(4*d/CV_PI); // dot diameter
dot(x,y,mil_u(d<16?16:d)); // display dot inside image
fprintf(out,"DOT %.3f %.3f %.3f %.2f %.0f %.0f\n",
u_mil(x),u_mil(y),z,d,v,t100us); //
}



2016-08-03 15:50 GMT, matt <ma...@mattbrocklehurst.co.uk>:

> You received this message because you are subscribed to a topic in the
> Google Groups "OpenPnP" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/openpnp/9275SyEGqBQ/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to

> openpnp+u...@googlegroups.com.
> To post to this group, send email to ope...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/openpnp/e8204f058ae51005e2d1091d0c74516c%40mail.mattbrocklehurst.co.uk.
> For more options, visit https://groups.google.com/d/optout.
>

[Cri S]

it's called: 
check(*str,r.size.width,r.size.height,mu.m10/mu.m00, mu.m01/mu.m00, r.angle,area,len)

this translate to:

check(char Type, double RotatedRectangleWidth, double RotatedRectangleHeight, double CenterOfGravity_X, double CenterOfGravity_Y,double angle, double Area, double CountourOutlineLenght)

All values are in Pixel and based on dpi values embedded inside image metadata are then translated to real values.

[Ray Kholodovsky]

Weighing in on a few earlier points... Once you hit 100 units of something decent sized (or I'm sure 400 of something smaller) - turnkey from one of many PCBA places in China suddenly becomes much more in reach.  Mark and I can make some recommendations. 

Of course, we are a "PnP Positive" type of group.  :)

There are many ways to get a Peter Betz type head made for you, all you have to do is ask :)

As it turns out I am completely sold out of Juki adapters from the first batch, if we get some more demand I will order another batch.  But I won't hijack this thread, please post in mine (Juki Nozzle Holder) if you are interested. 

Cheers,

Ray

[Wireb]

Ray, Yep saw your adapter and actually looking for the CAD files / dimensioned prints. I have a few chunks of 360 brass rounds I can use to make my own from. From the pictures looks like a easy lathe project with some finishing work on a rotary table. 

Yea if you have the budget and time is a cost China is a good options. In this case we got most of the parts already as part of a surplus pallet (expired by medical standards) and time is free (actually pays some of us back as it prevents us from moving on to the next project quite as fast :) 

Looks like I have a pair of Parker 4201 ball slides (3"x1.75"x0.75) with a travel of 2" in my parts bin. I think those will fit in the footprint of the current Z axis and have plenty of travel. More things to measure out next week when I get back there.     

[Ray Kholodovsky]

Mark posted the CAD files.  Link is here: https://groups.google.com/d/msg/openpnp/82hIAKQZnr0/bZiAFu3CCAAJ

I also just emailed you the drawings. 

If you lathe this yourself please please please provide video.  

We have a running joke about most of the tolerances being such that a monkey could make this on a barbecue rotisserie with a dull spoon, but I digress.  

--
You received this message because you are subscribed to a topic in the Google Groups "OpenPnP" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/openpnp/9275SyEGqBQ/unsubscribe.
To unsubscribe from this group and all its topics, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/a1bd69e8-e9d0-449e-b9a9-1ee48db60e9f%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

Ray

[Wireb]

Well I think you found your monkey. Don't have a working rotisserie but do have a bunch of spoons :)

Should be able to record and post some video. But will be a bit. Want to get all the mounting figured out and hollow steppers on hand (unless someone convinces me to go external jack shaft) before making the holders. 

[Paul Jones]

Hands up who’s done a 3+ hour manual placement run?? Even if my machine could only place caps and resistors I’d be jumping for joy!! It’s the painful area where it is too expensive to contract out, and not entirely sane to do manually too often.

 

Paul.

[John Socha-Leialoha]

Hi, after reading this thread, I'm realizing that I didn't name my machine well. I'm designing my machine to allow fully automatic operation. But I started with a machine that doesn't need programming, which I think might be the easiest to use when you're building only one board. But my plan for my "Assisted Manual" machine (I need a better name) is to support the spectrum from manual to automatic. Once I get the manual mode placing boards (this weekend), the next step will be to support semi-automatic mode. In this mode, OpenPnP will move the head to where it thinks the part should be picked or placed, and then you'll be able to use the control panel to adjust the position. I think this will perhaps be a better mode even for one board. From there, it's a very small step to fully automatic, as I'll just need to add a bottom camera.

