Forum Replies Created
-
AuthorPosts
-
aldenMember
I just made it through the entire file very quickly by setting the feed rates to F100 and F200. No crashes. It sounds like a flow control problem with XON/XOFF. Let me compare your Coolterm settings to mine. I’ll update. I’m using this forum more like an Instant messaging system.
UPDATE: Your coolterm serial settings show a Bluetooth serial port. This can;t be right. Have you installed the FTDI drivers? I ahve to beleive so, otherwise how would it work at all?
Interesting. You are running Coolterm 1.4.1 build 162. I am running 1.4.0b2 build 142. I wonder if something broke in the mean time? I worked with Roger to get this right, and even sent him a TinyG setup a year or so back. Maybe things went sideways? If you can still get the oder builds perhaps you could try that. I may try the current build, but not until I figure out how to recover to old one first in case it doesn’t work.
- This reply was modified 12 years, 3 months ago by alden.
aldenMemberI’m just starting to run your file now. I have run a lot of files over 500 lines of Gcode. Let’s see what’s going on here. A few questions.
– Is this in Inches on mm? Looks to be in inches but I’m not sure. I added a G20. If it’s in inches the initial Z lift of 10 inches is too high for my machine. I modified it. [UPDATED] I guess it’s mm as it’s a small watch movement. I seem to be drawing a Flava Flav sized watch. I’ll change to mm.
-Do you use the G43 tool offset? TinyG does not support this command. No harm done, as long as you don,t need it.
-The feed rates are very very slow. I see F1 and F2 being used. Is this what you have intended?
– What TinyG build are you running? Should be a number like 338.12
– What line number does ti stop on?
aldenMemberI have just wired up a v3 grblshield to a 42BYGHW811 which I got from Phidgets. It’s not identical to your motor, but it has the same current rating at 2.5 amps. Mine is a 200 step, I think yours is a 400 step. I have no issues and the chip is getting warm (as expected) but not hot. The sequence of wires on the connector is red-blue-green-black.
aldenMemberI’m pretty sure the drivers should have no problem with these motors. We have tested the grblshield with very many motors. I’ll look around and see if I have some of the Phidgets you have. Can you please provide some more information if you can.
– Try backing off the current as much as possible while the motors are still turning. The drivers should not be getting too hot with these motors, but expect them to get warm.
– How do you have the motors wired up? The green and black are paired and the red and blue are paired.
– What is the current setting? The pot should start in the middle of the range. Please confirm that the pots have not been forced past their limits (and broken).
– What axis or axes are the motors attached to? Are the LED lights flashing? Do you get the same behaviors on other axes?
– Do you have indication that grbl is functioning, and what commands are you issuing to grbl? The same as in your earlier posts?
– What is your grbl configuration? Has this changed from before?
– What is your power supply and is it adequate? Is it delivering consistent voltage when you try to move the motors? Sorry if you already asnwered this but I can’t see anything on page 1 of this thread while I’m typing on page 2.
-Alden
aldenMemberQuestions, questions. What version of Coolterm are you using, and is it Mac or PC? have you enabled XON/XOFF in the coolterm dialog? What does your coolterm config look like? I assume you have or you would not have gotten that far. Can you post the file you are sending somehow – as a dropbox, maybe?
aldenMemberThe normal way to do this is to issue a G92 command. Type G92 X0 Y0 Z0 to zero all axes. What’s actually happening is a bit more complicated, and involves the coordinate systems. The G92 is actually an offset to the currently active work coordinate system (G54 by default) – allowing you to specify the coordinates with which you would like to refer to the current tool position. More references can be found on our wiki:
http://www.synthetos.com/wiki/index.php?title=Projects:TinyG-Homing
Also, see the Kramer NIST spec for how G92 functions. This is what we coded to.
http://www.isd.mel.nist.gov/documents/kramer/RS274NGC_3.pdf
-AldenaldenMemberJan – we just lost about 3 days worth of posts discussing your issue. We’ve been fighting forum spammers and something got dropped. Don’t worry, your replacement board is on the way.
-Alden
aldenMemberYou can hang onto your current board until the replacement arrives. It might be handy to have them both there. Please use synthetos <at> gmail.com. If you have your order number or any other materials from your purchase please put that in the email so we can match up with your original order more easily. Sorry for your difficulties.
Alden
aldenMemberI’m out of ideas. We should send you another one. Would you be willing to send yours back once you get the other one? If there is a problem with the shield I’d like to be able to find the cause. Do you have a v2 or v3?
