Author Topic: Can Capacitance be Pulse width Modulated?  (Read 2777 times)

0 Members and 1 Guest are viewing this topic.

george65

  • Sr. Member
  • ****
  • Posts: 499
  • Country: au
Can Capacitance be Pulse width Modulated?
« on: March 05, 2017, 02:21:22 AM »

I have been playing with induction motors set up as generators by using caps for some years now.
The big problem with this is switching loads in or out and keeping the voltage from collapsing due to not having enough capacitance to maintain the field strength or overshooting on the voltage through having too much capacitance.

I'm wondering  If something like an Arduino could be used to monitor the output from an IMAG and control the amount of capacitance fed in according to the voltage level?

Given the Voltage rise and fall is slow unless very large loads are applied or removed, I'm sure the speed of something electronic would be able to handle it quite easily.  The "Normal" control for these setups is to have a controller which runs the output at a constant near full output and just divertys the amount of power required for the load to be driven from a dump load.

By increasing the amount of capacitance fed into the IMAG, I'm thinking ( if it's possible) this should work like an AVR and keep things within an acceptable and workable range.

Could it work or is it not possible to do this?

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: Can Capacitance be Pulse width Modulated?
« Reply #1 on: March 05, 2017, 02:53:36 AM »
I recall reading about a capacitance regulated induction motor generator.

Basically the capacitors were switched in and out of the circuit at the peak voltage of the sine wave, rather than the zero crossing. The capacitors when they were switched off had a diode and resistor keeping them charged at the peak voltage of the generator, so there were no current spikes during switching. the capacitors were some multiple of 1,2,4,8 uf iirc, i think there were 8 of them.


another option is a servo controlled variable transformer, but that is.. old school.
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

george65

  • Sr. Member
  • ****
  • Posts: 499
  • Country: au
Re: Can Capacitance be Pulse width Modulated?
« Reply #2 on: March 21, 2017, 10:38:43 AM »

Joe you gave me a heads up with at least what to search for.
Found one paper that suggests it is in fact possible in this application. They seem to suggest having a fixed capacitor and then having a controler to create a variable capacitor for the fine tuning as it were. From there as to actually how to do it goes right over my head.

Guess I'll have to get a PWM controller and a SSR and give it a try.
If it works then maybe I can work something out for an arduino or use one of the Chinese Voltage level boards and take advantage of their quick reaction times and bypass the mechanical relay with an SSR connecting another cap. 

Mary B

  • Administrator
  • SuperHero Member
  • *****
  • Posts: 3169
Re: Can Capacitance be Pulse width Modulated?
« Reply #3 on: March 21, 2017, 04:09:46 PM »
There are no voltage variable capacitors that would work for this, all have far to low of capacitance(varactors, used to tune radio circuits in one application). So switching large caps in and out is your choice.

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: Can Capacitance be Pulse width Modulated?
« Reply #4 on: March 21, 2017, 06:07:57 PM »
i might have an idea what document George is talking about.

if its talking about pwm'ing a split capacitor motor to get better starting torque and higher power factor, efficiency, i don't think it will work for this application.

i would seriously suggest a servo driven variac +capacitor, or go the 2,4,8,16uF route with switches.

if you don't have to switch the capacitors very often then hard switching them through a NTC (the type used for inrush limiting on power supplies) is probably very reasonable if the difficulty of timing the switching is too much, or if mechanical relays are used instead of triacs
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

george65

  • Sr. Member
  • ****
  • Posts: 499
  • Country: au
Re: Can Capacitance be Pulse width Modulated?
« Reply #5 on: March 22, 2017, 01:20:02 AM »

I tried Pulse width Driving an SSR and it doesn't work with the DC PM I have. Works fine with the AC one so they must be different frequencys.
I thought of using mechanical relays with a tight tolerance on the voltage parameters but then I also realised that the voltage controllers are all DC not AC.
 I then thought of using a voltage Divider to monitor the power but can only find them to about 30V DC as well.

Seems every idea meets a brick wall.

How would I hook up the variac to the capacitors?  Have the supply to the variac from the IMAG and the output to capacitors or ???

george65

  • Sr. Member
  • ****
  • Posts: 499
  • Country: au
Re: Can Capacitance be Pulse width Modulated?
« Reply #6 on: March 22, 2017, 01:24:54 AM »

This was the paper I found but it's over my head to work out what they are using or how they are going to control the capacitance.

http://lipo.ece.wisc.edu/1998pub/C131.pdf

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: Can Capacitance be Pulse width Modulated?
« Reply #7 on: March 22, 2017, 06:57:23 AM »
Yeah the variac is used to reduce the voltage to the capacitor. A 50% reduction in voltage makes a 100uf cap look like 25uf to the motor. Only problem is you sort of need 3 of them.

Anyhow i read that document some time ago and it seems simple enough. You need 3 ac switches across the capacitors and you short out the capacitor when it is at zero volts. The longer you short it out electrically it appears larger than it is. This might be fairly simple to program. But it generates harmonics in the output.
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

george65

  • Sr. Member
  • ****
  • Posts: 499
  • Country: au
Re: Can Capacitance be Pulse width Modulated?
« Reply #8 on: March 22, 2017, 07:44:19 AM »

Seems simple but all too hard.
I saw some pre built commercial IMAG controllers a while back but can't any more. Maybe they gave up on them as sales were too slow or something?

Is there any way an arduino can sense AC to output a PMW signal or are they limited to DC?

DamonHD

  • Administrator
  • Super Hero Member Plus
  • *****
  • Posts: 4125
  • Country: gb
    • Earth Notes
Re: Can Capacitance be Pulse width Modulated?
« Reply #9 on: March 22, 2017, 07:54:17 AM »
Put an opto isolator input in each direction across each of the three phases (with suitable drop resistors) and let the Arduino see the (6) outputs on its digital GPIO lines.

That would be my first attempt.

Rgds

Damon
Podcast: https://www.earth.org.uk/SECTION_podcast.html

@DamonHD@mastodon.social