Problem with GUI

208 views
Skip to first unread message

Jiri

unread,
Dec 14, 2013, 4:40:15 PM12/14/13
to cyclone-p...@googlegroups.com

Hi there,
having connected Cyclone hardware to my Windows7 comp (tested OK with Pronterface) I progressed according to the GUI Tutorial by Kobus.I got to the window showing the PCB design (attached), but in the Z-probe step I got stalled with the attached error messages (The last sentence in Czech says "The system cannot find the indicated path."). Any hint how to fix it, please?


figure_1.png
WinError.png

Kobus du Toit

unread,
Dec 16, 2013, 4:40:28 AM12/16/13
to cyclone-p...@googlegroups.com
Hi

That GUI is the very old version.  Please get the latest

Kobus

Peter Heim

unread,
Dec 16, 2013, 7:26:47 AM12/16/13
to cyclone-p...@googlegroups.com
Using the latest on a laptop but i cant re size the screen and there’s no scroll bar so i cant get to the next button


Using ubuntu 12.4 64 bit
Asus laptop

regards
peter

Kobus du Toit

unread,
Dec 16, 2013, 1:39:09 PM12/16/13
to cyclone-p...@googlegroups.com

You can set the scale of the app in the properties file or use ctrl+mouse wheel or shift+mouse wheel to change the scale temporarily, can't remember which combination changes screen and which the circuit view

--
Has recibido este mensaje porque estás suscrito a un tema del grupo "Cyclone PCB Factory" de Grupos de Google.
Para anular la suscripción a este tema, visita https://groups.google.com/d/topic/cyclone-pcb-factory/WLVuSiN_qK8/unsubscribe. Para anular la suscripción a este grupo y todos sus temas, envía un correo electrónico a cyclone-pcb-fac...@googlegroups.com.
Para publicar una entrada en este grupo, envía un correo electrónico a cyclone-p...@googlegroups.com.
Para ver este debate en la Web, visita https://groups.google.com/d/msgid/cyclone-pcb-factory/ccb9b01e-b14b-4289-aec9-6bef227d84f0%40googlegroups.com.
Para obtener más opciones, visita https://groups.google.com/groups/opt_out.

Jiri

unread,
Dec 17, 2013, 9:16:00 AM12/17/13
to cyclone-p...@googlegroups.com
Hi Kobus,
many thanks, will keep trying.
Jiri

Dne pondělí, 16. prosince 2013 10:40:28 UTC+1 Kobus du Toit napsal(a):

Kobus du Toit

unread,
Dec 17, 2013, 6:27:09 PM12/17/13
to cyclone-p...@googlegroups.com
Have you been able to get the new gui up and running?  There is a manual.pdf, but it was out of date 2 days after I wrote it

Peter Heim

unread,
Dec 18, 2013, 5:30:29 AM12/18/13
to cyclone-p...@googlegroups.com

I have made some progress with the GUI
I have a circuit from Kicad but it's placed at x53 and y-60 (or there a bouts)

some of the problems i encountered
  • visolate is not working in ubuntu 12.4 64 bit but works in 10.4 32 bit (not a problem)
  • after playing around it works well and i can shift the board to a location that is easyier to work from (lower left)
  • the drill file is a pain some times it works and other time it don’t this may well be lack of experience but i can't shift the start position
  • I using a GBRL with a arduino ( a nice cheap option) but it runs at 9600 i tried to change it to 115200 but it wouldn’t run is there a way to change the serial rate to 9600?
I was able to test the mill by using the file generated with visolate and using the "universal gcodesender" to do the milling
 

regards
Peter

Jiri

unread,
Dec 18, 2013, 6:06:55 AM12/18/13
to cyclone-p...@googlegroups.com
Hi Kobus,
I did not have so much time I would need but I can confirm it is "doing something". I experienced the same problem with size of Visolate window as Peter, but in Windows is possible to resize the window so that it fits on the screen. I've connected the hardware with Sanguinololu and got to Z-probing, but then had to interrupt it ( was too late in the night).

What is the recommended soft for PCB design which would give compatible Gerber files, please? I've tried to export something from Eagle, but was not successful with that...

