Z plunge depth too shallow

Hello everyone,

I finally finished my MPCNC Primo and it turned out great. I was able to make the crown test on first pass and decided to try a simple sign following the Vectric Tutorial. The sign turned out great on the x and y axis but the z axis didn’t plunge deep enough. I am assuming that the reason for this is because of the post processor. I used the included GRBL (inch) post processor to produce the gcode file. I am using the latest Jackpot controller and I used the built in FluidNC to run the Gcode file. All electronic components in the MPCNC come from the V1 shop. The material thickness is 0.965 inch and the depth the MPCNC goes to is about 0.25 inch on the final 4th pass. I am new to CNC so I am not sure why its doing this but I think its the post processor. I am using a Whiteside RD2100 1/4 down cut bit. 0.25inch plunge rate, 16mm/s feed rate on x and y axis. I am cutting foam so the feed and plunge rate is fine. I am not maxed out on the Z axis. Using Vcarve Desktop edition. Any help would be appreciated.

1 Like

Adding a bit more info to the post. I looked at the gcode for the cutout and it looks ok to me but I am a total newb when it comes to CNC so I am posting the gcode for someone else to look at.

It looks to me that its supposed to plunge to 0.2405 in line 12 but it doesn’t go that far down into the material. So I don’t think its a post processor issue. I think it maybe a firmware issue. I don’t know how the firmware controls the Z axis. The Jackpot board came preconfigured from the shop so I did not mess with anything on it. Maybe its not the firmware at all but I am not familiar with why its doing what its doing.

Cutout 0.25 deep_1-Cutout.gcode (1.6 KB)

There aren’t any probing commands in the gcode so how are you setting the initial Z height?

I see you are using G20 which means it’s in inches instead of mm. Most here use mm regardless of location.

I am just manually moving the spindle to the top of the material. I was my understanding that the 0 point is where the spindle is before i hit Play to run the gcode. Also I am not well versed with mm vs inches as to what I should set. I think I will redo everything in mm and see what happens. I think that maybe why this is happening. I have both mm and inch values in Vcarve set so I think I need to use one or the other and not both. I will set everything to mm.

That’s only true if you also zero the Z position. Typically, we do a G38.2 probe command with a touchplate to probe and set Z.

It’s probably worth looking through the milling basics.

I put the spindle where I wanted it before i turned on the machine. I assumed that would be the 0 for all axis. Like I said earlier a complete newb here. I do have a z touch plate but I have not wired it in yet. Will look into getting it setup.

1 Like

Sender has to be set to 0 also mine will hold the last start. Make sure x y and z are 0 in the sender

Ok I think I figured out the issue. There a couple of items I did not order from the V1 Store and one of them being the lead start screw for the Z axis. What I ordered was a 1 start TR8 screw instead of the 4 start that the shop sells. This explains why I am not getting the correct distance as I am only traveling 1mm instead of the 8mm I should be traveling for the same rotation. I don’t want to dissasemble the whole thing to replace the lead screw so I guess I will need to modify the firmware to correct for this.

Is there any benefit to using the 4 start vs 1 start as far a rigidity or accuracy is concerned?

As a follow up question. I just pulled the config files from GitHub and I think I need to modify this section of the config.yaml file to change the steps_per_mm from 200 to 1600 do I need to modify the other parameters in any way?

z:
steps_per_mm: 200.000
max_rate_mm_per_min: 900.000
acceleration_mm_per_sec2: 80.000
max_travel_mm: 300.000
soft_limits: false
homing:
cycle: 0
positive_direction: true
mpos_mm: 200.000
feed_mm_per_min: 300.000
seek_mm_per_min: 800.000
settle_ms: 500
seek_scaler: 1.100
feed_scaler: 1.100

That should be it.

1 Like

Um…

4 start lead scres is 8mm per revolution, times 200 steps per revolution times 8 for microstepping = 200 steps/mm.

Withough changing anything else, a 1 start screw is 2mm per revolution, which comes to 800 steps/mm.

Changing the steps/mm ought to be enough. Do keep in mind though that the motor itself will be turning 4x as fast for the same travel speed, and accelerating 4x as fast for the same travel acceleration. It is often recommended to reduce the maximum travel speed and acceleration for this reason, though I personally would not do so by a 4x factor.

For lifting, the 1 start screw you have will have a 4x mechanical advantage, reducing the load at the motor by about that much. As such, acceleration probably doesn’t really need to be reduced much if at all. Top speed might need curtailing, but not if it isn’t causing problems lifting the router. I’d take a “wait and see” with that, unless you will be using only very expensive materials for your testing. If you can afford to lose a piece or two in the first uses, try it and see.

one distinct advantage: Many people find the beam dropping when powered off. Going to a 2 start ore 1 start lead screw is a good way to mitigate4 or prevent this from happening, even if your machine is well lubricated. you are unlikely to have thois problem with the machine as it is.

Z speed is rarely a bottleneck in machining time, unless you are doing a lot of 3D carving, so this is a trade-off that I am personally willing to embrace. I have 2 start lead screws on my machine, for example, to prevent the beam from dropping when powered off.

1 Like

That maths, but they claimed that their 1 start screw is pitched at 1mm per revolution. :man_shrugging:

1 Like

Yeah 1600 was way too fast and stalled the motor. I will set it to 800. Also I did not have the newest version of FluidNC installed. I updated that and added UI3 so now I have the probe function that works great.

2 Likes

Okay. TR8 screws with the compatible nuts are usually 2mm though. I’d measure to be sure, even a regular ruler should tell the story if you go over a few threads.

Regular hardware sore M8 threaded rod with regular trapezoidal threads are 1.25mm pitch, I can’t really see the TR8 style ones being even finer than that, but hey, I’ve been wrong before.

2 Likes

Daryl, you are correct it is a 2mm. I have changed the steps to 800, acceleration to 50 and max travel to 150 and its much better. I was able to complete my 60x50mm test box with the correct depth and a speed that is ok for the test foam material I was using. Its a bit fast but for testing its fine for now. I will further dial in the speed as I make more test cuts. I now have a fully functional CNC machine that I built myself and I am very happy with the results.

Kudos to Ryan for an awesome design that is easy to assemble and works as expected. I am still in awe that I was able to build and have it working almost right away. I do have an IT background so building things is second nature to me but still there was no tweaking or fiddling required. The mistake was of my own making and I learned a bunch already along the way. Thank you all for helping me out. I can’t wait to make stuff on it.

3 Likes

See, that was crucial for me because of using Estlcam as a controller and doing a lot of carving. My Z speed is at 8000mm/min now. :sweat_smile:

2 Likes

Why does using Estlcam as a controller make Z speed a bottleneck in machining time?

My wife/girlfriend/partner used to complain about my shallow “Z plunge” until I learned this simple tri… Wait. Wrong forum again? Dang it, gotta wait until after the meds kick in to start posting… :grin:

2 Likes