Jackpot3 Stall Detection

Software / Firmware Jackpot Board / FluidNC
1

Hello folks!
I have a jackpot3. How can I detect it is stalling? I need to turn the stepper reverse a bit when it stalls. Note: I use jackpot3 for an automation build, not for cnc machines.

y:
steps_per_mm: 400.000
max_rate_mm_per_min: 1500.000
acceleration_mm_per_sec2: 100.000
max_travel_mm: 9000.000
soft_limits: false

motor0:
  limit_neg_pin: NO_PIN
  limit_pos_pin: NO_PIN
  limit_all_pin: NO_PIN
  hard_limits: false
  pulloff_mm: 1.000

  tmc_2209:
    uart_num: 1
    addr: 1
    r_sense_ohms: 0.110
    run_amps: 1.000
    hold_amps: 0.500
    microsteps: 16
    stallguard: 0
    run_mode: SpreadCycle
    homing_mode: SpreadCycle
    use_enable: true
    step_pin: I2SO.5
    direction_pin: I2SO.4
    disable_pin: I2SO.7
2

No, sorry.

Full specs when you click for more details. Jackpot3 CNC Controller - V1 Engineering Documentation

3

Diag Pin (needed for StallGuard detection) is not wired to ESP32 controller on the Jackpot 3 boards.

image
image909Ă—658 92.4 KB

Same for Jackpot 1, but, maybe you could wire-up the Stepper Diag pin (on the driver breakout boards) to an available GPIO pin on the JackPot 1 :person_shrugging:.

But, at that point, maybe you’re better off doing your own wiring setup using a regular ESP32 board…

breadboard
breadboard1920Ă—1869 522 KB

4

Initial efforts to implement stall detection on LR4/MPCNC class machines showed it to be unsatisfactory/unreliable and too machine/build specific to be of value.

So, it’s not implemented by default as noted above.

5

Thanks guys, I should have given a more complete answer. Way better than just saying no. Thanks again.

6

Now I’m curious…

Always neat to see JackPots used in other machines.

7

PXL_20260411_230026311
PXL_20260411_2300263111920Ă—3413 1.57 MB

It is a desktop plastic injection machine. I also used Fluidtouch display that Jason shared. I modified with Ai to adapt my build. I am asking for pellet auger feeding motor. when it stalls, i want it to turn reverse a bit.

8

I don’t think tmc stallguard actually works very well at low speeds you need ~2Rev/sec for it to work, and then you need it to kind hit a wall, not a slow decel, for it to work right.

I think you will need a force gauge of some kind.

Super cool project please share more when you can, we all love injection molding!

9

Neat project @eminfirtina ! Some context/ideas that may help (or not). Either way, good luck!

Because of the reasons Ryan mentioned, I was leaning towards updating my (non CNC) machine with stall detecting steppers, to instead use servo and/or geared DC motors and use current draw sensing to detect when near stall, or absolute stall. That was the next step for my project, was going to use a Maslow 4 PCB, or drivers like the ones it has. I got as far as forking Maslow 4’s fluidnc fork with minor edits for my setup so motors could be button activated.

That said… Getting rpm high enough for reasonable stall detection was possible for my situation using planetary gear speed reducers that were also helping to increase torque. Do you have gearing ratio like that already in this setup, if so maybe stall could be a hint to action on? Maybe I was using stall guard wrong, but I couldn’t get precise moment of stall detection. So, I’d probably rely on magnetic or optical encoders if precise positioning mattered.

10

You look really experienced with this stuff honestly. I’m still pretty new on the electronics/embedded side. I never really dealt with Arduino or ESP32 even for basic projects before this :grinning_face_with_smiling_eyes:

What pushed me toward the Jackpot board was the direct 24V outputs. It fit perfectly for my pneumatic solenoid valves, plus the integrated drivers for the NEMA17 made the wiring super simple.

At first I was using a regular NEMA17 and it was stalling constantly. Now I moved to a 5:1 geared NEMA17 and it’s MUCH better. It still stalls occasionally, maybe once every 5 minutes or so. The issue is plastic pellets getting stuck between the auger and the case. That’s why I’m trying to detect the stall and reverse slightly to clear the jam automatically.

Interesting what you said about StallGuard not being precise enough. It shakes vibrate when it stalls idk if this info would change something.
I’d actually love to hear more about your project too. What are the advantages of the Maslow 4 setup compared to others?

11

This is not the Maslow forum, but for the MPCNC and LowRider 4. :grinning_face_with_smiling_eyes:

12

It’s ok, we can talk about other machines, controllers, and software here.

We try to be objective.

13

I got annoyed with your message but I am trying to be a better person these days :smiley: I am a jackpot3 user and we are just sharing info about our builds. Don’t worry @Tokoloshe

14

Thanks Ryan and others. I appreciate it. I guess I will try to solve my problem with a mechanical solution. Maybe with an encoder or something else in the future.

15

Nah, don’t worry, I thought you mixed it up. Sure you can talk about the Maslow. :stuck_out_tongue:

16

Could it be as simple as a different funnel design? Wide open at the top and a tight fit on the auger?

I am not well versed in the design principles of these but I thought augers were too slow to fill a mold before things cooled too far. I have a million questions but feel free to skip them and just share some updates as you go. Seeing, maybe owning/building, and jackpot powered injection mold machine would be pretty sweet so I am definitely following along.

17

Even if you did use stallguard, I’m not sure you can reverse based on that triggering. It’s a really finicky thing mostly intended for homing.

Couldn’t you just move in a pattern like clockwise 20, reverse 5 or something like that? Or do you need to know the actual distance moved?

18

If it vibrates when it stalls, couldnt you use an auto knock sensor and reverse based on that?

19

Since the parts are not very large, it’s easy to feed them with an auger. Imagine a chamber: after the first injection, there’s still material inside, but it needs about 40 seconds to reach the right consistency. I use the auger to feed more material on top of the existing plastic inside. So, I have plenty of time(40s). I dont have design files, i may make a new feeder design. Also i think plastic pellets i bought are too small. I will keep you post it.
I have different plans for this project. I may add a Adafruit VL53L0X Time of Flight Distance Sensor to measure material needs to be added by the piston’s position.

20

Well I can try that by diving feeding value and adding small reverses. At least it will reduce the stall