[Elnino]

Hi all, i have some Vexta steppers and it seems that i'm just in need of a little more torque. Every now and then they seem to bind on the thread but it can't be by much so i'm wondering if i can run my 2.1A 'rated' motors a bit higer? say 2.5-3A?

I'm using Oatleyelectronics drivers with thier constant current devices as well so i can set whatever current i want with the abliity to go to about 3A each motor with some room to spare on my Power supply.

The motors do not even get warm at all as they are now. Is it safe to bump up the current a bit or am i going to fuse them??

[Markus53]

Its not a Good idea to Over-Current Vexta (stepper motors). It will cause the motors to run to Hot and this will shorten the lifespan of the motors and in the worse cause just burn um out. There are better and much more sensible ways to get (lots) more torque out of the motors. Cheapest solution is run um at slower speeds. Stepper motors loose torque the faster they go, also using a longer acceletaion ramp can get you better results. As you mention the 2.1 Amps i assume you are running them with a Unipolair steppermotor driver and a PK266 (or something like it) motor. By switching to Bipolair drivers you can get 30% more torque. Another step that will get you more torque is using a higher voltage with a Bipolair chopper driver. The coils in a stepper motor have to build up magnetic energy, by using higher voltages (with the same current setting) the buildup is faster allowing higher torque at the same rotation speed. Have a look at the Torque/rpm graphs on the Vexta website for your motor. You will see what i mean. A last important step is using micro stepping. Most of the new bipolair (and some unipolair) drivers can microstep your stepper motors. This means they can energize the coils in the motor partially, and thus forcing the motor to a assume a axial position between to physical steps. This gives a higher step resolution (up to a point) but also increases the torque at high RPM's.
I used to use my Vexta PK266 motors at half steps and when i switched to eighth steps (8 micro steps per physical full step) i got more than a 15% rise in top speed. With my Mills i have been able to get idle speeds of up to 4000mm/minute (157 inch/minute) using Vexta PK266-E2.0 motors and acme threads with a pitch of 4mm/rotation. I have used both TA8435H and A3977 based bipolair chopper drivers, with similar good results.

Hope this helps,

Greetz,

Markus

[altair]

Hi,
steppers are supposed to run very warm, almost hot, at their rated current.
I would say you just don't have the current you think you have in your motors. Check the settings on the drive.
Or maybe, they are just wired wrong, if they are 6-wire or 8-wire types.
Check the motor specs.

Guy

[pminmo]

elnino,
Couple of things. One, Guy is correct, if the motors aren't warm your proably not getting the current you think. Thus less torque. Two, where do you need more power, lower speed cuts or rapids? What power supply voltage are you using?

[Elnino]

Hmm, this is interesting. As i said, the motors are running cold - they do not even get warm after an hour of use.

At the moment i am using the oatleyelectronics drivers. They're not too well known but they do not have microstepping. I am also using thier constant current drivers which maintains the current in the coils at higher RPM.

I could get about 10x the RPM of the motors than at thier 'rated' specs with the CCD's. I'm maintaining 2.1A at up to 28v. (rated is 4.2v). When i built the machine i tested all the CCD's were set at 2.1A - it's possible they might need to be readjusted but they WERE at 2.1A

I'm only using 1.2mm pitch thread and i can maintain 1m/min. I have cleaned and cleaned the thread thinking that some metal shavings have got in there or something but i can't see anything that might even remotely cause it to bind. When it does, i can turn the machine off and manually turn the motor with NO feel of ANY resistance.

It even does it at 100mm/min but the difference being that it only loses 1 or 2 steps instead of a whole bunch. It is not because of the movement mech of the axis as i can not get anywhere near stopping the movement with my hand - it would take my hand OFF before it was likely to stop.

I also tried stopping the thread with pliers and it is almost impossible.

I'm possibly even putting this down to software and NOT the machine itself. I have been using Mach3 and i have heard of other reports from people of it 'losing steps'. I might have a shot at EMC2 and see if that has the same problem.

[kreutz]

I took a look at the oatley-electronics driver. It is an analog current controller and will behave as the linistepper without micro-stepping, so, the motors will work colder than with other current controlling drivers. Your problem could be related to mechanical binding. (I am supposing that your motors have the required torque to move and accelerate your load).

I don't think the problem is a MACH3 limitation unless your PC is too slow to keep at least a 25 Khz pulse rate. (since you are working in Full step "one coil at a time" mode).

Are you using the bipolar or the unipolar drive?

[tfmacz]

I have been told you risk demagnatizing the stepper motor if subjected to serious overcurrent.
That said My nema23 size steppers run very hot at the rated current.

My stepper motors are 8 wire interface. I am using them bipolar with the windings on each phase in parallel. The motor specs list 2.1A per phase so I was advised that with the windings in parallel I should set the motor current limits at 4.2A.

