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 Tue 17 February 2009, 16:09
CNCNow!
Just call me: Jorge
 
Sao Paulo - SP
Brazil
Selecting the voltage to the geckodrives

Quote:
Originally Posted by Richards View Post
MAX Voltage = 32 X SQRT(Inductance), so 32 X SQRT(1.5) = 39VDC.
Could you please clarify that formula ?

Why are you stating maximum voltage as a function of inductance ?

Why 32X the inductance square root ?

Jorge L.
Reply With Quote
  #2  
Old Tue 17 February 2009, 16:44
CNCNow!
Just call me: Jorge
 
Sao Paulo - SP
Brazil
Quote:
Originally Posted by martin77pl View Post
I am trying to figure out power supply for the geckos and my PK296AE-SG7.2.
Are you sure about the motor specs ?

Oriental Motor catalogue says:

2.1A, 2V, 0.96Ω, 6.0mH, two phases (bipolar)
Reply With Quote
  #3  
Old Tue 17 February 2009, 22:03
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Jorge, that formula was "discovered" by Mariss at Geckodrive about 2 years ago. We soon realised it this new formula gives the best results - a balance between motor heat and motor torque.

(The previous formula used for Geckodrives was 20 to 25 times the voltage stated by the motor manufacturer)

Remember, we are talking of the DC voltage supplied to the drive electronics. The drive supplies something completely different to the motor - that is a digital signal chopped into tiny sections.
Reply With Quote
  #4  
Old Wed 18 February 2009, 03:38
CNCNow!
Just call me: Jorge
 
Sao Paulo - SP
Brazil
Quote:
Originally Posted by Gerald D View Post
Jorge, that formula was "discovered" by Mariss at Geckodrive about 2 years ago.
Thanks Gerald !

Could you point any post by Mariss on the subject ?

I really want to fully understand the reasoning behind that equation ...
Reply With Quote
  #5  
Old Wed 18 February 2009, 04:14
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
http://finance.groups.yahoo.com/grou.../message/12995

plus

http://finance.groups.yahoo.com/grou.../message/12996
Reply With Quote
  #6  
Old Wed 18 February 2009, 05:29
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
Jorge,

There are two inductance ratings per 6-lead motor, one is bipolar series and the other is unipolar. In post #15, you quoted the bipolar series ratings. My quote (post #11) was in response to post #10, where the poster wrote that he was going to wire the motors half-coil. (To avoid confusion, I call it bipolar half-coil so that everyone knows that it is alright to use that type of connection with bipolar stepper drivers, like the Geckos.)

The PK296A2A-SGxx motors are BEST used with bipolar half-coil connections. Because of the gear boxes used on those motors, you get maximum usable torque and maximum speed when you connect the motors bipolar half-coil. The extra torque produced by bipolar series is wasted because the gear box cannot handle that much torque. Also, a motor wired bipolar series is limited to slower speeds when compared to a motor wired bipolar half-coil.

The formulas used were developed by Mariss at Gecko to help us get the maximum performance out of motors used with his Gecko products. PLEASE NOTE that the 32 X SQRT(inductance) formula gives the MAXIMUM voltage that a well designed motor can handle while keeping the motor's temperature from melting its insulation. That voltage will cause the motor to run very near its MAXIMUM temperature. I don't like to burn a finger when I touch a motor, so I de-rate that voltage to 90% or less of the computed voltage. In the case of the PK296A2A-SGxx motors, I run them at 35VDC, which keeps the temperature within my comfort zone.

When selecting components for a CNC router there are many options available. Some people want the highest possible performance. Some want the highest possible quality. Some want the longest possible usable life from the electronics. Others want the lowest possible cost.

The Oriental Motor PK296A2A-SG7.2 motors give good results at a moderate cost. They are good value. You can get higher torque at nearly the same level of cut quality by building a belt-drive transmission. Most likely just the cost of an individual transmission will cost as much as an entire PK296A2A-SG7.2 motor/gear box unit. (I've built several sets of belt-drive transmissions for my Shopbot, so I have some experience in that field.)
Reply With Quote
  #7  
Old Wed 18 February 2009, 05:50
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Mike, I have a problem with your "bipolar half-coil" wording because I have always called that "unipolar (half-coil)". The reason I choose to insert the term "unipolar" is because of Oriental Motor's spec sheet which gives two sets of ratings for 6-wire motors "bipolar(series)" and "unipolar". For our half-coil connections, we use the inductance & current ratings in the "unipolar" section of the OM data sheet.
Reply With Quote
  #8  
Old Fri 20 February 2009, 05:16
CNCNow!
Just call me: Jorge
 
Sao Paulo - SP
Brazil
Mike and Gerald, thank you very much !

