Page 1 of 1

Dibloff's eggbot/spherebot build

PostPosted: Mon Apr 21, 2014 2:14 am
by dibloff
Hello Folks. Here is my new project. It's been in the making for 3+ weeks now. I wanted to get this done by Easter (2014), but I failed. Well, it's just another failure, right -but anyways I want to get it done now.
The basic idea was to have a single piece printed frame for the eggbot. I though the eggbot era seems to be gone by now – since all the designs out there were done around 2011 - but for a newborn like me every joke is new :)
You can see the frame turned out nicely. I have not finished the design completely, but I wanted to get help with the electronics and I uploaded it to the Thingiverse hoping it'll get some traction and people might comment on it:
Well, in a few days it has gotten 500+ views and quite a few downloads. So maybe the topic is not completely dead.
Here is the basic design
2014_0414_EB 007.JPG

2014_0414_EB 009.JPG

It's intended to do eggs 56 x 41 size (in general). Maybe golf balls too. I planned to use the arduino uno, and stack an Adafruit motor shield on it. Well this was a completely bad choice. After 10 days of trying I have given up on the Adafruit motor shield. The main problem with it is that it's not communicating with the software's out there (Thunderbug, Zaggo) and If I want to make it run I'll have to re-write the code. The only thing I could make to run was the servo... The adafruit guys told me it's a bad path.
So I switched gears, got some breadboard and two SD8825 stepper drivers that I had plenty of (used them in Hadron builds) and wired it up according to the pleasant hardware's schematics ... 28-h563-no
The original schematics calls for two polulu stepper motor drivers – A4983. This is basically the only difference between their and my design. The A4983 has to have all three jumpers populated (MS1, MS2, MS3) to achieve 1/16 step micro-stepping.


Having populated in this case means that those pins have to be grounded to the “-” wire of the 5V bus.
For the SD8825 it is enough to have MS3(M2) on high only to achieve 1/16 micro-stepping


"Micro-stepping: On most controllers “H”(high) corresponds to a jumper in place and “L”(low) means no jumper installed."

All the firmware's call for 1/16-th microstepping.


Notice the last two numbers. 200 refers to the stepper motor having 200 steps/revolution, while the 16 refers to 1/16 micro-stepping

Here is the same thing from Zaggo

Notice that he is using less arguments to define a stepper motor. X_axis_ms1 and X_axis_ms2 and a couple others are missing.
At the end of the day all stepper motors can be defined by three arguments only. Number of steps it has to complete, the direction and enable.

So this is where I'm at right now. The motors grinds or jerks like a coffee-grinder. I've tried all the firmware's there. They all compile OK, and then I can upload them to the arduino.

The SD8825 stepper drivers have their pot set to VREF=0.5 V which corresponds to current limit = 2xVref = 1 A

Part of the stepper drivers being powered from the Arduino board (5V) and I've used a 12.6V 1 amp power supply (old battery charger) to power the stepper motor bus. I though this might be an issue, so I switched to an ATX power supply, but the contraption had the same issues.

I got stuck here. If I missed anything please advise.
Thank you.

Re: Dibloff's eggbot/spherebot build

PostPosted: Mon Apr 21, 2014 2:36 am
by dibloff
Here is a video of the motor grinding sound:

Re: Dibloff's eggbot/spherebot build

PostPosted: Mon Apr 21, 2014 5:11 am
by Enraged
did you check that your steppers are wired correctly? I had a stepper grinding issue and it was because I hadn't paired the stepper wires correctly so they were fighting against each other. It just sat there and vibrated.

Re: Dibloff's eggbot/spherebot build

PostPosted: Mon Apr 21, 2014 12:18 pm
by dibloff
- Enraged - Thanks for the suggestion with the motor wires. All wires double/triple/quadruple checked and the phases are connected properly. All motor phase resistances are in the 6.0-6.2 ohm range.