[Michael Anton]

I'm not trying to discourage anyone from building a PnP with that comment, but rather trying to point out the reality of using one.  It likely won't be faster from a user time perspective for a small run.  I'm talking about user setup time here, and not machine run time.  If it takes you as long or longer to set up the machine for a job than it does to place the parts by hand, it might not make much sense to run that job on a machine.  Sure it might be more fun to run it on a machine, but then you are not saving time in the process, which is really the point to the machine (or at least it is for me).

I setup and run jobs on a commercial machine all of the time, so I have a pretty good idea of where the break even point is.  Even though I don't like to populate by hand, especially since my eyes are getting old, I still do it often as it is just much more efficient if you only need a couple.

For stencils, I either cut them on a Silhouette craft cutter for pitch over 0.5mm or so, or have them laser cut in stainless from smart-prototyping.com.  I usually get PCBs from Smart-Prototyping as well, so getting a stencil from them at the same time is almost a no-brainer.

Mike

[Michael Anton]

I have, and I couldn't have set up the machine to do a run in the same amount of time.  Now if I needed a few more boards, it might have been a different story.

[Mark Harris]

I started placing parts around 9:30am this morning. I left the office at 22:30pm, once I was done reflowing boards and rewiring the 2 mistakes in the design. Client demo tomorrow. I spend around 4hrs a week on average hand populating prototypes :|

--

You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/SY3PR01MB1386E4BF1C857ACE5C152FCC9E070%40SY3PR01MB1386.ausprd01.prod.outlook.com.

[Michael Anton]

Here is where the tool chain can be found for driving a craft cutter from a gerber file: https://github.com/pmonta/gerber2graphtec.  This is the forum thread that discusses it in great detail: http://dangerousprototypes.com/forum/viewtopic.php?f=68&t=5341.  Pay attention to some of the enhancements that people have come up with in the thread, as these haven't always made it back into the code, and it is worthwhile to add them by hand.

Mike

[Michael Anton]

Now if you could standardize on parts feeders that could always stay loaded on the machine, that would speed machine setup greatly.  If the boards are all completely different, with not much in common, you might find it is still faster to populate by hand.  That is the downside to a machine that requires a bunch of setup (they all do).  If you are just running different revisions of the same board, you may be able to save a bunch of time, since they would be fairly similar.

It also makes a lot of sense to reduce the number of different components in the BOM, since that determines setup time.  It is better to place two resistors to create an oddball value, if the two resistors are already needed in the design, than it is to add another line to the BOM.

[Mark Harris]

I'm usually very constrained on size, so two resistors rather than 1 is almost never an option. I'm typically at around 60-80 line items for 150-300 parts with my current projects.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/4bef985c-e97b-4864-92c5-960a6032895e%40googlegroups.com.

[Michael Anton]

Wow, that's a lot of different parts!!  I don't think I've ever populated a board with that variety of components.  It will take one big machine with a lot of feeders to support that many different parts.

[Mark Harris]

My placement times tend to be high because I haven't seen the board in a few weeks by the time I get it, so i only have a vague idea of where to look on the board for the position - most of my placement time is looking on the assembly drawing for the damn part location haha.

I'm working on a 900x600mm bed, I can get 80 automatic feeders at 20mm pitch with front/back. Quite a few parts there are only 1 of (Microwave RF, so the part value matters a lot) - flat cut tape feeders on the bed would be more than suitable for these. Likewise, for ICs, i use quite a few in the $10 range in low volume, so I wouldn't ever buy full reels of. I'd probably only buy reels for common res/cap/ferrite values and the rest would be cut tape bought per project. The way budgets function at work, its better to buy 100 of a part each time for 10 projects than one reel of 5k of that part and use it on 10 projects, even if the reel costs less.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/4bc9240c-b840-46f9-83b6-0359cf07d875%40googlegroups.com.

[Michael Anton]

I'll be interested in seeing how your build works out.  I'm planning to do a 600x600mm CoreXY build, with feeders on at least 3 sides, at 12mm pitch for 8mm feeders.  Mostly I'm basing the design on commonly available linear rail sizes, but that might change.  This machine would be a bit bigger than the commercial machine I usually use, and since it never seems to have enough feeders, the new one should be a big improvement.

