MechMate CNC Router Forum

Go Back   MechMate CNC Router Forum > Common Assemblies & Parts > Motors & their mountings
Register Options Profile Last 1 | 3 | 7 Days Search Today's Posts Mark Forums Read

Reply
 
Thread Tools
  #1  
Old Sat 22 September 2007, 12:58
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Duty Rating stepper motors - setting the drive current

The conventional wisdom for setting a stepper motor's drive's current (eg. the resistor value for the Gecko) is to use the "Amp" figure published by the motor manufacturer. As an example, let's look at the following specs: (source)

Clipboard01.gif

Taking the very first motor on the list, it is a 2 Amp motor and the (left)graph shows the torque peaking at 21 kg.cm - same as the stated holding torque. The conventional wisdom is that if your Gecko's resistor is chosen for 2 Amp, this motor will give you a peak of 21 kg.cm torque.

However, I spotted something on those graphs which tells me that we will be very disappointed . . . . . . Look in the white block on the green graph: the last line says CURRENT = (squareroot)2 X RATED. That means 41.4% more current than rated! The 2 Amp motor was fed 2.8 Amps to produce the stated torque.

In a couple of cases where folk are telling me how great their steppers are performing, it now transpires that their currents are set higher than the motor specs. This subject is not really discussed in open forums......pumping more than the rated current into the motors. This thread is an attempt to bring this issue more into the open.

My take on the issue is that the current rating marked on the motor is for 100% duty cycle. ie. the motor can run at a steady speed and load which consumes that current for 24hours a day, without overheating. In our application of CNC routing, we don't run continuously at steady speeds and loads. Therefore, if our motors run cool all the time, we can increase the current . . . .

Most power tools have duty rating of much less than 100%. If you did run your PC router at the amperage marked on the plate, it will probably burn out within an hour. I am starting to think that the amperages marked on stepper motors are very "conservative"

The only way in which current can harm a motor is if it heats up the internals too much and the insulation breaks down. If the motor doesn't get hot, it isn't getting too much current. Steppers can run very hot and for most CNC routing it is probably okay to set the drives at more than the motor spec. . . . .

Comments?

Do all the motor producers manipulate the current value to produce their torque curves? Do they all use 41%? Do they tell us what current they used in the test?
Reply With Quote
  #2  
Old Sat 22 September 2007, 15:12
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
Gerald,
I think that you're correct in stating that HEAT is the limiting factor. For example, the Oriental Motor PK296A1A-SG3.6 motor draws 1A at 4.4V when wired bipolar. So 1A X 4.4V = 4.4W. The data sheet shows a recommended current of 1.5A at 3.3V when wired unipolar. That gives 1.5A X 3.3V = 4.95W. That's pretty close, but to get exactly the same heat (Watts) from a motor wired unipolar as a motor that's wired bipolar, we would need to do the math this way: 1A X SQRT(2) = 1.4A and 4.4V / SQRT(2) = 3.14V. Then running the math: 1.4A X 3.14V = 4.396W. Using those voltages and allowing the motor to draw that much current would produce the same amount of heat.

Of course, Mariss has used the rule-of-thumb that a motor can be run at voltages up to 25X its rated voltage or 80V, which ever is lower. So using that rule of thumb, and also using 80V as the maximum allowable voltage, the PK296A1A-SG3.6 motor can handle 80W bipolar (current limited to 1A) and 120W unipolar (current limited to 1.5A).

An interesting side note is that if we wire a motor half-coil, the motor will automatically pull 2X as much current because there is only 50% of the resistance to limit the current flow. That means that we need to decrease the Voltage to 50% of the bipolar voltage to limit the heat: 1A X 2 = 2A and 4.4V / 2 = 2.2V, then 2A X 2.2V = 4.4W.

Now, if we're running the motor at a really slow speed, and if that speed is constant over a long period of time, the motor will reach a much higher temperature than if it is mostly in STANDBY (at least if the G203v stepper driver is used that automatically reduces current to about 50% after a short period of inactivity). So, ultimate heat depends on Ohm's law, or Current X Voltage = Wattage and Wattage is our measurement for heat.

