Mesa 7i43 / 7i49 / 7i37 FIrmware Needed

More
21 Jun 2013 07:37 - 21 Jun 2013 07:37 #35859 by PCW
Can you do a HAL watch/halmeter on the resolver error and position pins?
if these are not right you are not going to get any further
Last edit: 21 Jun 2013 07:37 by PCW.

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

More
21 Jun 2013 08:15 #35860 by jsflanagan
From axis I opened the HALmeter results as follows:

hm2_7i43.0.resolver.error = FALSE
.position = -9.313226e-07 doesn't change when I move the axis
.velocity is all over the place +/-
.count = -1

-Scott

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

More
21 Jun 2013 09:08 #35861 by PCW
What is the resolver scale? or better still check the resolver angle when the shaft is rotated. If this is not right its likely a bitfile problem. I can make and check a new bitfile but it will be Tuesday before I can get to it

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

More
21 Jun 2013 09:13 #35862 by jsflanagan
I checked the resolver angle and it stayed at 1. Sounds like I need a new bitfile. I'll look forward to Tuesday. Thanks,

-Scott

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

More
21 Jun 2013 09:28 #35863 by jsflanagan
Actually I just read about compiling a bit file myself. Do I need to do anything special to fix this issue or just try to compile one by putting the correct name in the FIRMWARE.TXT when I compile? Something tells me it can't be that easy.

-Scott

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

More
21 Jun 2013 10:42 #35864 by PCW
You would have to use mesa's source to make a working resolver interface
and that source is not currently build-able with linuxcnc's scripts.
Its probably a lot easier to wait till Tuesday

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

More
21 Jun 2013 13:58 #35871 by cncbasher
the only line i can see you have differently to mine is
net z-pos-fb <= hm2_7i43.0.resolver.01.position => pid.1.feedback
i.e linking the feedback ...

building the fpga source is not for the faint hearted , although i'm not ruling out the bit file , the 7i43-4 version i posted is from a working system

i'll check the latest resolver build into my scripts and see how that builds

unfortunatly , i dont have a resolver to hand to test localy .

iv'e posted here the working files just in case you can spot a typo or something similar
i'm wondering if you have too much voltage swing across the resolver and it's swamping the input to the mesa card ..

i'll bow to PCW having the answer on this one ..

i notice your showing a resolver scale of 4000 , this is to me a bit on the low side , as you can see in my files their a lot higher
have you details on the resolvers you have
Attachments:

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

More
21 Jun 2013 19:06 #35879 by jsflanagan
Yes I didn't manually add that feedback line onto the Z axis, as I've been doing all of my testing on the X axis and pncconfig doesn't put that feedback in for some reason. But I do have that linked on the X axis. Also, I haven't set the resolver scale, just using defaults from pncconf and I was going to worry about that later.

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

More
12 Jul 2013 22:25 #36603 by jsflanagan
Wanted to check in and see if the newly compiled bit file was available? I'm still stuck on not being able to read resolvers right now. Thanks,

-Scott

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

More
13 Jul 2013 01:00 #36608 by PCW
Sorry I completely forgot about this
I'll try and get to this today

note: normal resolver scales tend to be very low (say 0.20000 for a 5 TPI ballscrew)

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

Moderators: PCWjmelson
Time to create page: 0.113 seconds
Powered by Kunena Forum