Why is there a voltage drop when I connect my motor to Mosfet controlled by PWM signals?

Discussion in 'The Projects Forum' started by Adarsh Shanmugam, Jul 28, 2018.

  1. Adarsh Shanmugam

    Thread Starter New Member

    Jul 27, 2018
    9
    0
    Hi! I'm trying to make a mini quadcopter. To control the speed of the motor, I'm using a mosfet in 'Low-side' configuration controlled by PWM signals from an arduino pro mini. The mosfet I'm using is 'SQ2310ES', an n-channel mosfet. The motor is a coreless dc brushed one, operating on 3.7V. The motor has a resistance of 1 ohm roughly. The battery I'm using is a '3.7V Li-ion 780mah 25C'. The motor works perfectly when I connect it directly to the battery and the voltage across the terminals measure '3.7V'. However, when I include the mosfet in the circuit, the voltage drop across the motor measures only '0.5V' despite the arduino producing a full duty cycle PWM (5V). The battery voltage does not drop and remains at '3.7V'. Gate voltage doesn't change either (5V). When I disconnect the motor from the circuit, the voltage across the motor slot increases to 3.7V. The problem is when I connect the motor, the voltage drops to '0.5V'. What am I missing? I'm new to electronics and it would be great if someone could help me.
    Here is the circuit:
    WhatsApp Image 2018-07-28 at 10.43.18 AM.jpeg
    Here is the datasheet of the mosfet:
    https://www.vishay.com/docs/67036/sq2310es.pdf
     
  2. crutschow

    Expert

    Mar 14, 2008
    20,007
    5,628
    That MOSFET is a logic-level device with a low Vgs threshold so I see no reason that it shouldn't be fully turned on.
    Try connecting 5Vdc to the gate.
     
  3. danadak

    Well-Known Member

    Mar 10, 2018
    1,710
    351
    What freq is PWM running at ? With a scope you see correct waveform at
    gate of MOSFET, which at 100% duty cycle should be 5V ?

    Regards, Dana.
     
  4. Adarsh Shanmugam

    Thread Starter New Member

    Jul 27, 2018
    9
    0
    Hi. I tried applying 5Vdc with the help of a SMPS. It still doesn't work. So I'm kinda sure the problem is not with the PWM.

    Hi. I don't have an oscilloscope but I read in the arduino website that the frequency of the PWM is approximately 490 Hz. I can't see the waveform but when I measure the voltage between the gate and the source, it measures 0.8 V while the motor is running.

    Update:
    The gate voltage did drop I guess. The gate voltage remains as 0.8V whether the motor is running or not. This causes my drain source voltage to raise to a value of 3.09V. When I measure the voltage between PWM pin 10 and GND of arduino, it shows 4.59V. But only 0.8V drops across the gate and the source. So should I lose my 200 ohm resistor? The voltage across the resistor measures about 3.71V. Will removing the resistor solve my problem?
    Thank you for your time.
     
    Last edited by a moderator: Jul 28, 2018
  5. danadak

    Well-Known Member

    Mar 10, 2018
    1,710
    351
    Normally a R of 10 - 50 ohms is all that is used. To suppress L transients
    when driving the gate.

    Your measurement with DVM is actually measuring AC at Gate I would guess.

    Free scope that might help -

    You can start with a PC sound card based scope for free. Will give you basically
    audio range scope, spectrum analyzer, and function generator all using your
    PC sound card.


    https://www.zeitnitz.eu/scope_en


    http://www.zelscope.com/


    http://www.ledametrix.com/oscope/


    http://www.virtins.com/downloads.shtml


    But first build a simple circuit to protect sound card inputs so you do not
    ruin from transients, overvoltage. Google "protect sound card input".


    For example http://makezine.com/projects/sound-card-oscilloscope/


    Regards, Dana.
     
  6. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    8,718
    2,003
    It should be next to nothing. Are you sure the resistor value is correct?
    It's possible the FET is faulty (gate-source leaky) or fake. Do you have another FET you could try?
     
  7. Adarsh Shanmugam

    Thread Starter New Member

    Jul 27, 2018
    9
    0
    Hi. Yes, I'm pretty sure it is a 200 ohm resistor. I have a couple of FETS lying around. I'll try with them and update the status. Thanks!!!

    Hey! I don't understand that 'AC' part. Do you mean the square wave function generated by the arduino as AC? I'll try to build that oscilloscope and find the waveform. I'll update the progress soon. Thanks! :)
     
    Last edited by a moderator: Jul 28, 2018
  8. danadak

    Well-Known Member

    Mar 10, 2018
    1,710
    351
    Yes, a PWM signal is an AC signal. looks like -

    upload_2018-7-28_8-33-49.png

    Regards, Dana.
     
  9. ebp

    Well-Known Member

    Feb 8, 2018
    1,984
    695
    If you are testing with DC applied to the gate it doesn't matter if the resistor is 2 ohms or 200k ohms. The gate-source leakage current spec for the FET is 100 nA max, which would result in a voltage drop across a 200k resistor of only 20 mV, or 20 µV for 200 ohms. The actual gate leakage at moderate temperature is going to be vastly less than 100 nA.

    If there is a voltage drop across the 200 ohm resistor of over 4 volts, the FET must be damaged.

    ===
    A PWM signal is normally NOT AC - it is DC signal whose amplitude varies with time. AC means that there is an actual change in the polarity. The gate current when a PWM signal is applied has an actual AC component because the capacitance of the gate means current will flow into the gate to charge the capacitance when the PWM signal goes positive and out of the gate to discharge the capacitance when the PWM signal returns to zero (in and out here refer to "conventional current" which is considered to flow from positive to negative).

