Hi,
I googled and found this:
It looks like you basically have 2 options.
#1 - Let the Amplifier close the servo loop and just send Step + Direction to the Amplifier
see page 3-31
#2 - Close the Position loop with KFlop+Kanalog
see page 3-32
Regards
Tom Kerekes
Dynomotion, Inc.
| Group: DynoMotion |
Message: 530 |
From: Jack |
Date: 7/27/2010 |
| Subject: Re: KFlop and Yaskawa motors |
Tom
Is it also possible to use speed control with the Yaskawa & KFlop?
Speed control seems to be the best choice for cnc applications.
Past discussion on this at the CNCZone.
Link:
http://www.cnczone.com/forums/showthread.php?t=39761&page=2
Jack
"It looks like you basically have 2 options."
Regards
Tom Kerekes
Dynomotion, Inc. |
|
| Group: DynoMotion |
Message: 532 |
From: Tom Kerekes |
Date: 7/27/2010 |
| Subject: Re: KFlop and Yaskawa motors |
Hi Jack, Yes speed (velocity) control would be the preferred option. You would need our KFlop+Kanalog board to output a +/- 10V analog signal to command the velocity to the drive. The drive would be configured to drive the motor (and load) at the speed it is commanded by the Vref input. You can test by applying a small voltage (ie 1.5V battery) to the Vref input. The motor should drive at 15% full speed. If you reverse the battery, the motor direction should reverse. The differential encoder signals from the drive would be connected to Kanalog so that KFlop can close the position feedback loop. With only simple Proportional Gain, KFlop will command a velocity proportional to the error (distance to target position). So basically the motor will be
commanded to go slower and slower as it gets closer and closer which results in an exponential curve.
Regards TK
| Group: DynoMotion |
Message: 536 |
From: Jack |
Date: 7/27/2010 |
| Subject: Re: KFlop and Yaskawa motors |
Tom
Thanks for the quick reply.
Seems like there is a lot of interest using the Yaskawa drives and motors with your products.
Jack
--- In DynoMotion@yahoogroups.com, Tom Kerekes <tk@...> wrote:
>
> Hi Jack,
>
> Yes speed (velocity) control would be the preferred option. You would need our
> KFlop+Kanalog board to output a +/- 10V analog signal to command the velocity to
> the drive. The drive would be configured to drive the motor (and load) at the
> speed it is commanded by the Vref input. You can test by applying a small
> voltage (ie 1.5V battery) to the Vref input. The motor should drive at 15% full
> speed. If you reverse the battery, the motor direction should reverse.
>
>
> The differential encoder signals from the drive would be connected to Kanalog so
> that KFlop can close the position feedback loop. With only simple Proportional
> Gain, KFlop will command a velocity proportional to the error (distance to
> target position). So basically the motor will be commanded to go slower and
> slower as it gets closer and closer which results in an exponential curve.
>
>
>
> Regards
> TK |
|
| Group: DynoMotion |
Message: 537 |
From: Jack |
Date: 7/27/2010 |
| Subject: Re: KFlop and Yaskawa motors |
|
TK
One more question.
Do you think this solution will handle Gcode that has thousands of short moves/line segments? Often this type of code can cause jerky movements with the cutter/spindle.
Jack
|
|
| Group: DynoMotion |
Message: 538 |
From: bouseman_73 |
Date: 7/27/2010 |
| Subject: Re: KFlop and Yaskawa motors |
|
Tom, thanks for your answer
But a have questions more...
1) Yaskawa servopack drive is using for control speed range +/- 2-10v (not 0-10v) . How can I setup KMotion for it?
2) And Servopack has A, B, C phase encoder output (for incremental), but KAnalog has A\B input signals only. What about C signal?
3) And what is the wiring for limit switch
thank you in advance
Slava
|
|
| Group: DynoMotion |
Message: 539 |
From: Tom Kerekes |
Date: 7/27/2010 |
| Subject: Re: KFlop and Yaskawa motors |
Hi Jack,
I think that has little to do with the drive or Step/Dir vs Analog control. That is more related to the G code interpreter and Trajectory Planner. I think KFlop does well compared to other systems. It can accept a minimum of 500 motion segments/second. A machine can't usually change direction that fast. I think our KMotionCNC handles small vectors better than Mach3.
TK
| Group: DynoMotion |
Message: 540 |
From: Tom Kerekes |
Date: 7/27/2010 |
| Subject: Re: KFlop and Yaskawa motors |
Hi Slava,
Regarding:
#1 - I think that means they can accept a range from -2V to +2V or a range from -10V to +10V, probably configurable in the Yaskawa settings. You could use the -2V to +2V range by limiting the Max Output on Kanalog. It would still work, but you would have a bit less resolution on the velocity command about +/- 400 DAC counts instead of +/- 2047 DAC counts. I would recommend setting the Yaskawa to use a +/- 10V range.
