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.

DRV8829 VCP resistor

Other Parts Discussed in Thread: DRV8829

Hi! I am currently designing a motor driver using DRV8829's for microstepping a NEMA34 and a NEMA23 motor.. I was checking out the PCB designs provided in the application notes as well as the one for the EVM and I noticed there's no sign of the 1MOhm resistor sugested in the datasheet for the VCP pin, so I would like to know what is this resistor's function and if it's a must under certain circumstances.

I also noticed there are 2 capacitors attached to the second VM pin, that seems to me are used as a filter... but I'm not sure.

Thanks in advance for the repply!

  • Hi Ivan,

    The 1MOhm resistor is recommended to bleed the charge pump capacitor. This will prevent the outputs from being shorted during power off or sleep modes, allowing the motor to turn freely. If you do not need this feature, the VCP cap will eventually discharge and the motor will begin turning.

    For more information, please refer to post http://e2e.ti.com/support/applications/motor_drivers/f/38/t/310106.aspx 

    The 2 capacitors connected to VM on the schematic are local decoupling. I am going on memory, but I believe one is place close to pin 4 and the other is placed close to pin 11. These capacitors minimize local voltage drooping.

  • Hi Rick,

    Thanks for the repply! I will take this into account for the design...

    best regards,

    Iván

  • Hi Rick,

    I have carefully designed a couple PCB's to test the DRV8829 taking the circuit from the EVM as an example as well as following the datasheet info and the advice you provided. Now that I am testing it with my motor I am experiencing some trouble. Whenever I power up the circuit the motor jolts a few time as if trying to rotate but it doesn't and inmediately after that both drivers enter into fault state.

    The drivers are being powered at 12V to microstep the motor at 2A peak using the internal current regulation. The motor is wired in serial configuration.  I have already checked the connections making sure each DRV8829 is properly driving its coil and also tried connecting it in parallel configuration with the same result. I am suspecting I am dealing with an OCP issue, due to the circuit responding (jolting of the motor and fault condition) again inmediately after a repeated power up or toggling the drivers reset pins. The PCB's don't even warm up so I don't think overheating is the problem.

    It would be really helpful if you provide your insight! Maybe I am not taking into consideration some other factor... 

    thanks in advance!

    Iván

  • Hi Ivan,

    It does sound like you may be experiencing some type of OCP issue. Have you examined the nFAULT pin? This provides an indication of OCP or TSD.

    If nFAULT is not asserting, take a look at VM at the pin. If it falls below UVLO, the device will reset

    What are the motor parameters? Can you provide your schematics and layouts?

    Scope shots capturing the change in current or voltage also helps.

  • Hi Rick,

    Thanks for your reply!

    The motor is a NEMA 34. In series configuration each coil is rated 0.96Ohm, 6.8mH, 3.18A and 3.05V. I am sending attached the Gerber files for the test PCB: 1362.pcb drv8829 - CADCAM.ZIP Please let me know if they are readable since it's the first time I generated this files for a PCB. If not, I can send you pictures of the different layers.

    I am using two PCB's and both of them are controlled by an atmel uC, which is providing the signals corresponding to the current level (I0...4), the decay mode (slow), and phase for each coil. Both fault pins are being monitored by the uC. The driver's sleep and enable pins are directly connected to 5V. The reset pins are being controlled by a switch. Vref is being provided by a pot conected to 3V.

    I have checked all the voltages with a digital multimeter and they seem to be stable. The fault pins show 0V inmediately after the motor jolts, which is why I believe there must be an OCP condition, unless the drivers are heating up really fast or there is in fact a voltage drop given that sometimes the uC resets itself as well.... Currently I don't have access to an osciloscope but I will definetely try to get some scope shots as soon as posible.  

    Best regards,

    Iván

  • Hi Ivan,

    Based on your connections, are you starting with I[4:0] at 0000 or 0%? Then when running you adjust the I[4:0] to create the sine wave? Does the fault occur immediately after changing I[4:0]?

    Would you check a couple of items?

    1) If you have a current probe, please capture the current when the DRV8829 is initially enabled or when I[4:0] is changed? This should confirm if it is an OCP conditions.

    2) Try mixed decay or fast decay (which ever is easier). You will see more current ripple, but this will provide a clue.

    Any scope shots of the current and outputs when the event occurs will be helpful.

    Thanks.

  • Hi Rick,

    To answer your question, yes the fault occured whenever the I[0...4] pins were changed for the first time. One of the coils started at 0% and the other at 100% following the sinusoidal pattern. The coil at 0% current didn't present the fault condition until it's current increased for the first time. The last time we tested the circuit we lost the power supply (an overhauled computer PSU) and both drivers apparently due to an overcurrent. Next week I am going to assamble new PCB's and perform the next round of tests at my university, where I have a reliable power supply and the scope. Any advice on lowering the startup peak current? I will let you know how the test goes and definetly post the scope captures as soon as I can.

    Thanks for your support!

    Iván

  • Hi Rick,

    Once I had the new set of PCB's ready I went ahead and tested them at the lab this morning... The results nevertheless were more or less the same. The drivers kept entering into fault state whenever the current value was required to change for the first time after power up. However this time I had a scope with me so I could come up with some interesting pics (sorry I couldn't get the captures).

     

    The most representative one is show above. Channel 1 is the current flowing through one of the coils and Channel 2 is the voltage at the drivers' VM pins. The motor was driven at 15V and 2A peak (Vref=1V, Rsense=0.1Ohm). The pic shows the moment when the driver tried to achieve a 70.7% current level. Inmediately after that, the driver's fault pin was driven low.

    According to my opinion, the driver is experiencing an overheating condition given that the current never surpasses the 6A barrier and the voltage drops aren't so long as to activate the UVP (assuming the voltage waveform shown is normal, due to the chopping perhaps). What do you think? Do you have another suggestion to increase the thermal relieve?

    Any advice would be helpful. Thanks again for your time and support!

    Regards,

    Iván

  • Hi Ivan,

    If you zoom out, what do you see the current and voltage do? Does the current begin to increase and the voltage begin to droop more?

    There are two items I saw that are of concern.

    First is the voltage at the VM pins. These voltage spikes are >10V which puts the device well below the UVLO threshold. If you can raise VM to 24V, this may isolate the problem. If it disappears, the problem is UVLO. If it still faults, there is something else.

    Overtemp could be an issue, but I am not sure. From the gerbers it appears as though the via size is 25mils. Is that correct? The datasheet recommends 12mils. Larger vias can allow the solder below the device to escape when mounting the device.

    Can you send the schematic of you board also? I saw the two local capacitors near the VM pins, but did not see the bulk capacitor that is usually shown on the TI EVM. Please note page three of the EVM schematic showing a 100uF bulk capacitor.