Quote:
Originally Posted by Richards View Post
PLEASE NOTE that the 32 X SQRT(inductance) formula gives the MAXIMUM voltage that a well designed motor can handle while keeping the motor's temperature from melting its insulation.
I have no time right now to elaborate further but I have to say that as far as I know the maximum voltage the motor can withstand is not related to inductance and heating is a function of current, not voltage (choppers are constant current sources, let´s not forget) ...

When someone like Mariss says something like that I can´t help but pay as much attention as I can but I must confess it does not make sense yet ... It seems to be an empirical formula that may be as good as any other. Or not ...

If someone could shed some light on it and rescue me from the darkness I´d be forever indebted.

Last edited by CNCNow!; Fri 20 February 2009 at 05:21..
Reply With Quote
  #9  
Old Fri 20 February 2009, 06:08
bradm
Just call me: Brad #10
 
Somerville(MA)
United States of America
Jorge, I think you're pretty much there. It's an empirical rule. Consider a given motor at a given voltage and a constant current. Assume it's not moving. All that power has to get dissipated as heat. Double the voltage, double the heat. I take Mariss' formula as a shortcut based on experience with the current designs of stepper motors used with his drivers. If you were to develop a heat sinked stepper motor, you could probably jack up your voltage more.
Reply With Quote
  #10  
Old Fri 20 February 2009, 07:03
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
I also agree it is an empirical formula developed by a conservative engineer. I have pushed the voltage a bit higher and everything was still okay. The current is the absolute limit, and there I don't push my luck.

My theory is that if I push the voltage too high, and get a motor that is too hot, then I can easily reduce the current setting.
Reply With Quote
  #11  
Old Fri 20 February 2009, 07:08
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
Jorge,
The formula works. I have tested it with more than twenty motors and all performed exactly as expected. When the motors are run hard (maximum load without loosing steps), at maximum computed voltage, their temperatures run very close to 80C (as measured with an Infa-Red thermometer. When the maximum voltage is de-rated 10% to 15%, the maximum motor temperature is about 60C.

On my test bench, I have a Variac, so I can easily adjust voltage going to a toroidal transformer. That means that I can dial in any voltage within the G203v's working range, 15VDC to 80VDC.

I've wired the motors half-coil, series, and parallel (for those motors that have eight leads). I've run the motors at speeds from 500 steps per second up to 45,000 steps per second. I run the motors without any load and with as heavy a load as I can consistently put on the shaft repeatably.

The formula saves a lot of testing. Before the formula was published, I started voltage testing a motor by giving it 15X its rated voltage and working upwards towards 25X of its rated voltage. Most of the time, a voltage 15X to 20X was all that I applied to the motors to keep the temperature within reason. Yesterday, I rewired a power supply to deliver 50.9 volts to a PK299-F4.5 motor wired half-coil. That is about 1/2-volt higher than the maximum voltage that the formula gives. I ran that motor for several hours as a B-axis (indexer) on my Shopbot to test some ideas that I have about running an indexer. The motor got toasty warm (but the shop was only 35F), so try as it might, the motor was not able to get the shop temperature up to my comfort zone.

By using the formula, you will be very close to the optimum voltage for a motor. Then, if you wish, you can adjust the voltage up or down until you get all the desired performance within the motor's rated temperature range.

Depending on the physical size of the motor, I usually run the motor at 80% to 90% of the maximum computed voltage. I run the PK296A2A-SGxx motors with a 35V power supply, which is 89% of maximum. I run the PK268-02AA motor at 50VDC, which is 83% of maximum. I run the PK299-F4.5 at 50.9V, which is 100% of maximum. All motors use half-coil wiring and all motors use current limiting resistors selected to give 90% to 110% of their rated current. (I don't always have resistors that exactly match the requirements and I just won't drive twenty-miles to buy a ten-cent resistor.)
Reply With Quote
  #12  
Old Fri 20 February 2009, 12:10
CNCNow!
Just call me: Jorge
 
Sao Paulo - SP
Brazil
Gerald, Brad and Mike, thank you gentlemen for your time and attention !

It´s getting increasingly interesting !

Mike, I love your detailled data !

