Forum Replies Created
-
AuthorPosts
-
aldenMember
Your issue is noted. Let me get to a spot where I can test all your files and see if I can reproduce the problem. This may take me a few days, unfortunately. Looks like you are running 0.92 – Crocs. Can you please post all the gcode files you ran?
Thanks
Alden
aldenMemberIt should be possible to drive TinyG from the Arduino RX and TX lines. There is a 3 pin female header right above the reset button that is labeled for Arduino connection. Follow the instructions on the board silkscreen and connect the RX to the Arduino TX line, the TX to the Arduino RX line, and the ground to the Arduino ground. You will also need to put a jumper on J17, which is the 2 pin male connector located to the left of the Arduino connector (in between the shrouded programming header and the FTDI USB chip). Remember that the board is running at 115,200 baud, so will need to set your serial lines to that baud rate.
aldenMemberActually, JSON was added in 0.93 ,Fanny Pack (release names are from bad fashion trends). You will need to upgrade. There’s info on the wiki about how to do this, but basically it involves an Atmel ISP MKII programmer and AVRstudio. Sorry if this was not clear.
Alden
aldenMemberHi.
Grblshield is a 3 axis stepper controller that runs the motors for the grbl softeare. Grbl runs on an arduino. Lookup github/grbl/grbl.
TinyG is a standlone motion controller that runs 4 axes. The specifications are different. See http://www.synthetos.com/wiki/index.php?title=Projects:TinyGBoth provide up to 2.5 amps per phase. Your motors should work but you should use heatsinks and fan cooling.
Lead time is typically about 1 week or less.
We ship USPS international.
aldenMemberIt looks like the Y axis is not working for some reason. You did all the tests we would. One question – What does the Y LED do? Barring a simple answer we may need to exchange the board. It’s difficult to diagnose without an oscilloscope and some test programs.
aldenMemberYou are correct. Pins D2-D8 are used as you stated. 5v and ground are also used but unlike pins D2-D8 you can use them for other purposes as well. Note that the V1 board uses the 3.3v supply instead of the 5v supply.
aldenMemberCheck the master branch. They should be up there.
aldenMemberI agree with you on the NC side – it’s more failsafe. This is one of the reasons I want to get to the SW parameter.
Many limit switches have both – i.e. are SPDT switches. Do yours?
aldenMemberThe limit switches are currently programmed for NO operation, not NC. Given your interest in different modes I can escalate the priority, but I’m reluctant to promise a date just yet. Greyhound is right – all the code for this is in the gpio.c/.h module if you want to give it a go in the mean time.
If you go for a hardware solution bear in mind that these are 3.3v inputs. Do not feed them w/5v or you risk burning them out. They are set to “totem pole pullup mode” as per the ATxmegaA data sheet (page 132 in Atmel’s doc #8077.pdf).
It may be as easy as wiring an NC switch across the input and ground and changing the interrupt edge for that bit in gpio.c
aldenMemberWhat terminal program are you using? I’m not sure this will work with the Arduino serial monitor – I’ve never tried it. Try using Coolterm (Windows or mac) at 9600 baud. Its available free on the web.
aldenMemberHere’s the link to the JSON spec.
http://www.synthetos.com/wiki/index.php?title=Projects:TinyG-JSONIt’s pretty much identical to the command line but there are a few differences.
– JSON can deal with all axis or motor parameters in a single command as opposed to multiple round trips
– It can also treat the motor or axis as an object, as opposed to a collection of unrelated parameters (see the spec to see what I mean)
– You can configure the contents of status reports using JSON, you can’t in text mode.On the flip side:
– Text mode has displays for “groups of groups” – like all axes, JSON does not. You have to get each axis separately. This is because there’s really not enough memory for arbitrarily large JSON strings.aldenMemberThe 0.93 dependency is the JSON code in TinyG. All communications between the GUI and the board is in JSON. This opens up new possibilities such as simplified synchronization of objects between the GUI and the board, and simplified GUI programming (serialize / deserialize operations replace knotty little parsers)
aldenMemberIt does sound like you are experiencing thermal shutdown. I’d look first at the mechanical system. This is usually the root of the problem – at least if the Y axis is fine and it’s running the same type of motor. Would this be the Proxxon Micro Mill MF70? How hard is it to hand crank the X axis? It could be that you need to loosen the jib screws on the X axis ways. Or perhaps it needs some way oil. Or perhaps the lead screw is binding. Does the friction seem worse at either end of travel and less in the middle? Could be a jib alignment problem. Finding the source of the mechanical issue is the best solution, if indeed it is a mechanical issue.
What stepper motors are you running – manufacturer, model number, and preferably a link to the spec sheet if you can find that. Roughly how far up do you have the current setting trimpots? All the way, 80%, 60%?
If none of this works then the heatsinks would help, but if you are pushing the envelope you will also want to fan cool them. They attach just by peeling and sticking – no soldering is required.
aldenMemberAs mentioned in the other post, you seem to have found the limit of the mechanical system. You can try to see if you can reduce friction in some way. If you have a “standard” mill then see if the ways adjusts are smooth and not binding (obviously they can’t be too loose or you get play). See if the screw and shaft assembly is true and in alignment, and of course the whole thing is well oiled. At the end of the day it’s all about the mechanical system.
aldenMemberExcellent. At the end of the day it’s the mechanical system that this must all be tuned to. You might take a closer look at the $4 and $5 defaults. These set the upper limit that the machine can run. Find that experimentally and set these. This will tell the software to limit inputs to these rates. That way if a Gcode file as feed rates that are too high for the board the system will still function properly – it just won’t achieve the requested feed rate in the file.
-
AuthorPosts