Now that everyone is nodding their head up and down, saying that they agree with Ohm's Law, or shaking their head side to side, saying that somebody has too much time on his hands, we get to Mariss's latest formula where he uses the motor's inductance rating, rather than the motor's resistance rating, to calculate the maximum voltage at which the motor should be run. So, using the same PK296A1A-SG3.6 motor, but using inductance rather than resistance, we find that we can heat up the motor to 80W when we use a full 80V at the bipolar rating of 1A. Mariss's formula, V < 1000 * SQRT(0.038H), shows that the motor could actually be run at 194V, IF the G20x stepper driver could handle more than 80V, BUT the G20x drive is limited to 80V. When we run the motor at 1.5A using half-coil wiring with the full 80V allowable by the G20x stepper driver, we see that the motor can handle 120W and still be within Mariss's safety margin. (1000 X SQRT(0.008H) = 89V, so 80V is about 90% of the maximum voltage that the motor could handle - in theory.) It's interesting to me that Mariss's rule-of-thumb of 25X voltage is a really good rule-of-thumb - according to his latest published formula.

My tests with PK292B2A-SG3.6 motors show that using Mariss's new formula and then limiting the maximum voltage to about 70% of the maximum computed voltage gives good results. Really good speed, really good torque and moderate heating. (The formula for that motor is: 1000 X SQRT(0.0015H) = 38.7V and 38.7V X 0.70 = 27V. I run that motor on an 18VAC toroidal transformer that gives 26VDC.)

So, the conclusion that we could probably reach is that we can pull as much current through the motor as we dare - as long as we stay within the motor's temperature rating of 80 C over ambient temperature. Personally, that's way too hot for me. I like room temperature or moderate heat when it comes to electronics. I start to get really uncomfortable when a motor's case temperature reaches 145 F (or 67 C) regardless of the ambient temperature.
Reply With Quote
  #3  
Old Sat 22 September 2007, 21:56
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Mike, isn't it interesting to you that everybody with Gecko powered CNC systems comment on how "cool" their motors are? For someone who cuts out kitchen cupboard doors all day, each of the axes either stands still or runs fast - both conditions being "easy" (cool) in terms of heating. No motor is crawling slowly for long periods. So, it seems logical that CNC folk could comment that their (gecko'ed) motors are cool.

It is also interesting to me that I've never read of anyone burning a stepper? Doesn't that also indicate that we have been applying very conservative currents?

I am very tempted to import a couple of $45 motors and boost their current to see when they cook . . . .
Reply With Quote
  #4  
Old Sat 22 September 2007, 22:06
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
PS. Those graphs above: It eventually struck me that the reason for testing with a current 41% higher than rated, is because the test was done with parallel connected coils.
Reply With Quote
  #5  
Old Sun 23 September 2007, 09:00
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
Let's take the last motor in the chart that has its rated current at 4A and its phase resistance at 0.95 ohm. We'll use those two figures to compute the motor's rated voltage: 4A X 0.95 = 3.8V. What we don't know is whether that represents the Bipolar, Unipolar, or Parallel data for that motor; however, let's assume that it is the Unipolar data.

To figure Bipolar data when we're given Unipolar data, we divide the Unipolar current by 1.4 to get the Bipolar current, and we multiply the Unipolar voltage by 1.4 to get the Bipolar voltage: 4 / 1.4 = 2.85A and 3.8V X 1.4 = 5.32V.

To figure Parallel data when we're given Unipolar data, we multiply the Unipolar current by 1.4 to get the Parallel current, and we divide the Unipolar voltage by 1.4 to get the Parallel voltage: 4 X 1.4 = 5.6A and 3.8V / 1.4 = 2.71V.

That gives us some basic data that can be used to start motor testing.
Reply With Quote
  #6  
Old Sun 23 September 2007, 11:53
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
I had to rush through that last post in order to get to church on time, consequently, I'm afraid that my mind kept wandering from important things during church to mundane things like stepper motors; however, I was able to verify that the Wattage (heat) produced by the Bipolar, the Unipolar, and the Parallel modes would all be the same (within a small round-off error). So, if I were testing a new motor and if I were going to use Gecko G20x stepper drivers, I would start the test by multiplying the Voltage by 15X. If that voltage was within the working range of a Gecko stepper driver, I would start testing the motor. If that voltage was NOT within the working range of a Gecko stepper driver, I would NOT use that motor - unless it was the only motor available and unless I was willing to overlook its reduced performance.

My testing normally consists of running the motor at all speeds at which that motor would be expected to run. Next, I would run some simple torque tests to verify that the motor could pull the load that it was expected to pull. Finally, I would step the motor very slowly - about 1/4 turn per second - until the motor's temperature had stabilized. If the motor's temperature stayed within whatever limits I had selected, then I would put that motor into service.

