Pololu/stepstick compatible stepper drivers without potmeter

I just posted this in the ESP32 Printer Board thread, but I thought it would be usefull to discuss this separately.

If you are building your own Pololu/RAMPS compatible stepper driverboard and want to replace the potmeter adjustment:

1. Set enable signal to disabled. This tells the driver to be in config mode
( not the driver chip itself, but an interface chip acting as digital potmeter )
The interface chip could be anything, like PIC or ATtiny

2. send steps and dir signal to be interpreted as " current up" or "current down". ( relative mode )
(eg. 1 step = 1mA) ( might require a new gcode or extra parameter )
2a. Same procedure, but steps are interpreted as absolute value.

Current adjustment would then look like: G1 S1 E1300 X500 Y500 Z1200 F[full speed], where S1 is the new adjust-mode parameter

Edited 1 time(s). Last edit at 05/23/2017 02:48AM by o_lampe.

yuck...

Accessing anything on a disabled signal just irks me... such a hack!

If you going to hack in a controller to replace the variable resistor on a pololo anyways, why not just break out the pololu vref pin and plug it into a spare d2a pin or a d2a device on a spi or i2c bus....

I wouldn't recommend changing current setting during a print, but it could be used before homing and when a stepper is in idle anyways.

I wonder why this idea creates almost no reactions? Isn' it worth to design plug in drivers with new features?

Edited 3 time(s). Last edit at 05/31/2017 02:43AM by o_lampe.

Two things:

1) Almost no one here develops firmware and controller boards for printers.

2) Even if you get rid of the pot, those modules are a poor thermal design.

You're putting lipstick on a pig, rearranging the deck chairs on the Titanic, trying to make a silk purse out of a sow's ear, dreaming about gorgonzola when it's clearly brie time, etc.

Edited 1 time(s). Last edit at 05/31/2017 07:27AM by the_digital_dentist.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Thanks for the honest answer DD, I actually had to laugh a bit...
You think of A4988 or DRV8825 and I'm with you.
But there are other chips like THB6128 ( RAPS128 ) which are the next best thing to onboard drivers like Duets TMC2660 to me.

I agree about hardware designers being rare on this board, but if we discuss such improvements here, the persons in question might see the opportunity to take advantage against the competition. Thousands of cheap Mega/Ramps/driver/LCD combos are out there and they can only conquer by price. Who's the first who offers next_gen plugin drivers?

What if, instead of individual boards for the plugin drivers, you made a larger board with several drivers on it, soldered to the ground plane as the engineers of those chips intended, and able to plug into the RAMPS board? If you used a large enough board it would solve the thermal problem. Then you could play with either digital pots or more robust, multiturn trimmer pots and maybe even put a couple clips on the board to hook a meter to set the current easily.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

gah annother layer?

But I have to admit I like this idea a lot better than hacking existing modules

Include a standard lcd connecter(s) and a decent 5v regular on the board and you might have a winner solution

The ( only?) benefit of plugin drivers is their replaceability. With a board hosting 4 or 5 drivers this is no longer true. Also the footprint would fit one specific controller only.
Maybe a sandwich board that goes between controller and drivers, but has separate sockets for them. You'd have to hack the potmeter of each driver then. The benefit is a simpler way to setup the current-control board, as it needs only one cpu for all drivers. It could even take care of microstepping. ( one thing that is really annoying with a RADDS board )

The main reason you need to swap those plug-in motor modules is because they are so unreliable. You're never really sure what the current is setting is. Errors in current setting and the poor thermal performance of the board leads to burning up the driver chips. The tiny pots are easily broken. Unless you invest in a ceramic screw driver, it is very easy to short the circuits on the driver board with the screwdriver and kill it.

As soon as you put the driver chips on a large enough board to dissipate heat, the problems start to disappear. Use quality pots and you won't be breaking them. If you want to be able to use different driver chips, just make a couple different boards with all TMC drivers or all Allegro drivers, etc.

What we're talking about here is a cheapskate's "solution" to the problem that keeps crappy 8 bit controller boards viable for the low end of the printer market a little longer. The best solution is to invest in a controller board that is properly designed - adequate heat dissipation, digital pots that set current, etc.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Quote
D_D
The main reason you need to swap those plug-in motor modules is because they are so unreliable. You're never really sure what the current is setting is. Errors in current setting and the poor thermal performance of the board leads to burning up the driver chips. The tiny pots are easily broken. Unless you invest in a ceramic screw driver, it is very easy to short the circuits on the driver board with the screwdriver and kill it.

That's why I want drivers with software adjustable current. No pots, no shortcuts and easy to keep cool(er).

When I started 3D printing, there were only A4988 drivers, the DRV8825 was brand new and widely unknown. Now we have RAPS128 and TMC2100 and maybe some more, I haven't heard off.
So speaking of a "dead end" is not entirely true. I'd say 8bit controllers will be on the market for the next 10 years. And 32bit controllers with plugin drivers too.
Whatever driver-evolution we will see in the future, the "crappy" plugin controllers will be the first to take advantage of them.

Quote
o_lampe
Whatever driver-evolution we will see in the future, the "crappy" plugin controllers will be the first to take advantage of them.

Not so. The TMC2660 drivers used on the Duet WiFi/Duet Ethernet are probably the best drivers available right now for most 3D printer applications. They have configurable microstepping up to x256, configurable chopper control parameters, over temperature status information, stall detection, short-to-ground detection, high current capability and more. But all these features need an SPI interface to configure them and report the data. This and the large package size makes them unsuitable for plug-in drivers.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

You are right dc42, but AFAIK they are available as plugin drivers for a specific Aztec board too. If they wouldn't need the SPI interface, they would be on every Ramps by now

There are the silent step sticks (Trinamics in a pololu format) but you either loose a lot of functionality this way or you have to manually route the SPI interface to the modules.

I think a bigger module with Step/Dir/Enable and SPI would be a good next standard (after Pololu). With bigger modules the chances for different stepper chips to fit the module are higher. Thermal design is easier and non trnamic steppers could use a PIC, 8051 or ATtiny between the SPI and their chip.

We should not forget that stepper modules allow for more flexibility. Most printers are OK with 4 Steppers, but with dual extrusion, color mixing and whatever else might come an additional stepper might come in handy. And Modules would also allow for the module to be placed close to the motor.

I don't think that the cheap printer sellers will come up with something new like this. My experience is that the cheapest sellers have no budget for research. They just copy whatever is out there and sell it in huge volume. And if they build something new the cheap folks don't make it open hardware.

Take a look at prices for the processors (Digikey, Mouser, Farnell,..) a Microchip Atmega2560 is much more expensive than a ARm Cortex-M0+ with 48MHz. You can even get M3 and M4 cheaper. But we still see cheap Arduino2560 clones with ramps used by the cheap printers. Only just recently the first closed source 32bit boards appeared,...

Therefore I think we need to come up with a solution and have a working board that uses the new modules. Then we can hope that the cheap sellers will start to copy that.

Back on topic: For a new Module standard I think a big factor is also the connector used. Connectors are surprisingly expensive. And you also don't want to waste too much board space beneath the module.