Still trying to make a basic motor ramp-up circuit

Discussion in 'The Projects Forum' started by summersab, Jul 30, 2016.

  1. summersab

    Thread Starter Active Member

    Apr 8, 2010
    132
    0
    So, I have had a few other threads on this topic. This is one of them:
    http://forum.allaboutcircuits.com/t...n-small-motor-is-connected-to-circuit.123530/

    However, for whatever reason, I could never get enough voltage going to the motor. So, I abandoned that circuit, looked at a basic capacitor charging circuit, and built the following:
    http://goo.gl/64XitO

    (In the circuit, the 12V and 6.5V inputs are actually boost circuits attached to a 3.7V li-ion battery. The 100ohm resistor is where the motor is attached).

    This works pretty well, and it ramps up the motor as expected. However, it causes the battery voltage to drop and the protection circuit to cut the supply. I measured the voltage from the battery once connected to the circuit, and it steadily drops to 3.2V before cutting out. What's happening?
     
    Last edited: Jul 31, 2016
  2. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,879
    371
    Your motor is drawing more current than the battery/boost circuits can supply.
     
  3. summersab

    Thread Starter Active Member

    Apr 8, 2010
    132
    0
    One would think that, but if I connect the motor to the battery directly, there isn't a problem. Heck, if I connect the battery to the step up converter and then the motor to that, there isn't a problem. It's ONLY when this circuit is involved. The motor is fairly small and draws 0.5A while running/2.0A stall.
     
  4. SLK001

    Well-Known Member

    Nov 29, 2011
    810
    224
    If your MOSFET is N-CHANNEL, then you are burning it up. The load should be above the FET. If the MOSFET is P-CHANNEL, then perhaps it is not sized properly, or you need a heat sink on it. Also, your capacitor is preventing the gate from going full on quickly (which it must do to prevent overheating).

    I still say that a PWM circuit is the way to go - in fact, it is the ONLY way to go.
     
  5. summersab

    Thread Starter Active Member

    Apr 8, 2010
    132
    0
    It's an N-channel IRL1004PBF. If I connect the circuit to my DC power supply (so it keeps running and doesn't cut out), it actually runs completely cool to the touch.

    Something I did notice, though. With the power supply, I bypass the 12V and 6.5V step-up converters and use those voltages directly. When the circuit runs and I check the reading for the 6.5V output on the power supply itself, the voltage drops (i.e. the actual needle on the supply drops). That's . . . not good, but I don't understand what is causing it.

    I'd agree with the PWM part, but I got to a point of frustration with building one. If there were a premade PWM that also had a ramp-up and hold feature (and it didn't cost $100), I'd buy it. I'm trying to keep things cheap and simple.
     
  6. SLK001

    Well-Known Member

    Nov 29, 2011
    810
    224
    A two channel opamp, a potentiometer, your N-Channel fet and some resistors and capacitors is all you need for your PWM control.
     
  7. SLK001

    Well-Known Member

    Nov 29, 2011
    810
    224
  8. summersab

    Thread Starter Active Member

    Apr 8, 2010
    132
    0
    You've seen my other threads in the past about this. I tried building this circuit:
    http://www.edn.com/design/analog/4314544/Integrator-ramps-up-down-holds-output-level (article)
    http://m.eet.com/media/1126960/12071-figure.pdf (actual circuit image)

    I could never get it to work right. I DON'T want an adjustable PWM (i.e. speed controller using a potentiometer as you suggested). I want a circuit that automatically takes a motor from off to full power and holds in the span of about 10-15 seconds. The circuit in my OP is something I found for ramping up LEDs, so I just added some higher-power parts. The smaller the better, though.
     
  9. SLK001

    Well-Known Member

    Nov 29, 2011
    810
    224
    Then all you need to do is replace the pot with a resistor and a capacitor. As the capacitor charges up, the motor will get more voltage (get faster) and once fully charged, will be at max output.
     
  10. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,671
    899
    summersab likes this.
  11. summersab

    Thread Starter Active Member

    Apr 8, 2010
    132
    0
    @jpanhalt That looks promising. I'm looking at the spec sheet, but I'm unsure how I would use it to do what I want (automatic soft-start/ramp up of my motor over 10-15s). Pointers?
     
  12. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,671
    899
    Here is a more explanatory application note: http://www.ti.com/lit/an/slit110/slit110.pdf

    In brief, the resistor from pin 4 to pin 7 and the capacitor from pin 7 to ground control the soft-start. That is, the time it takes for the controller to ramp up from 0% of the final PWM percentage to 100 % of the final percentage. Output is pin 11 (labeled GD for gate drive). Unfortunately, the pdf pages don't have numbers, but a complete description is on page 3. Here is a graph of what happens taken from that application note:
    upload_2016-7-31_19-52-16.png

    Generally speaking, you should settle on a soft-start time constant and not change it. Start with the recommended values. Then change the resistance or capacitance if you want a slower start or faster start. The final effect is similar to what you are trying to get with the RC on your gate in your original schematic, but rather than forcing the mosfet to operate in its linear region during the start, which causes excessive heat, the TI approach operates the mosfet always as a switch to minimize heat and ramps the PWM duty cycle. With that controller, you can also vary the PWM duty cycle with simple resistors.

