Hi Jason,
Do you want to check it with opencv or openpnpCapture driver?
--
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/ad0f3ed1-87c3-4c9d-8b7b-4d959980650c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Ok, tomorrow morning consider done ;)
--
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/cfd95705-6e15-41b5-a5d0-4f253b2d292c%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/20d3d604-4acf-4639-a671-3d0fdb13dff9%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/4f551c4f-0115-4730-b90a-75c1f3937c91%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/0047fa19-57bd-4d11-9193-e11f8e935380%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/7e26e891-2e15-40ec-bd51-0a0cb2c151a2%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/550015dd-53f1-44d5-a520-9555ff150d98%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/445c4422-10a2-4ef6-a0f2-fc009337e3e9%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/f14a436d-5b3e-4e21-b2cf-dee1f8222cd6%40googlegroups.com.
Hi folks,I've just pushed a new version of OpenPnP with a major update to OpenCV, which is the computer vision library we use. We went from version 2.4.11 to 3.4.2.I'd like to ask that a few people download the Latest version from http://openpnp.org/downloads/ and give it a try. In particular, it would be nice to get some Windows and Linux users, both 32 and 64 bit, if possible. I have personally tested Mac, and Windows 64bit.So, if you have a moment, please go to http://openpnp.org/downloads/ and download "Latest" for your platform. Install it, start it up and see if everything seems to be working. In particular, check that cameras are still initializing, and that your pipelines still work as expected.Let me know if you run into any issues.Thanks,Jason
--
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/1d391c8d-64f5-4869-997b-fdd4baa1f8da%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/6f110acd-8f52-41a0-b6f6-e0d9afe10136%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/CA%2BQw0jw7ikbN%2BKiveTguOFPopOX9fnVFKTovE7hemAb3ivjyJg%40mail.gmail.com.
Yeah.... No.
So our 1980s vintage machines have a repeatability of 0.02m and a speed of 72000mm/min
They have 1000 line (4000CPR) encoders and 20mm pitch screws
72m/min needs a screw speed of 3600rpm and that’s 240KHz per axis. Just to get to an old machine good for 1000-2000CPH.
0.02mm is just four encoder counts, I doubt you could get the following error much less than that. Besides, you must have more resolution than accuracy or you cant get repeatability. Our machines spec to 0.005mm resolution.
So this is a real world example of the kind of precision that was required in the 80’s to do the job. I bet todays machines are a notch or two better and faster.
On top of that, we’ve now run a few thousand boards in our OpenPnP retrofitted machine, so we’ve got a good feel for it’s limits.
As part of our current “lets revisit the retrofit and get it right this time” project, we’ve bought a KFlop board which looks like it will be fine, there’s a few show stoppers for us with the smoothie.
We’ve also gone to 450W brushless AC servo’s.
I’ll post a writeup when it’s all done and making money.
PK
Sent from Mail for Windows 10
--
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/c43d22f5-9b6e-4a0c-a138-111c122d5936%40googlegroups.com.
Well,
@PK
If you retrofit older machine - which was originally really good one ( casted iron or welded/heat processed), then indeed you can get high accuracy.
Usually such machines require fork lift :-) , I would not be able to place it in my office :-(
Yeah, real SMT lines are big things. We’ve just added a selective solder robot and stencil printer. Even with our little batch reflow ovens it’s starting to get tight in the assembly room.
But in such case step/dir control is a bottleneck ( well, you can always use something like that - https://dynomotion.com/KFLOP.html ).
I would use servos with CAN/RS485 direct control - you do not need step/dir controller for most of AC servos as they can be just told where to drive directly.
True, Kflop is a good middle ground though because it will take GCode over serial, someone has already done an OpenPnP implementation. It’s also reasonably priced.
PK
@Marek
Smoothie v1 and v2 use similar hardware to my CNC controllers - i can get 125kHz in case of v1 ( LPC1769) and 250kHz ( LPC4330). I could get even 400kHz - 250kHz is for 5 axes simultaneously. Just note that 400kHz usually means differential lines for step/dir control.
What kind of servo ( type) do you have ?
W dniu sobota, 9 lutego 2019 19:42:26 UTC+1 użytkownik Marek T. napisał:
Sure, typo, 100kHz not MHz.
It's not any overkilling. If I want get max speed of my motors I must use 100kHz steps and 2x mode for encoders (lower accuracy) or 200kHz and 4x mode. Sure can use x4 and 100kHz - and get it working very slow. Encoders are 1000dpr resolution. So or speed or accuracy. Don't forget there is leading screw by the way with its transmission rate.
I heard there are some controllers on the market with step frequency like 400kHz. So really 100kHz is nothing special.
Btw, Smoothieboard allows to run up to 145kHz at my settings, then begins hang up if set even 146. As it depends on different settings and somehow on the memory usage probably it is really not stable (also not recommended by smoothie creators) So I've tested it but didn't want to risk and left it set default 100.
--
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/b6cf6016-3001-4563-a769-6d80c4bed708%40googlegroups.com.
Well,@PKIf you retrofit older machine - which was originally really good one ( casted iron or welded/heat processed), then indeed you can get high accuracy.Usually such machines require fork lift :-) , I would not be able to place it in my office :-(But in such case step/dir control is a bottleneck ( well, you can always use something like that - https://dynomotion.com/KFLOP.html ).I would use servos with CAN/RS485 direct control - you do not need step/dir controller for most of AC servos as they can be just told where to drive directly.@MarekSmoothie v1 and v2 use similar hardware to my CNC controllers - i can get 125kHz in case of v1 ( LPC1769) and 250kHz ( LPC4330). I could get even 400kHz - 250kHz is for 5 axes simultaneously. Just note that 400kHz usually means differential lines for step/dir control.
What kind of servo ( type) do you have ?
W dniu sobota, 9 lutego 2019 19:42:26 UTC+1 użytkownik Marek T. napisał:Sure, typo, 100kHz not MHz.
It's not any overkilling. If I want get max speed of my motors I must use 100kHz steps and 2x mode for encoders (lower accuracy) or 200kHz and 4x mode. Sure can use x4 and 100kHz - and get it working very slow. Encoders are 1000dpr resolution. So or speed or accuracy. Don't forget there is leading screw by the way with its transmission rate.
I heard there are some controllers on the market with step frequency like 400kHz. So really 100kHz is nothing special.
Btw, Smoothieboard allows to run up to 145kHz at my settings, then begins hang up if set even 146. As it depends on different settings and somehow on the memory usage probably it is really not stable (also not recommended by smoothie creators) So I've tested it but didn't want to risk and left it set default 100.
--
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/b6cf6016-3001-4563-a769-6d80c4bed708%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
I’m on the road at the moment, let’s see if I can find a picture..
https://photos.app.goo.gl/8QQyge818fA1NsmP9
Made by Waterun, I visited them on a recent trip to China and took some boards with me. They lashed up a fixture and it did a good job.
PK
From: Michael Anton
Sent: Saturday, 9 February 2019 7:46 PM
To: OpenPnP
Subject: Re: [OpenPnP] Re: Request for Testing: Major OpenCV Update
@PK
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/ed26e150-2a21-402d-ace6-738679281e3e%40googlegroups.com.