Facing the table

My table currently has a 6mm mdf spoil board, I’m trying to cut quite thin materials and sometimes have to cut up to 5mm deep for a 1mm thick stock due to variance in the table.


I’m thinking I could put down an 18mm sheet of mdf, countersink the holes 8-10mm and then surface the whole table - is this recommended or is it likely the moisture in the air is going to warp the mdf and it’ll ultimately be useless?


On the surface (no pun intended) that’s a good idea. It’s a common practice.

The MDF should be very stable. On the low rider, the support of the table makes a big difference too. How have you built it?

On a 9x5 foot table, I wouldn’t be surprised to find 5mm error in Z. It doesn’t matter how thick your stuff is, that error will be the same.

If you base seems stable enough, then adding thicker MDF is going to be an upgrade in the first place. Surfacing it will make it parallel to the path of the bit, but make sure you’ve checked as much as you can first. It would suck to surface it, then fix something 5mm in the other direction. I would test it in a smaller area first. You can also glue/screw two MDF panels together to get it even flatter, but you have to do it on a very flat surface to begin with.

It would be great to see how this works out. I made some measurements of my torsion box construction. I haven’t seen anyone else’s results from a similar test.

I know mine isn’t perfect. Haven’t gotten around to surfacing it yet though. I’m currently using half inch foam for the spoil board, and am pretty sure surfacing that will be “messy”.

Ah! I swear I replied to this!

I’m currently using a ping pong table, so it’s definitely not going to be dead flat - and is possibly pliable.

I’ve been thinking about TMC2130 drivers, and if that could be used for auto homing, but then I found and just had a read of this https://www.v1engineering.com/forum/topic/surface-mapping-using-marlin-try-it-out/

And I’m thinking maybe I’ll give it a shot - Did you ever give it a go Jeffefeebeethree?

I think there’s an extra letter in there…

Nope. Haven’t tried it. My low rider is running grbl, and other than z not being perfect, I’ve had no issues for a while. My surface is smaller than most (maybe the smallest?) But it’s very flat (relatively). I wouldn’t do a chamfer on the whole work area though. Someday, after my MP3DPV2 and a bunch of other shop projects, I’ll go back to Marlin and maybe have trinamic drivers and do one driver per motor, and work on some endstops and touch plate stuff, maybe even some auto level stuff. That sounds like it’s far enough in the future to never happen though. Let me know how it goes for you!

For the record, it’s Jeff eee bee three. That’s how I say it in my head. I couldn’t get jeffe@aol.com, and when jeffeb got cluttered with spam, I went to jeffeb2 and then jeffeb3 and then they invented spam filters. Really a pretty boring origin story. Heh, the nineties.

1 Like

Lies, it’s El Heffe, or just Heffe when we aren’t being formal.

1 Like

Yeah, my billsey came from the first time I created an account on a Xenix machine… (Late 80s IIRC, well before my first internet exposure.) I typed my full name like I used on the BBSes I was on and the OS removed the space and truncated to seven characters. I’ve been stuck with it ever since.


What would be the benefit of the TMC2130 drivers? I don’t get how they would allow auto homing?

From what i understand they have a feature that can detect when they reach their limit and start missing steps, I’m pretty sure that’s how the new prusa printer does limit switching now.


From looking around it’s somewhat secretive?


I don’t think it’s secretive. Just not well documented.

You have to write them up with the SPI bus connected and there is a diagnostic pin that you connect to the endstop pin. You also need a chip select pin assigned for each driver, which makes it hard to keep the LCD.

It would work for x, x2 and y, but it would be hard for z.

The part that is a mystery to me, but is almost certainly open source, is the ability to detect missed steps during a print and recover. That would have to look different on a CNC machine. But it would be freaking awesome.

At any rate. The low rider works great without any endstops. I would recommend working without endstops first, until you get the hang of it. When you get endstops, the dumb switches are a lot easier to use than the tmc drivers. The tmcs are supposed to skip less often at the same current, have smoother motion, and the limits are software configurable. But they are not as easy as the drv8825s.


From a recent discussion I had I am not planning on going to the trinamics. They most certainly would require everyone that did not get a complete set from me to tune them individually. A bad endstop is easy to diagnose, an improperly set trimanic is not. On top of that they are not capable of as much current. They are still developing the firmware to use them so at this time there is no autotune or anything.

I am also not sure how they would deal with the forces on a cutting head vs a printer. Every step would seem like an endstop hit I would imagine, a skipped step should still be detectable though. I have a set, they work fine with the fancy stuff turned off but I decided not to spend too much of my time on them.

1 Like