• State Resolved
  • Locked Locked
  • Replies 15 replies
  • Subscribers 30 subscribers
  • Views 3698 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DRV110 for 1000Watts solenoid driver

Elangovan Mathaiyan86
Prodigy 140 points
Other Parts Discussed in Thread: DRV110

Dear Engineers,

Please let me know the below design will meet using the DRV110 solenoid driver...Our Requirement is,

Inductance of the solenoid = 25mH

Resistance of the solenoid = 11.5 ohm

Peak current reqd, Ipeak = 9 A

Required peak time, tkeep = 100 ms

I have choosen the below values based on the Calculations in the datsheet and EVM

Rosc = 0 Ohm (To set default freq = 20 kHz)

Ckeep = 1 uF (To set the peak time of 100 ms)

Rsense = 0.1 Ohm

Ipeak = 900 mV / 0.1 Ohm = 9 A ,  So Rpeak = 50 Kohm & Rhold = 50 Kohm

Vin = 15V, Vout =15V

Please clarify whether calculations are right else correct me. During the test, it seems no variation in the ON period of the PWM..It seems in 10% only between the 0W to 300Watts...

I have been using EVM by just adding the External Mosfet and current sense resistor..

Mosfet Spec are, Vdss = 600V, Id = 18.5 A, Rds(on) = 0.19 Ohm

Its seems the Mosfet get short between gate to source if i increase the watts...I am doing analysis here to find the issue.. I am looking great support from this forum..


Thanks in advance,

Elango

  • 0
    TI__Mastermind 26065 points
    Hi Elango,

    Our expert of this part will be noticed.

    Best regards,
  • 0
    TI__Genius 17380 points
    Hi Elango,

    Regarding the PWM, it only controls the frequency that the DRV110 checks the current. For these high current / high inductance cases , it can result in an effective frequency which does not match the PWM frequency because the DRV110 operates as a hysteretic controller. For small inductance cases where the minimum Ton is on the same order as the PWM frequency, the PWM resistor will control the inductor ripple frequency. For cases with large inductance / large current / small Rsense, the Ton will exceed the PWM frequency and so the actual frequency will be determined by:

    Vhyst = 5 mV (approx) so for Rsense = 0.1 ohm, Iripple = 50 mApk-pk.

    For a 50 mApk-pk, if you have a 0.7V diode drop and a 25 mH inductor --> 1.8 ms to decay 50 mA --> effective frequency of 560 Hz. So even though the DRV110 is checking every 50 us, it is likely immediately turning off unless the hysteresis point is reached, at which point it stays on for several cycles in order to bring the inductor current up the 50 mA.

    Now, due to noise on the node, the frequency is typically a higher than this, but it helps to explain why you probably aren't seeing 20 kHz.

    What is the capacitance of your FET? If the capacitance is too large you could be running into turn on / turn off issues (gate driver is 15 mA).
  • 0 in reply to Kevin LaRosa
    Prodigy 140 points

    Hi Kevin,

    Thank you for your quick reply.. Regarding the capacitance of the Mosfet,

    Input capacitance - 1137 pF @ Vgs = 0V; Vds = 100V : f = 1Mhz

    Output Capacitance - 68 pF @  Vgs = 0V; Vds = 100V : f = 1Mhz

    If we compare the Mosfet which is used in EVM means, the Input capacitance - 156 pF & Output capacitance - 23.5 pF

    From your statement we are understand below points. Please ensure the same.

    • Based on the current via Rsense, the frequency of the OUT will varry
    • Based on our requirment like current, inductance, we cant achieve the maximum ON time at 20Khz PWM

    You have calculated the Peak to Peak current of the solenoid using hys voltage, and Find Effective frequency..I could not able understand how you did this calc?

    From the effective freq result, whether my application can achieve only if PWM freq is 560Hz?

    As per the gate current, the Mosfet which we are using consumes only 150nA during the ON time..

    The exact question is, Whether can i achieve my applications using DRV110 itself?

    If yes kindly let me know the exact things have to follow and perform to achieve the same..

    Thanks in advance.........

