Brushless DC Speed Control

Discussion in 'General Electronics Chat' started by leahy268, Aug 1, 2012.

  1. leahy268

    Thread Starter New Member

    Aug 1, 2012
    11
    0
    Hi,

    Ok so heres the deal. New to doing this so bare with me if I sound like a N00b.. I am.
    I am currently working on an automation setup.
    To make a long story short.
    Part of what I need to do is vary the speed of a small 24vdc pump to keep the water level even.

    I had originally planned to just do this with PWM however I now find that brushless dc motors don't work so well with this.
    It does however happily change speed by varying the voltage..
    To use pwm to vary the voltage(which is the only way that I can work out how to do it dynamically) I need a low pass filter?

    Is that correct?
    If so does anyone have some good circuits for these? I would need something to do 0-24v.
    I can only seem to find circuits for this designed to do very low current low voltage stuff.

    Thanks

    Warwick
     
  2. russ_hensel

    Well-Known Member

    Jan 11, 2009
    818
    47
    First I think it is unlikely that the pump uses a brushless motor unless you built it yourself. Are you sure it is brushless, and how do you know?
     
  3. leahy268

    Thread Starter New Member

    Aug 1, 2012
    11
    0
    Definitely brushless.
    That was the specs from the manufacturer.
    Long life brushless motor.
     
  4. luvv

    Member

    May 26, 2011
    186
    31
    Could you just vary the time on and time off to keep water level?

    EX. pump full on one minute, off two,on one minute off two etc.
     
  5. leahy268

    Thread Starter New Member

    Aug 1, 2012
    11
    0
    Not sure if that would work.

    I guess I need to go further into what it is doing.
    Essentially the pump is recirculating hot water through a bed of grain.
    Grain, eventually making beer..
    The idea is that you don't want to pump too quickly or you create massive suction and the grain compacts. You also don't want too slowly as the water (or now wort) comes out of the grain and then gets pump through a heat chamber. Too slowly causes the wort to heat up too much and scorch. Basically you need to keep a solid inch on top of the grain at all times. This would be done using a float switch. i.e. when over 2 inches slow the flow. When under an inch speed up the pumping.

    Does that make sense?

    Stopping the pump every 2 minutes would leave wort heating up in the heat exchanger and burning(not desirable)..

    Warwick
     
  6. luvv

    Member

    May 26, 2011
    186
    31
    Oh I see, you are not so much trying to control level,but flow rate.

    Try a mechanical approach, restrict the flow w/ a adjustable ball valve.

    Along w/ the float switch to control level,may take a little tuning but maybe do the trick.
     
  7. mcgyvr

    AAC Fanatic!

    Oct 15, 2009
    4,770
    970
    Just use a 0-30V variable voltage power supply.
    Do you know how much current the motor needs?

    A mechanical approach is simple too.. You can just use a T and a valve.. However you will be dumping good water which is wasteful. Better than spilling the beer though.. That would be a total waste.. :)
     
  8. leahy268

    Thread Starter New Member

    Aug 1, 2012
    11
    0
    I think the variable voltage power supply sounds good.
    Current wise it's only 0.8 A ..
    Are you talking about one that could be controlled in software? (i.e. microcontroller)..
    I have only really seen ones for the old turn the dial..

    Working with a T would be spilling beer :p (or at least wort)..

    The trouble with electronic control of valves to restrict the flow is mostly the cost involved.
    High temperature food grade, variably actuating valves are big $$$..
    I know of people who have made their own but you also need quite expensive stepper motors as the valves take a fair amount of torque to close.
     
  9. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,013
    1,531
    A way to keep the grain from compacting would be to put the water into the tank from the bottom. This would make the water rise to the top and then be recirculated from the top level, also eliminating the need for a "level sensor".

    Your "heating chamber" Should be temperature controlled, so it can't over heat.
     
  10. leahy268

    Thread Starter New Member

    Aug 1, 2012
    11
    0
    I think you maybe misunderstand me. Recirculating from the opposite way doesn't work. This would cause the water to push the grain bed up. The grain bed is actually what filters the wort. (Ending in crystal clear beer). The heating chamber would be temperature controlled, however we still don't want the wort getting over a certain temperature. I would imagine from a software point of view this would be a PID situation. Eventually arriving at the correct flow rate for that beer.