What I expect out of a motor and what someone else expects might differ greatly. Consider the expectations of those who purchase airplanes. A commercial carrier like United Airlines would expect that the airplanes they purchase to be reliable, safe and long lasting. The military's expectations would be entirely different. They would place the ability to perform a specify type of mission over reliability, safety and longevity. My expectations about a stepper motor are more in line with the expectations that United Airlines has about the airplanes it purchases. Once I install a stepper motor and stepper motor driver into a piece of equipment, I don't want to ever have to even think about that stepper motor or that stepper driver again. I want it to work flawlessly until that piece of equipment is removed from service. That's basically what happened during the time that I designed process control computers for the photo industry. Of all the stepper motors and stepper drivers that I installed in Kodak-S printers, only one driver ever failed during the 15+ years that those machines were in service. Possibly, I could have used smaller motors than the PK268 motors or used higher voltages, or set the resistor on the stepper driver to allow more current through the motor, but I took pride in the fact that most of my customers never ever had to send for me to service those process control computers - or the motors and drivers that were part of those process control computers.

Most CNC routers are big machines with lots of room available for motors of any size; meaning that I don't have to worry about squeezing a large motor into a small place. My personal preference is to use the Oriental Motor PK299 motors, with a belt-reduction gearbox, rather than use the smaller PK296 motors with their built-in gearboxes. However, if I either didn't want to build my own custom belt-driven gear box or if I had easy access to PK296A2A-SGxx motors, I would NOT hesitate to use the PK296A2A-SGxx motors. I would run them at their factory current ratings, or, if necessary, at a current rating slightly above the factory recommendation (25% higher) if I needed more torque. I would feel comfortable doing that because I've tested those motors under those conditions and they meet all of my criteria.
Reply With Quote
  #7  
Old Sun 23 September 2007, 13:03
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Thanks for letting your thoughts wonder during church Mike - I might have to go to confession on your behalf, except that us Methodists don't have that.

I think that we have established that there are conditions under which the current settings can be increased beyond the spec sheet value with the overriding concern of keeping the temperature within limits.

Most of the guys building the MechMate don't actually know how hard and fast they want to cut. We are always going to have someone who finds that he he wants to go faster/harder, and then that person could experiment with different current settings (at own risk). The risk appears to be low if a careful eye (hand) is kept on keeping the tempature below 80C.
Reply With Quote
  #8  
Old Sun 23 September 2007, 21:24
Dirk
Just call me: Dirk
 
Alpharetta, Georgia
United States of America
I have found that using a higher current set is usually detrimental. A higher current set can give you more power, however the over current will create more resonance. Most steppers have a resonance band at certain rpms. Even though Gecko's compensate for this it's still there. With the right load conditions at this rpm the excess resonance can cause the stepper to stall.
I prefer to increase voltage as opposed to current to push a motor.
Dirk
Reply With Quote
  #9  
Old Mon 24 September 2007, 13:07
domino11
Just call me: Heath
 
Cornwall, Ontario
Canada
Quote:
Originally Posted by Richards View Post

Of course, Mariss has used the rule-of-thumb that a motor can be run at voltages up to 25X its rated voltage or 80V, which ever is lower. So using that rule of thumb, and also using 80V as the maximum allowable voltage, the PK296A1A-SG3.6 motor can handle 80W bipolar (current limited to 1A) and 120W unipolar (current limited to 1.5A).
Richard,
I dont believe the 80W and 120W figures are entirely correct. The power transfered to the motor is directly proportional to the duty cycle of the chopper driver at any given time. Yes, the higher voltage will make the current ramp faster in the motor, but when the current gets to the set threshold, the chopper duty cycle lowers to ensure that the power to the motor stays where you set it. So I beliieve that the formula you must use is voltage x current x steady state duty cycle. This could be measured with a digital scope when you are running at a static rpm and you should see that the power is a lot lower because the chopper is not on all the time. The higher voltages dont increase dissipated power with a chopper but just helps get the stepper up to the set speed faster. Like a bigger engine accelerates faster but doesnt necesarrily give you more speed.

You seem to be really digging into this topic, very enlightening
Reply With Quote
  #10  
Old Mon 24 September 2007, 15:54
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
Dirk,
You're absolutely right about increased resonance at higher than normal current. Every stepper motor seems to have a 'rough' spot somewhere in its speed range, usually at very low step rates, but increasing the current pulled by the motor extends that 'rough' area significantly. I fiddled with the current set resistors on a PK268 motor for a few hours as I ran it at various speeds. As the current went up, so did the resonance and so did the heat! Given the fact that a CNC router is expected to run well at any reasonable speed, increasing the current beyond normal could easily cause poor overall performance.