[matt]

Whats the commerical machine you use Michael?

Matt

On 2016-08-04 09:43, Michael Anton wrote:
> I'll be interested in seeing how your build works out. I'm planning
> to do a 600x600mm CoreXY build, with feeders on at least 3 sides, at
> 12mm pitch for 8mm feeders. Mostly I'm basing the design on commonly
> available linear rail sizes, but that might change. This machine
> would be a bit bigger than the commercial machine I usually use, and
> since it never seems to have enough feeders, the new one should be a
> big improvement.
>
> On Thursday, August 4, 2016 at 2:04:58 AM UTC-6, Mark Harris wrote:
>
>> My placement times tend to be high because I haven't seen the board
>> in a few weeks by the time I get it, so i only have a vague idea of
>> where to look on the board for the position - most of my placement
>> time is looking on the assembly drawing for the damn part location
>> haha.
>>
>> I'm working on a 900x600mm bed, I can get 80 automatic feeders at
>> 20mm pitch with front/back. Quite a few parts there are only 1 of
>> (Microwave RF, so the part value matters a lot) - flat cut tape
>> feeders on the bed would be more than suitable for these. Likewise,
>> for ICs, i use quite a few in the $10 range in low volume, so I
>> wouldn't ever buy full reels of. I'd probably only buy reels for
>> common res/cap/ferrite values and the rest would be cut tape bought
>> per project. The way budgets function at work, its better to buy 100
>> of a part each time for 10 projects than one reel of 5k of that part
>> and use it on 10 projects, even if the reel costs less.
>>

>> On 4 August 2016 at 01:42, Michael Anton <3d.m...@gmail.com> wrote:
>>
>> Wow, that's a lot of different parts!! I don't think I've ever
>> populated a board with that variety of components. It will take one
>> big machine with a lot of feeders to support that many different
>> parts.
>>
>> On Thursday, August 4, 2016 at 12:50:43 AM UTC-6, Mark Harris wrote:
>> I'm usually very constrained on size, so two resistors rather than 1
>> is almost never an option. I'm typically at around 60-80 line items
>> for 150-300 parts with my current projects.
>>

>> On 4 August 2016 at 00:26, Michael Anton <3d.m...@gmail.com> wrote:
>>
>> Now if you could standardize on parts feeders that could always stay
>> loaded on the machine, that would speed machine setup greatly. If
>> the boards are all completely different, with not much in common,
>> you might find it is still faster to populate by hand. That is the
>> downside to a machine that requires a bunch of setup (they all do).
>> If you are just running different revisions of the same board, you
>> may be able to save a bunch of time, since they would be fairly
>> similar.
>>
>> It also makes a lot of sense to reduce the number of different
>> components in the BOM, since that determines setup time. It is
>> better to place two resistors to create an oddball value, if the two
>> resistors are already needed in the design, than it is to add
>> another line to the BOM.
>>
>> On Thursday, August 4, 2016 at 12:20:00 AM UTC-6, Mark Harris wrote:
>> I started placing parts around 9:30am this morning. I left the
>> office at 22:30pm, once I was done reflowing boards and rewiring the
>> 2 mistakes in the design. Client demo tomorrow. I spend around 4hrs
>> a week on average hand populating prototypes :|
>>

>> On 4 August 2016 at 00:01, Paul Jones <pa...@pauljones.id.au> wrote:
>>
>> Hands up who’s done a 3+ hour manual placement run?? Even if my
>> machine could only place caps and resistors I’d be jumping for
>> joy!! It’s the painful area where it is too expensive to contract
>> out, and not entirely sane to do manually too often.
>>
>> Paul.
>>

>> FROM: ope...@googlegroups.com [mailto:ope...@googlegroups.com] ON
>> BEHALF OF Jason von Nieda
>> SENT: Thursday, 4 August 2016 1:47 AM
>> TO: ope...@googlegroups.com
>> SUBJECT: Re: [OpenPnP] Re: OpenPnP hardware conversion pitfalls to