    Doing it from software would be the way since this would change depending on what grains are being used and how much and so forth.

    Thanks

    Warwick
     
  11. spankey666

    Member

    Nov 30, 2011
    91
    3
    ive been fitting and using brushless dc pumps (bosch type) for years on race bikes , varying the speed as engine temp rises, all done by simple pwm ciruits. my experience is they work well.
     
  12. leahy268

    Thread Starter New Member

    Aug 1, 2012
    11
    0
    Hmm I was lead to believe that it wouldn't work...
    Or is this a matter of getting the pwm frequency high enough that the electronics on the pump don't notice it? Not sure if I'm even in the ball park here.
    Can someone give me maybe a more definite answer on whether that will work or not and then why/why not?

    Perhaps my understanding is lacking..
     
  13. strantor

    AAC Fanatic!

    Oct 3, 2010
    4,302
    1,988
    I've read several times that PWM won't work, but I don't know if that was backed by experience or if it was just assumption. Spankey666 is the first person I've ever heard say that it does work, and he says it's from experience. My conclusion that it probably works in a few scenarios but mostly not. You would probably be making a mistake to plan on it working.

    I think that the reason it's usually stated that PWM won't work going into the motor, is that the circuitry inside the brushless motor uses PWM. So you would be sending PWM voltage to a PWM circuit, or in other words, using a circuit that chops up voltage and sends pulses to a circuit that chops up voltage and sends pulses to a motor. If those 2 circuits are not perfectly synced (they are probably running at totally different frequencies) then you will likely have a lot of random "dead time" especially at lower speeds. At higher speeds the problem would be much less noticeable (maybe spankey is running at high speed, or in the upper ranges of PWM).
     
  14. leahy268

    Thread Starter New Member

    Aug 1, 2012
    11
    0
    Hmm,

    That was kind of what I originally thought.
    I don't know enough about electricity to answer this myself.
    But back to my original question..
    Is it possible to go from a pwm type output to a constant voltage?
    Or failing that is it possible to control a variable voltage power supply type circuit from an microcontroller.

    If either one can I be pointed in the correct direction?

    Warwick
     
  15. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    You can get a constant flow rate by gravity feeding water from a set height (head) above the apparatus (giving a fixed pressure) through a fixed orifice size.

    Then you can use a simple pump and float switch to keep the liquid level relatively constant in the head container. Provided the head height is a few feet, the small change in liquid height of 1/2" or so from the float switch cycling will be totally insignificant.

    Basically it's using reliable mechanical design to fix the flow rate, and a floatswitch+pump just to keep the top container full.

    Of course you can make the orifice variable (like a valve tap) to adjust the flow rate too once it is built.
     
  16. leahy268

    Thread Starter New Member

    Aug 1, 2012
    11
    0
    I certainly see your point with the mechanical method.
    However I would rather not leave the wort to cool in a chamber at the top.. That would kind of defeat the purpose of heating I would have thought.

    The other thing is are we saying that there is no method to provide dynamic but constant( as in not oscillating) voltage to be controlled by a microcontroller..

    If so then how do digital power supplies work?

    Warwick
     
  17. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,013
    1,531
    Yes I did misunderstand you. I thought you wanted to make wort not filter it. Sorry.

    But if the wort is temp controlled how will it get hot enough to scorch?
     
  18. strantor

    AAC Fanatic!

    Oct 3, 2010
    4,302
    1,988
    It's only .8a you can use a voltage regulator
     
  19. spankey666

    Member

    Nov 30, 2011
    91
    3
    i cannot find the circuit i used, all i have are assembly layouts, but from memory i believe the frequency is fixed and the mark-space ratio changes.
    there is lots of info on controlling brushless motors on google, and lots of controllers on ebay. these are the pumps i used http://www.daviescraig.com.au/Elect...R_PUMP_12V_SHORT___PART_No__9002-details.aspx
    and we get rotation from zero rpm to max with out any issues .
     
  20. leahy268

    Thread Starter New Member

    Aug 1, 2012
    11
    0
    I think this was more of my point.
    Yesterday I had a look at a couple of diagrams using LM317's but I wasn't clear on how to use them digitally from the data sheet.
    Perhaps someone here has done it?
    Perhaps I was looking at the wrong voltage regualtor as well.
     
Loading...