MechMate CNC Router Forum

Go Back   MechMate CNC Router Forum > Electrical & Electronic > 702. Power Supplies
Register Options Profile Last 1 | 3 | 7 Days Search Today's Posts Mark Forums Read

Reply
 
Thread Tools
  #1  
Old Tue 05 February 2008, 11:58
Kevin
Just call me: Kevin
 
Canton, NC (In the Smoky Mountains)
United States of America
I have a question about the size of the capacitor required for a power supply. I am using PK296A2A-SG7.2 motors and an Avel Y236801 500VA 25V+25V Toroidal Transformer. It is my belief that a 50 volt is a high enough rating but I am uncertain as to the mf size. I see that 10,000 mf is used by some. Is it a problem if I use a 22000mf 75V Computer Grade Capacitor? And can someone explain what is required to add a bleeder resistor to drain a capacitor if necessary...

Thanks, Kevin
(quietly working away in AZ)
Reply With Quote
  #2  
Old Tue 05 February 2008, 12:11
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
Kevin,
A 22,000uF 75V capacitor will work fine. I use 15,000uF to 20,000uF on my power supplies. Also, I use a capacitor whose DC rating is at least rated as high as the working voltage, which in your case would be 70VDC, so you're perfectly safe with both the capacitance and the voltage.

The bleeder resistor (if you use one) can be screwed to the capacitor's + and - terminals. All that it does is to drain or "bleed" off the electricity when the power supply is turned off. Capacitors can hold a charge for a very long time, so it's a safety feature. As Gerald has pointed out, with four stepper motors that act as bleeder resistors, a separate bleeder resistor is not necessary when used with multiple stepper motors. If you decide to use one, you can use a 47K ohm 1-watt resistor with a 70VDC power supply. That size resistor will still take about a minute to drain off most of the voltage and it will get warm, but it is a good compromize size. If you want the bleeder resistor to drain the capacitor faster, you could use a 4.7K 5-watt resistor. That would still take several seconds and that resistor will probably get HOT - so, be careful.
Reply With Quote
  #3  
Old Tue 05 February 2008, 12:14
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
A 25V+25V transformer will give you 50V Alternating Current (AC), when that is rectified to Direct Current (DC) it will be 1.41 times higher; ie. just over 70VDC. A 50V capacitor will pop (explode) if you put it with that transformer and rectifier. Don't forget the rectifier (that will make a big BANG). The 75V capacitor will just make it. 22000 uF is plenty.

If you have 4 drives connected to the capacitor, they act as a good bleed resistor. You don't need a separate bleed resistor unless you are just messing around on the bench - but then you can just use a loose gecko as the resistor.
Reply With Quote
  #4  
Old Tue 05 February 2008, 12:22
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Ah, Mike and I were typing at the same time - glad we more or less said the same thing!

Just a note on capacitors - they do get old and wear down. If it is just capable of 75V today, it will be capable of less in 20 year's time. This is one area where you avoid buying war-surplus of dubious age and origin.
Reply With Quote
  #5  
Old Tue 05 February 2008, 19:14
Doug_Ford
Just call me: Doug #3
 
Conway (Arkansas)
United States of America
Kevin,

Post some pictures of your work so other guys can learn and cheer you on as you progress.
Reply With Quote
  #6  
Old Thu 04 September 2008, 19:35
Robert M
Just call me: Robert
 
Lac-Brome, Qc
Canada
Send a message via Yahoo to Robert M Send a message via Skype™ to Robert M
Capacitors sources ?

