|
|
Post by vicpopo on Jan 24, 2017 22:11:43 GMT
new FW tested in motion with not ssi sensors and the result is extremly great .
|
|
|
|
Post by tronicgr on Jan 24, 2017 22:35:00 GMT
new FW tested in motion with not ssi sensors and the result is extremly great . Yes, while fixing the KLM switch function, I cleaned out the axis preparation section as well. Glad it works better now! Thanks for the feedback! |
|
|
|
Post by paulg100 on Jan 28, 2017 11:06:18 GMT
Progress update with SSI sensors, First the good news, all six sensors hooked up and after nearly a year of fighting with this... the jitter is gone!! FINALLY!, THANK YOU tronicgr Things to note with the new sensors: 1) very easy to mount 2) shaft is non contact so don't have to worry about lateral forces or wear and tear on shaft 3) Shaft dosent seem to be super fussy about alignment so again very easy to mount. 4) Robust picoblade connector so no more difficulties or worries about soldering to pot pins 5) seems very robust against emi especially when you consider the hot mess my cablings in at the moment for testing! All in all a big improvement over the old bournes pots But now on to the new problem  All motors are working and homing great with buttons from AMC, but the platform is not moving properly from BFF or from thanos stewart platform test program. All I get is a few random sudden movements. Was working OK apart from emi spikes with analog pots. Did the AMC code get tested when running from software when the SSI sensor option is selected Thanos? I cant think of anything else it could be other than the new AMC firmware? Thanks.    Files for mount and coupler here incase any use: www.thingiverse.com/thing:2068945Incase you wonder the little wrench tool on printer is a device for setting shaft spacer against pot bracket and coupler so magnet is in the correct position inside sensor. |
|
|
|
Post by tronicgr on Jan 28, 2017 14:26:21 GMT
Progress update with SSI sensors, First the good news, all six sensors hooked up and after nearly a year of fighting with this... the jitter is gone!! FINALLY!, THANK YOU tronicgr Things to note with the new sensors: 1) very easy to mount 2) shaft is non contact so don't have to worry about lateral forces or wear and tear on shaft 3) Shaft dosent seem to be super fussy about alignment so again very easy to mount. 4) Robust picoblade connector so no more difficulties or worries about soldering to pot pins 5) seems very robust against emi especially when you consider the hot mess my cablings in at the moment for testing! All in all a big improvement over the old bournes pots But now on to the new problem All motors are working and homing great with buttons from AMC, but the platform is not moving properly from BFF or from thanos stewart platform test program. All I get is a few random sudden movements. Was working OK apart from emi spikes with analog pots. Did the AMC code get tested when running from software when the SSI sensor option is selected Thanos? I cant think of anything else it could be other than the new AMC firmware? Thanks. Incase you wonder the little wrench tool on printer is a device for setting shaft spacer against pot bracket and coupler so magnet is in the correct position inside sensor. I haven't had the chance to try it on Ian's BFF software, i tried the xsim 6dof plugin with it. Can you try this firmware on it? www.dropbox.com/s/269d5arrxpfxe23/enc_6dof_AMC1280USB_v2.2_beta_block_plus_ssi_1024_180degrees_KLM_FIX.hex?dl=0Also increase the delay in Ian's BFF ini file to 4ms or 6ms, to stabilise the data stream. Thanks Thanos |
|
|
|
Post by paulg100 on Jan 28, 2017 15:39:49 GMT
OK I tried the firmware
Putting BFF aside for a moment I just tried the test program but still no luck, All I'm getting now is random sporadic movements that don't correspond to the test program.
Still works fine direct from AMC buttons.
Ive never had any issues on the software side before the new sensors... I checked the comport settings just incase there all fine.
Any ideas
Thanks.
|
|
|
|
Post by tronicgr on Jan 28, 2017 16:23:39 GMT
Can you write down a list of all the AMC1280USB LCD values you have?
You may have a setting that is different and causing it. I'll then try the same settings on my setup here to see if I can replicate it.
The only thing I can think of is some setting on the "Step" of getting back to home position that blocks the actual data. I use step 99%.
