Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: ddefocy on October 04, 2017, 07:53:17 PM
-
Hi ,
Trying something new , Mach3 controlled single axis (Nema-34 1600/oz) Press.
I thought with all these extra parts I have laying around I would try to fabricate a stepper driven press. Unlike most using the HF 20ton H-press/bottle jack method
I would try driving a 5-ton screw jack w/gear reduction. Using a 15amp 24vdc power supply and DQ860MA microstep driver running full 7.8amps I just need to drive this
single axis a few inches up/down and get enough tonnage to bend 16ga mild steel. My problem is the screw jack has a mighty 6:1 gear ratio and the screw needs to spin
16 revs to equal 1.0inch. This is where things get sketchy for me. So I figure the steeper motor would need to spin 96 revs to move the screw one inch. Thats ok but I'm new to mach3
and stepper drives , I'm looking for advise on what would be the best and most accurate way to figure out what would give me the most torque and get the desired accurate displayed motion.
My drive can run bewteen 400 and 51200 pulses/rev. Thank you in advance.
-
Hi,
two phase steppers which are by far and away the most common move 1.8 degrees per full step, or 200 full steps per revolution.
If Mach3 or some other device pulsed the Step pin of your drive the motor would turn 1 revolution. If your stepper driver can execute half steps,
and most can, then 400 pulses are required for 1 revolution. If your drive is set to execute '8 step microsteps' then 8 x 200 =1600 pulses are
required to turn 1 revolution.
Most of us would like to believe that because we can turn an 8 microstep motor just one pulse at a time it can achieve 1/1600 th of a revolution
per pulse or 0.225 degrees per step, that is a nice fine resolution. In fact it doesn't work that way at all but nice to think that it does. Resolution is
reliably increased at half step but at quarter step it loses its reliability and accuracy. So why does everyone microstep... because steppers are smoother
when operated at 8 or 16 step microstepping.
For your application do you require very fine control? Given the gear reduction of the screw jack even 1/2 step will give you plenty of resolution.
1.8 degrees x 1/2= 0.9 degrees. 0.9degrees/6 (gear reduction)=0.15 degrees. With a pitch of 16 TPI that results in a resolution of your presshead
of 0.02 thousandths of an inch per pulse. Yeah, that's right 50 pulses for 1 thousandth of an inch...plenty good!
Steppers lose their guts when going fast, they have something like half their torque at 500 rpm. Assume you want your press to have plenty of ompf
so decide a maximum speed of your stepper is 500rpm. The maximum speed of the presshead would be 500/96=5.2 inches per minute, not really fast
but useful. Trying to work out how much ompf it got is a bit of a guess because with a large gear reduction and fine thread a large part of the input
torque it lost to friction. I would make a stab and say your 1600 oz.in will be 800 oz.in at 500 rpm and half of that lost to friction ie the actual torque
available for useful work is 400 oz.in If the diameter of the screw jack is 1 inch then the tangential force at the outer periphery of the screw is 800oz
or 50lb. With the thread mechanical advantage of (2 x PI x 1/2(radius))/ 1/16(pitch)=50.2 Therefore your presshead thrust should be in the region
of 50 x 50.5 =2513 lb. DON'T stick your finger in it! Bit of a sketchy calculation but probably close enuf for you to decide if you wish to pursue the idea.
Just goes to show what decent mechanical advantage can do.
Craig
-
Craig,
Thank you for the very informative response. 2500lb's should do the trick , I was hoping to achieve 3T , but this powers the pursuit anyway. I will continue the build.
I was playing around with Mach3 last evening and it was difficult to find to the sweet spot of speed and torque without loosing steps. Has anyone played around with the
HT (High Torque) Nema-34 motors , I could boost the tonnage slightly with 1800+oz motor. running the servo drive at full current really got the motor hot !! little concerned
about that ?
Thanks again for your quick response.
-Dennis
-
Your motors have a current rating. Exceeding that continuously can damage the motor by demagnetizing it. It will still run but with way less torque. Steppers are continuous power devices so they consume the same power whether driving a load or not. So they heat up. Most decent stepper drivers reduce the current when not actually moving the motor.
