F-Error problems
23 Mar 2020 12:52 - 23 Mar 2020 15:43 #161205
by Clive S
Here are the ping times the highest is 0.35 ... Is the histogram OK as I see others without all the crap on the sides
.
IRQ coalescing disabled?.... Not sure how to do that
.
..
This is the f-error plot as it started a probe
Replied by Clive S on topic F-Error problems
.OK that has a Intel MAC, do you have IRQ coalescing disabled? (as mentioned in the hm2_eth man page)
A good check on this is pinging the Ethernet FPGA card and checking the ping times
Here are the ping times the highest is 0.35 ... Is the histogram OK as I see others without all the crap on the sides
.
IRQ coalescing disabled?.... Not sure how to do that
.
.
[AXIS_Z]
MIN_LIMIT = -82
MAX_LIMIT = -4
]# set to double the value in the corresponding joint
MAX_VELOCITY = 120
# set to double the value in the corresponding joint
MAX_ACCELERATION = 1500
# shares the above two equally between the joint and the offset
OFFSET_AV_RATIO = 0.5
[JOINT_1]
AXIS = Z
MIN_LIMIT = -82
MAX_LIMIT = -4
MAX_VELOCITY = 60
MAX_ACCELERATION = 750
TYPE = LINEAR
SCALE = -800
STEPGEN_MAX_VEL = 150
STEPGEN_MAX_ACC = 1875
FERROR = 1
MIN_FERROR = 0.25
DIRSETUP = 4000
DIRHOLD = 4000
STEPLEN = 4000
STEPSPACE = 4000
DEADBAND = 0
P = 1000
I = 0
D = 0
FF0 = 0
FF1 = 1
FF2 = 0.00015
BIAS = 0
MAX_OUTPUT = 0
MAX_ERROR = 0This is the f-error plot as it started a probe
Attachments:
Please Log in or Create an account to join the conversation.
23 Mar 2020 13:26 #161211
by tommylight
We had snow yesterday, my brain must have frozen solid !
Replied by tommylight on topic F-Error problems
I thought they were for the same joint, i will correct that now.
R: Tom
MAX_VELOCITY = 60 <<<<delete this
MAX_ACCELERATION = 900 <<<<delete this
Tom. can you please explain why I have to delete the above. I think that has to bee in a PlasmaC config
We had snow yesterday, my brain must have frozen solid !
Please Log in or Create an account to join the conversation.
23 Mar 2020 21:46 #161285
by Clive S
Peter. Have I done this correctly. As it has not effected the ping times.. Yes I rebooted
Replied by Clive S on topic F-Error problems
The ping times are _really_ bad
You can likely improve these by turning off the drivers IRQ coalescing
This is described in the hm2_eth manual page
man hm2_eth
The plot errors are not significant (+- 350 nM or +- 0.35 u)
Peter. Have I done this correctly. As it has not effected the ping times.. Yes I rebooted
Attachments:
Please Log in or Create an account to join the conversation.
23 Mar 2020 22:33 - 23 Mar 2020 22:53 #161290
by Clive S
Dyslexia. New screen shot. no difference. But I did not try sudo ethtool -C eno1 rx-usecs 0 as I have just seen it.
.
Edit: I have tried ethtool had to install it first. then got this:-
cnc@Brix:/etc/network/interfaces.d$ sudo ethtool -C eno1 rx-usecs 0
Cannot get device coalesce settings: No such device
.
Am I editing the correct file?
Replied by Clive S on topic F-Error problems
You have (AFAICS) an error in the interfaces file
( users should be usecs )
You should be able to try this live (no reboot) with this command:
sudo ethtool -C eno1 rx-usecs 0
but this will be lost when you reboot
(hence he need for editing the interfaces file)
Note, If this command fails the embedded one in the
interfaces file cannot work either so its always good to
try the immediate command
Dyslexia. New screen shot. no difference. But I did not try sudo ethtool -C eno1 rx-usecs 0 as I have just seen it.
.
Edit: I have tried ethtool had to install it first. then got this:-
cnc@Brix:/etc/network/interfaces.d$ sudo ethtool -C eno1 rx-usecs 0
Cannot get device coalesce settings: No such device
.
Am I editing the correct file?
Attachments:
Please Log in or Create an account to join the conversation.
Time to create page: 0.152 seconds