>> http://OSHStencils.com [1], stainless, same day laser cut - very

>> cool. Typically PCBCart/SEEED will do a stainless stencil for $15-20
>> for me. OSH pricing is very good, and the quality is much nicer than
>> what we've received from SEEED. We also had the stencils the next
>> day, we did a 2 day turn on the 4 layer boards for this project with
>> Amitron down in the USA due to the timeline, so couldnt get stencils
>> with the boards. These boards have several 0.4mm and 0.5mm pitch
>> parts which i'm told don't come out well with the craft cutters.
>>
>> If you're doing mostly 0805 and SOIC then the $50 craft cutters are
>> fine, but for 0.5mm pitch TSSOP or QFN/DFN you'll need something
>> laser or chem etched.
>>

>> --
>> You received this message because you are subscribed to the Google
>> Groups "OpenPnP" group.
>> To unsubscribe from this group and stop receiving emails from it,
>> send an email to openpnp+u...@googlegroups.com.
>> To post to this group, send email to ope...@googlegroups.com.
>> To view this discussion on the web visit
>>
> https://groups.google.com/d/msgid/openpnp/SY3PR01MB1386E4BF1C857ACE5C152FCC9E070%40SY3PR01MB1386.ausprd01.prod.outlook.com

>> [2].
>>
>> For more options, visit https://groups.google.com/d/optout [3].

>
> --
> You received this message because you are subscribed to the Google
> Groups "OpenPnP" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to openpnp+u...@googlegroups.com.
> To post to this group, send email to ope...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/openpnp/4bef985c-e97b-4864-92c5-960a6032895e%40googlegroups.com
> [4].
>
> For more options, visit https://groups.google.com/d/optout [3].

>
> --
> You received this message because you are subscribed to the Google
> Groups "OpenPnP" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to openpnp+u...@googlegroups.com.
> To post to this group, send email to ope...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/openpnp/4bc9240c-b840-46f9-83b6-0359cf07d875%40googlegroups.com
> [5].
>
> For more options, visit https://groups.google.com/d/optout [3].

>
> --
> You received this message because you are subscribed to the Google
> Groups "OpenPnP" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to openpnp+u...@googlegroups.com.
> To post to this group, send email to ope...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/openpnp/81fd453e-7f70-4502-8139-f689b5a02fc9%40googlegroups.com
> [6].
> For more options, visit https://groups.google.com/d/optout [3].
>
>
> Links:
> ------
> [1] http://OSHStencils.com
> [2]
> https://groups.google.com/d/msgid/openpnp/SY3PR01MB1386E4BF1C857ACE5C152FCC9E070%40SY3PR01MB1386.ausprd01.prod.outlook.com?utm_medium=email&amp;utm_source=footer
> [3] https://groups.google.com/d/optout
> [4]
> https://groups.google.com/d/msgid/openpnp/4bef985c-e97b-4864-92c5-960a6032895e%40googlegroups.com?utm_medium=email&amp;utm_source=footer
> [5]
> https://groups.google.com/d/msgid/openpnp/4bc9240c-b840-46f9-83b6-0359cf07d875%40googlegroups.com?utm_medium=email&amp;utm_source=footer
> [6]
> https://groups.google.com/d/msgid/openpnp/81fd453e-7f70-4502-8139-f689b5a02fc9%40googlegroups.com?utm_medium=email&utm_source=footer

[Paul Kelly]

I have, and I couldn't have set up the machine to do a run in the same amount of time.  Now if I needed a few more boards, it might have been a different story.

Me too. I’ve done marathon sessions with tweezers, assisted manual and fully auto PnP and everything Michael says is true..

 

The other thing to consider is that automated PCB assembly has some parallels with CNC machining in that the process often involves scrapping the first couple of parts that come out as you dial the machine in. (as opposed to manual process’ where it just takes longer to do the first one.

 

With PnP that means a couple of panels worth of components that you can only really wipe off the boards and hand place because they aren’t in their tapes, they have paste on them, or they’re somewhere under the machine. Now you do get better at it, eventually you’ll have first panels come out that just need a little nudge here or a part flipped there.  But we’ve been doing it for nearly 6 years and it still it doesn’t happen every time.

 

If its less than about 20 boards, we’ll hand or assisted hand place. Bear in mind that we have the absolute luxury of having every SMD part we use, on a dedicated feeder, on a machine, all the time. If you have to load feeders for each job then first panel yield drops away.

 

As Jason says, none of this is a reason not to build a PnP machine if that’s something that interests you. Heck, I’m pretty sure I got talked into building an injection moulding machine this morning!

The only point anyone is trying to make is that you are not going to be able to use your DIY PnP to assemble 1 off boards.

