This week was a very short one for our team, squeezing in as many bugs fixes as we could by weekās end to address some of the still outstanding issues that weād seen reported. This included:
Jogging causing gSender interface to freeze up
Interface inaccessible after job end
Interface freezing mid-job on large / long files
Computer sleeping causing issues in gSender
Check mode running incorrectly and quite slowly
We believe weāve now solved all of these problems!
If youāve experienced any of these issues previously then please give this new release a try and let us know if youāre still experiencing them. I also have yet to receive much feedback on some of the new features from the previous build, including:
Start from line
Macro descriptions and reordering
Nearly doubled speed of file information processing (dimensions, cut time, etc.)
New refactor of visualization processing which should speed up visualization as well as possibly prevent visualization freezing up
Added fixed input on console tab
Joystick support
Are yāall liking these? Letās discuss!
Moving forward weāll be taking a week off of gSender development to give the team a much needed break and will be coming back to it soon for more fixes and features
Thanks for the update, will be trying the new version this weekend.
Iām interested in the joystick control, although Iām not sure how to configure it and didnātā see any settings for a joystick. Would this be a USB joystick to the computer or does it need to be wired to the LongMill directly?
Hey Jim, the joystick control can be found in the setting area near the keyboard shortcuts. Itās for supporting USB joysticks that are plugged into your computer that then can be used to control the LongMill or other CNCs. You should see that you can detect button presses and then assign actions to those presses similarly to how keyboard shortcuts are set up - weāre working at some documentation for the new features to get out in the near future
Itās hard to keep up with docs since gSender just keeps adding more functionality
Thanks for the work and look forward to trying this build. My household is trying to get over a horrific battle with Covid19. Iām not sure when Iāll have my strength or lungs back, but hope to get back in the shop soon.
Chris,
Not sure where to report bugs - after setting a button for jogging the z axis on my joystick, pressing that button causes the z to move continuously and it wonāt stop without pressing the stop button on the LongMill controller.
gSender 0.7.1
@Jimbo Jim: Does it do the same thing regardless of the jog mode selected - precise, normal, rapid? Can you map a different button on your joystick to jog Z and see if the problem persists, please?
Hi - the jogging is starting to be rather unpredictable. The most annoying aspect of this is when the router travels farther than I asked - I had several incidents today where a 1mm Z jog suddenly moved 10mm - which is nearly catastrophic when traveling downā¦
Other aspects seem to be pretty reliable.
Iām back in the shop and downloaded the new build. (Thanks for the good thoughts) I noticed it doesnāt run when you when you click finish while having the box checked to run program. Minor but thought Iād mention it.
Iāve also noticed some jogging issues too. It seems to jog a bit longer than I intend for it to. I believe it has only been on the Z axis.
I am still a little foggy from covid so this all could be me too.
I tested this further: The problem exists on all 3 speeds. I also reassigned buttons to be sure the issue wasnāt associated with particular buttons. Regardless of which button or speed, jogging the z with the joystick causes continuous movement which can only be stopped with the āoopsā or stop button on the control module.
Seems to be specific to joystick functionality then? I hope we havenāt broken the jogging through regular means (button clicking & keyboard shortcuts)
Actually, Iāve broken 2 2mm end mills in the last 2 days when the āpreciseā Z- jog suddenly moved roughly 10mm down hard into the stock. I consider fixing this thoroughly and quickly a very high priority. Not being able to rely on the jogging is a killer.
Iāve been doing the jogging by clicking directly on the UI. I have an application called ātouch portalā that can send key stroke sequences to gSender, but it gets less reliable with each revision to gSender. Iāve temporarily stopped using it, waiting for some stability to return.
I thought that perhaps the mouse was at fault, and switched from a wireless mouse to an older wired mouse. That had no effect - I broke 1 end mill with each one.
Frank, we havenāt been able to replicate this issue on our end. Me and the team just brainstormed a bit on this but if you have a video of the issue occurring (obviously Iād recommend not having a bit in the router while youāre replicating it) with the Console tab being visible as well, then that would go a long way toward us understanding what could be happening
On a separate note, Iām curious what the application of the touch portal is fulfilling for you
Iāll see about trying to replicate it. I canāt have the console visible at the same time as the router as theyāre about 5 feet apart. Iāll focus the camera on the console while I jog with no end mill in place; however, I donāt know how long it might take to catch something.
The touch portal āhostā running on the controlling computer communicates with an iPad mini running a second copy (via wifi), and I have custom pages set up with buttons for the operations I want to have gSender carry out - jog right, go to precise modem,etc. I can hold the iPad directly in front of the router while I jog it without switching attention between the computer and the LongMill. Itās analogous to people using a console controller, only I have more flexibility.
I replicated the problem, and have video of the console. Thereās a comment at the front, and the actual problem is at the end. Itās about 1.5 minutes:
The final jog movement goes from Z-0.1 to z-1000.
Something about pressing the mouse button a little harder than usual seems to be the culprit. I have to have a VERY light touch with the mouse.
The behaviour of the button is to send either a single jog command or a continuous jog command based on how long the button is held down. Right now the threshold is 250ms - if you hold it down longer than that, it will assume you want to continuous jog and send a larger movement.
For the next build weāre going to look at tweaking the values specifically for the the Z axis to make it a little less touchy - weāll look at doubling the activation time for continuous jog to 500ms and also change the movement speed on Z to be a fraction of whatās set for more precise Z movement. 10mm movement overshoot is probably not super noticeable when moving in the X or Y axis on continuous jog, but on Z where youāll hit the workpiece it becomes a problem - weāll see if these new values help in that regard.
So itās meant to overshoot on the X and Y when doing a continuous jog? I got sick with Covid for a little over a month and didnāt do any work on the longmill. When I got back in the shop I thought things felt different with the jogging. I had no idea if it was me or if my brain was still on sick time.
Itās not meant to overshoot - thereās just a delay when the movement is stopped via a jog cancel command to where the head decelerates to a stop - the movement finishing isnāt instant nor can it be. Nothing has changed wrt how jogging works.