Considering that motor heat comes from losses [ Joule´s effect, Eddy currents and hysteresis ] and none of them are directly voltage related, you´ve got me puzzled !

I want to try to replicate your tests and I´ll get back to you with my findings as soon as possible ...

Hopefully an old dog can learn a few new tricks, huh ?
Reply With Quote
  #13  
Old Fri 20 February 2009, 13:10
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
Here's a portion of "Step Motor Basics", a white paper that Mariss wrote several years ago. Most of us have read it over and over. Some of the paper has become dated, but it's always a very good read.

MOTOR HEATING AND POWER SUPPLY VOLTAGE
There are two major causes of motor heating; copper losses and iron losses. Copper losses are the easiest to understand; this is the heat generated by current passing through a resistance, as in the current passing through the motor’s winding resistance. Often this referred to as “ I squared R” dissipation. This cause of motor heating is at a maximum when the motor is stopped and rapidly diminishes as the motor speeds up since the inductive current is inversely proportional to speed.

Eddy current and hysteresis heating are collectively called iron losses. The former induces currents in the iron of the motor while the latter is caused by the re-alignment of the magnetic domains in the iron. You can think of this as a “friction heating” as the magnetic dipoles in the iron switch back and forth. Either way, both cause bulk heating of the motor. Iron losses are a function of AC current and therefore the power supply voltage.

As shown earlier, motor output power is proportional to power supply voltage, doubling the voltage doubles the output power. However, iron losses outpace motor power by increasing non-linearly with increasing power supply voltage. Eventually the point is reached where the iron losses are so great that the motor cannot dissipate the heat generated. In a way this is natures’ way of keeping someone from getting 500 hp from a size 23 motor by using a 10,000 volt power supply.

At this point it is important to introduce the concept of overdrive ratio. This is the ratio between the power supply voltage and the motor’s rated voltage. An empirically derived maximum is 25:1. That is to say, the power supply voltage should never exceed 25 times the motor’s rated voltage.
Reply With Quote
  #14  
Old Fri 20 February 2009, 17:10
dragonfinder1
Just call me: Dave #49
 
Astoria, Oregon
United States of America
wire gauge

Mike

I'm sure this has been covered before but I can't find it. What should the wire gauge be for wiring the motors? I'll have 45v DC and 3.5 amps. The electricans where I work suggest 18 gauge. That seems a little big, They really don't have much experence with motors that small. Their reason is that the DC voltage will drop off fast and cause the motors to over heat. The wire length will be about 30 feet.

Dave
Reply With Quote
  #15  
Old Fri 20 February 2009, 17:16
smreish
Just call me: Sean - #5, 28, 58 and others
 
Orlando, Florida
United States of America
18 Gauge stranded copper is the recommended motor wiring.
Reply With Quote
  #16  
Old Fri 20 February 2009, 17:24
dragonfinder1
Just call me: Dave #49
 
Astoria, Oregon
United States of America
Quote:
Originally Posted by smreish View Post
18 Gauge stranded copper is the recommended motor wiring.
Thanks for the quick reply. i'll get on order

Dave
Reply With Quote
  #17  
Old Fri 20 February 2009, 21:13
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Dave, please read this:
Using quotes when replying to a post - only use if essential please
thanks
Reply With Quote
  #18  
Old Sat 21 February 2009, 13:42
dragonfinder1
Just call me: Dave #49
 
Astoria, Oregon
United States of America
Gerald

July of last year you stated that you were going to start using 21 gauge screened wire for your motors. Have you had any problems using that wire?

dave
Reply With Quote
  #19  
Old Sat 21 February 2009, 21:13
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Dave, we purchased it, but havn't assembled those machines yet.
Reply With Quote
  #20  
Old Mon 08 July 2013, 09:38
IPenev
Just call me: Ivan
 
Varna
Bulgaria
Quote:
Originally Posted by Gerald D View Post
Jorge, that formula was "discovered" by Mariss at Geckodrive about 2 years ago. We soon realised it this new formula gives the best results - a balance between motor heat and motor torque.

(The previous formula used for Geckodrives was 20 to 25 times the voltage stated by the motor manufacturer)

Remember, we are talking of the DC voltage supplied to the drive electronics. The drive supplies something completely different to the motor - that is a digital signal chopped into tiny sections.
I was so confused between the two, thank you.
Is this formula discovered experimentally or there are some calculations behind it, I would like to understand it better
Thanks
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 09:28.


Powered by vBulletin® Version 3.8.3
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.