Question Intermittant HF350 control.

I'm using Gcode to start and stop an HF350 spindle during tool changes. Occasionally it fails to start when commanded, so a rapid Emergency Stop is called for to save the drill from obliteration.

There seems to be no easy way to monitor the start up, so perhaps a gentle start up is called for followed by a stepped acceleration to final 18000rpm. Any other suggestions would be welcome. Maybe the controller is defective in some way.

Control configuration is as follows:-

Hi Peter, I used your settings to control my HF500 through UCCNC which worked fine but then started not necessarily stop/starting in the right order.I have admit I intended to take a closer look at what I had done wrong but never got round to and lost a bit of confidence in it.

As we both mentioned earlier a section for UCCNC would be a great help, I just can't get my head!

You seem to have a pretty good understanding of UCCNC can tell me how set a keystroke to operate the tool length probe?

UCCNC master class anyone?

Thanks Hal

Hi Hal,

That is one of the things on my "to do" list!! Need to make a tool sense digitizer and learn how to programme a suitable macro first. This would probable lower the spindle fairly fast until first contact, then back off and approach slowly for second contact. At this point the measured and stored height of the digitiser would be entered into Z and drilling/milling recommence.

I'm using the Gcode generated by Pcb-Gcode to stop and start the HF350 for a tool change. This raises the spindle to allow access for tool change. Once done, Z is lowered manually (at the moment) to the correct starting point and the axis zeroed. Cycle start then sets all axes to starting point if not already there, then waits for a further cycle start before powering up the spindle (hopefully) and proceeding.

The following is an extract from the generated code - it might be of some use to you (comments are mine):-

N00750 G82 X-12.7000 Y20.3200
N00760 G82 X-12.7000 Y17.7800
N00770 M05 ;Stop spindle
N00780 G00 Z40.0000 ;Raise spindle
N00790 G00 X0.0000 Y0.0000 ;Go to X/Y reference point
N00800 M06 T03 ; 0.9000 ;Stop for tool change to #03 (0.9mm drill), then restart cycle
N00810 G01 Z0.0000 F250 ;Goto Z==0 if not already there
N00820 M06 ;Stop and wait for cycle start again
N00830 G00 Z2.0000 ;Raise spindle for drill clearance of 2mm
N00840 M03 ;Start spindle
N00850 G04 P5.000000 ;Allow plenty of time to hit the OMG button if spindle stalls!!!!!
N00860 G82 X-28.7020 Y27.3050 Z-1.8000 F250 R2.0000 P1.000000 ;Carry on drilling
N00870 G82 X-28.7020 Y29.8450
N00880 G82 X-28.7020 Y32.3850

How about calling the topic "UCCNC for Dummies"? The trouble with user manuals written by experts is that they understand the subject too well - what is obvious to them is not obvious to the un-initiated like me.

Peter

Hi Guys - the M03 commands from UCCNC/UC100 are extremely reliable. We have never seen an M03 issue with commercial machines. With larger three phase spindles running with VFD's the VFD itself has a power up curve. So the spindle reaches max RPM after X seconds - this can be programmed.

Then you use the UCCNC to delay the next line after M03. The M3 delay on after M03 is in ms so 3000ms = 3 seconds. This is the best way to give yourself some time to see if the spindle comes on.


The STEPCRAFT HF spindle controller is not a VFD type and does not have the capacity to provide such a delayed ramped startup cycle.

Must ask STEPCRAFT what may be causing the M03 spindle on in some cases to not work.

Hi Rory,

To add a little more information, when the fault happens the spindle kicks briefly at M03 as the "On" buttton indicator illuminates and the display shows the programmed speed. However the spindle remains stationary through to the next M05.

At this point the indicator extinguishes and the display shows "0" AND the audio "power on" tones sound. If the M03 through to M05 sequence operates correctly there is no sound, otherwise everything else appears to be the same.

The next M03 to M05 sequence usually operates correctly without any remedial action apart from cancelling the "OMG" status and restarting the Gcode at the missed M03 command.

The speed demand happens to be quite high ("88") - perhaps I should try to repeat the test for a lower speed demand - "10" perhaps?

I have a delay programmed for "OMG" action

(N00850 G04 P5.000000 ;Allow plenty of time to hit the OMG button if spindle stalls!!!!!)

Purists note:- for "OMG" read "JC" if you prefer it!!!

Peter

I have had similar problems with HF500, except when the spindle does not start, no further start commands will work until I turn off the stepcraft and back on again. I.e. once it fails to start, then I have to manually start it every time until I power cycle. Even pressing the Spindle start on the UCCNC screen does not work.

Turning power off and on of the spindle controller itself, does not fix the problem, and neither does restarting UCCNC. Turning power off from the stepcraft seems to be only way, and it is something I do not want to do after every tool change...

After the spindle fails to start, I can still manually start the spindle by pressing the button on the spindle controller.

I have not had enough time to really debug this, so I just need to be very quick to reset the UCCNC when I notice that spinde did not start, which I usually do by making sure that my XY coordinates are not very close to the starting point when I hit the start cycle, so I have time to stop the machine while it is moving to start location. Hmmmm I might want to make those delays longer just in case anyways to provide me more time to stop machine before it starts cutting. After the first fail it is easier, as then I just start it manually for rest of the job...

This will be much bigger problem when the automatic tool changer gets here, as then you might not be next to the machine while it changes to the tool...

Still suffering with failed automatic start up on my HF350 - brought into focus yesterday when that caused breakage of a couple of carbide drills. My fault for forgetting to implement my own workaround.

I know that the problem has been highlighted to Stepcraft and that a possible solution has been proposed - a pity that this hasn't been actioned yet. A case of "digitum extractum" !!!

I've had ro resort to manually starting the spindle whilst the Gcode is halted for a tool change - this always seems to work (so far), and the manually set speed is corrected to the Gcode programmed value as soon as the cycle is restarted.

Same problem for me with the HF 500. Stepcraft doesn't care. Or at least they don't comment on an solution.

I've done an extended test on my workaround and found that I just need to press the "Engine ON" button (to quote Stepcraft) to get apparrently normal start at M03 command. No need to actually set the spindle turning.

Went through around 100 tries at MO3 startup with this fix - with no failures. So it seems there needs to be a longer delay between whatever happens when M03 is asserted and the start of the spindle accel process. There is certainly no problem in the UCCNC side of things.

There's probably an MCU controlling the whole process, so a proper fix would probably require a software modification. Maybe that's what is stalling Stepcraft (sorry for the pun!). Nevertheless it needs fixing !!!!

Are there just the three or four of us suffering with this - if not please make it known on the forum. Maybe, just maybe, our collective moan will then be taken seriously.

Stepcraft - COMMMUNICATE!!

I have the Stepcraft-2 840 make anything package and have similar problems. The spindle shuts off and stops moving randomly while traveling along the X axis. It only happens traveling to the positive along the X axis. It's been 2 months or better and support has not given me a fix for the problem. I am using the HF500 spindle. They said it was a software issue as there was an added apostrophe in the g-code view screen in UCCNC. I resolved that issue by doing a complete factory reset of the computer and deleting everything not needed to run the machine. However, that did not solve the problem of the spindle shutting down. So far the US support has been virtually useless.