Any one knows some « Online » good sources for capacitors ?
Don’t laugh, I can’t seem to find any even after some Goggle searching
Thanks
Reply With Quote
  #7  
Old Thu 04 September 2008, 20:51
domino11
Just call me: Heath
 
Cornwall, Ontario
Canada
Robert
Try www.digikey.com , www.mouser.com, www.newark.com , www.e-sonic.com. But my favorite place in montreal is www.addison-electronique.com 514-376-1740 8018 20e ave Montreal. They are fantastic for electronic parts.
Reply With Quote
  #8  
Old Fri 05 September 2008, 07:07
Robert M
Just call me: Robert
 
Lac-Brome, Qc
Canada
Send a message via Yahoo to Robert M Send a message via Skype™ to Robert M
Thanks Health…
Tried Addison a fee weeks back… Believe it or not, it took over 4 clerks to understand (immigrants) & know what the hell I was requesting… only to find out they could not find or order most items I was looking / requesting for.
They’ve gone more into cheep Audio and low quality commercial overstock. Not the good old Addison electronics store I knew back in the 80’s
Robert
Reply With Quote
  #9  
Old Fri 05 September 2008, 08:32
domino11
Just call me: Heath
 
Cornwall, Ontario
Canada
Robert,
What size capacitor are you looking for, I might be able to fix you up.
Reply With Quote
  #10  
Old Fri 28 November 2008, 05:57
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
Copied from another thread:

. . . . . . a suitable capacitor; however, finding adequately sized capacitors with screw terminals is becoming more and more difficult, since most designers have switched to the "snap-caps" that are generally soldered onto a circuit board. (I recall seeing some photos of snap-caps being hot glued to a thin piece of MDF. That would work although I would prefer drilling holes through the MDF and securing the snap-caps with nylon ties.)
Reply With Quote
  #11  
Old Fri 28 November 2008, 07:27
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
"Snap-caps" versus screw terminals

For soldering, Snap-in capacitors ("snap-caps"):





Screw terminal capacitor:


Terminals have female threads - screws not shown.
Reply With Quote
  #13  
Old Wed 14 January 2009, 07:48
Nikonauts
Just call me: Nikonauts
 
Johore
Malaysia
in case nobody notice, thread starter said 22,000mF.

I made a mistake by buying 15000uF, only to later realise that gecko recommends (80000*I)/V minimum. So for my config (50V, 14 A) it's 22,400uF minimum. I have a spare caps 1600mF 400V, given by a friend. Worry it might be too big.

Is it too big?
Reply With Quote
  #14  
Old Wed 14 January 2009, 08:44
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
You PROBABLY have a 1,600uF 400V capacitor. It was common practice several years ago to use either uF or mF to mean 0.000001 farad. The lowercase letter "m" is used for micro. The uppercase letter "M" is used for mega. An MF is 1,000,000 farads. An mF is 0.000001 farad.

It's my understanding that before the U.S.A. even admitted that there was a "metric system", that mF was used interchangeably with uF.

The 1,600 mF capacitor is too small, but the 15,000 uF capacitor will probably work just fine. I use 11,000uF capacitors on various portions of my test bench to drive four stepper motors. The motors run just fine. Remember that a stepper motor draws more current the heavier that it is used. Most of the time, most steppers are drawing far less than their maximum current, so MOST of the time, the drain on the capacitor is less than maximum. IF you are going to ALWAYS run your machine at its maximum load, THEN you might need to use a larger capacitor. Capacitors can be connected in PARALLEL to increase the capacitance.

It's good practice to use a capacitor that has a working voltage about 2X the output voltage of the AC transformer because the capacitor will be expected to handle the PEAK voltage, not the average (RMS) voltage. The peak voltage is 1.414 X the average voltage. So, using a capacitor with 2X the voltage rating of the transformer gives you a safety margin. Using a capacitor that has a higher voltage rating than 2X the output AC won't hurt, but unless you just happen to have that size capacitor on hand, it will be larger and more expensive than necessary. Also remember that capacitors have a limited life. The dielectric inside a capacitor will eventually dry out. Unless a capacitor is overheated, that process takes years. It's good practice to examine the capacitors occasionally to see if they're leaking or bulging. It's also good practice to read a power supply's AC ripple (with the power supply under load). Most volt meters can read the AC ripple on a DC power supply by simply using the AC setting. If the ripple starts to go higher (more volts) and if the load on the power supply hasn't changed, then it's time to start thinking about getting new capacitors.
Reply With Quote
  #15  
Old Wed 14 January 2009, 10:19
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Small m, or lower case m, is the metric abbreviation for milli, being 0.001, or 1/1000th. Whenever a metric based person sees m in front, that is 1/1000th. . . . . . undoubtedly, no exceptions.