----------

Heath,
You'll also absolutely right about how a chopper drive works and how much current is really being pulled by the motor. The PWM circuit (pulse width modulation) or other type of chopper circuit is usually designed to use as much voltage as is available to force as much current as the motor is rated to withstand in as short a time as possible. Once the motor is pulling that current, the voltage is reduced through the chopper circuit to keep the motor from pulling too much current through its coils - and overheating. So, in reality, multiplying the motor's rated voltage by the motor's rated current would give a much better idea of the motor's Wattage rating. One thing that seems certain is the fact that a motor running in its rated voltage/current range should last for years, but a motor that is overdriven and run at too high a voltage or at too high a current level is going to get overly hot and have a shorter life.
Reply With Quote
  #11  
Old Tue 25 September 2007, 07:11
domino11
Just call me: Heath
 
Cornwall, Ontario
Canada
Mike,
Sorry for calling you Richard in the last post, was doing two things at once. The voltage you drive a motor with a chopper type circuit is not that detrimental as long as you dont exceed the rated current. When a chopper drive is being used the motor sees the driver as a current source, either on or off not as a voltage source. Older drivers that use a ballast resistor to limit current appear to the motor as a voltage source and in this case you can more easily damage your motor from overpowering it. The main thing is as you mentioned is not to exceed the current rating of the motor windings. The coil resistance and parasitic inductance along with the current you are running it at determines the power being dissipated in the windings. The chopper drivers excels at this as it just switches on and off very quickly to dynamically limit the current into the windings. The voltage we feed the driver just dictates the speed at which we can accelerate the motor up to its rated current. Saying all this I would not run a stepper over its rated current as this would induce a larger power dissipation, and this is probably in my thinking why people have noticed higher resonance issues in higher current modes. The stepper makers have tested their products and have found the sweet spot where torque, speed and parasitic losses are at their best compromises.
Reply With Quote
  #12  
Old Tue 25 September 2007, 15:39
Mike Nash
Just call me: Mike Nash
 
Bessemer, Alabama
United States of America
There was a thread here: http://www.cnczone.com/forums/showthread.php?t=29262
in which Mariss may have answered this question regarding drive current vs motor current. See post 32.
Reply With Quote
  #13  
Old Fri 08 August 2008, 10:23
Greg J
Just call me: Greg #13
 
Hagerman, New Mexico
United States of America
After operating my MM for some time now, I’m finding that the motors are generating too much heat for my comfort level.

I had posted earlier that the temperature leveled out around the 120’s F after an hour of operating. My projects are now getting bigger and the MM is operating continuous for longer periods.

Using an IR Temp. gun, I’m seeing temperatures in the high 130’s F after 3 to 4 hours of continuous run time. I have never reached 140 F yet.

Here’s my equipment setup:

Oriental Motor: PK296A1A-SG7.2, Unipolar Connection

Gecko G203V Drivers with 12.5 Kohm current set resistor

Power Supply: AnTek PS-3N70R9, 70V, 300 VA

Is it as simple as increasing the current set resistor and limiting the current to the motors?
Reply With Quote
  #14  
Old Fri 08 August 2008, 11:16
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
If the temps are still below 140 F, the motors are still happy

But, if you want to drop the temps (and drop torque & max speed) you can easily reduce the resistor value.
Reply With Quote
  #15  
Old Fri 08 August 2008, 11:45
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Some good reading:

http://www.orientalmotor.com/automat...20Systems.html

Temperatures also discussed.
Reply With Quote
  #16  
Old Fri 08 August 2008, 12:24
Greg J
Just call me: Greg #13
 
Hagerman, New Mexico
United States of America
Thanks Gerald,

I like good, lunch time reading.
Reply With Quote
  #17  
Old Mon 08 June 2009, 11:46
kaartman
Just call me: Koning #20
 
Abu Dhabi
United Arab Emirates
Reduce current set resistor by how much?

I have used these motors http://www.kelinginc.net/KL34H295-43-8B.pdf with UNI Polar connection, with the resistor calculated at 74.8K, these motors run so hot that I cannot hold my hand to them for longer than 5 seconds, at what incriments should I reduce the resistor value
Reply With Quote
  #18  
Old Mon 08 June 2009, 12:52
domino11
Just call me: Heath
 
Cornwall, Ontario
Canada
Koning,
If you are using the 203V drives, you cannot wire up the steppers in a unipolar setup as you only have 2 terminals for each winding. You need to use either half coil or series (full winding). See the 203v reference manual. If you are using full winding, blue/green and brown/white then your resistor should be a 35K resistor. If you are using half coil, blue/red and brown/black, or yellow/green and orange/white then your resistor need to be a 74.8K sounds about right. What is your power supply voltage? For half coil you should not be any higher than 58VDC (Mike would recommend only 75% of that)
Reply With Quote
  #19  
Old Tue 09 June 2009, 01:23
kaartman
Just call me: Koning #20
 
