Welcome, Guest
  • Page:
  • 1
  • 2

TOPIC: Communication problem via RS232 over SMI and Step/Dir problem.

Re: Communication problem via RS232 over SMI and Step/Dir problem. 6 years 8 months ago #751

  • csearcy
  • csearcy's Avatar
  • OFFLINE
  • Moderator
  • Posts: 460
  • Thank you received: 77
  • Karma: 23
Step and direction signals are 5VDC. You should take every precaution you have available to reduce noise on these signals... as they can take in a very high frequency... noise and spikes included.
Things you can do...
-Separate the 5VDC wiring from anything that could induce noise.
-Add an EMI/RFI filter on the AC power that feeds the DC supplies used to power the SmartMotors... KB Electronics.. Corcom.. and others sell these for $30-50.
-Use a good Single Point Ground in your panel. Wire all chassis ground wires directly to SPG. Shields may or may not need to be wired to SPG depending on the system. ...Sometimes they are better tied to their own supply (which may be "floating" from everything else).
The administrator has disabled public write access.

Re: Communication problem via RS232 over SMI and Step/Dir problem. 6 years 6 months ago #777

  • Xavierov
  • Xavierov's Avatar
  • OFFLINE
  • Fresh Boarder
  • Posts: 7
  • Karma: 0
Hi,
I started to rewire my whole setup, it got better but I still have much to do here.
Added chokes, a clean star-ground, bought a second power-supply.

I tested the setup with some simple 50mm up and down moves of the axis with a 10kg load.
The problem here is that it looses or shoots over some encoder steps on every up and down move.
The SMI Interfaces shows no errors but I see the Motor1-com1 position changing slightly on every up and down move.
My question now is, did I miss a command that allows the program to be slightly out of position? Something like an error tolerance? (If not, it can only be my wiring)
My actual PID tuning is: P=700 I=1400 D=1300 L=1000 KS=1 KG=ON
Could this very suboptimal tuning be also a reason for the errors?

Thanks!
The administrator has disabled public write access.

Re: Communication problem via RS232 over SMI and Step/Dir problem. 6 years 6 months ago #778

  • csearcy
  • csearcy's Avatar
  • OFFLINE
  • Moderator
  • Posts: 460
  • Thank you received: 77
  • Karma: 23
What is the firmware revision of your motor? If it's 4.78 or higher... you have an option for TOB (Trajectory Overshoot Brakeing). If the revision is lower than 4.78... you may be able to solve the problem with tuning... and Accel/Decel values from your Step/Dir controller. The firmware revision should be on the motor sticker... with Mfg date... and full part number.
Please send me all information you can find about your motor... then I can offer better advice.
To establish a baseline... you can try removing all PID values... letting all values go to default.
The default values for your motor should be..
KP=42
KI=28
KD=550
You can verify those with RKP..RKI..RKD.. after removing them from your program.
-Chuck
Last Edit: 6 years 6 months ago by csearcy.
The administrator has disabled public write access.

Re: Communication problem via RS232 over SMI and Step/Dir problem. 6 years 6 months ago #779

  • Xavierov
  • Xavierov's Avatar
  • OFFLINE
  • Fresh Boarder
  • Posts: 7
  • Karma: 0
Sadly the version is 4.15b, pretty old.
Ser. E14478
Model: SM2320SQ
Mounted is a Parker Daedal Linear Actuator with 200mm travel and 5mm pitch. 10kg load.
30V power Supply
I made some more tests, the axis always wanders down about 0-10 encoder steps. Every few up/down motions. Changing the accelaration or speed only changes the error slightly. (And randomly)
This error would be no big deal, but the problem is that it accumulates.
The SMI chart shows no constant errors but changes the position slightly.
The zero positions in a test row were:
1
8
0
17
18
18
35
51
88
73
94

If this can't be a programming/firmware error it must still be a signaling error.. I just can't figure out how there can still be problems. (There are also stepper controllers on the equal ports)

Anyway thank you for your help! I'll continue trying to trace the error.

Edit:
I found out that the error also occurs when just moving without any hard accelaration
When moving 1cm = ~3 lost steps 2cm= ~ 14 10cm=~ 156
Last Edit: 6 years 6 months ago by Xavierov.
The administrator has disabled public write access.

Re: Communication problem via RS232 over SMI and Step/Dir problem. 6 years 6 months ago #780

  • Xavierov
  • Xavierov's Avatar
  • OFFLINE
  • Fresh Boarder
  • Posts: 7
  • Karma: 0
Ok,
I shut the rest of the machine down and connected it to a seperate parallel port without optical isolation etc,.
Voilá, zero error.

So its "just" and EMF problem.
Are there any circuits to protect the inputs from Noise?
I think one problem is that the Motor has an RS232 cable a step/dir cable and a signal cable for error and limit outs. All with their own GNDs.
I have no idea how to properly connect this
Are there any easy but good working circuits for protecting the inputs from EMF?
There is VFD in the system, I guess its the one producing the most noise.
Another problem might be that the Step/Dir "connectors" aren't shielded.
Last Edit: 6 years 6 months ago by Xavierov.
The administrator has disabled public write access.

Re: Communication problem via RS232 over SMI and Step/Dir problem. 6 years 6 months ago #782

  • csearcy
  • csearcy's Avatar
  • OFFLINE
  • Moderator
  • Posts: 460
  • Thank you received: 77
  • Karma: 23
The best protection from the VFD is to not be connected with it on the DC side of the Smartmotor power supply. Use a EMI/RFI filter on the AC side of the Smartmotor power supply. Float the DC side of the power supply (don't connect it to ground in your panel). Use twisted sheilded wire for connections to the Smartmotor Step and Direction, as well as I/O. The shields may (or may not) need to be tied to ground in the panel (sometimes it works best to tie them to DC common... if it's floated). Step and Direction do need to have a common wire to the Smartmotor GND for reference.
The administrator has disabled public write access.
  • Page:
  • 1
  • 2
Time to create page: 0.151 seconds
Powered by Kunena Forum