-
Notifications
You must be signed in to change notification settings - Fork 113
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
General discussion #28
Comments
Hello really nice project. IMHO one of the best that i saw for GRBL control. I have a few ideas how to improve program:
|
Hi, Thanks!
|
Hi there martin2250! I just found this wonderful project and was wondering if you could answer a question of mine: I have an Arduino Mega2560 with a Ramps 1.4 board and plan on flashing GRBL on it. My question now is: The readme states that the probe should be connected to the A5 pin of the Arduino. Would it be possible to change the pin to the default Z_MIN pin of the Ramps board which would be D18? |
Hi sirsensor, yes absolutely! you can adjust the probe pin to any free pin on your 'Mega by changing these lines in 'cpu_map.h' :
you'll probably need to change other pins (and peripheral registers) as well, I think there is a template cpu_map for the AtMega somewhere in the grbl repository. |
Ok after like 4 1/2 hours I was finally able to flash a proper GRBL version on my arduino mega2560 and set up everything correctly. At this I just want to say: "Thank you for this f****** nice program". The UI is flawless. Both the 3D viewport and the sidebars look so goddamn good. The only thing I need now are a better engraving bit and some DRC fine tuning so my traces are not hairthin. |
Hi! Using "pcb-gcode.ulp" I use this parameters: resulting in something like that: You can view the resulting traces as preview within "pcb-gcode" and can check whether all spacings are sufficent and well done. And: Yes of course, I use Martins wonderful tool for that ;-) Harald |
Additional info: The design is two sided, smallest structures are 0,2 mm (see SMD footprint top right) and the following pic: Such small structures are impossible to mill without a hight adjustment as done with Martins super duper tool. Drawback of those fine bits is, they crack very very fast and do not stand for long (if not cracking ;-) Harald |
Nice!! Very good!!!
<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
Livre
de vírus. www.avast.com
<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>.
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
…On Mon, Oct 2, 2017 at 1:00 PM, deHarro ***@***.***> wrote:
Additional info:
I once used bits with 0,1 mm tip and 10° to realize this layout:
[image: grafik]
<https://user-images.githubusercontent.com/26298406/31086519-5682b262-a79a-11e7-8603-11cf0ff812ae.png>
The design is two sided, smallest structures are 0,2 mm (see SMD footprint
top right) and the following pic:
[image: grafik]
<https://user-images.githubusercontent.com/26298406/31086620-95036ec8-a79a-11e7-82f3-ba81eee0f86a.png>
Such small structures are impossible to mill without a hight adjustment as
done with Martins super duper tool.
Drawback of those fine bits is, they crack very very fast and do not stand
for long (if not cracking ;-)
Obviously milling times rise dramatically since every trench between
traces has to be cut several times to get the spacing (you don't want to
have such low spacings of 0,2 mm between each and every trace on your
layout, soldering would be awful).
Harald
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#28 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AFKGbqlnKhPxUMvKyE4MdaPXULUMvttrks5soQilgaJpZM4NFfEl>
.