  • 0 in reply to Elangovan Mathaiyan86
    TI__Genius 17380 points

    Hi Elango,

    Sorry for the delay, I was out last week on vacation.

    I'll detail my explanation below but I've got a concern that came from recalculating today. When you say: "

    Resistance of the solenoid = 11.5 ohm

    Peak current reqd, Ipeak = 9 A"

    Is that correct or was there a typo? With a Vin of 15V, the maximum current you can get is going to be 15 V / 11.5 ohm = 1.36 A. In my calculations below I assumed it was a typo and meant to be 11.5 mohm and thus neglected it.

    I should have been more clear. To get to that point I used the rough estimate:

    V = L * (di/dt) where we can simplify di/dt to Δi/Δt since we are on small scale.

    When the FET is off, Tdown, V = voltage across the inductor = voltage across the diode (roughly 0.7V though actual results my vary significantly here), Δi = ΔVsense / Rsense where ΔVsense = 5 mV.

    So, 0.7V = 25mH * 5 mV / (Δt * 0.1 ohm). Solving for Δt = 1.8 ms.

    When the FET is on, Ton, V = voltage across the inductor = 15 V - I * Rsense ≈ 14 V, with all the other values the same leads to 89 us

    Thus, total time is approximately 1.8 ms + 89 us ≈ 1.8 ms --> 560 Hz. Note that the actual frequency will probably vary significantly as there were many neglected variables.

    Have you been able to take a look at scope shots of the FET Gate, source, and drain? Also, did you modify the diode? It is only rated up to 2A, so it could be causing issues when you try to force 9A through it.

  • 0 in reply to Kevin LaRosa
    Prodigy 140 points

    Hi Kevin,

    Thanks for your reply..There is no typo here.The actual specifications of the solenoid is same.

    Resistance - 11.5 Ohm

    Peak current - 15A

    Operating time - 100ms.

    Yes i just replace the diode by MUR460. As per my calculations also the frequency reaches near 555 Hz only.

    As we have captured the waveform at drain and gate using isolated probe but it is not that much clear.

    But one thing we can say is now the concept is working, It works when i set the oscillator freq around 20KHZ and if i reduce the frequency i cant able achieve the concept working..

    Actually the coil voltage  is 110VDC and DRV110 Vcc is 15V.. In our requirement the peak current may reach near 37A for 50ms @ 200VDC...

    Kindly let us know will be there any issue if we operate such higher current rate?

  • 0 in reply to Elangovan Mathaiyan86
    TI__Genius 17380 points
    Hi Elango,

    I don't see anything that should cause this to have an issue, though as you mentioned you may need to experiment some with the oscillator frequency due to the high gate capacitance.

    My only concern at the higher temperatures is for the FET power dissipation, but even then I don't see too much risk. I calculate (by t = CV / I = 1100 pF * 15 V / 15 mA ≈ 1 us) that the rise time should still be small relative to the PWM frequency so I don't see too much issue. If you find any particular concerns, we can help you out with them, just be sure to upload the scope shots so we can see what is going on.
  • 0 in reply to Kevin LaRosa
    Prodigy 140 points
    Dear Kevin,

    Thanks a lot..Your answer has given a big confident to us..Let us know if any particular issue arises during the upcoming test and evaluation...

    Thanks..........
  • 0 in reply to Kevin LaRosa
    Prodigy 140 points
    Hi Kevin,

    Please clarify below point.

    The default level of the EN pin is High (Internal Pull up). But my requirement is to keep the default level of LOW and make the pin high when ever needed to drive the solenoid.
    The Limit switches going to control the EN pin (Mechanically).

    Please find the some ideas..

    Regards,
    Elango
  • 0 in reply to Elangovan Mathaiyan86
    TI__Genius 17380 points
    Hi Elango,

    The internal pull up is 500 kΩ (shown on p.7 of datasheet), so as long as you drive with something with less than 50 kΩ, you should have no problem overriding the internal pull up.
  • 0 in reply to Kevin LaRosa
    Prodigy 140 points

    Hi Kevin,

    Thanks for your valuable inputs. Please find the values of Rpeak, Rsense, Rhold, Ckeep calculated for our requirements.