If NikoN says he has a 1600mF 400V capacitor, that is 1 600 000 (1.6 million) uF, or 1.6 Farad, which sounds like it needs a forklift to move around. If that "1600mF" capacitor can be carried in one hand, it is probably a 1600 microFarad, particularly if it is 400V rated.
Reply With Quote
  #16  
Old Wed 14 January 2009, 10:36
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
A snag with the metric system is that the correct symbol for micro is µ , a character not found on any common keyboard. The nearest character is the u.

Here are all the metric multiplier names and symbols: http://en.wikipedia.org/wiki/Micro-
Reply With Quote
  #17  
Old Wed 14 January 2009, 12:28
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
Gerald is right. I should have said that the lower case letter "m" WAS commonly used by capacitor manufacturers in the USA to represent "micro" back when we rebels (USA metric holdouts) were largely ignorant of the metric system.

I had to dig deep to find a capacitor that had the old style marking, but I found four in the power supply of my first computer, a 1977 IMSAI. The caps are soldered in, so I couldn't see the complete value, but "mF" was clearly visible followed by "MFD".

(That computer was built sometime before Jerry Pournelle, a writer for the defunct BYTE magazine, wrote in amazement that Bill Godbout was offering power supplies in his CompuPro brand computers that featured a bank of capacitors totaling 0.8 FARARD. Bill Godbout was thee premier computer manufacturer pre-Apple and pre-IBM P.C. My first trip to California was around 1980 to visit the Godbout factory to get some help setting up some CompuPro computer cards. If I remember correctly, the theory behind having that much capacitance was due to the fact that the S-100 bus computers used an unregulated 8VDC to 12VDC power supply. Each card in the computer had one or more 7805 type voltage regulators to convert the unregulated voltage to the required 5VDC. Since dropping voltage across a regulator produces heat and heat was the largest cause of chip failure - at the time - Bill Godbout built power supplies that furnished 7VDC so that less voltage had to be dropped across the regulator; therefore the regulators would run cooler and he would have fewer boards returned because of heat-caused stress failures. The 7805 chip will not work if the AC ripple is too great, after all, the regulator works by blocking extra voltage, not by producing voltage when the supply is insufficient. Because Godbout could safely use 10V capacitors, the capacitors' mass was less than you would expect and the idea worked for a very short time until Apple and IBM introduced switching power supplies to the computer world. Shortly before Apple and IBM took over the market, one switching power supply manufacturer offered a multi-voltage power supply that was specifically designed to produce regulated voltages high enough so that the S-100 bus computers, with their on-board regulators could be used with that power supply.)

Most of that is totally off-subject, but it might explain how things worked back in the wild-woolly days when personal computers were first being introduced to the masses.

Edited: To those who would wonder why you shouldn't just use a regulated switching power supply: A switching power supply works best when its load is constant. A stepper motor creates a load that is constantly varying. The variation can be extreme. When a stepper is coasting along, it may draw 25% of the current that it requires when it has to really work. Some switching power supplies cannot cope with that type of load.

Last edited by Richards; Wed 14 January 2009 at 12:38..
Reply With Quote
  #18  
Old Wed 14 January 2009, 12:48
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
"If you car lights dim when your car audio system produces a deep bass note, then your amp will be greatly helped with a car audio capacitor." <---source

Those awful booms emitting out of cars are typically powered by huge capacitors at supposedly around 2 Farad (example). However, the voltage ratings are low, and it is doubtful whether the Farad rating values are genuine, or hyped up for the car audio industry.
Reply With Quote
  #19  
Old Wed 14 January 2009, 18:41
jhiggins7
Just call me: John #26
 
Hebron, Ohio
United States of America
I've noticed the terms "computer rated" and "audio" used to describe capacitors. If I have two 10,000 microfarad, 100 volt capacitors, and one of them is described as "computer rated" and the other is described as "audio," is there any real difference?

