Newb Help needed

Home Forums TinyG TinyG Support Newb Help needed

Viewing 10 posts - 1 through 10 (of 10 total)
  • Author
    Posts
  • #7888
    bvi8370
    Member

    Hi Guys,

    I’m having issues testing my Tinyg. When running a gcode test of a circle in tgFX, the stepper fields are collapsing. The test starts out fine, but the steppers slowly lose power and stop after about 20 seconds. They let off a high pitch at that point, and my computer barely responds until I pull the power(or usb) to the Tinyg. I went with the 3amp steppers since the Tinyg can supposedly handle them with proper cooling.

    Any suggestions? bigger power supply? lower amp steppers?

    Here’s what I have:
    Tinyg v8
    24v 8.3 amp power supply
    3 – nema 23 3.0amp steppers (model KL23H2100-30-4B at automation tech)
    stepper specs:
    4-wire
    2 Phase
    Bi-polar
    Holding torque 382 oz-in
    1.8°/200 Steps Per Rev.
    3.0 Amps Current Per Phase
    Inductance: 7.0 mH

    And the test code I’m using in tgFX:
    G17 G20 G90 G94 G54
    G0 Z0.25
    X-0.5 Y0.
    Z0.1
    G01 Z0. F5.
    G02 X0. Y0.5 I0.5 J0. F2.5
    X0.5 Y0. I0. J-0.5
    X0. Y-0.5 I-0.5 J0.
    X-0.5 Y0. I0. J0.5
    G01 Z0.1 F5.
    G00 X0. Y0. Z0.25

    As a side note, when I connect a 12v .45amp PC case fan, the tinyg becomes unresponsive and Windows reports that it has malfunctioned. I tried two different fans and had the same result. The fans blow fine while connected.

    Thanks for the help!

    Barry

    #7891
    cmcgrath5035
    Moderator

    Barry
    If I had to guess, you may be experiencing thermal shutdown of the drivers, but that is somewhat a wild guess.

    First up
    1. We need to see your full parameter set, the results of a $$ command. Here is the most useful way for you to post it:

    2. Your parameter set will report what Firmware build you have on your tinyG. If not already, you will want to upgrade to fw 440.14.

    3. Since you mention tgFX, I’ll assume you are running from a Windows PC?
    tgFX is no longer supported and likely has issues interoperating with fw 440.14. If you want to run a GUI, I’d suggest you give ChiliPeppr a good look

    As a side note, when I connect a 12v .45amp PC case fan, the tinyg becomes unresponsive and Windows reports that it has malfunctioned. I tried two different fans and had the same result. The fans blow fine while connected.

    Are you powering the fan from the tinyG fan header? Have you set the voltage jumper for 12V? I wonder if the 12 v fan is actually being fed 24V and dragging down that power supply. Otherwise, I can’t explain the behavior.

    Does your job run in air (not cutting)?
    Perhaps you are milling too fast, although F2.5 is rather slow already.
    What are you milling?(wood, AL, foam, ?)

    • This reply was modified 9 years, 4 months ago by cmcgrath5035.
    #7894
    bvi8370
    Member

    cmcgrath5035,

    Here’s the link to the dump file. The firmware version is 440.14. I ran the same g-code in putty, and the results were the same. I did try Chilipeppr before tgFX, but there was no movement from the motors. I probably wasn’t doing something right. The fan was connected to the board, but it’s not a big deal for me to wire it to the extra posts on the power supply. The job is running in air (the tinyg and motors are not mounted to a machine yet).

    Thanks for the help!

    Barry

    #7898
    cmcgrath5035
    Moderator

    So what OS are you running? The $$dump is really hard to read with the funky spacing, but I could edit it to be more useful.

    I suggest you turn homing off until you have a machine connected.
    $xsn, $ysn, $zsm and $asn = 0.

    No way should you get thermal shutdown in your current setup – unloaded steppers require very little current to spin when unloaded. So that is not the cause.

    Try this:
    Reset your tinyG, it will come up in mm mode ([gun] default gcode units mode 1 [0=G20,1=G21])

    From command line, enter
    G0 X125
    Your X motor should turn 100 revolutions
    G0 X-125
    Your X motor should turn 100 revs in the other direction.
    Likewise play with Y and Z.

    Just trying to make sure motors are connected correctly.

    No need for fans.
    To be honest, the tinyG shutting down with the fan connected makes no sense at all.
    Can you monitor your power supply voltages while this is running?
    Could be a bad PS, I suppose.

    If you try using Chilipeppr, you could issue these commands in the Serial Port Console (lower left) or you could use the Jog widget

    #7900
    bvi8370
    Member

    I moved the setup to another room to isolate it from the rest of my computer and network gear. I was testing on Win7 x64 but switched to my laptop which is Win10 tech preview. I also installed Teraterm to see if the feedback from $$ was formatted better than Putty – it’s not.