Dne středa, 18. prosince 2013 0:27:09 UTC+1 Kobus du Toit napsal(a):

Kobus du Toit

unread,
Dec 18, 2013, 6:27:05 AM12/18/13
to cyclone-p...@googlegroups.com

Hey

I used Diptrace and then you can use gerbv to convert it to something visolate can read. Just select the converter on the first Page. When gerbv opens the file, save it and it is converted

Kobus

Jiri

unread,
Dec 18, 2013, 7:20:41 AM12/18/13
to cyclone-p...@googlegroups.com
Kobus,
thanks. I really appreciate your support and responsiveness.
Best regards from Prague
Jiri

Dne středa, 18. prosince 2013 12:27:05 UTC+1 Kobus du Toit napsal(a):

Kobus du Toit

unread,
Dec 18, 2013, 7:22:25 AM12/18/13
to cyclone-p...@googlegroups.com

I am just happy someone is using it :)
Funny thing is I have not even used it yet, got side tracked to a dlp 3D printer design

Kobus du Toit

unread,
Dec 18, 2013, 7:28:49 PM12/18/13
to cyclone-p...@googlegroups.com
Please try the new version.  I added a config in the properties file so that you can change the speed.  You can change the contents of the property file in the system.  Just might need to restart after the change.  I don't have a board with me now so I couldn't test that the speed is actually working, I tested with the emulator and the code doesn't seem to crash

Peter Heim

unread,
Dec 19, 2013, 5:27:16 AM12/19/13
to cyclone-p...@googlegroups.com
Hi Kobus
No luck using the GBRL board with the GUI when the connect message comes up it displays garbage
if you have a arduino you can load the program on that you don’t need the controller shield
http://blog.protoneer.co.nz/grbl-arduino-library/ best part about this board is it's cheap($25) and simple to get started

Keep up the good work on the GUI I will be using it to prepare the board at least

Peter

Kobus du Toit

unread,
Dec 19, 2013, 5:47:56 AM12/19/13
to cyclone-p...@googlegroups.com

I will have a look Monday, I am off next week. Please give  me all your files for the  circuit and drilling so long

Jiri

unread,
Dec 19, 2013, 4:37:57 PM12/19/13
to cyclone-p...@googlegroups.com
Hi Kobus,
I have reinstalled my computer as I found that the GUI has some collisions with the symbols od decimal separators. In Czechlands we use decimal comma (,) instead of decimal point(.). So having my Windows set for CZ, the GUI crashes on "Create data points" (more precisely: it crashes even on "Drill GCode preview", but this can be skipped)- see the attached listingCZ. When I changed Windows settings to decimal point separator, I got up to  Z-probing, actually to the first contact of Z-probe / see attached listingEN. I was not sure what is wrong there (?maybe just the word  "echo" when only floating decimal is expected?). In the Emulation mode all works OK, with the Sanguinololu/Marlin does this problem.
To be sure I decided to do a fresh clean installation of Windows7.  Did not help unfortunately.

Dne čtvrtek, 19. prosince 2013 11:47:56 UTC+1 Kobus du Toit napsal(a):
listingEN.txt
listingCZ.txt

Kobus du Toit

unread,
Dec 19, 2013, 5:16:11 PM12/19/13
to cyclone-p...@googlegroups.com
Can you please give me your circuit and gcode files.  I will change my machine to use the same formatting as yours and I can then test


--
Has recibido este mensaje porque estás suscrito a un tema del grupo "Cyclone PCB Factory" de Grupos de Google.
Para anular la suscripción a este tema, visita https://groups.google.com/d/topic/cyclone-pcb-factory/WLVuSiN_qK8/unsubscribe. Para anular la suscripción a este grupo y todos sus temas, envía un correo electrónico a cyclone-pcb-fac...@googlegroups.com.
Para publicar una entrada en este grupo, envía un correo electrónico a cyclone-p...@googlegroups.com.

Kobus du Toit

unread,
Dec 19, 2013, 5:22:00 PM12/19/13
to cyclone-p...@googlegroups.com
You said it also crashes on the drill preview.  Please then also send me the drill file

Can you also send me the output from the log on the right hand side of the GUI?

Jiri

unread,
Dec 20, 2013, 3:06:36 AM12/20/13
to cyclone-p...@googlegroups.com
Kobus,
for testing, I would not dare to use other inputs than those you provided :-)  (all files attached). I even created folder C:\Temp\brain :-)
But the problem of decimal separator I've solved (I believe) by change of Windows regional setup.
What is the obstacle now is the unfinished probing... The response from hardware seems to differ from the response of emulator-see attached screenshot.
Can you direct me to the current version of Firmware?
brain.rar
Probing Stalled.png

Jiri

unread,
Dec 20, 2013, 5:58:31 AM12/20/13
to cyclone-p...@googlegroups.com
Hi Peter,
I downloaded the program, modified the Baud rate to 115200, compiled in 1.05 IDE, uploaded to Arduino Uno and tested. Works just fine (no endstops connected, of course). Attached is the printscreen showing start of probing and the libraries folder with modification. Good luck!
Jiri

Dne čtvrtek, 19. prosince 2013 11:27:16 UTC+1 Peter Heim napsal(a):
GRBL.rar
GRBL.png

Kobus du Toit

unread,
Dec 20, 2013, 6:02:24 AM12/20/13
to cyclone-p...@googlegroups.com

Jiri I will change the code tomorrow so that it doesn't parse the results so we can see what comes back from your controller. I am using Marlin from Carlos.

