And now for something completely rediculous

[eloid]

Great job on this build, I always like to look back a see what other have built, Any updates changes to this build since then.

How the ATC holding up any info where sourced it from?

[mactec54]

How the ATC holding up any info where sourced it from?

As for the ATC he made it all the mechanical parts, software integration himself, the Spindle Part was made in China, which was very well made

[eloid]

any info on the vender of the spindle or the brand name for the spindle?

[mactec54]

any info on the vender of the spindle or the brand name for the spindle?

Check to see if this company has it, Kirk, at Majesticmachinery.com

[eloid]

Has anyone come across the drawing or schematic showing how the atc is controlled for tool changes, with a pneumatic solenoid and values & relays etc?
And how it wired to a bob?

[hacker7]

I'm with eloid any info on the ATC workings macro's etc..........

[pupdogg]

The servos will indeed accelerate them faster, as well as achieve almost double that speed. Problem lies with the pid tuning. Unable to get stable results with the accell set that high. Jerk is definitely a problem with that silly gantry. Servo continuous torque is 3.18nm. The highest torque I am using right now is approx 150%. The servos are rated for 800% for 3 seconds. Continuous speed is 3000 rpm, max speed 600 rpm (this would be 4000 in/min on my machine. Also, gearboxes and pinion are rated for 3800 rpm continuous, 6000 rpm cyclic. Torque on the gearbox is limited to 100NM peak (estop), 50NM accel, 30NM nominal. Having said that, I think I need to derate the drives. Crap...

Although I did achieve stability at the higher accell rates, the it came at the cost of accuracy and stiffness of the system. As in, you could push the axis with about 20 lbs, and it would move .125".

If I was able to go back to position mode(stepper mode for the galil), the servo drives can certainly handle those accels. With the adaptive tuning, it would sound a lot better too, while maintaining the best accuracy available from my hardware. Currently, the pulsing functions on my galil are toast. Don't know why, but I confirmed it with galil. No time right now to send it back, so I will stick with this mode until a later date.

What kind of machine have you used that has better accel rates? I would love to see/read more about it. One machine tool builder told me he was achieving 150 in/sec squared, but after talking with other integrators, I do not believe it.

Sorry to dig this post up from the graveyard but I have a question for @spoiledbrat: From the pictures and your setup, it looks like your machine should've been plenty rigid to be able to achieve higher acceleration rates but reading some of your comments on motor tuning and the jerky behavior of the machine, it seemed a bit confusing as to why you could not achieve that without sacrificing quality. Did you ever figure anything out on this subject matter? Did it end-up being the tuning procedure for your Yaskawa servo-drive/motor setup or something else?

[mactec54]

Sorry to dig this post up from the graveyard but I have a question for @spoiledbrat: From the pictures and your setup, it looks like your machine should've been plenty rigid to be able to achieve higher acceleration rates but reading some of your comments on motor tuning and the jerky behavior of the machine, it seemed a bit confusing as to why you could not achieve that without sacrificing quality. Did you ever figure anything out on this subject matter? Did it end-up being the tuning procedure for your Yaskawa servo-drive/motor setup or something else?

Yes he did get it fixed he changed the control

[Spoiledbrat2]

Mactec,
I miss you. You are good people!

Pupdogg, I switched over to Industry Controls (a smoothstepper based board, from Kirk Brown. Once I did that, I was able to put the drives into Position mode. The Yaskawa drives have an autotuning feature, that calculates inertia of the system live, and sets appropriate gains/filters for best performance. I was able to get my machine up to 2000 ipm in the x and y.

Oddly enough, I am revisiting the tuning of the X axis. It seems to studder after small bump moves using the keyboard, and was showing evidence of under-damped response in some of may cuts (cut my first cabinet with it recently).

Problem is, autotuning will not work for a gantry, it is one servo at a time. I did learn, however, that Sigmawin+ software was updated, and it now has an "Autotune with reference" feature. When I tried to use it on my x axis (mach3 was jogging the gantry back and forth for me), the gantry motion started to sound very, very good. Crisp and solid, for lack of better description. However, before the autotuning sequence completes, the drive saturates (too much torque command), due to fighting the slower, not-so-well-tuned slaved axis.

I am about to attempt tuning again from scratch. I had my inertia set to 1000 (10:1), but after speaking with an engineer at my vendor, I was informed to start out at 0.325 (1045 lb gantry, divided by 32 ft/s sq, then reduced by square of gear ration (10:1 in my case).

If anyone on the zone has any insight into a solid process to go about tuning this beast (Yaskawa 1kw AC Servos), please do revive this thread and chime in. I know this machine is capable of much, much more performance than I had previously settled for.

Cheers!

Rob

[spoiledbrat]

back again, awaiting mod approval.

Sent from my SM-G892U using Tapatalk

[mactec54]

back again, awaiting mod approval.

Sent from my SM-G892U using Tapatalk

Good to hear are you grounded

[spoiledbrat]

So, Mactec, always good to hear from you.

I had been running the machine on an MCB-1, in position mode. since it is impossible to use autotuning using Sigmawin+ on a gantry, i basically found a setting that "worked" and left it be for a few years. now that i am cutting cabinets, i am retuning them now.

Turns out, i had the inertia ratio set to 10%, and i am now achieving far better resultd at 2.5%. I intend to do some experiments/bracketing to nail down the best inertia ratio (autotuning on single axis does this perfectly for Y and Z axis), and then proceed from there. I want my gantry to sing like my carriages.

Anyone having any advise for tuning sigma 5's in a gantry configuration, please chime in. I will post process and results.

Cheers!

Rob

Sent from my SM-G892U using Tapatalk