A problem I had with the steppers "stalling" turned out to be a PC problem. I am running EMC2 and a video card was causing the PC to miss steps. We verified this by creating a tester with a pizo type speaker on each channel step pin on the PC parallel port. You could hear "clicks" in the "squeal" as each axis moved. at certain speeds the missing steps would cause the stepper to stall.
I Installed a different video card and the problem went away. Sorry can't remember what make the video was???
Try removing/replacing any extra cards one at a time.

This stuff is always interesting...
Good luck....
Ted

[Elnino]

Mine is the '6 wire' one. I can never remember whether that is uni or bi polar....

Machine is a P4 1.8 and i have tried different frequencies without change. It is however a 'brand name' box - perhaps it's parallel port is not overly 'standard'.

I agree with you that it seems like mechanical binding but it just does not make any sense. Once it 'binds' i can just take my finger off the button and put it back on and off it goes again. And like i said, i shut the machine down and manually turn it from where it bound up and i feel no resistance.

It does the same regardless of rate of movement - 100mm/min or 1000mm/min. the only difference is how many steps it loses.

[Elnino]

I have been told you risk demagnatizing the stepper motor if subjected to serious overcurrent.
That said My nema23 size steppers run very hot at the rated current.

My stepper motors are 8 wire interface. I am using them bipolar with the windings on each phase in parallel. The motor specs list 2.1A per phase so I was advised that with the windings in parallel I should set the motor current limits at 4.2A.

A problem I had with the steppers "stalling" turned out to be a PC problem. I am running EMC2 and a video card was causing the PC to miss steps. We verified this by creating a tester with a pizo type speaker on each channel step pin on the PC parallel port. You could hear "clicks" in the "squeal" as each axis moved. at certain speeds the missing steps would cause the stepper to stall.
I Installed a different video card and the problem went away. Sorry can't remember what make the video was???
Try removing/replacing any extra cards one at a time.

This stuff is always interesting...
Good luck....
Ted

Hmmm - SERIOUSLY interesting that!! (the latter part anyway). See, what i have noticed is that it does seem to 'pulse' as it transverses the table. it's hard to explain but it sorta chages pitch high, then low every second or so. The binding appears to coincide with the pitch change....Maybe that's it!!

Maybe it's just some stupid IRQ confict happening or something! It does make some sense.

Oh, and on your 2.1A vs 4.2A - that seems wrong to me - If you parallel the coils, it halves the resistance thus you should halve your current, not double it. But then again, i have limited experience with steppers (read: 'Nothing before building my CNC machine')

edit - Now that i think of it, 4.2A does make sense halve the resistance and the current will double but here we 'set' the current rather than having it as a result of voltage through resistance. Ignore my comment above....

[Vern48]

I might suggest going to the Mach3 site.
They do suggest that a mother board integrated video card is a bad idea.
Also if you are running Windoze, they have a down load txt of all the supid system programs running in the back ground that you should turn off and how to do it. ie. Windoze security update, etc.

[Me2]

G’Day Elinino,
I had the Oatley Stepper drives and was losing steps intermittently. I had the old drivers that used the surface mount opto-isolators on the board. I found they where playing up so I asked Oatley to send some more down. Soldered them in, same problem.
I noticed that the new boards no longer use them but instead use transistors. I removed the opto-isolators from the board and basically made up an add on board using their new arrangement. No more problems.
If you are using the old opto-isolated boards it is worth looking at.
Have you contacted Oatley direct?

Cam

[Elnino]

I might suggest going to the Mach3 site.
They do suggest that a mother board integrated video card is a bad idea.
Also if you are running Windoze, they have a down load txt of all the supid system programs running in the back ground that you should turn off and how to do it. ie. Windoze security update, etc.

i have done that and i have disabled all the crap that is not needed. My machine does not have integrated video either....

I'm going to have a bash with another 'generic' machine and see if that makes a difference.

G’Day Elinino,
I had the Oatley Stepper drives and was losing steps intermittently. I had the old drivers that used the surface mount opto-isolators on the board. I found they where playing up so I asked Oatley to send some more down. Soldered them in, same problem.
I noticed that the new boards no longer use them but instead use transistors. I removed the opto-isolators from the board and basically made up an add on board using their new arrangement. No more problems.
If you are using the old opto-isolated boards it is worth looking at.
Have you contacted Oatley direct?

Cam

Mine are the new ones. I have not spoken to oatley's yet - i was assuming that the problem was with my machine and not thier boards....

I'll see how it goes with a new machine dual boot with Mach3 and EMC and compare.

[Warpspeed]

I would monitor the motor winding drive current on an oscilloscope. Make sure you are reaching the full 2.1A (or whatever), and also that the waveform is regular with no hiccups or variations during a long traverse.

That will at least indicate if it is the computer, the stepper driver board, or possibly some mechanical problem.

[derekj308]

Hi Elnino

Check out this website

http://www.orientalmotor.com/support...connection.htm

It summarises the correct settings for all combos of unipolar and bipolar 4, 6 and 8 lead.

You say you are running a 6 wire setup so if you are connecting all leads you are running unipolar and should tune your current controller to the name plate current rating on your motor.