#2 - The C signal is a index pulse that occurs once per motor rev. It isn't usually needed. It can be helpful for homing very accurately, but you still need a home switch since the index happens every rev so without the switch you wouldn't know what rev you were at. Homing and such is performed in software by using a KFlop User Program written in C. This is very flexible and can handle many homing techniques but does require you to write or adapt an existing homing program. So for example your homing process may be something like: jog a medium speed until the home switch is detected, then keep moving at slow speed until the index pulse is detected, then stop. The C signal can be connected to any of the unused 16 A or B differential input pairs.
#3 - Limit switches are also configurable for any spare input. Kanalog provides 8 opto isolated inputs that can be directly connected to 5 to 24V signals.
Regards
TK
| Group: DynoMotion |
Message: 610 |
From: amir |
Date: 8/26/2010 |
| Subject: Re: KFlop and Yaskawa motors |
|
i have Panasonic Servo Drive model:ADKA100GVTAAA(100W) and ADKA400GVT(400W) .and all are analog. only speed command.
they are very old( about 20 years ) and made with Yaskawa company.
Their appearance is exactly like Yaskawa.
i run them with Dc voltage from 0~10 V,and all thing is ok.
Whether i can use Kflop and kanalog ?
Amir
|
|
| Group: DynoMotion |
Message: 612 |
From: Tom Kerekes |
Date: 8/26/2010 |
| Subject: Re: KFlop and Yaskawa motors |
Hi Amir,
That should work. You must have encoder signals as well, preferably differential signals
TK
| Group: DynoMotion |
Message: 613 |
From: amir piry |
Date: 8/26/2010 |
| Subject: Re: KFlop and Yaskawa motors |
thx Tom
please see my schematic and wiring:
http://i33.tinypic.com/adchls.jpg
i have a problem with this driver ,when i set "speed command" to 0V ,motor was Motionless but not locked , in practice i can turn shaft with hand.
are you can help me?
Amir
--- On Thu, 8/26/10, Tom Kerekes <tk@...> wrote:
From: Tom Kerekes <tk@...>
Subject: Re: [DynoMotion] Re: KFlop and Yaskawa motors
To: DynoMotion@yahoogroups.com
Date: Thursday, August 26, 2010, 8:00 PM
Hi Amir,
That should work. You must have encoder signals as well, preferably differential signals
TK
| Group: DynoMotion |
Message: 614 |
From: Tom Kerekes |
Date: 8/26/2010 |
| Subject: Re: KFlop and Yaskawa motors |
Amir,
Do your motors have tachometers or encoders connected to the drive?
Normally if you command 0V to the speed command input to an amplifier with tachometer feedback the drive will attempt to maintain at 0 speed. The motor will NOT be "locked" (servo'ed) to a position. Encoder Position feedback is needed to do this. With just tachometer feedback the motor should feel hard to turn like there is a lot of friction (viscus). The faster you try to turn the motor the harder it should be to turn.
I hope this helps
TK
| Group: DynoMotion |
Message: 615 |
From: amir piry |
Date: 8/26/2010 |
| Subject: Re: KFlop and Yaskawa motors |
thx, very good explain.
with regard to your comments, i think is tachometer feedback.
please see my motor and schematic :
http://i38.tinypic.com/xzk7d.jpg
http://i36.tinypic.com/svm35y.jpg
I'm disappointed for using them.
However,can i run them with kflop\kanalog ?
Amir
--- On Thu, 8/26/10, Tom Kerekes <tk@...> wrote:
From: Tom Kerekes <tk@...>
Subject: Re: [DynoMotion] Re: KFlop and Yaskawa motors
To: DynoMotion@yahoogroups.com
Date: Thursday, August 26, 2010, 11:52 PM
Amir,
Do your motors have tachometers or encoders connected to the drive?
Normally if you command 0V to the speed command input to an amplifier with tachometer feedback the drive will attempt to maintain at 0 speed. The motor will NOT be "locked" (servo'ed) to a position. Encoder Position feedback is needed to do this. With just tachometer feedback the motor should feel hard to turn like there is a lot of friction (viscus). The faster you try to turn the motor the harder it should be to turn.
