Forum Replies Created
-
Posts
-
Hi Ray,
Is it the case that (A) you sometimes get a good connection, but lose it when the cable is wiggled, or (B) you never get a good connection, but always get the “no module named serial” error?If (A), you might consider trying a different USB cable if you have one handy, and please contact our store directly about getting a replacement cable or EBB if necessary. If (B), please say so. :)I’m pretty sure that a motorola phone charger can’t supply 550 A of current. As for regulated, I would *guess* so. Does it have a roman numeral in a circle, say “IV” or “V”? If so, that would indicate that it’s a modern switching (regulated) supply.
I’m not sure what power supply that could be, but please make sure that it has *regulated* 5 V output. That is necessary for proper operation.
As for the issues that you are describing, it sounds like there may be multiple soldering problems. Double check that all of your components are in the correct locations and correctly oriented, and check to see if you might have any accidental connections between neighboring pins on the bottom side, or any solder joints that look dry, blobby, or cracked.It sounds like you’re at a stage where you could take the data generated by the FFT program and make it so that you could put its data– instead of the data from the camera –in the data array that gets sent (1) to your screen and (2) to the Peggy in the Mirror Peggy sketch. And, that would be the most straightforward way to do so.
You said in the other thread that you had this working now. Do you know what the problem turned out to be?
“redMatrix” is one of several internal arrays for storing the colors on the screen.
The colors are pre-defined colors that can be used in the game, referring to them by name rather than numbers.Did you move the power jumper from “battery” to the “ac adapter” position?
Were you not able to get any of the programs working, or is this still in RoboPaint?
Well, that’s partial progress at least. :(
This is the first report that we’ve had of this kind of a problem on a Mac; we’ve heard one similar report on a Windows machine, apparently due to a bug in the underlying library that RoboPaint uses to connect to the serial port. I have added this report to our bug tracker ( https://github.com/evil-mad/robopaint/issues/127 ), and may possibly have some ways to follow up a little later in terms of gathering additional logging data. Can I ask what Mac OS version you’re using?In the mean time, I’d suggest trying* A different USB port on your computer.* The other two WaterColorBot control programs, RoboPaint RT and our Inkscape extensions.Hmmm… maybe more than one issue here.
First off: I’m sorry to hear about the damaged wire– do you think that the wire is possibly broken other than just the loose end being broken off? (If necessary, we can send you a new motor, but you’ll be up and running much sooner if the problem can get sorted out with the motor that you already have.)In any case, it sounds like you’re still not getting a good connection on one or more of your motor wires. The two motors should give identical resistance when they’re hooked up correctly and energized. I’d suggest removing all four wires from the terminal block and inspecting the ends. Make sure that the metal are cleanly exposed on each, and inserted to such a degree that the terminal block will make a good connection. Once inserted, pull gently on each wire, to make sure that they’re held in place.If you still can’t get a good connection, you might try (A) stripping that green wire a little further back and trying again and/or (B) swapping the wires between the two motors. If the problem moves with the motor, then that would confirm that it’s some sort of an issue with the wires (or less likely, the motor itself). If it does not move, then that would indicate that it’s actually a problem with the EBB.Separately, I would have expected that if you were trying to print, you should still have one motor moving, even if the second were disconnected. So, there may still be something amiss. If Robopaint appears to be moving *very quickly,* it may be in “simulation mode,” which is what it does when it does not have a connection to the WaterColorBot. (When you launch RoboPaint, it will indicate very clearly if it cannot find the WaterColorBot– if you haven’t seen this, disconnect the WaterColorBot’s USB cable and restart RoboPaint, so you can see how it reacts to this situation.) You can also test this by asking it to lower and raise the brush. If it has a good connection and power, it will comply.When we designed in the part, it was an active production part, easily found in distribution, and not remotely overkill. Unfortunately (1) they have now been discontinued in the through-hole package, and (2) most of the comparable parts do not have the same pin order.
The actually important spec– the reason that we were using this type — is the saturation voltage. If the board needs to be able to run from 4.5 V (as do the Peggy 2 and Meggy Jr RGB), and the LED drivers drop a certain voltage, and you need to be able to drive blue or white LEDs that take up to 3.6 V to run, then there’s only about 0.5 V (or less…) available for the high-side driver to take up.So, your high-side driver needs to be able to handle roughly 25 LEDs * 25 mA = 625 mA (25 LEDs for a Peggy 2 row, or 24 LEDs for a Meggy Jr RGB row), and have low saturation voltage, hopefully spec at < 0.5 V at 1 A. And, if you search for those two requirements (plus packaged in a through-hole package), you’ll find very few options. One is the “big expensive brother” of the 2stx2220, the stx790A (also now discontinued). Another is the BC640TA, which might be a good choice for you.There are some bugs in the underlying libraries that we use, which can occasionally cause a freeze, particularly at high generation number. So far as I know, those have not been fixed. We’ve been thinking about rewriting the software with a different voronoi generation library, but have not had time to do so yet.
You may find it helpful to download Processing and run directly from source, so that at least you’ll get an error message when the core crashes.And, you may find it helpful to save output at lower generation numbers. We’ve generally found things to begin to converge at around generation 40, and look best after generation 100.@emu: unrelated issue. Please adjust your motor current setting on the EBB, as described on the documentation page about “making your first plot.”
This may not be what you’re looking for, but we normally recommend StippleGen– a utility that we built for this very purpose — which can convert jpg files to black and white stipple or TSP drawings that can be drawn very well by an Eggbot: http://wiki.evilmadscientist.com/StippleGen
Two good examples from other folks as well:
http://www.flickr.com/photos/lodefink/8602880115/in/pool-eggbot
http://www.flickr.com/photos/75390883@N06/8559930536/in/pool-eggbot