--
Sds
L.Abbadia
|
Thanks everybody! |
Thanks! :-) |
Hi everyone, got a issue with the Mega2560, @sirsenor can you help me on thread : #32 ?? |
Would it be possible to use the software for woodworking? I would love to engrave something into wood, but the wood may have irregularities and the height at point A may be different than at point B. I already saw that the heightmap is stored as an xml file and thought that measuring out a grid of points and storing them in the xml file might work out perfectly fine. Would that be possible? |
Yes, that would be absolutely possible, but why bother? You could just as well use a switch as a probe (you'd need to re-zero Z, but that would be trivial) or move a small metal plate to each location as it probes. |
Thank's both, it works well ! |
Hi Martin,
|
Hi Harald, 1: yes, it just changes the initial size when opening the 'create height map' window to the size of the current toolpath. 2: weight in this case is not the 'mass' kind of weight. One new feature is that the probing pattern is not fixed (aka not always a meander like before). rather OpenCNCPilot always searchea for the nearest point (relative to the current work position) that has not yet been probed. the new setting is used to weigh the distances in X and Y direction differently. (aka a point that is at X1Y0 is thr same 'distance' as X0Y2,l when the X axis weight is 2). I added this so I can tell OpenCNCPilot to move my X axis preferably and minimize the number of Y moves (moving my machine in Y jerks the entire table, X not quite so much) Bent axes and crooked tables are already compensated by the HeightMap Cheers |
Hi Martin, Again on bent axis: The difference to the already implemented functionality is, OpenCNCPilot would have to overlay this overall HeightMap only partly (only the part of the table which is used by the given design). For my setup (my LCF-1 with bent X axis) only a few (lets say 10..20) points have to be sampled along the X axis (in theory). Ok, it should be a mesh, so 10..20 on X axis by 10 on Y axis (my table is 50x32 cm, usable area roughly 30x22 cm). Q1: Assuming a perfectly even table (as it should be with my new table), would such a functionality result in an evenly planed surface, even if the axis are bent? ... I think, yes, it would. |
You can always save a height map of the entire surface and open it later. I don't think adding this functionality would be a good idea, for reliabilty reasons. With the current method of applying the height map to a complete file, you get a preview of the modified toolpath. also modifying the toolpath on the fly could lead to some unexpected behaviour. I'd rather have OpenCNCPilot only do exactly what it's told to do.
you also can't assume that your material will have the same curvature as your table, especially when it's concave (on a convex surface with clamps at the border this could work) Q1: disregarding that the tool is not perpendicular to the surface and that your axes will probably deform unter load, yes. I would not rely on it though |
As in the past I worked mostly with POM plates, I assume strongly, that those plates are as flat as the table :) The parts I machined yet were not so big, so the curvature had no big influence, and the one or other a little bit bigger part was made in such an early state, that I had no idea that my mill may be not as accurate as I imagined :-) But all in all I get your ideas and can follow. Ok, was just an idea. |
hello maybe you can implement some things from here https://github.com/Denvi/Candle as options; |
Hi Ilie, |
aaa you can use keyboard to move then is ok. |
I have a doubt, don't know if this is the best place to ask... please forgive me if it's the wrong place. What does the bellow message means ? "the toolpath is not containned in the height map" When I click to apply the heighmap I got it. best regards, |
probably it means that you made a wrong hmap with other origin or something. so the positon where you need to touch with tool is outside of the area of hmap |
To avoid this I normally choose bigger coordinates for my pcb and always
even numbers.If my pcb is 77x 68 i choose 80 x 70 and it always works fine
Em 25 de jan de 2018 11:35, "ilie321" <notifications@github.com> escreveu:
… probably it means that you made a wrong hmap with other origin or
something. so the positon where you need to touch with tool is outside of
the area of hmap
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#28 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AFKGbi5RicBxsr21zFs-u-Bo50Ne-oxAks5tOGbxgaJpZM4NFfEl>
.
|
millimeters...
Em 25 de jan de 2018 11:44, "Luiz Abbadia" <luiz.abbadia@gmail.com>
escreveu:
… To avoid this I normally choose bigger coordinates for my pcb and always
even numbers.If my pcb is 77x 68 i choose 80 x 70 and it always works fine
Em 25 de jan de 2018 11:35, "ilie321" ***@***.***> escreveu:
> probably it means that you made a wrong hmap with other origin or
> something. so the positon where you need to touch with tool is outside of
> the area of hmap
>
> —
> You are receiving this because you commented.
> Reply to this email directly, view it on GitHub
> <#28 (comment)>,
> or mute the thread
> <https://github.com/notifications/unsubscribe-auth/AFKGbi5RicBxsr21zFs-u-Bo50Ne-oxAks5tOGbxgaJpZM4NFfEl>
> .
