NEMA 23, weird velocity results.. need help

More
19 Nov 2013 18:34 #40973 by fannlow
Am thinking of changing to rack pinion...
If it moves too slow..

Power supply rated at 36v, which is lower than the 48v u mentioned.

Any references for the motors nema you guys are using and if I am getting the right speed for the motor size that u have?

I seems to run faster on the mach3 program if I recall correctly?

Hmm, or can I import that into linuxCnC?
Didnt really like windows as its too costly for my small production and each machine will need a license.

In my country tht will be an additional costs of about 2000 my money.
The material cost is only 1400 for my whole frame.. hahaha...
Motors and boards about 1000.

So thts my point.. and spending that 2000 doesnt realky justify that 5 or 10mm/s speed increased...

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

More
19 Nov 2013 18:46 #40974 by cncbasher
if it is running slower then check your tuning , it should run just as easily , it's a question of tuning the results
but the first question is to actualy get it all running properly .and remove any mechanical binding or problems first , once it is running correctly then and only then adjust the max velocity and acceleraton for better running , changing too many values at once , will have you running in circles .

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

More
19 Nov 2013 19:52 #40975 by fannlow
Ok tried the settings recommended.

Steps/rev 200
uStep 4
Pulley 1:1
Pitch mm/rev 1.75
Max vel 8mm/s
Max acc 20mm/s2

This is the final result without load.
Is this suppose to be the result I should be getting for a Nema23 425ozin?

I have attached a screenshot.
What is maxstep rate for 2nd screenshot?
My max pulse rate now is 3657.1 Hz

Thanks
Attachments:

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

More
19 Nov 2013 23:49 #40978 by cncbasher
i notice you seem to have a high base period , so check your period jitter and perhaps ammend
for example i have lower than 5000 base period , so this allows a max step rate of up to 50,000

take some time and play with the values , you may improve them drasticly

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

More
20 Nov 2013 00:02 #40979 by fannlow
CncBasher:
did you meant from the 2nd screenshot where the max base jitters are (33858)?

I have no idea, how to reduce this jitters... :(

i read all the rest of the docs in wiki, but its a HP to begin with, running 2gb DDR2, 160GB SATA, GeForce gCard.

i have also disabled hyperThreading on BIOS, and know of nothing else at this stage to further reduce this..
I got the 33858 after having the machine run itself over 23hours (using if I remember correctly "cell" screensaver).

any recommendations?

what PC are you using? do you have a spec, i can perhaps source my local store for something similar?

thanks

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

More
20 Nov 2013 00:12 #40980 by ArcEye

i have also disabled hyperThreading on BIOS, and know of nothing else at this stage to further reduce this..
I got the 33858 after having the machine run itself over 23hours


It is not likely that you have 33K max jitter without spikes

Read this FAQ and do the realtime latency test from a terminal and you should get some idea when the spikes are occurring
www.linuxcnc.org/index.php/english/forum...-the-latency-problem

If the spikes are regular see this FAQ
www.linuxcnc.org/index.php/english/forum...me-latency-solutions

Then do lspci -vv in a terminal and check the chipset number and revision against the list of those requiring the smi fix

At the same time check what video driver is being used by the video card

If it is one of the non-free ones, it could also be another source of problems, try removing the packages and substituting the open source driver

regards

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

More
20 Nov 2013 00:30 #40981 by fannlow
Hi ArcEye,
I didn't quite get the "spikes" that you are referring to...

will run a terminal check tomorrow, and update.

The video card driver... I didn't install it from the CD-ROM that came with it.
I bought the GeForce from a local store and just plug it in...
Linux detected it and used it right away..

so I ain't quite sure what it is now or how I can know where it is, since I'm not too familiar with Linux.. :dry:

any help is greatly appreciated :)

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

More
20 Nov 2013 00:51 - 20 Nov 2013 17:06 #40982 by ArcEye

I didn't quite get the "spikes" that you are referring to...
will run a terminal check tomorrow, and update.


The dialog based latency test just shows the highest jitter figures in a cumulative total.
So if you have 1 spike of 33k and 23 hours of 5K, it will still show 33K

The video card driver... I didn't install it from the CD-ROM that came with it.
I bought the GeForce from a local store and just plug it in...
Linux detected it and used it right away..


If you have selected use non-free drivers during the installation, it will use the proprietary drivers automatically.

The print from lspci will show what drivers are in use for the video card

If you find that you get spikes when opening openGL programs, or moving windows around etc etc ie video based
the proprietary drivers can be removed using synaptic and the alternative open-source ones will be substituted instead

regards
Last edit: 20 Nov 2013 17:06 by ArcEye.
The following user(s) said Thank You: fannlow

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

More
20 Nov 2013 07:28 #40986 by fannlow
Hi CncBasher:
I saw you said I should have a 48v power supply to get my drive stronger ya?

then I read on the power selection which is Ohms x Resistance = Voltage?
I went back to check on my driver and my motor.

Driver DM542A, takes 18VDC to 50VDC, while
motor Nema24HS9430, ohms 1.4, 3.0 max A = 4.2V

so to say, for 3 motors, I only need around 12.6V?

that's where I am lost for the 36V or 48V powersupply "ideal" that you mentioned...
I'm looking to understand where I have gone wrong, so I can properly re-do my math?

attaching the Nema23 Step Signal here for Step/Dir Setup and Hold, etc...
see if that helps? this is from the manufacturer...
www.longs-motor.com/productinfo/detail_12_25_116.aspx


another one is for the Driver DM542A, also from the manufacturer..

many many thanks :)
I'll run and check for hardware as well, and maybe can lower my base jitters
www.longs-motor.com/productinfo/detail_12_80_131.aspx
Attachments:

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

More
20 Nov 2013 16:23 #40998 by cncbasher
connect the stepper drivers as per sheet 5 of the dm542a pdf

forget about ohms law , this is all taken care of by the stepper drivers they work by limiting the current to the set value , also remember you have 2 coils per motor not 1

your motors are rated at 3A per phase , so setting your stepper driver to 3.31A or 4A peak should be fine , the easy way to check is to see if the body of the motor is extremley hot to touch
after about 1 minuite , and if so reduce the current settings , if the motor is running cool , then increase , think of the voltage is giving you speed , and the current the torque is a very simple way to remember
also it is expected stepper motors run hot as they are energised all the time , but not to the point of burning your skin ,
there are a number of posts on the forum where i have explained this in a bit more detail .

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

Time to create page: 0.196 seconds
Powered by Kunena Forum