-
Hi,
2500 lb is just an informed guess. The actual stall pressure could well be more, much more.
If using half stepping the motor motion will be pretty harsh. The main reason for using microstepping is for smooth motion and the
illusory sense of increased resolution is very seductive however false it may be. You could try using 8 or 16 microstepping, the increase in smoothness
of motion may mean that its easier on the ear and may also be cooler running.
Steppers do run pretty warm, I back the current off on my motors to have them run a little cooler but they still get to 50-60 degrees C on the outside
and I imagine closer to 100 degrees inside....doesn't seem to do any harm. Whether they would stand that 24-7 as in industrial use I suspect not
but they have given me about 1000 hrs a year for several years so can't complain.
Craig
-
and it was difficult to find to the sweet spot of speed and torque without loosing steps.
The most torque (the rated torque) is when they are not spinning. The faster you go, the less torque they will have. Those motors probably lose their torque faster than any other motors you can buy, due to their high inductance.
To get more torque at higher speeds, you need to increase the voltage. To get the most out of those motors, ideally, you'd need well over 120V. 24V will give very poor performance at all but the slowest speeds.
-
Thank you for this input. I currently have a 15amp 24vdc supply source at the stepper driver. What your saying is go to say a 48vdc supply voltage and I can achieve more RPM's and torque ?
Cheers.
Dennis
-
Yes, but if possible, I would go to 60-72V. The more the better.
And you only need about a 6 amp supply for the one motor.
-
The stepper has max 1600 oz-in torque.
More volts will make it accelerate faster .. but will never pass the stall torque.
1:6 gear reduction = 1/6" lift on a screw, be it acme or ballscrew.
Acme/etc == 30% efficient.
So you get about 500 oz-in / 1/6", around 25.4/6 = 4.23 mm screw rise.
This is what I had:
At 3 Nm on 1:3 on a 5 mm ballscrew 90% efficient thrust is == 600 kgf, or 6000 N.
About 1500 lbs of force.
Thrust force is about (efficiency, 30% on acme or gears-your case) x rise x torque, all numbers normalised to same units.
I use Nm / rise in mm, etc. in metric.
On 500 oz-in == 4 Nm effective, and 4.2 vs 5 mm effective, about 1.3 x 1.2 == 936 kgf force, 2000+ lbs lift force.
-
I actually do have inventory to try , located 80vdc 6amp supply.
The Joycedayton screw jack specs:
Screw Dia:1.5" Not ACME actual ballscrew and Nut.
thread picth:1.0"
Grear Ratio : 6:1
worm shaft turns for 1.0" = 6turns
This by no means will be a quick press , My target was 3Tonnage , I would be happy if this setup yields 2T.
Ideal speed 10ipm , realistic will likely be in the 5-7ipm ?
I appreciate everyone's input.
Thanks
Dennis
-
The ballscrew will make a huge difference in speed and force. Since you are using a stepper I suggest you implement advance speed, press speed, and retract speed. Hydraulic presses all do this. Possibly a button that causes the advance speed when held, press when released, then when full stroke is reached you get fast retract to a settable starting height. This gives both high force and high speed.
-
Eventually that is the plan. I'm new to Mach3 so maybe someone could point me in the right direction regarding these features of changing the speeds using I/O.
The press screw jack has 19.0" inches of travel , I'll only be using 8-10 inches of that travel. So retracting at a higher speed would be very desirable.
Couldn't I just write the federate into the NC code for the stroke ? like G0 (rapid traverse) ?
Thanks
Dennis
-
Code like this:
G0 Z.25 (rapid approach)
F5
M1 (pause until start is pressed)
G1 Z-.25 (press stroke)
G0 Z2 (rapid retract)
M30 (rewind program top wait for start)
-
Quick Update on the press build. Sorry its been awhile. So many other projects.
So after a lengthy build I attached the coupler to the 1600oz Nema 34 stepper and attempted to bend a 1.0in wide 16gage mild steel piece of material 90 degree.
The machine drives smooth and accurate but when it came to the actual grunt work of getting the tight radius in the bend it stalled and the stepper skipped steps. (big problem) !!!