Regards,
John
Reply With Quote
  #20  
Old Wed 14 January 2009, 20:00
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
I suspect the major difference is quality?

http://www.tedss.com/electronic-part...de-capacitors/
Reply With Quote
  #21  
Old Thu 15 January 2009, 06:17
Nikonauts
Just call me: Nikonauts
 
Johore
Malaysia
Quote:
Originally Posted by Gerald D View Post
Small m, or lower case m, is the metric abbreviation for milli, being 0.001, or 1/1000th. Whenever a metric based person sees m in front, that is 1/1000th. . . . . . undoubtedly, no exceptions.

If NikoN says he has a 1600mF 400V capacitor, that is 1 600 000 (1.6 million) uF, or 1.6 Farad, which sounds like it needs a forklift to move around. If that "1600mF" capacitor can be carried in one hand, it is probably a 1600 microFarad, particularly if it is 400V rated.
thank you, Gerald. It's the size of a Red Bull can. So i think it's 1600uF mislabeled as 1600mF.

So if i later want to add one more capacitor to the 15000uF, i guess it'd be in parallel to the one now?
Reply With Quote
  #22  
Old Thu 15 January 2009, 06:42
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Yes, in parallel.
Reply With Quote
  #23  
Old Sun 15 February 2009, 07:05
kanankeban
Just call me: Hector #89
 
Monterrey
Mexico
Bleed resistor and charge indicator?

Hi...
Im starting to think in my power supply...Im kind of paranoid with the charge of the capacitors I dont remember in full, but I think the paranoid factor was aquired when I was a child, and messed around with a capacitor .
My questions are,
1. why we dont need a bleed resistor if the geckos are conected?
2. although question 1 can we add a bleed resistor for extra protection?
3. can we add a kind of LED or warning light to inidcate that the capacitors are charged and warn not to play with them?
4. What are the logics of the bleed resistor and how is it installed on to the capacitors?
Thanks...
Reply With Quote
  #24  
Old Sun 15 February 2009, 08:06
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
1. The Gecko stepper drivers act like bleeder resistors, i.e, the coils in the motors are connected to both + and - on the power supply, just like a bleeder resistor.

2. Yes.

3. You could add an LED; however, an LED uses Current to work, about 10mA. To get 10mA at 70V would require a 700 ohm resistor at 7W! You would want to use at least a 25W resistor to keep the heat down and a 100W resistor would be much better.

4. Connect one end of a bleeder resistor to the + (plus) terminal of the capacitor and the other end of the bleeder resistor to the - (minus) terminal on the capacitor. A 4.7K 5W resistor would work for a 70VDC power supply. A 1.5K 5W resistor would work with a 35VDC power supply. Just keep in mind that the stepper drives will have already drained the capacitor before the bleeder resistor does its job. The only time that the bleeder resistor would be necessary is if you have fuses installed between the power supply and each stepper and if all the fuses blow.

The value (resistance) of the bleeder resistor will determine how long it takes to empty the capacitor. The lower the resistance, the faster the capacitor will be drained. The lower the resistance, the more heat generated by the resistor. I like to use a resistor that is at least 5X larger (wattage) than required, to keep the heat reasonable.

A manually operated power dump circuit might be an option. If you use multi-pole E-stop switches, one of the Normally Open poles could be connected to a large transistor or FET. Then, when you pushed the E-stop switch, the N/O pole would close and the transistor or FET would quickly drain the capacitor. When you released the E-stop switch, the capacitor would be allowed to charge.