I hope this helps
TK
| Group: DynoMotion |
Message: 616 |
From: Tom Kerekes |
Date: 8/26/2010 |
| Subject: Re: KFlop and Yaskawa motors |
Hi Amir,
Those appear to be 3 phase brushless motors with encoders. The schematic notes 1000ppr (4000 counts/rev). You will need to determine the encoder signals. I see there are 14 encoder wires. Probably 6 differential signal pairs plus power and ground. The 6 encoder signals are probably 3 signals used by the drive to do the motor commutation. The other 3 are probably A, B, and Index signals. You would need to find the A B differential signals (4 wires) and connect them to Kanalog (in parallel with the existing connections to the drive). If you have an oscilloscope it would be easy to determine. Otherwise trial and error would work. It should all be possible.
Regards
TK
| Group: DynoMotion |
Message: 619 |
From: amir piry |
Date: 8/27/2010 |
| Subject: Re: KFlop and Yaskawa motors |
Hi Tom
thx for support.
you say :"The motor will NOT be "locked" (servo'ed) to a position" and will attempt to maintain at 0 speed.
my question :
Whether in this type of motors ,If the pressure load axis was high. will not cause Rotate the shaft motor ?
I hope you understand what I mean.
Amir
--- On Thu,
8/26/10, Tom Kerekes <tk@...> wrote:
From: Tom Kerekes <tk@...>
Subject: Re: [DynoMotion] Re: KFlop and Yaskawa motors
To: DynoMotion@yahoogroups.com
Date: Thursday, August 26, 2010, 11:52 PM
Amir,
Do your motors have tachometers or encoders connected to the drive?
Normally if you command 0V to the speed command input to an amplifier with tachometer feedback the drive will attempt to maintain at 0 speed. The motor will NOT be "locked" (servo'ed) to a position. Encoder Position feedback is needed to do this. With just tachometer feedback the motor should feel hard to turn like there is a lot of friction (viscus). The faster you try to turn the motor the harder it should be to turn.
I hope this helps
TK
| Group: DynoMotion |
Message: 620 |
From: Tom Kerekes |
Date: 8/27/2010 |
| Subject: Re: KFlop and Yaskawa motors |
Hi Amir,
No it has nothing to do with the type of motor. All motors will work this way. They need encoder feedback such that the servo loop will increase the torque to whatever is necessary (within limits) to move to or hold the comanded position.
TK
| Group: DynoMotion |
Message: 625 |
From: amir piry |
Date: 8/31/2010 |
| Subject: Re: KFlop and Yaskawa motors |
Hi Tom
thx for very good support.
I will use kflop/kanalog for plasma machine(with mach3 software) .
There is no limit to it?
Can I use the THC?
I do research about smoothstepper .
With some restrictions, including THC and
backlash ...
Amir
|
|
|
| Group: DynoMotion |
Message: 626 |
From: Tom Kerekes |
Date: 8/31/2010 |
| Subject: Re: KFlop and Yaskawa motors |
Hi Amir,
Well I don't know what THCdoes in Mach3 (Torch Height Control ?) Is it documented somewhere? What do you need it to do exactly?
I think such things are possible. One user with a CO2 laser system developed a Z focus control. Basically the Z axis had two modes.
#1 was normal where the Z axis servoed to an absolute height based on encoder position.
#2 was a dual position loop where the Z axis servo was commanded to follow an analog LVDT (height sensor) so Z would track some relative height above the stock.
M codes were used to switch modes. I think it worked well.
After many many requests for backlash compensation we finally added it a couple months ago. To my knowledge no one has tried it :}
Regards
TK
| Group: DynoMotion |
Message: 627 |
From: amir piry |
Date: 8/31/2010 |
| Subject: Re: KFlop and Yaskawa motors |
Hi Tom
link about THC :http://www.centricut.com/New_Lessons/lessons_10.html
in fact, with sensor board (thc) we can measure voltage between nozzle and plate ,and then Z axis change to Maintain constant height.
thx for all things
Amir
--- On Tue, 8/31/10, Tom Kerekes <tk@...> wrote:
From: Tom Kerekes <tk@...>
Subject: Re: [DynoMotion] Re: KFlop and Yaskawa motors
To: DynoMotion@yahoogroups.com
Date: Tuesday, August 31, 2010, 9:46 PM
Hi Amir,
Well I don't know what THCdoes in Mach3 (Torch Height Control ?) Is it documented somewhere? What do you need it to do exactly?
I think such things are possible. One user with a CO2 laser system developed a Z focus control. Basically the Z axis had two modes.
#1 was normal where the Z axis servoed to an absolute height based on encoder position.
#2 was a dual position loop where the Z axis servo was commanded to follow an analog LVDT (height sensor) so Z would track some relative height above the stock.
M codes were used to switch modes. I think it worked well.
After many many requests for backlash compensation we finally added it a couple months ago. To my knowledge no one has tried it :}
Regards
TK
|
| | | | | |
|
| | | |
|
| | | |
|
| | | | | | | | | | | |