I have heard of high torque steppers that have as much as 1800oz torque but I'm not sure this is my answer. ? I lowered the motor tuning velocity and accel to the min for max torque and
the results were better but not worthy of calling this machine a press. The downside to the settings change was a predictable slow as molasses pace.
I'm hoping someone has some insight or suggestions here please. Hi-Torque ? Servo or DC motor , still needs to be programmable. Hoping to stay with the mach3 software as I'm really found of it now.
Thanks in advance.
Dennis
-
Hi,
steppers are really good for torque at slow speeds, much better than a servo of the same size. The servo
comes into its own at speed.
1600oz.in is 8.3 ft.lb or 10.4 Nm. You are going to need a bloody big mega expensive servo if you wish to go that way.
The only other way is to gear down your stepper and take the speed hit.
I have little 23 size steppers running through 10:1 low backlash planetaries and they max out at about 5 ft.lb or
960 oz.in....really really impressive for little motors...tad slow though!
Craig
-
Have you calculated how much torque you actually need?
-
A 1.5-2Kw AC servo, with a 10:1 gearbox might work.
DMM sells a 1.8Kw servo with 11.5Nm torque. With a 10:1 gearbox, that would give you about 16,000oz-in at 150rpm. And up to 40,000 oz-in of peak torque for a short time.
About $700 for the servo and drive, and probably more for the gearbox.
-
Hi Ger21,
thats pretty good buying really.
I had in mind a angle gear-reducer. I got one off Ebay 19.5:1 reduction, 2min arc backlash, 380 Nm output, including
airfreight to New Zealand $330NZD or about $250USD.
Craig
-
Hi , Thanks for the replies.
Current set-up is 6:1 gear reduction with 1600oz Nema 34 @ 75vdc. I was looking at the Screw jack with the single input shaft and thought what if ?
So I contacted Joyce Dayton and the friendly customer support guy said that I can remove the single input shaft and make my unit a double input shaft.
Now if I go with two HI-torque 1841oz Nema-34 steppers with 6:1 gear reduction that's twice the HP I may get the desired results (speed and power)
The addition of the two new steppers , another 80vdc power supply and one more stepper driver along with the Screw jack mod and coupler I'm looking
roughly $700 extra dollars. Its a tough decision but I think this should do the trick.
Not sure why the pic is rotated , sorry.
-
Hi,
that project looks good. The equipment the back ground of your pic suggests that you are an avid builder.
one more stepper driver along with the Screw jack mod and coupler I'm looking
roughly $700 extra dollars. Its a tough decision but I think this should do the trick.
Seems like a lot of gold to spend without being it will work. The advantage of the servo that Gerry mentioned is that it will hit its rated torque at 1500rpm,
no guessing involved.
Craig
-
So looking into the servo options. I have some experience retrofitting a laser system with ClearPath motors and looking at there SD Nema-34 motor the torque spec looks like
its peak is at the RPM were I'll be happy with the speed of the machine. This is roughly the same cost as adding the Hi-torque steppers and additional bits. Here is the data from there site.
I like that these motors are so quiet too.
https://www.teknic.com/model-info/CPM-SDSK-3441S-RLN/
-
Hi,
I think the 1.8kW DMM servos will eat those Clearpath types, to start with the DMM types are powered by 230V or 320VDC link voltage and 11.5Nm works
out to over 1800 oz.in, through a 10:1 gearbox you're talking serious grunt.
Craig
-
I think the 1.8kW DMM servos will eat those Clearpath types, to start with the DMM types are powered by 230V
Yes, an AC servo will easily outperform a Clearpath, and they cost about the same.
A lot of people seem to be jumping on the Clearpath bandwagon lately. They are great, if you want a drop in replacement for a stepper motor. But if starting new, AC servos provide far more power.
Be careful with the voltage on the DMM;s, though. I just got one back today from being repaired. It blew as soon as I powered it up, apparently from overvoltage. They claim they should be safe up to about 260V, they tell you not to provide over 240V. I have about 245V here at home. I'm going to get a transformer to lower the voltage down before I fire it back up.