I built one of the Oatley bipolar drives and current controller a few years ago when I first started experimenting with hobby CNC. I had similar problems with stalling and missing steps and never worked out what it was. I can only assume it was midband resonance which the Oatley drives do not compensate for. In the end I bought industrial stepper drives which do have midband resonance compensation and I haven't had any issues since. They weren't that much more expensive and I got the performance I was expecting.

tfmacz - If you are running 8 wire bipolar parallel at 2 x the nameplate current I would expect you would get your motors pretty hot under high duty cycle conditions or prolonged full stall current. Oriental say to run 1.414 x rated current for bipolar parallel. Try running at 1.414 x and see if you motors run cooler (not necessarily cold).

Cheers

[Elnino]

Well, i loaded up another machine with EMC (from the live CD) and.....It works FINE!

Same rapid speeds and it's perfect....

Now the question is - Was it the other machine or is it EMC???

I'll let you know tomorrow....

[idgara]

I've used the Oatley drivers.
They are full-step bipolar only. Don't expect much in the way of rapids.
Crank up the current, your steppers should be running hot. Just about too hot to keep your hand on for long.
You may also need to fan cool the board to keep the current sense resistors from melting the solder.

[Elnino]

My drivers are the unipolar ones, not bipolar. As someone mentioned before, there are 'old' versions as well. Mine are the new ones.

The motors barely get the chill taken off them but they do the trick on my machine and rapids are adequate. My table is only 300mm Sq. and with rapids of 1000mm/m it's fine.

Well, under windows with Mach3 is was 'ok' but not as good as with EMC2. Still useable but not perfect like EMC was.

I also noted that to get rid of the 'pulsing' i had to enable the IRQ for the parallel port in windows device manager. It still done it on the new machine.

So - Perhaps i put this down to incompatible parallel port on the machine i was using? Not sure but that's about all it could have been. Odd indeed.

I'll still look at the current running through the motors and see if perhaps it has dropped a bit. One thing i did not test was the current when all 3 motors were in use... I only done it with each board connected seperately.

thanks for the input people - it has been much appreciated!

[laheyth]

Go ahead and raise the current, say 10%, and see if you can detect any temp rise, ...keep raising it until happy, but I would also be looking at your screw and see why it is binding

[laheyth]

to clear up the current thing, if you look at the winding spec, and find that coil can take 2.1a and then parallel them, you want the current to go to 4.2, so each coil see 2.1a

[erlyrisa]

I have vexta motors TOO!!
-I also have the oatley electronics setup

-the oatley elctronics stuff is only good for fairly small motors.... hence why your motors are cold (they are supposed to have at least some heat coming out of them)
-as the other guy said 'hot to touch' , but not egg cooking hot. - the better controllers switch off the current when the motor isn't being used, so as to not cook it.

--as for sticking - I presume you have a ballscrew drive (or thread drive) and the motor is attatched to the thread.

-I made a machine like that, with vexta motors, but with proper 10micro step controllers. - it still sticks!!

The ballscrews problem (and the thread an nut problem especially) is that the thread acts like a guitar string, with the nut acting like your finger on the guitars fretboard. You want the finger to only be loosly on the string... but the harder you twank the string it doesn't matter how loosly you put your finger in the fretboard you still get a note other than E!

You will notice that if your thread is supended from both ends that the worse palce for sticking is in 1/3rds than 5ths and at other harmonic positoins along the thread.

the cure:

the nut should be 'loose' as possible in the transverse wave direction "up and down" .... consider moulding it into a block of silicon.

place dampers on the thread... on your gantry place some oily rubber of felt near the exit and entrance to your nut, about 5cm+ away (the further the better) that push down or up on the thread... giving you a nice neat section of 'straighter' thread, no matter what harmonics a playing along it.

-or simply replace thread with ballscrew- they need some really big amplitude in the harmonics in order for them to get stuck.

Or make your own thread bsed ballscrew (what I did)

ballswcrews aka RECIRCULATING ball screw can be made with a pair of nuts , rubber and some transmission oil.
...
drill a hole in two seperate nuts.
get a tube the nuts can fit into.(snuggly)
shove the nuts into the tube fixing with builders bog.(execpt where the holes are.
Seal ends with rubber capping.

---the idea is that the incompressible trans oil wil flow through the hole, re-oiling the thread in the direction of travel.
--- a better tip is to drill the nuts with a 45degree hole, so that the two holes meet.---even better add more nuts with holes drilled sideways to make a long tube connecting the two end nuts.
--- a good tip for creating a good seal is to expoxy the end nuts onto the thread with a very thin coat of oil on the thread (builders bog works better) - this way the only oild that escapes will be a thin film, and your accuracy will be about that much too!
---last tip... recyle the oild that escapes on to the thread by placing a pan under the thread (may want to just encase the whole thread in a tube) and running the oild down to a resovoir that replenishes the transoilnut. - or just place a grease nipple onto your transoilnut and replenish when necessary.

-hope this helps