@Tegbert Understood (finally) 
I’ll be surprised if the results are any different with UGS. The gcode will be the same. If you can export a dxf from CC, I can see what happens in VCarve, if you like. If the design is proprietary, you can PM me.
Tom, I couldn’t have said it better, I haven’t had time to respond for the last few days and was going to say the same thing. I too have the same results if I use center as my start point. I have g-code that ran perfectly on USG and they end up with a Y+ offset when I use Gsender.
I have made jigs for small parts and I need to have a constant starting point to reference from.
Grant, Sorry for the confusion with the .crv file and it’s dimensions. I ran the file with a piece that was 12x12 and had the results that I first mentioned. The second piece ( the .crv posted) had the slightly different dimensions that were in the file that I posted, but the offset was still present on the second finished jig.
@braehill No worries, Len. I am at a complete loss. If you can run that gcode in UGS, ensuring that the material dimensions in VCarve match the dimensions of the actual piece, then clearly something is going on in gSender on your machine. I ran it in gSender with the correct material dimensions and it looks perfect. I can only conclude then, that gSender per se is not the issue.
I’m hoping that someone will see what I am missing. I’ll bow out by suggesting that you may want to flash the firmware on your controller. You lose nothing and it may solve the problem. Other than that, I have nothing. If you do solve this, please post the solution. I need all the education that I can get.
Len, I am wondering if you also have issues with your probe? I can only probe in the z axis and xyz axis, cannot probe xy or x or y independently…Just curious since we seem to be having the same issue!
Thanks
Tom, My only other post on this forum is about not being able to probe in imperial settings but does fine in metric. I find my edge the old fashion way with a piece of .001 shim stock rather than irritating myself with the probe for now. I run into problems in the Y-Axis where it throws Alarm #2 , something about an unexpected movement before 38.2 and some other stuff that I don’t remember. Somebody from Scienci asked me to post my console next time it happens and I haven’t used my machine since to post it.
One thing that I might add is that I run my machine on a Rasberry Pi 400.
@Tegbert @braehill I apologize if I am stating the obvious, guys, but for the first time today, I noticed that when using the touch plate to set either XY or X or Y only, you don’t start with the bit above the circle on the touchplate. I may be the only guy here who didn’t know that. I have never used the plate to set only XY or X or Y. However, could this be part of the problem setting your axes?
Grant, I use a touchplate from another manufacturer. I set the measurements in the settings tab for it as it’s slightly different size than Sienci’s .
I have a couple different macros for my probe plate that were written for USG but I haven’t taken the time to try them in Gsender yet. I’ll post my findings when I do.
1 Like
It’s never just a lightbulb, you have to rewire the whole house.
1 Like
The old saying (highly paraphrased) is, “Spit into one hand, guess into the other. See which one fills up faster!”
Give yourself some time and literally, in a step-by-step manner, write down everything you do to demonstrate the problem that you are having. Post the steps.
Somewhere in that process, someone here will recognize some anomaly.
Otherwise, there’s only guessing and anecdotal information.
You have to use your machine a lot to get used to all the procedures for successful CNC.
Hello all,I have bee away from my machine for a couple of days but wanted to update progress. I downloaded latest version of UGS and set axis’s manually and the resulting job file ran as designed in CC. I am beginning to understand that all of the programs have there bugs though, UGS would run out of control at times when jogged or I would not be able to rapid jog and had to upload several times to complete initial evaluation. I will try probe with UGS and advise of results.I like the appearance and features of g-sender and would prefer to use it?
Well, my post before yours applies to your situation with gSender, too. Anecdotal information is not of much use in troubleshooting situations.
Are you no longer interested in using gSender?
@braehill Len: I just read your post on the gSender board and it appears that this issue was resolved by an eeprom flash and loading gSender 1.0.5 onto a Pi. If that is the case, please post here.
Tks.
I am very new (as in only 4 very small projects) but my background is programming and a very good instinct. My most time consuming project has been a 3 layered sign I just did last night. If I had this kind of problem it would have definitely showed up on that one.
I would say that you don’t have a software issue.
Before you change software you should run the alignment diagnosis procedures in g-sender. I thought mine was square too until I did that. Now I am only 1.3mm off which is good enough for me.
Once that is done, set your x y & z again like you normally do to a piece of wood. Move the router else where just for the fun of it (I like to be in control!) than type G20 in your command line. Make sure you use a large bit to visually see homing issues. If it doesn’t home to the EXACT center of the bit to your EXACT corner of your wood than maybe your procedure to zero is wrong?
Try the paper test first at least.
-
Use the paper method to set the Z height
-
Touch the edge of the material while turning the tool by hand. When you feel the tool touch, stop and set zero
-
Touch the other edge of the material in the same manner and set zero 4. Raise the tool above the Z surface
-
Enter in the command line “G20” (if you use inches), “G21” (millimeter) and hit enter
6.Divide the tool diameter by 2 (example…0.25 tool is 0.125 when divided by 2)
→ 0.125 is used in the example below. Make sure to enter your divided number -
In the command line enter “G0 G90 X0.125 Y0.125” and hit enter
-
Set X and Y to Zero on your controller.
@iwannaapple Welcome to the group.
With respect, the advice in your first paragraph is misleading. Typing G20 into the console will do nothing WRT to moving the Mill. That command only tells the controller to use imperial units. G21 is for metric units. Neither command actually moves the Mill at all.
Your step by step instructions show that I’m not telling you anything that you don’t already know. I’m bringing this up only because @Tegbert may not know that entering G20 into the console as set out in your first paragraph will not home the “exact centre of the bit to the exact corner of the wood”.
Yes, I skipped a lot of stuff. Knowing my mind skips I printed the list.
Sorry about that and thanks for the correction.
@iwannaapple No worries. We welcome the input. That’s what makes this an educational place. Take care. Have fun.
I got a chance to run my machine for a simple facing job the other night . I was having all sorts of trouble so I flashed GRBL and uploaded 1.05 onto my Raspberry Pi and the probe function worked fine. The constant Y+ offset seemed to go away for now also.
I got a little time to run the machine again. This time I ran the surface function on both sides of a piece of HDPE (same piece as talked about earlier) and then ran the program that I had trouble with the offset. It’s a simple program that would show the offset even at a glance and it seems to be gone.
I’ll try some other things later to verify my results but that is all that I had time for today.
1 Like
Hello,
Thank you Len for your update, I have been away from my mill for awhile but took the opportunity to follow your fix and have gotten the same results! Thanks to everyone who took the time to help troubleshoot this with me!
Regards
1 Like