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.

DRV595EVM fault

Other Parts Discussed in Thread: DRV595, DRV595EVM

Dear DRV595 expert!

I have purchased the DRV595EVM module and was inteding to use it to drive a TEC module.

When testing the module seems to go into a fault state. It  shut-s off for approcimately 1.2s.

The device is not warm since I can pit my finger on it and keep it there.

If I disconnect the JP3 it will remain in the off state. Based on that my understanding is that it is the over current detection happening.

I find this strange since I am far from the 7.5A trip point in the application.

I have two TEC's in series and test them by supplying a set of supply voltages.

The EVM module is  configured as out of the box. I have connected the IN- to GND and IN+ is set so that I will get maximum current.

Starting at 27VDC I can get 4A.

When decreasing the voltage the current will also decrease however the failure keeps happening.

At 21VDC I have 3A but at the module still module shut's off.

At 17.7VDC I have 2.5A but unit still behaves intermittent and shut's off. Hower the failures seems to happen with less intensity.

At 14.2 VDC I have 2A and still have the problem.

partnertechonline-my.sharepoint.com/.../guestaccess.aspx

 

Could you please advice?

Another question:

EVM module is stated to handle 3A but device 4A. What is the limeting factor on the EVM board since the device can meet 4A?

Regards Anders

.

  • Hi Anders,

    Our experts have been notified and should reply soon.
  • Hi Rick!

    We actally bought two EVM modules. I shipped one to a collegue of mine.

     I asked him to set-up and test the unit. His peltier element is a more  efficient version and he tested at lower currents.

    The problem also occured on this second set-up.

    There seems  to be some conditions that we do not fullfill to make these units work reliably.

    Changing operating frequency does not remove the problem.

    Checking the PWM output at C6 with a 1Mohm probe/15pF increases the intensity of the problem.

    Reducing the input so that the NMOS transistors no logner is in 100% PWM the problem seems to dissapear.

    Increasing again so problem apears then removing the probe from C6 the problem dissapears. Iassume that this moves the PWM from 100% to something below.

    Putting the probe on C7 no obvous change in behaviour due to the probe loading.

    Capacitive loading on C6 seems to be a problem. Running at 100% Duty seems to be a problem.

    I do not see any reation to the triggering of the Fault pin and the problems found.

    Removing C11 from the evalboard as suggested in an old thread do not fix the problem.

    Measuring at C6 removing the TEC load running at just beow 100% duty the voltage on C6 switched up to PVDD. When setting 100% the voltage drops.

    Could you please explain what is triggering the fault pin and what is shutting down the device?

  • I missed the datasheet information that states PWM duty <96%.
    Will the fault pin become active if duty is driven beyond >=96%?
    It seems that the device enters a intermittent state where it operates most of the time but somethimes shuts off and triggers the fault.
    Could you please explain what is driving this behaviour?
  • Hi Anders,

    Sorry, I did not realize our experts had not responded. Let me ping them again.
  • Hi, Anders,

    Based on your comments this seems to be an Over temperature, Over current or a combination of both problem. Can you please provide the input voltage used at IN+ when the issue appears?.

    The max recommended operating voltage for this device is 26V, so the use of 27V may damage the device as the PWM spikes generated due the PWM output switching could overpass the operation limit of the amplifier. 

    One thing you might want to check is that the internal BST diodes of the device are not damaged, this can be done easily with a DMM, measuring between BST and GVDD pins.

    Best Regards,

      -Diego Meléndez López
       Audio Applications Engineer

  • Hi Diego!

    Thanks for your reply.

     The problem occurs over a range of input voltages. My intention was to get maximum current out of the device. This I achieve from about 1.4V on IN+(IN- at GND) and PVCC 24VDC. However, then I also get the fault intermittently. The device is not warm I can easily put my finger on the part and not feel any heat. Can it really be an overtemp problem then?

    I power the device below 28V which is the over voltage trig point. Can it be some spikes that triggers the overvoltage detection and generates the fault?

    Diodes are 0.56- 0.6 on pins 17- 32 except the GND pins.

    I have used two TEC parts in series and needed the high voltage to create the needed current. The problem seems to disappear when I run the device at a duty less than 100%.( lowering the voltage on the IN+ pin)

    Removing one TEC device I can deliver the needed current at a duty below 100%.( Lower IN+ voltage neede for the same current)

    Increasing the PVCC voltage above 28V but below 30V I do not get any over voltage shutdown(28.5V I did not dare to go higher) 

    Lowering the PVCC voltage with one TEC so that the duty reaches 100% the problem occurs again. It seems that one explaination can be that I was violating the <96% duty requirement. However, there is no information on that this will cause the fault pin to be activated.

    Is my conclusion correct?

    In the device datasheet header, it is claimed that the part can handle 15V/+-4A. I assume that this is across the stated -40 deg C to 85 deg C.

    Is this correct understood?

    There are no specified data in the datasheet that defines these parameters.

    I assume that it is the DUTY, overvoltage, over temperature, rdson all together with the layout that gives the stated features.

    Is my understanding correct?

    In the application I am looking at, the environment is max 40 deg C. I therefore expect that I can use a higher voltage than the 15V.

    I will be needing max 4A but this can happen at maybe 20 volts across the TEC. However, the ambient conditions are not as server as in your claimed performance.

    My understanding is that as long as I fulfill the recommended operating conditions and keep margin to the absolute max operating conditions  I will not violate the parts reliability

    Is my understanding correct?

  • Am I on the right track or is there something else causing the issue?

  • Hi, Anders,


    Sorry for the delayed response. Your assumption is correct, as long as the conditions from the datasheet are met, the device reliability won't be compromised.

    After reading your test process, I can see that the issue could be effectively related to the duty cycle as the BST capacitors are charged during the high side of the switching cycle. These capacitors hold the GS voltage high enough to ensure the high side MOSFETS are fully turned ON. If these MOSFETS are not fully turned ON, the device might trigger a false overcurrent event, entering into fault state.

    Best Regards,

      -Diego Meléndez López
       Audio Applications Engineer

  • Just a complementaty question with regard to the implementation.
    From the DRV595 system block diagram it is clear that the return of the switch transistors are separated from the other GND pins.
    There are however three GND pins on the ringht side of the chip.
    Are all three floating from the rest of the chip?
    My plan is to measure the current through the TEC through the floating GND pins.
    Is this possible?
    My interpretation of the block diagam is that if the thermal pad is going to be referenced to the GND plane of the board. It needs to be connected to the GND pins not associated with the switch transistors.
    Is this correct understood?