-Alden
aldenMemberI stand corrected, you are right. You do not need to make the Z axis mod. I got this confused with another issue.
That said, perhaps it is good to try to ramp up the motor speed and see if it fails at lower speeds. It could be that the erratic behavior is due to operating on the edge of the maximum speed (seek rate) Try the following sequence of tests:
G1F50Z10
G1F100Z20
G1F200Z30
G1F400Z40
G1F500Z50
G1F600Z60
G1F700Z70
G1F800Z80
G1F900Z90
G1X1000Z100
G1X1100Z110
G0X120It is important that the grbl settings do not ask grbl to deliver more than 30,000 steps per second. Can you please post your grbl settings? You might try the tests with the default grbl settings which are something like this (I do not have board here to test from, so I’m going from memory – I may be able to edit later when I get back to my bench) – UPDATE: Edited to agree with grbl defaults
$0=755.91
$1=755.91
$2=755.91
$3=30
$4=500
$5=1000
$6=0.1
$7=28
$8=10 <– This is normally 50, but if you are having problems due to acceleration let’s drop it down to a slower acceleration
$9=0.05Also, can you tell me if you are running a v2 or v3 board? It says so on the silkscreen on the top. This shouldn’t make a difference, but I’d like to know.
Thanks
– Alden
aldenMemberJan, You need to make the Z axis mod to get the axis to 2 microsteps. Here is a link for that. If you do not want to do that yourself we can do this for you if you return your board. But it’s really very easy and takes about 5 minutes with an exacto knife and soldering iron.
http://www.synthetos.com/wiki/index.php?title=Z_Axis_Mod_for_ShapeokoOnce you have done this please use these grbl settings for 400 step motors from DrRob’s calculator:
X/Y belt pitch (inches): 0.08
X/Y motor (steps/rev): 400
X/Y pulley teeth: 18
X/Y microsteps: 8
Z threaded rod pitch (mm): 1.25
Z motor (steps/rev): 400
Z microsteps: 2
$3 Step pulse width (µs): 30
$4 Feed rate (mm/min): 750
$5 Seek rate (mm/min): 1100
$6 Arc segment (mm): 0.1
$7 Step port invert mask (binary): 0
$8 Acceleration (mm/s²): 9.8
$9 Cornering junction deviation (mm): 0.05You should get something like this:
$0=87.48906386701663
$1=87.48906386701663
$2=640
$3=30
$4=750
$5=1100 (and this might even be too high if you Z axis is not very well aligned)
$6=0.1
$7=0
$8=9.8
$9=0.05With your current settings (which you posted on the Shapeoko forum) you may be trying to run grbl faster than its maximum pulse rate.
aldenMemberSome further questions:
1. Did you get or have you performed the Zaxis mod to make the Z axis work at 2x microsteps?
2. Are you running 200 step or 400 step motors on X, Y and Z? (200 step = 1.8 degrees per step, 400 step = 0.9 degrees per step)
3. Is your motor part of an assembled Shapeoko or is this behavior occurring when the motor is sitting by itself (unloaded)?It looks like whatever the case your grbl settings are off. See DrRob’s calculator at: http://noblesque.org.uk/ShapeOko/grblcalc/
For 400 step motors, X and Y at 8x microsteps and Z at 2x microsteps the settings should look more like this:
$0=87.48906386701663
$1=87.48906386701663
$2=640
$3=30
$4=750
$5=1100 (and this might even be too high if you Z axis is not very well aligned)
$6=0.1
$7=0
$8=9.8
$9=0.05If you are running 200 step motors change the values and see what you get from the calculator
aldenMemberIs you Z axis actually in the machine or is it just a motor hanging off the Z axis. If it is in a machine I would look at the friction on that axis.
aldenMemberPerhaps it’s your Gcode commands. Issuing G0x100 once will move X to 100. Issue it again and X is already at 100, so it doesn’t move (when in absolute mode, which is the the default). Issue a G0x0 and it will move back from 100 to 0.
aldenMemberSounds like your power supply is not sufficient. What are you using? You should have a supply that provides at least 4 amps (general rule of thumb). Do you have another supply you can try? I’d also recommend 24 volts as opposed to 12v. There is no correlation to $0, $1, $2 that I’m aware of. Also, try a slower move, like g1f300x100y100 and see if that works.
-
AuthorPosts