PK

 

 

--

You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/0e4aff7e-1c9a-4bab-991c-ce97075cdca2%40googlegroups.com.

[matt]

All true... and its the same with larger scale commerical PnPs too not
just these low-end ones.

I used to work with a iPulse M3 and the amount of time it takes to get
the feeders setup, program dialed in, checked, corrected because the cad
program output differs to the real world, run off the first PCB, do any
rework, then go back and correct the program again. Its a great deal of
work. And even when youve got it setup and doing PCBs, you still need to
rework some of them, misplaces, feeders jamming etc.. and this was with
a £150,000 machine, automated stencil printer, automated conyeors and
oven.

> FROM: ope...@googlegroups.com [mailto:ope...@googlegroups.com] ON
> BEHALF OF Jason von Nieda

> SENT: Thursday, 4 August 2016 1:47 AM
> TO: ope...@googlegroups.com
> SUBJECT: Re: [OpenPnP] Re: OpenPnP hardware conversion pitfalls to

> http://OSHStencils.com [1], stainless, same day laser cut - very cool.

> Typically PCBCart/SEEED will do a stainless stencil for $15-20 for me.
> OSH pricing is very good, and the quality is much nicer than what
> we've received from SEEED. We also had the stencils the next day, we
> did a 2 day turn on the 4 layer boards for this project with Amitron
> down in the USA due to the timeline, so couldnt get stencils with the
> boards. These boards have several 0.4mm and 0.5mm pitch parts which
> i'm told don't come out well with the craft cutters.
>
> If you're doing mostly 0805 and SOIC then the $50 craft cutters are
> fine, but for 0.5mm pitch TSSOP or QFN/DFN you'll need something laser
> or chem etched.
>

> --
> You received this message because you are subscribed to the Google
> Groups "OpenPnP" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to openpnp+u...@googlegroups.com.
> To post to this group, send email to ope...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/openpnp/0e4aff7e-1c9a-4bab-991c-ce97075cdca2%40googlegroups.com

> [2].
> For more options, visit https://groups.google.com/d/optout [3].

>
> --
> You received this message because you are subscribed to the Google
> Groups "OpenPnP" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to openpnp+u...@googlegroups.com.
> To post to this group, send email to ope...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/openpnp/5DAA68E3962947118782EB4687BB47BF%40HomePC
> [4].

> For more options, visit https://groups.google.com/d/optout [3].
>
>
> Links:
> ------
> [1] http://OSHStencils.com
> [2]

> https://groups.google.com/d/msgid/openpnp/0e4aff7e-1c9a-4bab-991c-ce97075cdca2%40googlegroups.com?utm_medium=email&amp;utm_source=footer
> [3] https://groups.google.com/d/optout
> [4]
> https://groups.google.com/d/msgid/openpnp/5DAA68E3962947118782EB4687BB47BF%40HomePC?utm_medium=email&utm_source=footer

[Michael Anton]

I sent a version of this to Matt directly, but I thought I'd post this here as well.

I use an old ECM 93 machine made by MDC in Japan.  It runs control software that was written in compiled BASIC running on DOS, so that should give you some idea of the age (we got it in 1998 or so).  It is a belt driven system using servos, so it does move fairly fast for what is pretty much a bench top machine, and at the slow speed that I run it at, places about 1200 parts/hour.  It doesn't have a vision system on it, but does have an on the fly laser measurement system (CyberOptics LaserAlign) that determines if the component is the correct size, what the angle of rotation is, and where the center of the component it, so it places parts fairly accurately.  The vision system in OpenPnP may be a lot nicer (I hope).  We only have 12 auto feeders, as they were $600 each when we bought the machine.  Though it does support cut strips, and a vibratory feeder for tubes as well.  Getting 40-50 different parts loaded is a real chore though, and requires some creative part mounting.  The auto feeders are air actuated, and completely mechanical, but they work very reliably.

We wrote our own programming software for it, which generates machine files, so we don't have to use the terrible machine interface for setting up jobs.  Instead we can just type the feeder list into excel, and the software combines it with the component location report from the CAD software, and generates the file set for the machine.  This works far better than the import system that the machine software uses.  This also helped us to work around the many machine limitations, and implement the machine setup in a more sensible fashion.