Let me know and I'll try it at once.
Thanks
Thanos
|
|
|
|
Post by tronicgr on Jan 28, 2017 16:40:03 GMT
Also can you post the STL files for the mounting of the sensors on the gearbox? This mounting solution looks great!
Thanks
Thanos
|
|
|
|
Post by paulg100 on Jan 28, 2017 16:45:30 GMT
Sure Here, Also added thingy link to previous post for mount above.  Only settings I changed compared to old sensors is lowered max motor speed, oversample from 2 to 1 and obviously selected 12 bit sensor. Thanks. |
|
|
|
Post by tronicgr on Jan 28, 2017 21:17:49 GMT
Sure Here, Also added thingy link to previous post for mount above. View AttachmentOnly settings I changed compared to old sensors is lowered max motor speed, oversample from 2 to 1 and obviously selected 12 bit sensor. Thanks. Fabian's utility does not show other options like the "motion data only" option added to handle communication issues with Ian's motion software sometimes: See where to set it here: youtu.be/KRU755Q1JugTry it and let me know. Thanks Thanos |
|
|
|
Post by paulg100 on Jan 28, 2017 22:58:47 GMT
OK that setting seems to have done the trick, motion in bff is starting to look good!
The "motion data only" setting seems to have disabled the motion in 6dof test program though, is that expected? Should other software like xsim and sim tools still work with that setting On?
Now to tidy up cables and ground sheild etc and hopefully I can start tuning and get some videos at long last!!
|
|
|
|
Post by tronicgr on Jan 28, 2017 23:47:57 GMT
OK that setting seems to have done the trick, motion in bff is starting to look good! The "motion data only" setting seems to have disabled the motion in 6dof test program though, is that expected? Should other software like xsim and sim tools still work with that setting On? Now to tidy up cables and ground sheild etc and hopefully I can start tuning and get some videos at long last!! Are you talking about the xsim 6dof plugin? motionsim.blogspot.com/2013/05/6dof-software-plugin-for-x-sim3-is.html?m=0I used this offline version to test the new functions and "motion data only" works fine on that! Actually to be more specific the offline version of it v3.4 of hexpod here: hexpod.xyz |
|
|
|
Post by paulg100 on Jan 29, 2017 10:41:52 GMT
Yes that's the plug in.
Thanks for the link to hexapods version, ill take a look at that once ive read how to set the parameters correctly.
Just wanted to confirm again the platform is working great now with BFF!! finally like it should, smooth and no jitter,
I can finally start focusing on the software and tuning side and start using the sim!
Thanks again for all your help and putting the work in to supporting the new sensors.
more updates soon.
|
|
|
|
Post by vicpopo on Jan 29, 2017 18:19:00 GMT
Hi paulg100,
Happy for you and now I will update my sim rig with the new ssi sensors .You gave me the confirmation that the result is much better ;-)
|
|
|
|
Post by hexpod on Jan 30, 2017 1:10:36 GMT
OK that setting seems to have done the trick, motion in bff is starting to look good! The "motion data only" setting seems to have disabled the motion in 6dof test program though, is that expected? Should other software like xsim and sim tools still work with that setting On? Now to tidy up cables and ground sheild etc and hopefully I can start tuning and get some videos at long last!! Are you talking about the xsim 6dof plugin? motionsim.blogspot.com/2013/05/6dof-software-plugin-for-x-sim3-is.html?m=0I used this offline version to test the new functions and "motion data only" works fine on that! Actually to be more specific the offline version of it v3.4 of hexpod here: hexpod.xyzThe new, probably licensed version is supposed to come out in the near future. Trying to make it bug free and more user friendly. You can expect dynamic center of gravity, corrected combined kinematics, unified mm units and couple of interesting new settings.
cheers
|
|
|
|
Post by paulg100 on Feb 12, 2017 22:38:40 GMT
Finally something resembling proper motion!
Was worried that the motion might not be fast or detailed enough but glad to say its looking really good so far and things are still smooth and dialed way back in BFF here.
|
|