>
|
I have though it has something to do with the "borders" (limits), but I created the heightmap from the gcode size, i.e. the software defined the boundaries. If I click on "simplify" I can apply the height map without any problem. Anyway, thank you both for the hints. best regards, |
Hi Alexandre, the error "toolpath is not contained" shouldn't appear anymore, even if the HeightMap is smaller than the toolpath. If you still have them, can you please upload the gcode file and height map to pastebin and open a new issue? Greetings |
Just started to use the latest version of OCP (v1.5.10) and have just noticed that the 'Size from Gcode' button is missing from the 'New Heightmap' window that pops up when using the probing Panel. Is this by design, or has it been omitted inadvertantly when making the changes during the update from v1.5.9? I only ask as I can't quite see the point of changing something that worked flawlessly in the previous version, and from my personal point of view will be sorely missed. Am now changing back to v 1.5.9 as I much prefer the probing setup as it was - too much of a faff having to input all the coordinates each time.... Just my take on it, others may prefer otherwise of course. |
Hi happycncer! This one is on me (look here #141). But the feature is still there: If you leave "Margin" at zero, nothing is changed against the former solution (besides the text on the button ;) Harald |
My fault entirely - didn't realise that the 'Apply' button would do anything without a positive number in the Margin Box, duh! I was just clicking on the OK button because I didn't want any border line - played around trying to get what I wanted by putting in the coordinates and decided it was too time consuming - hence my comments above. That was why I went back to 1.5.9 because I knew it 'just worked' as they say. Age is getting to me..... |
I was puzzled, too, whenI got the "done" message from Martin and had a first look at the new dialog. |
Just as an aside, is there any way of increasing the retraction height when using the Probing routine to create a heightmap? Have made myself a microswitch type probing 'bit' and would like it to raise it up a little further than it does at the moment - but can't seem to find anwhere within the settings that seems to address this. Perhaps my eyesight is failing along with everything else? As usual, any help would be greatly appreciated.... willow52 |
On the window "Machine" go to "Settings" and then "Probing" tab there you
will be able to change the parameters
Em qui., 29 de out. de 2020 às 10:16, happycncer <notifications@github.com>
escreveu:
… Just as an aside, is there any way of increasing the retraction height
when using the Probing routine to create a heightmap? Have made myself a
microswitch type probing 'bit' and would like it to raise it up a little
further than it does at the moment - but can't seem to find anwhere within
the settings that seems to address this. Perhaps my eyesight is failing
along with everything else?
As usual, any help would be greatly appreciated....
willow52
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#28 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AIQT2SGIXH5CAPFFJQ64GPLSNFTMBANCNFSM4DIV6ESQ>
.
--
Alexandre Benson Smith
www.thorsoftware.com.br
Santo André - SP - Brasil
(11) 4992-3496
(11) 9 9641-1808
(Enviado via webmail)
|
Boa tarde ao todas em Brasil! Alexandre, I did try that this morning, and alteration of the standard 5mm setting to 10mm made no difference at all. Thought at first it was because I hadn't shut down and re-started the program, but no, have now done that and it still makes no difference. My switch 'bit' works perfectly, but although the safe height is set at 5mm, it only retacts while moving from one point to the next by no more than 2mm - and 0.5mm of that is to release the microswitch. Actual clearance above the workpiece is only a little over 1mm, hence my desire to lift it a bit higher..... which I can't seem to do? willow52 |
Hi willow! Give Martin the chance to correct it... :-) Harald |
Even though I'm not really active here ATM, I read all posts... eventually. Mention this in an issue or submit a pull request, and I'll fix it soon™ :) Cheers! |
It's about time we all take this opportunity to correct some of the wording in OCP to make it more meaningful. Kind regards, |
Hello Harald, The CNC milling machine is The spindle nuts are custom-made from POM plastic and are set free of play I had all along maintained that trapezoid lead screws required anti-backlash nuts. Cheers, |
Hi Hanspeter, My idea back then with gurus Z-axis was, that with his very heavy spindle and sled he will never have problems with backlash during probing. The sheer masses of spindle and Z sled will pull the Z axis down reliably during the probing process. As I wrote later on, it is a totally other story when he is actually milling. Then there are forces against the gravity and backlash will occur. (Actually a very old topic you are referencing, don't you agree? ;-) Harald |