Note that they still sell a version of this machine, with full vision support.  I believe it is even available in kit form, and I've been tempted to get one, but the price is still in the $10k range.  You can find their website here: http://mdc-smt.co.jp/goodsindex3.htm.

Mike

[matt]

Hi Mike

I just read your direct reply.. so didnt notice it was direct :)

I've read quite a bit about that laser align - do you think its as good
as having cameras?

Matt

> find their website here: http://mdc-smt.co.jp/goodsindex3.htm [1].

> --
> You received this message because you are subscribed to the Google
> Groups "OpenPnP" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to openpnp+u...@googlegroups.com.
> To post to this group, send email to ope...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/openpnp/1fd596b3-d123-43c0-b234-240b383cbace%40googlegroups.com

> [2].
> For more options, visit https://groups.google.com/d/optout [3].
>
>

> Links:
> ------
> [1] http://mdc-smt.co.jp/goodsindex3.htm
> [2]
> https://groups.google.com/d/msgid/openpnp/1fd596b3-d123-43c0-b234-240b383cbace%40googlegroups.com?utm_medium=email&utm_source=footer
> [3] https://groups.google.com/d/optout

[Michael Anton]

I just did it again, and replied directly...

They work pretty well for the most part, but might be more fiddly than a well tuned vision system, though I don't have any experience with the latter.  Given that modern machines seem to use vision instead, I can only assume that it is either superior, or cheaper.  I guess the big advantage, is that laser is done easily on the fly, so no time is lost using it.  Probably a combination of laser align for smaller less critical parts, like passives, and vision for finer pitch would be pretty ideal.  Or, just use an on the fly vision system, but those are probably pretty difficult to DIY.

The laser systems are set up to give you a measurement of the width of one axis of the component, with the center in the axis.  The component is then rotated to find the minimums, which gives the measurement of both axis, centers, and the component rotation.  They also move the component vertically, to measure the Z height, which can be used to determine if the component was picked up tombstoned, or is just the wrong part.  It's pretty slick as far as determining if the part was picked correctly, but can be a pain to get set up correctly for some parts.  SOT23 parts have always been a bit difficult to get scanned properly.

Now that I have a good library of component definitions, setting up jobs is really pretty easy.  I can usually have machine files generated for a board within a couple of hours (not counting feeder setup on the machine), and usually the first board will come out just fine, as long as I wasn't sleeping when I set up the job.  I'm hoping I'll be able to do something similar with OpenPnP, but I honestly haven't played with it too much.  There are crazy limits on the ECM machine, like it only supports 50 different part definitions, and those are machine specific, rather than job specific, so we regenerate that file for each job, as it was a severe limitation.

Mike

[Graeme Bridge]

I remember dangerous prototypes showing one of the chines machines and a program they used to export the placement data from eagle, is there something similar for open PNP? 

[alex]

[Jason von Nieda]

Mike,

The thing that folks don't consider when making the "it's not worth it" argument is that there are other factors involved. There are many people who, due to age or illness, can no longer place parts that some of us would consider "easy". Setting things up for a couple hours to run the machine for a couple minutes might be the only way that someone can continue to enjoy their hobby. This is more common than you may think. I get a good number of people telling me these kinds of things at Maker Faire every year.

I am in a similar position. The reason I started this project 5 years ago is because I was hand placing a couple boards a day for a small widget I had designed. I couldn't afford to do a run of a hundred of more so I was placing them as orders came in. It made no financial sense to do this - I wasn't really making any money - but I was really enjoying seeing something I designed being bought and used in the real world. Unfortunately, due to some back and wrist issues it was very difficult for me to sit hunched over a desk for even an hour building boards. And now here we are.

There are a lot of folks on this list who see PnP purely from a commercial or financial perspective. "Does it make sense to spend X to get Y?" And that is fine, but it ignores a whole host of other reasons to do it. Many makers and hobbyists are not driven by a business and are just doing things for the heck of it, because they enjoy it, because they want a challenge, or just because it's fun to watch a robot do your work for you.

Jason

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/87ceb7cb-7375-4523-97fd-b2123c26bbf4%40googlegroups.com.

