[billsnogo]

I have an MDI with SEM servos that I have retrofited for mach3. When I run a test program with just the axis jogging, it runs well. Actually, when I run the test with the axis and spindle it runs fine, but today for a first time I started cutting some plastic, and the servos stuttered. I am running this with a Hitachi VFD, and am wondering if my 20amp service is maybe the problem now that there is actually a load? I need ideas guys? Anyone know what could cause this problem?

[billsnogo]

no ideas?

[davy182]

Servo Tuning is probably needed.

- - - Updated - - -

Have you checked how many amps its pulling.

[billsnogo]

I have not. I do not have a clamp on meter unfortunatly. Could be servo tuning as I don't truelly understand what I am seeing when using the tuning software. thanks

[billsnogo]

Anyone tune the dugong DG3S-16035 drivers with a sem servo? Put the machine up for sale locally, don't know why it is doing this, and kind of tired of troubleshooting. Still will keep it if I can get the bugger running right.

[mactec54]

billsnogo

How well did it run, without the spindle running, did you run the same program without the spindle running, It sounds like you have a noise problem, some photos of your VFD wiring Etc would help to see if it is noise or not

[billsnogo]

Without the spindle running it was fine. It also was fine when the spindle and axis were running, UNTIL I started to cut plastic. That did it in. I will need to tomorrow take pics how I have my vfd wired in and try and run a program again without cutting plastic, and with again to see if it changes again. Just pretty frustrated as I am not electrically inclined, but am so close to being "done" if there is such a thing.

[Iron-Man]

I have an MDI with SEM servos that I have retrofited for mach3. When I run a test program with just the axis jogging, it runs well. Actually, when I run the test with the axis and spindle it runs fine, but today for a first time I started cutting some plastic, and the servos stuttered. I am running this with a Hitachi VFD, and am wondering if my 20amp service is maybe the problem now that there is actually a load? I need ideas guys? Anyone know what could cause this problem?

Billsnogo can post video and pictures of your set-up? Please describe what you mean by "stuttered" These are step/direction drives? Please check and do the following:

********** Use a note book to record all test values for the P I D ************
*********** Write down your current P I D values.
********** Use a free motor, not connected to your mill **********************
********** As a additional note describe the response of the servo, i.e. "Stiffness of shaft at 0 speed, etc ***********
********** When properly tuned, the servo will hold its position at 0 speed ****************************
Try setting the P value first. Say P 25, I 5, D .023. Try low values at first. Adjust P and note response of system.

1. Encoder cables should be shielded and separated from any high voltage.
2. The SEM servos will need to be tuned (P I D) to your Dugong DG3S-16035 servo drives.
3. In Mach, you will need to set step width, acceleration etc for final tuning.
4. Now that you have tuned the motors, you will need to make a final tuning for machine inertia etc.

This is over simplified but will get you going in the right direction.

Hope this helps,


Iron-Man

[awerby]

As well as what Ironman said, you should make sure there's a braided shield over the spindle leads, grounded at the VFD end, since they will spit out a lot of Radio Frequency Interference otherwise.

[billsnogo]

Thanks guys. I will try to get a video this weekend. The studder is it will move and ever so brefly stop and start again. It happens so fast that you don't notice unless you are looking for it, or you can feel it in the table the sudden stop/start again.

Oh ,and what are PID values? Yes, I am VERY green. Sorry......

[Iron-Man]

Thanks guys. I will try to get a video this weekend. The studder is it will move and ever so brefly stop and start again. It happens so fast that you don't notice unless you are looking for it, or you can feel it in the table the sudden stop/start again.

Oh ,and what are PID values? Yes, I am VERY green. Sorry......

Your motors are not tuned. You will need to download the software; CNCdrive - motion controls and follow the directions. PID info; http://www.pmdcorp.com/downloads/Tuning_Servomotors.pdf

Do not give up. When you finally get them tuned, you will be able to tune other motors and systems with ease.


