Cincinnati Milacron Arrow 750 Retrofit

More
13 Apr 2021 12:57 #205788 by stegrg
Yes, I will be commutating based on the encoder (in absolute) and it is a brushless permanent magnet motor.

I didn't realized that "bldc" hal component can be set with offsets between encoder index and motor zero. I will look into that further.

Do I need to establish motor zero by applying +ve to servo U terminal and -ve to servo V & W or does bldc hal do it another way? (I don't get consistent motor zero when I do it this way)

Please Log in or Create an account to join the conversation.

More
13 Apr 2021 13:24 #205791 by andypugh
bldc can be set up to initially use the +U -VW orientation, and then switch to a more accurate orientation based on an offset from the index once it sees the index.

To determine the offset, one approach is to find the offset where the motor runs at the same speed for the same input current in both directions.

Please Log in or Create an account to join the conversation.

More
13 Apr 2021 13:46 #205793 by stegrg
Running the motor at the same speed for the same input current in both directions would hold the servo stationary at motor zero locations?

Please Log in or Create an account to join the conversation.

More
13 Apr 2021 14:01 #205795 by andypugh

Running the motor at the same speed for the same input current in both directions would hold the servo stationary at motor zero locations?


No, it is a way to check that the response is symmetrical, which is a hint that the offset is correct.

Please Log in or Create an account to join the conversation.

More
13 Apr 2021 14:45 #205801 by stegrg
Oh, I see... Finding the correct offset is a trial and error process where you electrically spin the motor back and fourth and adjust the offset until you get the same response in both directions.

So to initially get the servo spinning using 8i20 cards, I could simply estimate the motor zero using +U, -VW and use the estimated offsets in hal "bldc" to get things started, correct?

Please Log in or Create an account to join the conversation.

More
13 Apr 2021 15:05 #205804 by andypugh
Yes.
Bear i mind that a hall-sensor-commutated motor will regularly be 15 electrical degrees out of phase as there are only 6 steps to the cycle. So it isn't _all_ that critical.

Please Log in or Create an account to join the conversation.

More
13 Apr 2021 15:42 #205806 by stegrg
Do you mean during initialization an accurate motor zero it isn't that critical, or after the index is read, or both?

Please Log in or Create an account to join the conversation.

More
13 Apr 2021 21:59 #205830 by andypugh

Do you mean during initialization an accurate motor zero it isn't that critical, or after the index is read, or both?


Both, really. It's nice if it is spot-on, But the lead-angle of the two fields is typically 90 degrees, so +/- 5 degrees on that isn't a massive problem.

Please Log in or Create an account to join the conversation.

More
14 Apr 2021 09:22 #205866 by stegrg
Ok, thanks for the information!

I'm going to piece together a Hal file and try to get this servo spinning.

Please Log in or Create an account to join the conversation.

More
20 Apr 2021 18:01 #206551 by stegrg
I managed to get the servo spinning using the 8i20 card but suddenly the servo stopped and I can't seem to get it going again. I have a red fault illuminated on the 8i20 card when the power supply is connected. I checked the fault bits in Hal Configurations and the fault.overcurrent bit is high. If I unplug the power supply and restart Linuxcnc, the fault.overcurrent bit remain low until I plug in the power supply. It then suddenly faults again. I tried unplugging the servo from the 8i20 card and the same thing happens so I don't think the servo is causing the overcurrent fault. Does anybody have any ideas or experience with this fault?

Please Log in or Create an account to join the conversation.

Time to create page: 0.083 seconds
Powered by Kunena Forum