Jiri

unread,
Dec 20, 2013, 7:42:14 AM12/20/13
to cyclone-p...@googlegroups.com
Kobus,
that's perfectly OK, take your time. I just did not want to be the showstopper :-))
I am using firmware from Carlos'es github with very minor modification- in order to have correct logic of endstops. I just wanted to be sure I have the current version of it, unlike the GUI originally.
Reaction to G30 command seems to look like this:

ok Z:-5.52
echo:endstops hit:  Z:-5.52

(I guess the second line is redundant and actually making problems to the parser?. I tried to eliminate it in firmware but without success.)

Kobus du Toit

unread,
Dec 20, 2013, 1:38:29 PM12/20/13
to cyclone-p...@googlegroups.com

Nice, that I can fix easily

Kobus du Toit

unread,
Dec 20, 2013, 7:05:09 PM12/20/13
to cyclone-p...@googlegroups.com
I copied a new version of the app.  Hopefully it sees both lines as one line, it seemed that way looking at the error.  it will then take the last part of the line after the colon as the probing value


--
Has recibido este mensaje porque estás suscrito a un tema del grupo "Cyclone PCB Factory" de Grupos de Google.
Para anular la suscripción a este tema, visita https://groups.google.com/d/topic/cyclone-pcb-factory/WLVuSiN_qK8/unsubscribe. Para anular la suscripción a este grupo y todos sus temas, envía un correo electrónico a cyclone-pcb-fac...@googlegroups.com.
Para publicar una entrada en este grupo, envía un correo electrónico a cyclone-p...@googlegroups.com.

Jiri

unread,
Dec 21, 2013, 6:07:53 AM12/21/13
to cyclone-p...@googlegroups.com
Hi Kobus,
works fine, we are almost there... The next problem is that during probing the Z axis does not lift up, i.e. now behaves like this:
1. getting to the first probe XY position
2. slowly decreasing Z until probe gets contact
3. moving to another XY position with the same Z coordinate

Obviously, step is missing between 2 and 3 to lift up probe a little, otherwise the probe (tool) would scrap the PCB or even break itself. I do not know whether the lift should be part of firmware G30 command (but doubt it), so should be probably solved by this host?

Dne sobota, 21. prosince 2013 1:05:09 UTC+1 Kobus du Toit napsal(a):

Kobus du Toit

unread,
Dec 21, 2013, 6:10:07 AM12/21/13
to cyclone-p...@googlegroups.com

Cool, I will look at it tomorrow quick. Glad you are able to get it so far

Kobus du Toit

unread,
Dec 22, 2013, 2:24:30 AM12/22/13
to cyclone-p...@googlegroups.com
Please see if the code does the correct thing now.  The probe screen has lift height now and it is by default 2.  If 2 isn't the best value I can change the default to something else.  After each probe it is supposed to move to position 2 before it moves to the next spot

Jiri

unread,
Dec 22, 2013, 3:00:01 AM12/22/13
to cyclone-p...@googlegroups.com
First try: works fine.

Dne neděle, 22. prosince 2013 8:24:30 UTC+1 Kobus du Toit napsal(a):
probing.jpg

Kobus du Toit

unread,
Dec 22, 2013, 5:04:05 AM12/22/13
to cyclone-p...@googlegroups.com

So is 2 A good value?

Jiri

unread,
Dec 22, 2013, 5:27:13 AM12/22/13
to cyclone-p...@googlegroups.com
It works but I believe it could be lower. Currently the move is about 5mm, I believe even half of it should be fully sufficient.

Dne neděle, 22. prosince 2013 11:04:05 UTC+1 Kobus du Toit napsal(a):

Kobus du Toit

unread,
Dec 22, 2013, 5:29:12 AM12/22/13
to cyclone-p...@googlegroups.com

How is the gcode that it produced for you? I will make the default 0.5 tomorrow, maybe even make it a property in the file

Jiri

unread,
Dec 22, 2013, 5:39:57 AM12/22/13
to cyclone-p...@googlegroups.com
I haven't done any thorough analysis (my wife delegated to me some pre-Xmas tasks :-) ) but I did a "dry run" of the resulting gcode and seems ok.
Attached for comparison and analysis.
Dne neděle, 22. prosince 2013 11:29:12 UTC+1 Kobus du Toit napsal(a):
gcode2analyse.rar