Yup, just collect your suggestions in the issue :) |
Yes Boss, I do. |
Hello, is it possible to modify the connection to allow using a TCP port? I can through an ESP8266 make a transparent COM port between TCP and serial easily, I just need the program to support this connection. If it is not possible to implement for you, I will try to do it in due course, I will share the code if I can make it work correctly. |
Hi, Remember OCP connects directly to the USB port of Arduino Uno or Nano. If your aim is to connect to GRBL over long distance, then get yourself RS485 converter, Cheers, |
Using a tcp connection would allow the machine to be connected using the installed local network, regardless of the distance, or simply a WiFi connection without installing cables, so the control pc does not have to be near the milling machine, avoiding dust and vibrations that in the long run they will end up damaging the pc. I am already implementing the connection using TCP, I need to test that the machine behaves correctly using the new connection, from what I saw in the programming, martin2250 had considered using other data sources, perhaps, or perhaps it is just part of the modeling program that he decided to use, he only added a new type of connection and what was necessary to master it in addition to the new parameters. |
In that case consider using USB Extender 165ft Kit with 4 USB 2.0 Hub you can source from amazon. You may also take a look at this url https://github.com/gnea/grbl/wiki/Grbl-v1.1-Interface Cheers, |
Another reason to do it is that the pc that I am dedicating to this, has its USB ports in not very good condition, considering that it is a pc that I rescue, I am not complaining. But, finally, I only intend to give more freedom to the program, which supports an ethernet connection to link the machine opens many paths, like the one I need with my machine. |
If I recall correctly OCP connects to Arduino via USB, if you want OCP Gcode ported to ethernet then you should get If your USB on your PC is doggie get a new PCI to USB card. If it happens to be a laptop with faulty USB, get yourself Cheers, |
As I said, although it may not have been understood by the quality of the translator, the modification is already implemented, martin2250 only has to decide whether to add the changes to the main program or not. |
Hi @gonzalocargut, |
I have translated the changes into English, it is not my natural language, so sorry if something looks strange. Review the changes to open the pull request, this is my first time doing this process, so hopefully it goes well. |
Hi Martin, I have installed a tool length sensor switch, similar to your setup in the video, and I have learned how to pause a program to allow me to use G53 to move to the Sensor successfully. What I cannot seem to do is to get the OCP program to exit a 'Hold' situation which occurs as soon as the M0 has been called. Only thing that appears to cancel the 'Hold' is a soft reset. Is this right? It is not really clear from your instructional video what the correct sequence of operational commands are to allow me to carry on and run a second toolpath. You do/did say during the video you would be putting up details of the macro on the WIKI soon (without any promises!) but I have looked everwhere I can for help but can't find anything more helpful anywhere. Can you give some further guidance on this at all please.... |
Hi Richard, you should be able to cancel the 'Hold' by pressing the cycle start ('Start') button (between 'Hold' and 'Soft reset'). I usually export each tool as individual file, so I don't have to pause in the middle of a file. By checking the "Pause" checkbox in the file tab, you can also export multiple tools as one file. I do have to admit I forgot about the wiki entry, not sure when I'll be able to get around to it. Cheers |
Thanks once again Martin, think I've got it now. I was trying to do it running two files saved at the same time - adding an M0 between the two separate parts of the combined file. Not now surprised that it wasn't working as expected. I can now confirm that I can re-start OCP as you outlined and I should be good to start using this now. Looking forward to using this quite a lot in the future and it was definitely well worth the time spent setting this up after all! |
Hi everyone, I just enabled discussions for the OpenCNCPilot repository. Feel free to share your thoughts and projects over there from now on! https://github.com/martin2250/OpenCNCPilot/discussions |
as the title states.
please only use this thread for questions and discussion and open new feature requests for actual issues with OpenCNCPilot.
Martin
The text was updated successfully, but these errors were encountered: