Help needed for PWM windmill controller

Discussion in 'The Projects Forum' started by jbrols, Jun 30, 2013.

  1. jbrols

    Thread Starter New Member

    Jul 29, 2011
    18
    0
    Hello!
    After a couple of years of searching I finally found a controller, that might suit my needs. It has its own (old) discussion thread PWM Heating load controller based on frequency, but for reasons unknown to me I'm unable to complete registration at that site, so will seek for help here, if you don't mind.
    It has no complete circuit drawing, just a PCB, which can be opened in ExpressPCB only, but can't be exported to different format, so I could only print-screen it for viewing.

    An allmost whole look at the PCB
    [​IMG]

    A little closer look at the LM2917 part
    [​IMG]

    Can someone, please, confirm, if I made a correct conversion to a DIP-8 version of LM2917? Couldn't find a DIP-14 version.
    [​IMG]


    I'm very far from being an electronics expert (not proud of it :(), so I have some questions for ones, that are ;).
    First- How to adjust the usable frequency (RPM) range to fit my particular wind turbine? There is one 10k potentiometer, but I'm not really sure, if it is for lower or for higher end of the RPM range.

    Most of the parts are already coming to my address, but before experiencing magic smoke I would like to have some helping hand, please.

    Thank you, good Samaritans :)
     
  2. Dodgydave

    Distinguished Member

    Jun 22, 2012
    4,980
    744
    you need to draw the circuit out using the pcb and components first, then see what you can do to modify it.
     
  3. jbrols

    Thread Starter New Member

    Jul 29, 2011
    18
    0
    Ok, here is the most important part of the schematic, created by the original author of this circuit:
    [​IMG]

    Hope now it gets clearer to understand. The question still remains the same- how to adjust the RPM range in which the PWM grows from 0% to 100%?

    Regards
     
  4. Dodgydave

    Distinguished Member

    Jun 22, 2012
    4,980
    744
    Have a look at the datasheet for the LM2917, by altering the resistor and capacitor values on pins 2, 3
     
  5. jbrols

    Thread Starter New Member

    Jul 29, 2011
    18
    0
    Thanks, I have all the datasheets, but don't have a good understanding, how these op-amps or comparators work.
    The output switching frequency at this configuration is supposed to be 18kHz, but I wonder, if it will lead to increased inductive or some sort of losses in a power line due to the fact, that the controller will have to be at the turbine, but the heating element will be some 100 meters away. So I was thinking, that maybe decreasing the frequency to, say, 10kHz would improve efficiency?
    That leads to another question, which I can't figure out by myself- how to change the operating frequency of this controller?

    Sorry for bugging you all
     
  6. jbrols

    Thread Starter New Member

    Jul 29, 2011
    18
    0
    Hello, is there anybody out there :) ?
     
  7. wayneh

    Expert

    Sep 9, 2010
    12,103
    3,038
    by efficiency, do you mean heat into the heater coil? Some folks use these as a way to dump excess power and thus aren't too concerned with efficiency.

    That length of wire with a substantial current will indeed have some DC resistance, which you can estimate. I'm not real sure about the effect of frequency and any inductance effects. It does sound like you'll be in the broadcasting business at 18kHz with 100 meters of "antenna".

    Oh, and the comparators are simple enough; they have two output states, high or low. The output state is determined by the relative voltages on the two input pins. Whenever voltage on in+ exceeds in-, the comparator output goes high. The low output is an open collector transistor to the low power supply rail. A current sink. The high output is just open, with that transistor turned off. It must be pulled up with an external resistor if you need a positive output voltage.
     
  8. jbrols

    Thread Starter New Member

    Jul 29, 2011
    18
    0
    I'm not using any kind of electricity accumulation (no batteries), but instead am willing to transform all the power into heat. For about two years I was using simple, but unreliable and unhealthy for generator relays, that turned the heating elements on and off after the turbine got to speed. I always wanted something more reliable, like this PWM controller, that would allow me to connect resistive load powerful enough to load the turbine at high winds without the fear of owerspeed and yet not to be too hard at low wind speeds.
    At quite powerful wind breezes I've seen near 3kW from the generator at the heater end of the power line (measured by multiplying DC volts and amps).
    I'm just concerned, if I'll manage to tune this controller to my turbine correctly, so that the PWM duty cycle wouldn't go to 100% too soon or too late.
    There are, of course, quite big power losses in the long wires to the heater, but I'm afraid, that they might grow even bigger, when carrying relatively high frequency DC signal instead of a "pure" DC from just a rectifier. That's why I would consider to decrease this frequency. But maybe I'm totally wrong about this.
    I'm sure, there are some wiser guys than me ;)
     
  9. crutschow

    Expert

    Mar 14, 2008
    13,003
    3,232
    Why are concerned about losses in the wire? It just adds to the power you are trying to dissipate in the heater.
     