    Moving everything seemed to work briefly but each motor would stop part way through the iterations using G0 X125, G0 Y125 and G0 Z125. The coordinates continued to increment in Teraterm even though the motors had stopped abruptly and were letting of a high pitch(tried all three axes separately and together). At least the Tinyg was reporting the entire test back to Teraterm. However, the reporting stopped after a few tests, and Windows started losing connection to the Tinyg every time I run a test. Windows reports the device has malfunctioned, and I have to pull the power from the wall for everything to reconnect. The fan is not connected.

    The power supply was steady at 24.4V through every test – even when the motors halted. I’m not sure enough voltage is making it to the motors. I tested on the black and green and only saw 2.5V. The motors are 5V. I’m not sure if the 5V is across both windings for a total of 5V.

    I’ve uploaded a couple photos of the board with connections and the PSU.

    Beyond frustrated at this point. Maybe I bought some of the few motors that just don’t work well with Tinyg. Or maybe my Tinyg is a lemon.

    #7902
    cmcgrath5035
    Moderator

    I am a bit stumped too.
    I don’t think it is your motors.
    Something is borked in the internals of your tinyG application, or perhaps tinyG hardware.
    Since we are hacking here, try this:
    Step 0 – The next step will wipe out our parameter set, so make sure you have a copy of your current settings.
    By the way, how did you first enter your tinyG parameters?
    Step 1 – Start up tinyG and terminal. If you can, capture your terminal session to a log file
    Step 2 – I typically enter a $ or ? command just to verify good interface behavior, you get some short status/parameter listings.
    Step 3 – Enter the command $defa=1 and enter.This step will perform a ‘factory reset’ of your tinyG, restoring a set of default compiled in parameters to EEPROM
    Step 4 – Run $$ command, then use command interface to reset your parameters to the values you have been testing with and try again.
    Step 5 – Post your terminal log if you made one

    #7910
    bvi8370
    Member

    I followed your steps and reset the board. I’m guessing the board had a default config on it, as I never made any changes to the existing config.

    I did a comparison of the previous setup($$) to verify the file did in fact change. Nothing notable was different, just some json feedback parameters. All the feed rates were the same.

    The first test G0 X125 halted after about 5 seconds with the high pitch I’ve come to fear so much. Also, the USB disconnects as soon as the code is executed. That is starting to sound like a power issue.

    I lowered the feed rates for each motor from 800mm/min to 300mm. Pretty much the same result except slower.
    I lowered the feed rates to 100mm and that’s when the PSU took a dump. There was a small pop from the PSU during the G0 X125 test. No smoke but it’s KIA. Maybe the PSU was on death’s door the whole time. I hope the tinyg was unaffected.

    Off to Mouser.com I guess. 8020 extrusions arrived today; linear motion comes tomorrow so it’s kinda late to turn back now!

    By the way, of the two Meanwell PSUs suggested here, which one might be best for my setup?

    Thanks again for your help!

    Barry

    #7911
    cmcgrath5035
    Moderator

    OK, after some digging I realized that you probably did have default settings, $xtr=1.25mm is the key for me, that is a default value. So your would not expect to see changes after the $defa=1. I had you do the defa=1 since it reloads some non-user settable (not visible with $ or $$)internals as well that might have been contributing, based on the experience of others.

    Lowering the feed rates was a good experiment, but frankly the motors require very little current to spin when unloaded.

    The fact that tinyG fw was misbehaving (disconnecting from PC) is also a good indicator that PSU was drooping, causing logic level voltages to excessively droop on-board, likely causing tinyG to reset. It seems a common failure mode in the supplies is for the current limiting detector to fail, resulting in full output at low amps output but output voltage droops (fails) under load. Since steppers demand current in very short pulses, this can be difficult to diagnose with just a voltmeter.

    Since you are NEMA23s, go with the 350W supply. 14A is a bit of overkill, but 6A (the 135W) would be marginal.

    If you want to play with your existing supply and it has an “output current adjustment POT”, you could (off line) crank that pot back and forth a couple times, then leave it set to max. A common failure mode is a ‘dirty pot’, sometimes this will yield short term improvement, but frequently the improvement does not last long.

    #7944
    bvi8370
    Member

    I received the Meanwell 350 watt PSU today, and I’m happy to report the Tinyg and NEMA23 steppers work flawlessly with it.

    I spent the weekend cutting, threading and assembling 8020 extrusions for the frame and gantry. Now to get a project box to mount the electronics and pick out a spindle. Here’s a photo of the project so far.

    FLA100 build

    I have a local steel fabricator cutting the rails and leadscrew to length. My original plan was to just buy the frame from FLA, but building everything myself has been a lot more gratifying. Nate has the plans and BOM readily available for the FLA100 on his website.

    Thanks for the help troubleshooting!

    Barry

    • This reply was modified 9 years, 4 months ago by bvi8370.
    • This reply was modified 9 years, 4 months ago by bvi8370.
    • This reply was modified 9 years, 4 months ago by bvi8370.
    #7948
    cmcgrath5035
    Moderator

    Cool, good luck with the build!

Viewing 10 posts - 1 through 10 (of 10 total)
  • You must be logged in to reply to this topic.