Personally, I just use the stepper motors as bleeder resistors. If I have any worry that the capacitor is still charged, I take a reading with my meter. If there is a voltage present, I grab my largest flat screwdriver with the thickest plastic handle and place the blade of the screwdriver across the capacitor's terminals. THAT drains the capacitor in a hurry!
Reply With Quote
  #25  
Old Sun 15 February 2009, 10:18
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Quote:
Originally Posted by kanankeban View Post
3. can we add a kind of LED or warning light to inidcate that the capacitors are charged and warn not to play with them?
The LED's on the geckos serve that purpose - do not play with anything while there is light from those LED's.
Reply With Quote
  #26  
Old Sun 15 February 2009, 15:49
kanankeban
Just call me: Hector #89
 
Monterrey
Mexico
Quote:
Originally Posted by Gerald D View Post
The LED's on the geckos serve that purpose - do not play with anything while there is light from those LED's.
Thats a good advise for those like me with some electric & electronic disabilities , Thanks...
Reply With Quote
  #27  
Old Mon 16 February 2009, 05:02
javeria
Just call me: Irfan #33
 
Bangalore
India
Quote:
Originally Posted by Richards View Post
, I grab my largest flat screwdriver with the thickest plastic handle and place the blade of the screwdriver across the capacitor's terminals. THAT drains the capacitor in a hurry!
Well - dosen't this shorten the life of the capacitors? good practice or bad or just to use when in hurry?

RGDS
Irfan
Reply With Quote
  #28  
Old Mon 16 February 2009, 05:34
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
So far, I've never blown all four fuses going from the power supply to the Gecko stepper drivers, so I've never had to drain a capacitor manually.

Too many times new users keep asking the "what if" questions. After being told that the stepper motors would act as bleeder resistors, some ask, "yeah, but what if they don't"? That leads the explanation of adding a bleeder resistor. When they see how large and unwieldy that bleeder resistor is, then they ask, "but what if you don't have a bleeder resistor and the stepper motors don't drain the capacitor"? That leads to the screw-driver.

The answer to the question is that when more than one stepper motor is attached to the same power supply, you don't need to add a bleeder resistor. Period.
Reply With Quote
  #29  
Old Wed 08 July 2009, 07:49
KenC
Just call me: Ken
 
Klang
Malaysia
Choosing Reservoir Capacitor 101

I build Audio amp in my other spare time... hope my 2 cent worth could be helpful.

select Cap voltage rating > or = 2 x Vac.

Calc Min. Farad.

Case.
Transformer secondary Voltage, Vsec= 40Vac,
Required DC current rating, Idc=12A
Max. final ripple voltage of DC, Vripple , (typically 5%)

Vpeak= sqrt(2) x Vsec= 56.56 V
so,
Vripple(pk-pk)= Vpeak x 5% = 2.828V(pk-pk)

For 50 hz mains,
Cmin. = 0.01 * Idc / Vripple(pk-pk) = 0.01 * 12 / 2.828 =0.042434F = 42,434 uF##

For 60hz mains,
Cmin. = 0.00833 * Idc / Vripple(pk-pk) = 0.035360 =35,360 uF

Vdc =Vpeak - (Vripple/2) - Voltage drop accross bridge = 56.56 - 2.828/2 -1.4V= 53.746 V

You can decide on what type of ripple voltage you want and size the minimum cap size accordingly.

Hope this help.

Last edited by KenC; Wed 08 July 2009 at 07:51.. Reason: correct mistake
Reply With Quote
  #30  
Old Wed 08 July 2009, 08:15
KenC
Just call me: Ken
 
Klang
Malaysia
Parrellel coil or 2 rectifier? Series coil? Inrush current?

From the above, I'm sure you could see the advantage of using separate rectifier, using 2 bridge, the current involve in each filter circuit halved and so does the cap size.

One can determine if the coils output voltage will equal when the measured resistor of coils are identical...

Richard is right, forget about the bleeder resistor.

Inrush current upon cold start can happen when cap is large... reduce the cap size or connect a NTC (negative temperature coeficient) thermister in series with the primary coil. The NTC thermister has high resistance when cold and zero resistance when reach operating temp.

The safest way to "series" up coils is after it is rectified and filtered. then you don't have any risk of connecting the coils in opposite phase. The only way to determine the phase of the coil is by looking at the AC waveform in a expensive oscilloscope.... now who want to do that!

Last edited by KenC; Wed 08 July 2009 at 08:20.. Reason: spell...
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 10:52.


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