  10. jbrols

    Thread Starter New Member

    Jul 29, 2011
    18
    0
    The heater is in a room, but the ~100m wire is outside the room, so it makes totally wasted heat.
    If only I could put the controller close to the heater, there would be no problem, but then I would need to change the cable from one phase (two wires) to three phase (at least three wires) cable. That would be outside of my budget's limits :(.
     
  11. jbrols

    Thread Starter New Member

    Jul 29, 2011
    18
    0
    I finally realized, that I have a talent to start topics, that no one really is interested in :cool:...
     
  12. deco75943531

    New Member

    Jul 2, 2013
    5
    0
    Looks so complicated, I do not know how to design, to learn about
     
  13. Dodgydave

    Distinguished Member

    Jun 22, 2012
    4,980
    744
    Its not that were not interested, we need the full diagram and what conditions its going in, to alter the freq rate of the LM2917 ( its a frequency to Voltage converter),are set by the R1 and C1 values as per the datasheet.

    Do you want us to make it for you!
     
  14. jbrols

    Thread Starter New Member

    Jul 29, 2011
    18
    0
    Was that a question or rage :eek:?

    As far as LM2917 goes, it may be clearer for me now, but how to change the PWM operating frequency, no one has told me yet. Might not be that easy even for wise guys :D.
     
  15. Dodgydave

    Distinguished Member

    Jun 22, 2012
    4,980
    744
    To set the frequency as per datasheet Formula ; VOUT = fIN × VCC × R1 × C1,


    So lets start with what frequency do you need, and what is the supply voltage?
     
  16. jbrols

    Thread Starter New Member

    Jul 29, 2011
    18
    0
    To my understanding this formula VOUT = fIN × VCC × R1 × C1 is for adjusting the input frequency (generator RPM) value to the output DC voltage level of LM2917, which goes to LM339.
    I guess, the LM339 is the one, that actually dictates this 18kHz (as stated by the original author of the circuit) mosfet driving frequency and the duty cycle of it. This 18kHz PWM signal then goes from pin-1 of 339 to mosfet driver TC4422. http://forum.allaboutcircuits.com/showpost.php?p=623420&postcount=3
    Please, correct me, where needed.
    I was thinking of decreasing this 18kHz frequency to maybe 5~10kHz in case, if there were too great power losses in long cable to the heater.

    The most important thing is to be able to adjust this controller that way, that it, for example, starts to PWM (1%) at 60 turbine RPMs and gets to 100% duty cycle at 300 RPMs or higher.
     
  17. wayneh

    Expert

    Sep 9, 2010
    12,103
    3,038
    It seems to me that R6 and C12 control the oscillation frequency of the upper comparator. Just make C12 a bit larger to slow the oscillation. The voltage on pin 5 of the LM2917 controls the duty cycle.

    Solving your problem is all about knowing voltages. You need to know what the LM2917 is putting out your critical rpm points, and then configure the LM339 to provide the duty cycle setting you want at those points.

    I like the arrangement of using the LM339 to vary the duty cycle - I think that could be a solution for other folks coming here looking for ways to adjust LED brightness with voltage-controlled PWM like this. I'd be curious to see how well it works.
     
    Last edited: Jul 3, 2013
  18. jbrols

    Thread Starter New Member

    Jul 29, 2011
    18
    0
    Thank you, will give it a try, when completed. Would it be correct to assume, that the 10k potentiometer sets the start point of PWMing? But how to set the point, at which the PWM becomes 100%?

    Apologies for so many simple (for experts) questions, but, while searching, I've seen others looking for similar controllers with no luck, so this might help others too.
     
  19. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,007
    1,530
    I've got a dumb question, where are you getting the frequency from your generator/turbine? I'm assuming your making DC? If so why not just compare the output voltage from the gen to a reference voltage then use that to control the PWM?
     
  20. jbrols

    Thread Starter New Member

    Jul 29, 2011
    18
    0
    The generator is made from a modified 3 phase electromotor with a permanent magnet 12 pole rotor. It gives 3 phase AC current, which is then rectified for use as a DC. Frequency (RPM) signal will be taken from one phase through a small transformer, as it is shown in a schematic (post 3).
    I'm no expert, but for now this type of controller seems the most promising, if only I could tune it correctly.
    Comparing DC to a reference voltage might be suited mostly for battery charging (maintaining constant output voltage) and deflecting the remaining power into a dump load. For me the voltage can vary, as long as it doesn't exceed, say, 230V and gets converted into usable heat.
     
Loading...