    The exact specifications are,

    The Nominal Input voltage = 120V DC

    The Resistance of the solenoid at 25 C = 5.489 ohm

    The peak current required, Ipeak      = 120/5.489 = > 21.89 amps

    The Hold current required , Ihold = 5 Amps (Approx)

    The maximum Peak current Duration = 50 ms (Approx) 

    As per datasheet and Application Note please refer the value selection of Rsense & Rpeak for external settings

    Rpeak                                                 Rsense                                                     Ipeak

    50 Kohm                                              1 ohm                                                      900 mA

    50 Kohm                                              2 ohm                                                      450 mA

    200 Kohm                                            1 ohm                                                      300 mA

    200 Kohm                                             2 ohm                                                     150 mA

    100 Kohm                                             0.59 ohm                                                  1 A

    Based on the above comparison chart fro TI  and Formulas, We have calculated the below values.

    Rsense = 0.05 Ohm, 5W

    Rpeak = 50 Kohm

    Rhold = 50 Kohm

    Ckeep = 0.5 uF

    Please correct me if my understanding is wrong and let us know the exact values for Rsense, Rpeak, Rhold required to meet our requirement or exact way to choose those values.

    The peak current set here is approx 20 A. When solenoid current goes above this value, DRV110 have to sense and turn OFF or reduce the Width of the output PWM. 

    Please let me know if you are required any more info.

    With Best Regards,

    Elango

  • 0 in reply to Elangovan Mathaiyan86
    TI__Genius 17380 points

    Hi Elango,

    Based on the equations, your values will provide approximately 18 A peak and 3 A hold. To get 22A and 5A, you would need to decrease Rsense to 0.03 Ohm and increase Rpeak to 90k. I've attached a calculator which can help.

    Note that these values are approximate and may vary based on your actual board design, so we recommend validating and fine tuning as needed. DRV110 Parameter Selector.xlsx

  • 0 in reply to Kevin LaRosa
    Prodigy 140 points

    Kevin,

    Great..We will proceed and follow the calculations and let u know the feedback on the result.

  • 0 in reply to Kevin LaRosa
    Prodigy 140 points

    Kevin,

    We have found an issue (High freq like 2kHz oscillation) in the EN when we break the high signal to EN pin. Please refer the Schematic & waveform captured at EN pin,.

    Hope you will be ok about design and solenoid specifications from the above conversation.

    We have testing the circuit in two modes.

    1. Powered ON the entire circuit for 100ms (120VDC high pulse applied in connector P1 across CLOSE & COM for 100ms)

    2. The high pulse to the EN pin is break up on 30ms & OUT also. Because mechanism operates, change over the position of limit switch from CLOSE to OPEN (But Remaining circuit will be powered ON for another 70 ms)

    No malfunction @mode1. But in Mode 2, After break the high pulse to EN pin there will be an high freq noise Like an oscillation for few ms before PIN voltage becomes LOW. Separate schematic to this part also have been shared. We have placed an filter capacitor across EN pin to Gnd, Noise has been arrested but Turn OFF delay of EN pin increases (Seems not good)

    The another doubt is, Once we have fixed the peak current, hold current and Peak time (Let us consider peak time is 100ms), What will be the result if EN pin keeps HIGH for long time like as 1sec. Whether solenoid operates in HOLD current mode or some other happen.

    Figure 1 : DC Control Board Schematic (S1 is Limit switch)


    The components are in the EN pin side is, S1, R13, D15 & R12 (20K).

    Figure 2 : The EN pin behavior during the mode 2 operation (Green -  EN pin & Yellow - OUT pin)

    Fig 3: Enable pin Behavior after a filter capacitor 2.2nF between Starting of R12 to Ground.


    If we consider the figure 3, when we add an filter capacitor Noise in the EN pin has been reduced but Slow Turn OFF.

    Still there is an noise interruption in the OUT signal
    .

    We are assuming that, is due to noise in the HV Gnd. Please share your thoughts with us.

    Regards,

    Elango

  • 0 in reply to Elangovan Mathaiyan86
    Prodigy 140 points
    Kevin,

    Please find Schematic & waveforms in separate tag below

    Regards,
    Elango