Abu Dhabi
United Arab Emirates
Hallo Heath
OOOOps I made a mistake, went back to the papers, double check and tested, the wires are Parallel connection, Power supply at 58VDC and the current resistor at 309K, this is the Max current resistor, seems to me that the answer is on the papers,(above #14) will you agree that I can try to reduce the resistor value by 25%
Reply With Quote
  #20  
Old Tue 09 June 2009, 07:39
domino11
Just call me: Heath
 
Cornwall, Ontario
Canada
Koning,
Ok for parallel connection, your 309K is as high as you would want to go. To go to 75%, the setting would be for 4.5 or 5 amps (or in that area.) The closest standard resistor for these setting are 84.5k for 4.5A and 118K for 5 amp. I would probably try the 84.5K and see how the performance and heat is to your liking. You can always move up a notch or two from the 84.5K. Let us know how it worked for you.

PS how is the weather up your way, we have some guys from work heading to your area of the globe next week.

Last edited by domino11; Tue 09 June 2009 at 12:47..
Reply With Quote
  #21  
Old Thu 11 June 2009, 00:36
kaartman
Just call me: Koning #20
 
Abu Dhabi
United Arab Emirates
Morning Heath
I will let you know in a few days, have not been home from work yet for the past few days, It is hot here, midday temp to the tune of 45 degC (113F), we are bessed though, the humidity has not been that high yet, the avg humidity is about 65-80%. the down side of this climate is that 99% of the people think that the indoor temp of the buildings should be 18degC (64F) ---my terms FREEZING and that is why this time everybody suffers head cold and sinus problems.
Enjoy the day
Reply With Quote
  #22  
Old Sun 21 June 2009, 00:07
kaartman
Just call me: Koning #20
 
Abu Dhabi
United Arab Emirates
Good morning
I had a word with the boys at the electronic shop in search for the resistors, now I am confused, the resistors are in K ohm @ half, quarter or one Watt. What happens with the value of the 4.5 A 0r 5A. To get the value of 84.5 or 118K I will need to place various resistors in series, is this correct, do I take half ,quarter or 1 watt resistor?
Regards
Reply With Quote
  #23  
Old Sun 21 June 2009, 02:45
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Koning, various people have found that parallel connections make motors hot and give no extra performance for CNC router work. (The extra torque shown on the data sheet is at very slow speed, where we have little load). Strongly suggest you try the motor "half-coil" (remove yellow,green,orange,white wires and insulate them individually).

Your 58V DC supply will be good for the 3.3mH inductance of half-coil.

To pick a Gecko resistor for 4.3Amps limit current, it calculates out at 75 kiloOhm

A very common resistor is 68 kiloOhm, but 75 kOhm resistors are also quite common. Either one will do, but check them with your multimeter

Geckodrive says to use 0.25Watt power rating for the resistors.
Reply With Quote
  #24  
Old Sun 21 June 2009, 07:37
J.R. Hatcher
Just call me: J.R. #4
 
Wilmington, North Carolina
United States of America
Send a message via Skype™ to J.R. Hatcher
I'm thinking "Father's day " in universal. I would like to wish every father on this forum ..... Happy father's day ..... and to the father of the MechMate a special thanks for the time and effort (as seen in the last post) given daily, ....Happy father's day Gerald.
Reply With Quote
  #25  
Old Sun 21 June 2009, 09:43
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Thank's JR. Sean took me (and family) to a great lunch at Suikerbossie . . . I suppose he did that on behalf of all you MechMaters

I certainly am not alone in all of this; for example we have JR as the father of the grinding skate, Nils as the father of screwed on V-cap rails, Greg J as the father of gantry untwisting, Javeria as the father of doing everything the long way around , etc., etc. . . . . . . too many to mention all. Thanks to ALL of YOU!
Reply With Quote
  #26  
Old Sun 21 June 2009, 22:54
kaartman
Just call me: Koning #20
 
Abu Dhabi
United Arab Emirates
Thank you Gerald, i hope to be home this weekend and do some changes on the connection and resistors, take care
Reply With Quote
  #27  
Old Tue 23 June 2009, 05:07
javeria
Just call me: Irfan #33
 
Bangalore
India
Ah! Thank you G'
Reply With Quote
Reply

Register Options Profile Last 1 | 3 | 7 Days Search Today's Posts Mark Forums Read

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -6. The time now is 02:09.


Powered by vBulletin® Version 3.8.3
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.