-
Hi ger21,
with due respect 1.8kW servo and driver for $700 is very cheap. That they may be a bit suspect under modest overvoltage conditions is hardly suprising.
The real test is how they perform/survive emergency de-acceleration when they 'generate' and will if uncontrolled push the DClink voltage sky-high and blow
something.
Craig
-
Budget limitations at this point means that any change to the motor configuration would need to be "plug and Play" meaning no changes to the controller
step and direction and Nema-34 frame size only. The Clear Path servo motor may be the only viable option to gain the extra torque. I agree looking at the specs the DMM
servo would be the better option but at this time I'm either going with the Clear Path servo motor or 2 hi-torque steppers. I measured the tonnage last night with a load cell and I'm at exactly 1-ton and need to be at least 2.5 ton to get the desired results and material thickness range. That's with one Chinese 1600oz stepper at 6:1 gear reduction and I needed to slow slow slow it way down to get that 1.0ton load pressure. looks like the torque range is peak till about 125rpm at 12nm on those CP servos. I'm going to get a quote from them later today.
-
Hi,
those CP servos are good for 500 oz.in if you expect to run them at their peak torque they will fault out within seconds.
Fundemental fact about motors: torque s proportional to current. Therefore to run those servos at three times rated torque (1500 oz.in) will require three times
rated current.....and that level of current might be sustained for 10 seconds but not much more.
Those DMM servos require Step/Dir signals only, they have their own 230V power supply, they require less of your controller than the CPs. Don't see that you should
have to do anything to your controller at all.
Craig
-
Worth giving those DMM's a look.
The concept of this machine is to mimic (as close as you can) a hydraulic piston operated press.
The peak current (max load) comes into play at the coining stage when your piston or in my case the screw runs
into the bottom of the die. Stalling at this point cannot happen , The idea would be to hit max peak and stop. Then
that is what the machine would be rated for tonnage. That little 1600oz stepper stopped and stalled at 1.0ton. Not nearly
good enough. If the DMM servo's frame is equivalent in size to Nema-34 I will definitely consider them.
Quick look at the 1.8kw motor and the frame size is 130mm , I would have to really hack the frame up to shoehorn that motor in.
-
Hi,
best rated torque for 80mm frame (34 size) DMM servo is 2.4 Nm, not enough.
The 11.5 Nm (rated) DMM is 130mm frame, quite a bit bigger.
Large torque requires a large rotor diameter...
Craig
-
Hi,
don't know if you've had much to do with AC servos but almost all of the offer 'torque mode' usually with analogue input.
You dial up 8V and the servo will produce 8/10 x 11.5 (Rated torque)=9.2Nm through your reduction and screw to give 1.3 t say. Doesn't matter if it
stalls it still produces the force that you wanted. If you want more force....turn up the knob a bit more.
Caig
-
Hi ger21,
with due respect 1.8kW servo and driver for $700 is very cheap. That they may be a bit suspect under modest overvoltage conditions is hardly suprising.
The one I have is actually 400w, and the fact that it's very cheap is the main reason I bought it. ;D
I wanted to do some testing with it, before I buy the two more that I need. I was a bit disappointed that it blew up the second I applied power to it. But they fixed it for free, so I'll have another go at it when I lower my voltage.
-
Hi ger21,
yes I've looked at them too. Given that I live in New Zealand the actual cost by the time it gets here including airfreight and tax is closer to $1000USD.
I needed a servo for a spindle and was proposing to direct couple and really wanted a bit more speed than the 1.8kW DMM (1500 rpm).
On Trade-Me, a NZ auction site, I got a 1.8kW Allen Bradley servo (6.2Nm @3500 rpm) and drive for $800NZD including freight or about $600USD.
It is of course second hand so cant really be compared to a new DMM but so far it KICKS ANUS!!! I can twist 8mm carbide tool off like a carrot and talk
about making chips....my little mill can't really keep up, it starts to flex alarmingly and it leaps around, I had to tie it to the wall!
I recall that Hood had used several of them over the years and recommended them and I can too. EBay has a fair selection of them.
Craig