Welcome! Log In Create A New Profile

Advanced

TOS-100 stepper driver (autodiagnostics, up to 1/256steps etc.) why nobody tried?

Posted by smilem 
TOS-100 stepper driver (autodiagnostics, up to 1/256steps etc.) why nobody tried?
March 29, 2014 09:05PM
Hello, I wonder why nobody tried the very adcanced TOS-100 stepper motor driver.

TOS-100




[www.motioncontrol-community.org]


Should work wonders with RADDS
[www.max3dshop.de]

And a delta design printer smiling smiley

Edited 1 time(s). Last edit at 03/29/2014 09:10PM by smilem.
Re: TOS-100 stepper driver (autodiagnostics, up to 1/256steps etc.) why nobody tried?
March 29, 2014 11:42PM
Smilem: I would guess the main reason is that it's USD $40 for one controller board.

As for the chip on the board, the TMC260 is about USD 8.70 from Digikey, and won't fit in the usual standard Pololu form-factor.

That said, if this DICE form factor takes off, it should be possible to put it on one of those easily enough.
Well, 40$ for controller is not cheap, but it's more than 32 microsteps too. Doesn't overheat either.

Peopple pay like 1400Eur for spiderbot 3d printer and are happy, it uses simple RAMPS board!, 3steppers, and 3Igus linear rails, some aluminium etc. Overall cost cant be more than 450EUR. You can get like 10steppers from ebay for 80$ !

Could I connect this to RAMPS, or megatronics instead of the usual motor stepper driver? I'm waiting for RADDS because delta printers need the additional precision.
Re: TOS-100 stepper driver (autodiagnostics, up to 1/256steps etc.) why nobody tried?
March 30, 2014 06:34PM
I see no reason why not. The board is designed for mounting on top of an Arduino by itself though, so you'll have to mount it separately and run wires to it from the Pololu sockets on the RAMPS. You will also need to supply it with a logic supply (+5V and Ground), apart from just the step/dir/enable lines.

Notes:

Each time you double microstepping, you need to send twice as many step/dir pulses. If you want to keep the same speed as with 16x drivers, this means the Arduino needs to do almost twice as much work (in a set time period). Given that 32x seems to be pushing the existing speed envelope (from what I've heard from people who have pushed these limits) you may find that using step/dir control with higher microstepping values using an Arduino Mega requires that you slow down your moves.

Also note that the more microsteps you use on the extruder, the slower it will extrude (depending mainly on the extruder type, filament size & nozzle size, as that determines the volume of plastic that comes out the end for any given extruder movement), so you might need to play with your microstepping to get plastic extruding out the nozzle fast enough to keep up with your moves. If you can't extrude plastic fast enough to keep up with moves, your object will not print right.

Given you're doing a Delta, you may find it OK, but you do want to look at something faster (like a Due) for control. I don't know if this shield is 3.3V compatible though (which you'll need to be if you want to run off a Due, RADDS doesn't change the logic levels, nor does RAMPS-FD).

Note: For X/Y Cartesian bots, Z really doesn't need that great a microstepping, as you're driving it with a screw-thread.
Quote
smilem
Well, 40$ for controller is not cheap, but it's more than 32 microsteps too.

Meh...Measure it with a micrometer, mark it with a crayon, cut it with a chainsaw.....
Re: TOS-100 stepper driver (autodiagnostics, up to 1/256steps etc.) why nobody tried?
March 31, 2014 10:29AM
Quote
smilem
but it's more than 32 microsteps too
With the frequency most of the Arduinos run at, I thought I saw the max step frequency to be around 4000 per second. That works out nice with a 1.8 degree stepper (200 steps per rotation) with 1/16th microstep as 3200 leaves plenty of wiggle room for a rotation per second. Going up to 32 you can't do a full rotation in per second. You'd have to slow down the rotational speed to ~1/14 what you could with 1/16th stepping. So while your microstepping accuracy goes way up, your speed goes way down. And I'd honestly have serious questions as to the accuracy of 1/256 microsteps with the NEMA 17 motors we usually use.
Re: TOS-100 stepper driver (autodiagnostics, up to 1/256steps etc.) why nobody tried?
April 03, 2014 06:52AM
If this is like other TMC chips, sure it uses step and direction pins, but you have to set it up first via spi, something thats not n the current reprap firmwares..

So It will never be a drop in replacement for a pololu as you need to connect the spi bus.

But as mentioned the DICE standard does have SPI on its connectors...
Hi,

I developed this board. I may can help you. First of all you can get something much better than this: [www.indiegogo.com]

It has 3 of the TOS-100 for a fraction of the price - but with additional motion controllers to overcome the step speed.


The micorstepping can be adjusted from 1/1 to 1/256th - so no problem here. You only have to insert the initialization code from the TOS-100 library in your firmware … so after that it is an drop in replacement.

But honestly, try the T-Bone winking smiley

Marcus
some news.. Stepstick-like-Trinamic TMC2100

SilentStepStick
Re: TOS-100 stepper driver (autodiagnostics, up to 1/256steps etc.) why nobody tried?
January 21, 2015 04:49PM
According to one of the videos the silent mode is pretty useless unless you're going at slow speeds if torque is of any concern. So I guess it would be great for your homing and you don't mind missing steps if it meets any resistance, but if you want to print fast, it's not of much use.
Re: TOS-100 stepper driver (autodiagnostics, up to 1/256steps etc.) why nobody tried?
January 28, 2015 02:29PM
cdru : I don't fully agree, that video you mention - Thomas's right ? - also says the other mode - spreadCycle - is still less noisy than standard pololus to some extent, while being usable for a reprap.

I am definitely going to try those out to figure for myself if they can soften the noise of my Prism.


Most of my technical comments should be correct, but is THIS one ?
Anyway, as a rule of thumb, always double check what people write.
Re: TOS-100 stepper driver (autodiagnostics, up to 1/256steps etc.) why nobody tried?
January 28, 2015 02:54PM
Quote
DeuxVis
cdru : I don't fully agree, that video you mention - Thomas's right ?
Yes, that video. I'm not sure how you can not fully agree. I just basically rephrased exactly what he says. At about the 2:15 mark he talks about the quiet mode and describing it as "gently suggesting that now maybe it should move to that spot and if it doesn't, the driver isn't going to take any extra efforts to get it there" and then follows it up demonstrating minimal torque by stopping and pushing the x-carriage the opposite direction intended, and then just flat out saying it makes it pretty useless for our use with 3D printers. He even shows in the datasheet where it says only for low and medium velocities. Seems pretty cut and dry.
Re: TOS-100 stepper driver (autodiagnostics, up to 1/256steps etc.) why nobody tried?
January 28, 2015 03:52PM
That stealth chop mode is described as useless for us yes. But not the spread cycle one.


Most of my technical comments should be correct, but is THIS one ?
Anyway, as a rule of thumb, always double check what people write.
I'm glad there is interest in the Trinamic divers over the standard ones.

However the T-bone:

You have 2PWM putputs,
1 for 1st extruder
2 for heated bed
3 for secondary extruder missing
4 for heated chamber missing


Please add missing features, and make board rectangular !
@smilem You requested some features and a rectangular board ? look here my Board It also comes with a powerfull CPU,..

Any features missing?
Sorry, only registered users may post in this forum.

Click here to login