Hope this helps,

Iron-Man

[billsnogo]

Okay, now I know what you are talking about. I used the software when I first hooked up the servos, and burned one up and let it oscillate too long and burned it up. Since then I am hesitant to mess with them, even more so since I don't really know what the heck all those mean, and when I have achieved a correct tune......
The guy who makes the drives was very nice and gave me good email support and got it where there was good holding at a standstill, and little to no oscillating (still does occasionally do it when I make small moves).
I will post what values I have and still try and get a video going, been busy, so may be a few days.
thanks again, I need all the help I can get

[Bob La Londe]

What type of job were you cutting in plastic? How did you generate the code. For example a high resolution 3D in Exact Stop mode will look like your machine is having a seizure. Even constant velocity with a very small rounding factor can do that. You can get similar results with 2D operations if you have lots of short line segments in your polylines.

Anything that has short point to point moves will accelerate away from the point and decelerate to the point rather violently if the machine has a high acceleration rate set.

Air cut the actual code you were cutting on the plastic. Not test code. If its the same result it has nothing to do with a cutting load, and everything to do with your code and Mach3 setup.

Sorry, but Mn is little to far for me to just pop over and take a look... or put it on my truck.

P.S. I am curious to see how your situation works out. I am considering the DuGong drives myself.

[billsnogo]

Okay, things are busy now and have little time to touch the bridgeport, but went out and tried it again. Now it stutters when not cutting. It runs smooth without the spindle turning, but as soon as the spindle is on, stutter. So, looks like voltage interferance? My servos are using the same wires that it left the factory with, but the control wires for the VFD are shielded cat 6 cable. Not sure how to go about finding where this voltage is coming from, and best options. I guess time for some searching/reading. thanks again so far guys

[Bob La Londe]

Sounds like VFD noise. VFD and motor and control cable shield should be grounded at the main ground bus inside the machine and NOWHERE ELSE. Same with all other shields.

[mactec54]

Sounds like VFD noise. VFD and motor and control cable shield should be grounded at the main ground bus inside the machine and NOWHERE ELSE. Same with all other shields.

This would be a disaster if someone was to ground their Shields, in the way you have said

[mactec54]

Okay, things are busy now and have little time to touch the bridgeport, but went out and tried it again. Now it stutters when not cutting. It runs smooth without the spindle turning, but as soon as the spindle is on, stutter. So, looks like voltage interferance? My servos are using the same wires that it left the factory with, but the control wires for the VFD are shielded cat 6 cable. Not sure how to go about finding where this voltage is coming from, and best options. I guess time for some searching/reading. thanks again so far guys

Your VFD main Power cables need to be Shielded, & the Shields need to be Grounded correctly, that is why I asked for some photos was to see how you have everything wired

[billsnogo]

Your VFD main Power cables need to be Shielded, & the Shields need to be Grounded correctly, that is why I asked for some photos was to see how you have everything wired

I will get pics this weekend! I need all the help I can get, and pics are worth 1,000 words

[Bob La Londe]

.. and yet it works for all kinds of shields. Sound systems, control systems, etc. All the shields are tied back to a single central ground. Perhaps I didn't express it well, but its the norm to eliminate ground loop interference and shield from other noise. How exactly is it disastrous to star ground your shields?

Note: there does need to be isolation between the control computer and the control. This is done by optical isolation so there is (theoretically) no electrical potential between the two. The problem is that unless you place your machine on rubber pads and don't ground it there is always a common ground (sometimes quite weak). And yes I am aware of some devices that say they need to be grounded at the signal source and the device. You are not actually grounding at the device, but tying the device to ground.

[Al_The_Man]

At the risk of sounding like a broken record, the control, if it is a PC, its P.S. is earth grounded and so is the VFD through the grounded neutral, so there is no galvanic isolation between PC and spindle motor, in spite of opto isolation etc.
Al.