eggbot first run calibration issue

[Rainer M. Engel]

Dear egg-botter,

I built up my egg-bot as soon as it found its way to Berlin. My issue
is now that all seems to work properly except plotting correct. I
tested the servo and motors (distinct steps) via interface within
inkscape. All seem to work fine.

But when I want to print even the simplest and smallest path from
inkscape the Pen Arm (MOTOR I) kind like falls to the side and moves
against the chassis. By that I get a line round the egg, since MOTOR
II turns.

I'll provide more details to this (software version and driver are the
most recent), but maybe one of you knows what I've done wrong.

regards,
Rainer

[Brian Schmalz]

That sounds a lot like an adjustment issue with the current going to the stepper motors. You may want to try adjusting the little current set pot on the circuit board as you run your plotting and see if increasing the current to the motors will allow your pen arm to stay where it's supposed to stay.

See the first section of this page: http://wiki.evilmadscience.com/Eggbot_Troubleshooting

*Brian
 

Home www.schmalzhaus.com
Work www.logicpd.com

---

From: Rainer M. Engel <brigade...@googlemail.com>
To: EggBotUser <eggbo...@googlegroups.com>
Sent: Tue, May 24, 2011 10:38:22 AM
Subject: [EggBot] eggbot first run calibration issue

--
You received this message because you are subscribed to the Google Groups "EggBotUser" group.
To post to this group, send email to eggbo...@googlegroups.com.
To unsubscribe from this group, send email to eggbotuser+unsub...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/eggbotuser?hl=en.

[Bill Culverhouse]

I agree, what you want is enough current to hold the arm steady, too much current

will cause the motors to over heat. You can engage the motors with the manual

control tab, and then adjust the pot until the arm holds its position. This also

increases the voltage to the egg motor.

 

-b

To unsubscribe from this group, send email to eggbotuser+...@googlegroups.com.

[Brian Schmalz]

Bill (and all egg-botters):

Just to be clear about the heat issue: even if you crank the current limit pot all the way up, and deliver maximum current to the motors, they will not over-heat, in the sense that they will not become damaged by the heat. They WILL get hot enough that you won't want to touch them! And when electronics gets that hot, it starts smelling like 'hot electronics' which many people dislike.

So it's always best to keep the current as low as you can, while still maintaining proper holding torque on the arm. But rest assured that you can't break anything (at lest I haven't been able to) by turning the pot too far. The driver chips have excellent thermal cut-outs, which will give very jerky motion (or none at all) if they are getting too hot. This mechanism prevents them from ever being damaged by over-heating. And the motors are very sturdy and won't die from the heat either.

Also, as you increase the current to the motors further and further, you will find that they loose resolution. What I mean by this is that plotting will still work, but lines will take on a more stair-stepped look. This is because, at some point as you turn up the current limit, the driver chips can't generate all of the correct micro-steps (steps within a step). Another good reason to go as low with the current as you can get away with.

There is a sweet spot with that little pot, and it's not too hard to find with some experimentation.

*Brian

 

Home www.schmalzhaus.com
Work www.logicpd.com

---

From: Bill Culverhouse <bill.cul...@gmail.com>
To: eggbo...@googlegroups.com
Sent: Tue, May 24, 2011 1:03:56 PM
Subject: Re: [EggBot] eggbot first run calibration issue

[Random Info Net]

Interesting.  I'm new to Egg-Bot and was surprised at how stair-stepped my first few attempts have been.  My first thought was that the stepper motors didn't seem to have the resolution I was expecting.  Maybe I need to turn the current *down* on mine.  Sounds like there's no harm in fiddling with it.

Might slippage also produce stepped results?

Completely different topic:  Anybody know why there's a Forum at Evil Mad Science *and* a Google Group?  They aren't synchronized, are they?

[Windell H. Oskay]

On May 24, 2011, at 6:23 PM, Random Info Net wrote:

> Interesting. I'm new to Egg-Bot and was surprised at how stair-stepped my first few attempts have been. My first thought was that the stepper motors didn't seem to have the resolution I was expecting. Maybe I need to turn the current *down* on mine. Sounds like there's no harm in fiddling with it.

True indeed, no harm in trying.

> Might slippage also produce stepped results?

Not typically, just "backlash."

> Completely different topic: Anybody know why there's a Forum at Evil Mad Science *and* a Google Group? They aren't synchronized, are they?

The forums are for all Evil Mad Science products, Eggbots included. There's also a Yahoo group. ;)

[Random Info Net]

From the perspective of a new Egg-Bot owner, I think it would be useful
to have something like a"Egg-Bot tuning flowchart", complete with
associated test patterns and information about how to read the plotted
patterns. E.g. "When you plotted the spiral test pattern #3, does it
exhibit steps, as in the diagram below? If so, you might try reducing
current to the stepper motor. " And "Does your test pattern exhibit
backlash, as in the diagram below? If so, you might try <what?>"

Heck... I don't even know what "backlash" is, in the context of
Egg-Bot. What would it look like, for some given test pattern?