[Cri S]

If there is one user, that do it on regular basis, then setup time and
component (part setup)
optimize with time and is lower. My hesitation was as it's for fablab
and for new
user parts need to be zeroed and part library cannot grow in that case.
As you have written, you have shapeoko.
Cheapest is to add downlooking camera, pickup vacuum pen with pump to
Z axix with
(solid state) relay that actuate the vacuum pump
and connect external rotating stepper on bed with closed loop pipeline.
It's slower, and normally if you make board,
you can place strip feeder on the direction you need the component on
board, but it's
really cheap doing it with cnc.
on ReferenceBottonVision.java change findOffsets, and
after this line add the two other:
MovableUtils.moveToLocationAtSafeZ(nozzle, startLocation);
nozzle.place();
MovableUtils.moveToLocationAtSafeZ(camera, startLocation);

and at the end instead of "return offsets;"
MovableUtils.moveToLocationAtSafeZ(nozzle,startLocation.addWithRotation(offsets));
nozzle.pick();
nozzle.moveToSafeZ();
return startLocation.derive(0.0,0.0,0.0,0.0);


This are basically the modification to use CNC adding 16-25$ pen
vacuum pump and additional
10$ stepper. The pen need have straight needle and the 1/3/5mm plastic
suction cups.
With that you could assembly from 0603 up, and having cheap vga cam
(<10$) you can
do 0402 up to 33mm component, just check that field of view is 35x46mm
or lower and
select one webcam with 6 leds, that works great.
Just the needle is not ideal for 0402, but should work, it works best
with 0603 and up.





2016-08-04 14:05 GMT, Jason von Nieda <ja...@vonnieda.org>:

>>>> <https://groups.google.com/d/msgid/openpnp/4bd40515-ef0f-4c89-a2df-8855a177d6fc%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>> .

>>>>
>>>> For more options, visit https://groups.google.com/d/optout.
>>>>
>>> --
>> You received this message because you are subscribed to the Google Groups
>> "OpenPnP" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to openpnp+u...@googlegroups.com.
>> To post to this group, send email to ope...@googlegroups.com.
>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/openpnp/87ceb7cb-7375-4523-97fd-b2123c26bbf4%40googlegroups.com

>> <https://groups.google.com/d/msgid/openpnp/87ceb7cb-7375-4523-97fd-b2123c26bbf4%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .

>> For more options, visit https://groups.google.com/d/optout.
>>
>
> --

> You received this message because you are subscribed to a topic in the
> Google Groups "OpenPnP" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/openpnp/9275SyEGqBQ/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to

> openpnp+u...@googlegroups.com.
> To post to this group, send email to ope...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/openpnp/CA%2BQw0jzk_UrgB%2BswMJ0%3DfseMTYQ%2Bb%2B_Ea9HkjN%3DoXb7db1Xj6A%40mail.gmail.com.

[Rich Obermeyer]

Some very good information found at MDC.  Thx for sharing.

--

You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/1fd596b3-d123-43c0-b234-240b383cbace%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Michael Anton]

I agree with all of this, but do think it is worthwhile pointing out that having a machine does require setup which takes time.  That shouldn't be a deterrent if you need to have the functionality, or have some physical limitation that would prevent you from doing a manual job, but if everyone thinks it will make their life easier, and free up their time, then that is simply not the case for small runs.

I come at this with lots of real world commercial experience, and there are others on this list with the intention of using these machines for commercial use.  It seems to me that there should be a practical point to this, and many people without the experience may not understand the many pitfalls.  I'm just trying to put some of those out there.  If is not considered constructive to the project to also point out the setup requirements of having a machine, then I'll just bite my tongue instead.  Again my point was not to discourage, but rather meant as a bit of a reality check.

[Paul Kelly]

That’s nothin!

This thing: http://soledigital.com.au/hibeam.html has more than 500 mount points on two sides (OK, so a lot of them are LEDs) we use both machines to speed it up...https://www.youtube.com/watch?v=PBWvSFipJ1w&ab_channel=PaulKelly

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/4bc9240c-b840-46f9-83b6-0359cf07d875%40googlegroups.com.

[Jason von Nieda]

Hi Mike,

I'm not trying to stifle your input or anything, I'm just trying to provide an alternate viewpoint. I feel that too often people see these machines from a purely commercial interest, and I want to remind folks that there are lots of other reasons to own one besides making things faster, or making money.

Basically, I want people to know that it's okay to build a machine even if you don't have what other people think are good reasons for doing it :) 

Jason

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/2B2AFB30BEE34222BC197680C9FE40DC%40HomePC.