5i24 substitute using ethernet?
- Nitram
- Topic Author
- Offline
- Elite Member
Less
More
- Posts: 210
- Thank you received: 15
10 Oct 2019 09:21 - 10 Oct 2019 09:37 #147621
by Nitram
Replied by Nitram on topic 5i24 substitute using ethernet?
Thanks for your help on this Peter.
I will have to take a pause on further testing just for now, as I need to install linear scales. Once that is done I hope to come back to this and keep testing to try to pin down this difference between the boards. For now, whilst installing the scales I'll stick with the 5i24.
Just to be clear, the reason I keep referring to 30 rpm, is that this is the RPM that the tool orient cycle takes place at, as a pre-cursor to a tool change. Much faster than this and the orient lock does not engage correctly to lock the spindle for the tool change.
Both boards will run the spindle at 30 RPM, it is just that the 5i24 will start it spinning up immediately whereas with the 7i80 it takes approx 4 seconds to start spinning up. Hence as I don't know if there is any correlation, the query on the differing board frequencies and PWM frequencies earlier.
It may be that I need to tune the spindle PID or adjust the scale to suit. I haven't tested this yet though, but it seems counter intuitive if everything else in the mill config is precisely the same.
One thing which you may be able to enlighten me on is that input voltage to the electronic potentiometer is 0 to+12 volts, whereas the spindle VFD command input threshold is 0 to +10.
From a scale perspective this would be 10/12 or .8333 and given the voltage testing done above which confirms 1000 rpm per volt (as well as max spindle rpm being 10,000 rpm which corresponds to the max VFD command voltage of 10v) this looks to be about right. If you have a look at the scale setting below it is reasonably close (in numerical terms) at 8800 but at this figure is several orders of magnitude off. I have been wondering about the implications of this and whether there is a fundamental PID/scale setting which is confounding the issue...
I do hope that it might be OK to revisit this with you Peter, when I get the chance to get back to this and progress the migration of control of this machine from 5i24 to 7i80.
Again, thank you!
From the ini file:
#********************
# Spindle
#********************
[SPINDLE_9]
P = 5.07
I = 22.05
D = 0.02
FF0 = 22.35
FF1 = 1.0
FF2 = 0.0
BIAS = 0.0
DEADBAND = 0.00001
#MAX_OUTPUT = 8.0
ENCODER_SCALE = 2000
OUTPUT_SCALE = 8800.0
OUTPUT_MIN_LIMIT = -10000
OUTPUT_MAX_LIMIT = 10000
I will have to take a pause on further testing just for now, as I need to install linear scales. Once that is done I hope to come back to this and keep testing to try to pin down this difference between the boards. For now, whilst installing the scales I'll stick with the 5i24.
Just to be clear, the reason I keep referring to 30 rpm, is that this is the RPM that the tool orient cycle takes place at, as a pre-cursor to a tool change. Much faster than this and the orient lock does not engage correctly to lock the spindle for the tool change.
Both boards will run the spindle at 30 RPM, it is just that the 5i24 will start it spinning up immediately whereas with the 7i80 it takes approx 4 seconds to start spinning up. Hence as I don't know if there is any correlation, the query on the differing board frequencies and PWM frequencies earlier.
It may be that I need to tune the spindle PID or adjust the scale to suit. I haven't tested this yet though, but it seems counter intuitive if everything else in the mill config is precisely the same.
One thing which you may be able to enlighten me on is that input voltage to the electronic potentiometer is 0 to+12 volts, whereas the spindle VFD command input threshold is 0 to +10.
From a scale perspective this would be 10/12 or .8333 and given the voltage testing done above which confirms 1000 rpm per volt (as well as max spindle rpm being 10,000 rpm which corresponds to the max VFD command voltage of 10v) this looks to be about right. If you have a look at the scale setting below it is reasonably close (in numerical terms) at 8800 but at this figure is several orders of magnitude off. I have been wondering about the implications of this and whether there is a fundamental PID/scale setting which is confounding the issue...
I do hope that it might be OK to revisit this with you Peter, when I get the chance to get back to this and progress the migration of control of this machine from 5i24 to 7i80.
Again, thank you!
From the ini file:
#********************
# Spindle
#********************
[SPINDLE_9]
P = 5.07
I = 22.05
D = 0.02
FF0 = 22.35
FF1 = 1.0
FF2 = 0.0
BIAS = 0.0
DEADBAND = 0.00001
#MAX_OUTPUT = 8.0
ENCODER_SCALE = 2000
OUTPUT_SCALE = 8800.0
OUTPUT_MIN_LIMIT = -10000
OUTPUT_MAX_LIMIT = 10000
Last edit: 10 Oct 2019 09:37 by Nitram.
Please Log in or Create an account to join the conversation.
- PCW
- Away
- Moderator
Less
More
- Posts: 17868
- Thank you received: 4776
10 Oct 2019 10:46 #147630
by PCW
Replied by PCW on topic 5i24 substitute using ethernet?
The OUTPUT_SCALE should be set to be the expected RPM at 12V
Have you tried using a larger I term? I suspect your current I term will take several seconds to correct any offset errors.
Have you tried using a larger I term? I suspect your current I term will take several seconds to correct any offset errors.
Please Log in or Create an account to join the conversation.
- Nitram
- Topic Author
- Offline
- Elite Member
Less
More
- Posts: 210
- Thank you received: 15
10 Oct 2019 11:01 #147631
by Nitram
Replied by Nitram on topic 5i24 substitute using ethernet?
I'll will go ahead and test that Peter.
Once I get these scales in I'll revisit the testing here.
In the interim, thanks again!
Once I get these scales in I'll revisit the testing here.
In the interim, thanks again!
Please Log in or Create an account to join the conversation.
Moderators: PCW, jmelson
Time to create page: 0.052 seconds