    From what I understand of your project, you will want the "manual" mode. The TPIC2101 seems to have been developed for older automotive applications. (I was surprised it is still available.) Both its target duty cycle and soft-start are controlled by simple resistors and/or RC circuits. I used it to provide three speeds, each with soft start, for a 12-V, high-powered winch.

    After you have had a chance to review that section in the app note, if you still have questions, I will try to dig up the schematic for that project. That was done well before I even touched a microcontroller, and it worked well. Today, I would use a microcontroller and a dedicated gate drive, if I needed more current drive for the mosfet(s).

    John
     
  13. SLK001

    Well-Known Member

    Nov 29, 2011
    810
    224
    This is what I was talking about. It should work for you. Adjust C2 or R5 to get the timing that you want. With these values, the total ramp time is about 18secs.

    SOFTSTART.jpg
     
    summersab likes this.
  14. summersab

    Thread Starter Active Member

    Apr 8, 2010
    132
    0
    @SLK001 - that's basically the circuit below, correct? If so, I've built that and fiddled with it, but for whatever reason, using a 12V supply to the circuit and a 5V supply to the transistor powering the motor, I never seem to be able to get more than 3.3V or so to the motor. That's why I'm more than a little interested in the TPIC2101 - if there is a purpose-built component that can do what I need, I'm on board.
    http://m.eet.com/media/1126960/12071-figure.pdf

    @jpanhalt - I'm intrigued by this IC. My immediate concern is that it states a required 8V minimum input for Vbatt. Since I'm trying to power this with a 3.7V Li-ion cell and a step-up converter (adjustable), that is certainly feasible, but I would rather use lower voltages. I wonder if 5.5V would work since that is what the chip regulates for the V5P5 output.

    I'm also looking at the second document you provided and trying to figure out how to configure the device to do what I want. Let's just start with putting the IC in manual mode. I understand that it requires 5.5V to the MAN pin (via the V5P5 pin, most likely), but then I'm lost with the underlined part and everything that follows. Huh?
    To assert Manual mode, the MAN pin is pulled high, and the voltage difference between the MAN and AUTO pins, 0 V to 2.2 V, determines the IC’s output PWM drive signal of approximately 18% to 100% depending on Vbat.

    After (hopefully) getting it in manual mode, the capacitor between the SPEED and INT pins will let me control the time of the soft start. The GD output goes to the gate of a FET which in turn powers the motor. Hooray!

    Then there are the pins that I'm not sure are absolutely required or what they do:
    • ROSC
    • COSC
    • CCS
    • AREF
    • ILS
    • ILR
    What are the oscillator pins all about? The last four seem to be related to current limiting, but are they required for getting the IC to work?

    Thanks a lot for your help!
     
  15. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,671
    899
    Sorry, I didn't pay enough attention to your voltage requirement. The principle remains the same, ramp on/off states rather than a partially on state if the problem is heat.

    Can you use a microcontroller?

    John
     
  16. summersab

    Thread Starter Active Member

    Apr 8, 2010
    132
    0
    "The principle remains the same, ramp an on/off state rather than a partially on state if the problem is heat."
    I'm not sure I follow what you're saying, here. Also, heat hasn't been an issue since this is a really small motor.

    I don't see why I couldn't use a microcontroller, but I don't have any experience with them. I was hoping to get this working with a rather basic circuit and then pursue a microcontroller down the road.
     
  17. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,879
    371
    I think that's unlikely to produce a satisfactory result in this case, given the problems with more simple systems.
     
  18. summersab

    Thread Starter Active Member

    Apr 8, 2010
    132
    0
    This is seriously all I'm dealing with:
    IMG_20160801_165936 (1).jpg

    This exercise is clearly proving to me that I suck at circuits, but I don't understand why it is so difficult to take a motor that small and ramp it up from 0V to 5V in 10-15s. I'm wanting to keep the circuit relatively small and simple since I'm trying to create a wearable device. If there is a COTS circuit that will make this work, I'm all down, but I haven't found one.
     
  19. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,671
    899
    I wouldn't call that a logical answer.

    John
     
  20. SLK001

    Well-Known Member

    Nov 29, 2011
    810
    224
    I looked at your circuit - I don't know WHAT it it trying to accomplish. And "NO" they are not the same. It looks like the ramp in that circuit is really a ramp, which is bad for motor control (the transistor will overheat while operating in the linear region).

    What my circuit does is gives you a PWM pulse train that increases from 0% duty cycle to 100% in about 18 seconds. The 12V is not critical - it should also work at 5V.
     
Loading...