Hello,
Thanks for all your replies on my encoder question, now I have another. How am I to build a power supply to support 3 24VDC 40A (peak) Servo motors? We've got 120VAC and 220VAC coming into our house, but I'd rather use 120 if possible. I've seen some how to's, but they seem to go up to only 10A or 20A. Can anyone help?
Thanks,
Brooks
The secret is this: Use a toroid transformer with the output rated at 1000 watts and 17 volts. Use rectifiers rated at least 50 Amps. The capacitor(s) should be rated at 50 VDC. When the output of the toroid goes through the rectifier converting the AC power to a pulsating DC voltage and the pulses are filtered by the capacitor(s), the DC voltage will be 24 Volts DC.Originally Posted by bprzybyl
17 Volts X 1.414 = 24.038 Volts.
1000 Watts divided by output voltage (24) = 41.666 Amps.
24 VDC times 41.666 Amps = 999.98 watts or 1 Kilowatt.
Use one of your "how to's" to build a 10 or 20 amp supply with the above rated components to build a 40 Amp supply.
You will need a power supply for each motor if you intend to run them at the max ratings. Question: Where are you going to get the servo drivers as they normally are rated at 25 or 30 amps?
The truth is you only need about a 20 amp supply for each. Any excessive current will pull down the voltage, but the encoders (required on servo systems) will keep the machine position regardless of the voltage.
Hope this helps.
I've been looking at them, and I see a VA rating. Is this a wattage rating? If so, then I would need a 1000VA 17V Transformer, correct?
Also, where did you get your 1.414 number when changing 17v to 24v.
Thank you very much,
Brooks
Yes- Volt Amp rating and watts is the same thing.
Yes -Just like my previous post stated. You need a 1000 VA 17 Volt transformer.
The 1.414 is a constant that is used when using a full wave bridge rectifier to convert the AC voltage to DC. It has to do with the peak DC voltage value when averaged out (or Filtered) by the capacitor.
.
1/Actually the reason VA is used for an AC component like a transformer and not watts is the VA does not neccessarily mean the same thing as Watts because VA is 'Apparent' power, there is a thing called power factor comes into play as with a capacitive or inductive device on AC the point of peak current and peak voltage may not coincide, therefore you cannot use the W=IxV equation.
Do a search for Power Factor if you need more information.
2/When you measure an AC voltage it is usually measured and indicated as RMS (Root Mean Square) value, This is a AC Sine wave value that provides the same 'heating' or power effect equivalent to a DC value,
When rectified and smoothed by adding capacitance, the resulting voltage is now a product of the peak AC which is Peak = to 1.414xRMS.
Or the RMS is .707 x peak.
All the above falls into the realm of AC theory, and tends to get a bit more complex than pure DC, where simple Ohms law formula applies.
Al.
CNC, Mechatronics Integration and Custom Machine Design (Skype Avail).
“Logic will get you from A to B. Imagination will take you everywhere.”
Albert E.
You don't need that big of a transformer - That power would assume demagnitizing current being supplied to all three axis at once.
You NEVER, repeat - NEVER want to bring your servo to the 'PEAK'. The motor will over heat and demagnitize in short order. Short runs over the continous current rating of the motor are fine as long as the average over any short time (2-3 minutes) is less than the continuous. The motors have a high thermal time constant and will take some time to heat up and cool down. But these short runs should not approach the published peak current.
With this is mind, the transformer is also rated at continuous current and can take short runs above it's rated capacity without damage. You will need to size your capacitor banks to handle the ripple current to get decent life out of them either way and you can pretty much assume the delivered current is the ripple current for sizing
Also, the transformer voltage will sag at load and will not deliver the full spec power into a bridge rectified supply unless specifically called out. This is an addition to the voltage you loose through the bridge diodes and the controller FETs. The amount of voltage sag is highly dependant on transformer geometry and should be listed by the manufacturer as a table of voltage vs current. You will need to compensate for this drop by increasing the transformer output voltage without going over the maximum for your controller
The 1.414 is actually the square root of two and comes from calculating the Root-Mean-Squared value of a sinusiod.
If you can post more info on your motors, we may be able to give you a more realistic estimate for your power needs.
At 1000VA, I would suggest a power factor corrector to keep VA = Watts but this would most probably beyond the capabilities of most hobbiests and is probably not needed in this case
Aaron
I'll post everything I can find here, but in case you need any more information, it can be found here http://www.pennmotion.com/pdf/lcm_bulletin.pdf. Here are the specs-
M14437A105
Motor Data-
Cont Torque- 50 oz-in
Peak Torque- 410 oz-in
Motor Constant- 13.1 (oz-in)/(W)^(1/2)
No load speed- 3211 RPM
Friction Torque- 2.20 oz-in
Rotor Inertia- 6.7x 10^-3 on in^2
Electrical Time constant- 1.5 ms
Mechanincal Time Constant- 5.5 ms
Viscous Damping(Infinite Source Impedance)- 0.25 oz·in/krpm
Viscous Damping(Zero Source Impedance)- 127 oz·in/krpm
Max Winding Temp- 311 °F
Thermal Impedance- 41.0 °F/watt
Thermal Time Constant- 32.3 min
Motor Weight- 54.5 oz
Motor Length- 5.703 in
Winding Data is shown below.
That's everything I've got as far as information goes.
Thanks for all your help,
Brooks
(50 oz-in)/( 10 oz-in/A) = 5A continous current
You really only need a 15 Amp supply. This will cover most situations because you normally would not be moving all three axis at max current.
You may want to figure in extra capacity for a 4-axis now if you can fit it in the cost.
Take a look at page LCM-17, you want to stay under the red line at all times and don't want to venture far from the darker shaded area.
To figure the voltage needed to get a certain torque at a certain speed
1 - get the needed current from line 31 ( torque/10 )
2 - figure voltage to drive the current through the motor resistace (answer 1 * 0.59)
3 - Figure Back EMF using line 39 ( RPM/135.32)
4 - Add 2 to 3
There are several free motion control software estimators that can give your peak torque needs
Aaron
Last edited by pastera; 08-25-2006 at 12:03 PM.
Sorry for taking so long to reply (I HATE WORK).
So one 15A PS for an entire machine? I do plan on adding a 4th axis, but it will be a stepper, and I believe it draws only 3 amps. So I would be OK either purchasing or making a ~20A 24VDC power supply? But what happens if the motor starts to draw 40A? Does that burn out components? If anyone has any questions as to the size of my mill, it will be a larger (not much) version of swede's mill (www.5bears.com).
Thanks for all your help so far,
Brooks
Lots of good info here- thanks for posting!
Do you guys know anyone in the Zone/Ebay community who deals in custom power supplies?
I'll need one for my servo system.
Thanks!
They are simple to build, I often get Toroids off ebay from this guy, he sells all configurations.
Item number: 270092840898
You just need a Bridge rectifier and Suitable Electrolytic Capacitor to complete.
The advantage with a toroid it is relatively easy to modify to the voltage you want if you get one that is close and add or subtract windings to tailor to the voltage you want.
Al.
CNC, Mechatronics Integration and Custom Machine Design (Skype Avail).
“Logic will get you from A to B. Imagination will take you everywhere.”
Albert E.
Al, shouldn't a bleeder resistor be included as well? How do you calculate bleeder resistor size?
Posting Permissions
LinkBack URL
About LinkBacks