    Sometimes the voltage applied to a FET's gate is truly AC. This typically would be the case if the gate were driven through a transformer.
     
    Last edited: Jul 28, 2018
    shortbus likes this.
  10. crutschow

    Expert

    Mar 14, 2008
    20,007
    5,628
    That's a large problem.
    If the PWM is set to 100% then that voltage should be 5V.
     
  11. crutschow

    Expert

    Mar 14, 2008
    20,007
    5,628
    I beg to differ.
    It's a matter of semantics, but the generally used meaning of AC is that the voltage varies with time, not that it actually changes polarity with respect to ground.
    Electricity is only defined as AC or DC.
    DC doesn't vary with time. Any variation is AC.
    In practice it's what an AC meter would read.
    By your definition you would have to say that an AC meter doesn't always measure AC.

    So a PWM signal is AC with a DC offset.

    Running a signal through a capacitor doesn't convert it from DC to AC.
    For an example a signal going through an AC amplifier is considered to be AC all though the amp, even if it is offset by the amplifiers DC bias as it goes through the amp.
     
    Last edited: Jul 28, 2018
    absf likes this.
  12. Picbuster

    Active Member

    Dec 2, 2013
    804
    99
     
  13. MaxHeadRoom

    Expert

    Jul 18, 2013
    16,109
    4,714
    That has always my definition of the term, as in a service supply, the current changes (alternates) in direction through the conductor.
    Max.
     
    shortbus likes this.
  14. ebp

    Well-Known Member

    Feb 8, 2018
    1,984
    695
    As soon as you define AC to be anything that is time variant, DC becomes notional only and no longer exists in reality. You can no longer say that a primary cell or a voltage regulator or anything produces DC, except by special pleading. Change in direction of current flow is a true and identifiable physical phenomenon.

    The definitions and explanations of AC and DC found on the web are all over the place - inconsistent and quite remarkably unhelpful. The fact that the terms conflate ideas of potential ("voltage") and current certainly doesn't improve matters.
     
  15. ericgibbs

    Moderator

    Jan 29, 2010
    5,902
    1,110
  16. crutschow

    Expert

    Mar 14, 2008
    20,007
    5,628
    That is strictly true.
    But in practice we take anything below an arbitrary frequency as DC, depending upon the situation.
    For example at one extreme, I've seen the DC drift in op amp offset with time defined as very low frequency 1/f noise.
    Usually we select some arbitrary frequency (normally unstated but usually in the μHz or lower region) as the point below which the voltage is considered DC, such as for a power supply or battery.
    It certainly is.
    That doesn't make it a good definition for "AC".
    And typical AC measurement instruments pay no attention to that.
    To me if an AC meter measures a voltage, or it looks like AC on a scope then that's an AC voltage.
    To define it otherwise leads to a too restrictive definition of AC.
    It can be AC with a DC offset but it's still AC (even if it doesn't reverse the current direction or go around zero volts).
    The restricted definition of AC leads to the bizarre situation where there is AC going into an amplifier with no DC offset, but it's not AC inside the amp until it appears at the output with no DC offset.

    Perhaps that is an engineering viewpoint, not an academic one, but it is certainly more practical.
     
  17. danadak

    Well-Known Member

    Mar 10, 2018
    1,710
    351
    Interesting point you made. I think of it as a measurement with respect to
    a reference point.

    So if reference is ground then I concur.

    But if reference is Vdd/2, a gross approximation to CMOS threshold, then
    polarity does change in the PWM example, but not independent of what it
    is driving. Your point about C load. Reminds me no matter how much I have
    learned I still know very little :)

    Interesting discussion point.

    Regards, Dana.
     
  18. Adarsh Shanmugam

    Thread Starter New Member

    Jul 27, 2018
    9
    0
    Yeah man. I guess the FET is damaged as others said. I'll try the circuit with some news ones and see how it goes. Thanks!!!
     
  19. Adarsh Shanmugam

    Thread Starter New Member

    Jul 27, 2018
    9
    0
    UPDDATE:
    So, I tried using "P55NF06" mosfet in the same circuit because it had a minimum threshold gate voltage as 2V. There were some improvements but there is still some drop in voltage across the motor when I connect it.
    However the improvements were:
    1. The gate-source voltage doesn't drop and remains as 5V when duty cycle is full.
    2. Voltage drop across the 200 ohm resistor has become negligible. (Approx 0V)
    3. The drain-source voltage measures 0.18V
    4. The voltage measure across the motor when connected to the circuit is 2.77V. So, that is definitely an improvement but still I'm missing out 1V approx. The battery voltage doesn't drop and is stable at 3.76V.

    When I disconnect the motor, the voltage across the motor slot rises from 2.77 to 3.76V. The voltage across drain-source measures 0V. So, this must mean that the MOSFET is actually conducting perfectly until I connect the motor, right?

    I feel like I'm close to solving this problem but still can't figure out why there is a drop in voltage across the motor even though the gate-source voltage is 5V. To make sure the problem is not with the PWM, I tested the circuit by applying 5Vdc across the gate and source with the help of a SMPS. The results remain the same. Is mosfet still the problem? Thanks in advance!

    Here is the datasheet of the mosfet:
    https://www.st.com/resource/en/datasheet/stp55nf06fp.pdf
     
  20. crutschow

    Expert

    Mar 14, 2008
    20,007
    5,628
    So that means the MOSFET is working fine.
    The problem appears to be a voltage drop somewhere from the battery to the MOSFET.
    Measure the voltage drop between the battery and the motor, and between the motor and the MOSFET drain.

    What wire size are you using for the connections?
     
Loading...