GRBL and TFT screens (and some WiFi too)

Do you have the link to the web_UI for the esp8266? I’ll try pushing it to the wifi adapter and see what happens.

Sorry, this is the firmware for the esp to connect with prettymuch any gcode machine via serial: https://github.com/luc-github/ESP3D

This is the web ui: https://github.com/luc-github/ESP3D-WEBUI

1 Like

https://github.com/luc-github/ESP3D-WEBUI

@Atom Thanks but I have done it now. I am retired so time is not an issue, besides that is one of the fun parts of all this messing around. I already had the ESP32 DEVKIT V1’s on hand and that is a sight easier to design a board around than a bare ESP32… The machine it will probably be going on is limited in space for the electronics, fitting another uP to the Nano just to handle wifi is a non-starter, much easier to just order up 5 PCBs for $2 (plus shipping) and use the DEVKIT to run GRBL-ESP32 and add the display on a separate UART

IMG_20210131_172925

2 Likes

Yeah, and since Marlin doesn’t use $ as part of a command like GRBL does, it probably won’t work correctly with the current firmware on the ESP. Oh vell. :man_shrugging:

Well that worked out great!

Thanks!

1 Like

And also…

So , unfortunately, with this adaptation, you can’t run the WiFi and/through the TFT.

1 Like

I’m glad you got it working so easily. Also I would imagine that it is a one or the other situation since both communicate via serial connection. But you could use an old phone and just leave it always connected via the web ui. Then you have a tft :wink:

I found a post from a French guy with a firmware revamp to give some nice icons although the functionality appears unchanged.
MKS-TFT-24-CNC-icone-en-couleur.zip (960.5 KB)

and

MKS-TFT-24-CNC-icone-thème-sombre.zip (822.1 KB)

although I suggest you keep your config file as his are the Chinese ones…

Another bit of news - if you have homing enabled and the option set that forces an alarm state on boot then the screen is unable to get past it as it does not have any way to send either an $X or a $H so you will have to add at least one of those functions to the configurable icons in the ‘settings’ page. I also find there is a 15 second delay to the nano accepting serial inputs from the screen if the USB port is ‘live’ (but not if the USB is just powered using a power only cable ie. no data).
@Strider_Matic you were right…the wifi adapter is next to useless unless you are using a MKS Robin controller board…oh well!

1 Like

I really like the Couleur theme, those icons are pretty detailed. I also like the carve icon in the Sombre theme, has a more industrial look to it, like a bigger, badder laser head. Yeah, that power-on locked state I disable before compiling.

I was able to get the firmware @Atom recommended loaded to the wifi board and when wired directly to the controller worked OK, but after a while I started getting erratic behavior with the communications and having to power-cycle the controller to reconnect. Plus, you need separate virtual serial port software to make it work “correctly” anyway. Although I never did try using the web interface to upload gcode to it, so maybe that works well. Seems like the best way to go so-far for me is cabled, or using octopi like I use on my ZenXY running GRBL.

1 Like

I am at the same place, the problem I have with the display driving the workflow is the lack of feedback (ironically). The ‘alarm’ condition being the case in point, along with positional data to a lesser extent. Also I just love LaserGRBL software and, as the laser machine is portable, I move it within range of the USB lead. Transfering files by memory card is long winded when you are used to file transfer via RJ45. I use Simplify3D for 3d printing, which has a network file transfer facility with Repetier Server running on an OrangePi Zero to connect to my 3 printers - that just works flawlessly. MKS wifi adapter consigned to the scrap box.

Amen to that. Even if it’s just across the room, it seems an unnecessary step, so to speak. :laughing:

1 Like

OK…got a bit further, I have got a ESP32 Devkit V1 (DOIT) running now on my cantilever laser board.

I can connect to it over wifi fine but I am still awaiting a tf card holder so I cannot check the file upload option just yet but I can control the board from GRBL_WEBUI in my browser. Jogging works fine, as does the command line but I can’t seem to get the override panel to work. I also have a MKS TFT24 connected to the ESP32 and can issue commands from there ok except the power command to control the laser is not proportional at all, in fact it is all over the place.

Worse still is that connecting over the USB to LaserGRBL doesn’t appear to work either, it connects but if you send any command over USB it promptly disconnects. I will have to try re-compiling the firmware without wifi enabled, that way I may still be able to connect over USB and use LaserGRBL…enough for one day…more experimentation needed.

2 Likes

Mike, do you have any of those boards for sale? I would just need the bare board as I probably have everything else around here somewhere. I’m going to be grabbing some of these anyway…
https://www.amazon.com/Organizer-Development-Dual-Mode-Microcontroller-Integrated/dp/B08FR5Y85W/ref=sr_1_15?dchild=1&keywords=esp32&qid=1613287420&sr=8-15

@Strider_Matic… PM sent… Those are actually the RIGHT boards too! :slight_smile: There are a few Devboards, all with different pinouts. I had a problem trying to get ESP32_GRBL onto the V1. In Arduino IDE there are two options for programming the V1, either use ESP32 dev module with the following settings


or there is an entry for a DOIT ESP32 DEVKIT V1 without all these options. If I used the DOIT ESP32 DEVKIT V1 option it would fail compile saying that 1807554 bytes was too big a program to fit and if I used the generic entry with those options it would load but the device would only bootloop. I eventually found if I ran the ESP32 flash download tool to erase the flash then it would program correctly using the generic entry and the options in the picture.

So…a little further again, If I disable the bluetooth and the wifi and compile it then the software does work a lot better, it stays connected, it all seems to work ok, but then I discovered LaserGRBL has had some upgrades…There is a telnet option under Grbl/settings/connection protocol, once that is selected you can enter the ip address of your ESP32 and specify port 23 (192.168.1.24:23) and it works under wifi (once you have re-enabled wifi and recompiled!). I’m not sure of the speed yet as I have not yet integrated it into my laser machine, but it appears to work fine.

2 Likes

I’m definitely looking forward to building a couple of these. It looks like they would be great for the ZenXY tables also.

1 Like

So the microsd card readers arrived yesterday and, having soldered it in place, the card can be used to hold .nc files ready to be used with the display. However…the speed of lasing over the wifi seems to equal that of USB or sd card - or very close to it, so there is little downside to ‘over wifi’ operations and (for me) LaserGRBL, I just need to try some longer files to prove wifi reliability.

Display mounted on the carriage and the esp32 board ready to do Tux.


2 Likes

@dart1280 That board is working out great! Thanks again! I managed to design an enclosure for it with a friction-fit lid and using the ESP3D-UI to send jobs to the SD card works great. I have some TMC2209s driving the laser engraver now (nice and quiet :shushing_face:). I’m using a parallel motor splitter for the Y and everything is staying cool.

2 Likes

Why parallel rather then wiring them in series? It’s my understanding you will get better performance out of them if they are wired in series.

Well, one because I had it laying around and I’m a little bit lazy :sweat_smile:, and two, these motors are rated at 10v. So when I had them connected in series for another project they would not work correctly. Since there is almost no load on the axes, I can keep the current down because the X axis is the only one I move quickly, so far :wink:

1 Like

It halves the current to wire them in parallel. But on a smooth axis like that, with no load on the tool, I agree it should be fine.

I am not sure I agree with the diagnosis that the motors won’t work in series, but I won’t pick a fight :slight_smile:. If it works, it works.

2 Likes