Thanks for your input!

[Dan Newman]

On 24 May 2011 , at 8:03 PM, Random Info Net wrote:

> From the perspective of a new Egg-Bot owner, I think it would be useful to have something like a"Egg-Bot tuning flowchart", complete with associated test patterns and information about how to read the plotted patterns. E.g. "When you plotted the spiral test pattern #3, does it exhibit steps, as in the diagram below? If so, you might try reducing current to the stepper motor. " And "Does your test pattern exhibit backlash, as in the diagram below? If so, you might try <what?>"

Some of these would be easy to demonstrate with photos such as staircase steps.

Some problems can look the same; for example, a pen which was a little loose at first
vs. the egg cam'ing out a little before settling in. [Tighten the pen; wiggle the
egg a little, turn it a bit, try wiggling, turn it some more, try wiggling, eventually
it will seem to be solid in the egg cups.]

Some distinctly different issues will all have the same net effect. For example, if a
horizontal line of length 3200 pixels is drawn, the ends line up horizontally but don't
meet, then at least three culprits can all be present and contributing: pen friction,
too tight of spring compression, or too weak of motor current. If the ends
didn't meet horizontally either, then some combination of precession, cam'ing, or
a loose pen also occurred. [And, BTW, don't discount pen friction: for me at
least, it's a real issue.]

Documenting each of these separately sounds like a good idea and I may
consider putting a wiki page up with them. But hey, it's a wiki so feel
free to start the page yourself!

A diagnostic flow chart might not pay the same sort of dividends. A new user
making their first plot or two may well see a problem which is the combination
of several different issues. It may be difficult for them to correctly
identify each of the individual issues which contributed. I know it would
be a challenge for me at least. And my knee jerk reaction would be to go
back to basics ("improving plotting precision" steps) first so as to try to
eliminate as many problems first.

> Heck... I don't even know what "backlash" is, in the context of Egg-Bot.
> What would it look like, for some given test pattern?

One effect which I think we here tend to refer to as "backlash" or "psuedo backlash"
is caused by "wind up" in the egg cups. When the egg motor turns the egg, some
of the work will be stored as elastic energy in the egg cup itself: the egg cup
will twist a tiny amount. Only so much of this twisting will occur as you
continue to turn the same direction (i.e., it doesn't continue to accumulate
like winding up a rubber band). But, when you go to turn the opposite direction,
you'll release that twist and possibly turn farther than intended in the opposite
direction. This is the opposite effect as mechanical backlash in gears where you
lose some motion when you change direction. It may well be that the stepper
motors are subject to this more traditional form of backlash.

When you have wind up only in one direction, the result is another case of a
line not drawing as long as it was expected too.

Regards,
Dan

[Bill Culverhouse]

I find that "stair step" is reduced by reducing speed. It seems

the vibration of the motor at higher speeds causes a jitter in the

pen. Also shortening the length of pen reduces pen wobble.

By shortening I mean move the pen arm closer to the writing

point of the pen. I've noticed the even tightly held by the thumb

screw a pen will still wobble a bit in the holder if it sticks through

the pen arm very much.

 

My $0.02.

 

-b

[Dan Newman]

On 25 May 2011 , at 5:46 AM, Bill Culverhouse wrote:

> I find that "stair step" is reduced by reducing speed. It seems
> the vibration of the motor at higher speeds causes a jitter in the
> pen.

How fast is fast? That is, I'm curious as to what speeds you've
seen this at. And, I'm curious as to just how fast people do
plot and get satisfactory results. I know I tend to be a slow driver
as regards the Eggbot, never going over a self-imposed speed limit of
500 steps/second for pen-up or pen-down operations. (I compensate
for this with my S. California freeway driving ;) I have seen any
number of videos of folks appearing to get good results and cruising
along at what I think are faster speeds, but I've never been sufficiently
curious to try myself. (Which is to say that I don't have any good
reason for why I don't go above 500 steps per second.)

Dan

P.S. I generally plot at around 250 - 400 steps per second for pen down,
and 500 steps per second for pen up. My choice of speed tends to have
more to do with how much ink I want to lay down (thicker/thinner lines
and more/less color saturation). Different pens behave differently as
do different types of eggs (less predictable).

P.P.S. And I've always been curious what the rough goose eggs do to John
Deckert's pen nibs!

[Random Info Net]

I tried a couple of the tips from this thread, last night.

I started by reducing all the speeds, but that didn't make a noticeable
difference. However, I will keep the speeds low, just to be safe.

Then I tried the current. If I remember the directions correctly, I
turned the potentiometer a little counterclockwise and that gave an
equally bad result. Turned it further counterclockwise and the arm just
collapsed. Turned it clockwise past the point where it was when it came
from the factory, and there was a HUGE improvement. Counter = lower
current and clockwise = higher?

I'm now satisfied that the unit is performing as it should.