Jiri

unread,
Dec 23, 2013, 6:23:10 AM12/23/13
to cyclone-p...@googlegroups.com
Hi Kobus,
some new findings. I moved closer to the final hardware configuration (pcb vise, conductive tool, probe clips). What I did is that I increased artifically the zero Z level before 1st probing by a piece of paper 0.28mm thick, made the probing and then run the resulting gcode "milling" (but switched off spindle). As expected, the tip of the tool remained everywhere above the plate (maybe at the expected distance 0.18mm, but I did not measure it). So I believe it works in principle.
There certainly are some minor issues to improve, for example:
- in case that the value of probe is negative, probing works fine. In case the value is positive, the probe is not lifted but lowered (!) before going to next XY  probing coordinates. In fact this moves up all the spindle at the tip of the tool! Luckily, an easy workaround is changing the value of Z homing so that the probes are always negative.
- the pattern of probing. It would save moves and time if probing pattern would be snake-shaped, rather than returning from Ymax to Ymin before probing next X row.

I am sure it would be easier for you to fix the code minor issues in case you could observe the machine in reality. Any chance you put together your own soon?

Kobus du Toit

unread,
Dec 23, 2013, 6:35:40 AM12/23/13
to cyclone-p...@googlegroups.com
Thank you for the feedback
I will see what I can do about changing the probing to snake, might be a bit more work than the other fixes.
I will probably not get back to building the machine anytime soon, spending every bit of time I have on the DLP printer now.  I am completely happy to fix up the software for you guys

I assumed the Z value will always be negative so I wrote the code that way.  I can remove the multiply by -1 and we can see what will happen.  I copied a new version without the -1

    double findZValue(final Coordinate coordinate, final ArrayList<ProbingData> probingList) {
        double result = 0;
        for (ProbingData probing : probingList) {
            if (probing.contains(coordinate.x, coordinate.y)) {
                result = probing.zValue / ZVALUE_COMMA_CALC * -1;
                break;
            }
        }
        return result;
    }

Jiri

unread,
Dec 23, 2013, 8:02:14 AM12/23/13
to cyclone-p...@googlegroups.com
Kobus,
slightly better:
- works fine for negative value of probing sample
- does not neither lower ( :-) ) nor lift ( :-( ) for positive values.
But as I've written: when value of Zmax is set up in firmware the way that probes can only be negative, this is not a problem.

I willl give you a rest now. First - Xmas is coming. Second -you can progress with the DLP printer while I will study Diptrace :-)

Wishing you peacefull Christmas!
And many thanks for sharing results of your work.
Jiri

Kobus du Toit

unread,
Dec 23, 2013, 8:06:47 AM12/23/13
to cyclone-p...@googlegroups.com

You can change the size of the screen in properties, there are scale properties. You can also use ctrl+mouse wheel or shift+mouse wheel to temporarily change the scale.

I know now for next time not to use grid layout, it causes these kinds of issues

Jiri

unread,
Jan 1, 2014, 4:47:04 AM1/1/14
to cyclone-p...@googlegroups.com
Hi Kobus,
all the best for 2014!

Another question to GUI2.
The size of working area is currently limited in Carlos'es firmware to 135x84mm:
// Travel limits after homing
#define X_MAX_POS 135
#define X_MIN_POS 0
#define Y_MAX_POS 84
#define Y_MIN_POS 0
#define Z_MAX_POS 50
#define Z_MIN_POS -50
it would not be a problem to modify this, but I believed it is a Cyclone standard...
It seems that the GUI2 probing is actually considering different (bigger) area?


Dne pondělí, 23. prosince 2013 12:35:40 UTC+1 Kobus du Toit napsal(a):

Kobus du Toit

unread,
Jan 1, 2014, 4:57:49 AM1/1/14
to cyclone-p...@googlegroups.com
HI

Good year to you also and all the guys on the forum :)

I didn't know there where limits.  My code looks at the size of the PCB and then creates the probing datapoints for that size of PCB.  When you select the speed it takes the width and divides it into parts and it takes the height of the PCB and divides it into parts





--
Has recibido este mensaje porque estás suscrito a un tema del grupo "Cyclone PCB Factory" de Grupos de Google.
Para anular la suscripción a este tema, visita https://groups.google.com/d/topic/cyclone-pcb-factory/WLVuSiN_qK8/unsubscribe. Para anular la suscripción a este grupo y todos sus temas, envía un correo electrónico a cyclone-pcb-fac...@googlegroups.com.
Para publicar una entrada en este grupo, envía un correo electrónico a cyclone-p...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages