DRV8306: IC DRV8306 Lot problem

Part Number: DRV8306

We have tested DRV8306 with two lots. We found the phase commutation logic inconsistent. Please refer attachment.

The phase connection U, V & W does not work in New lot (V W & U works).

Please justify the above. All our circuitry and components are fixed, then why same IC behaves differently? 

Regards,

Avinash M.

  • We have tested the IC on the same PCB with the same motor and connection. We found both ICs are logically inconsistent. It seems that the phase connection within the IC package has been changed. The phase output in our Old lot is UVW, however, in a new lot, it is VWU, remaining all setup and connections are the same.

  • Hello Mechtex,

    Can you please send an image of the top of the package for each lot of devices?

    Thanks,

    Matt

  • Hi Mechtex,

    Thanks for sharing this information - we are following up on this accordingly within our teams, and will keep you updated. 

    Best Regards,

    Andrew

  • Hi Mechtex,

    For the problem being described, can we have some more details on the scope of the problem?

    1. Do all the devices within the new lot (1149298CD1) have the same problematic behavior? 
      1. or is it a small qty within the new lot?
      2. how many units of the new lot were tested using the same PCB?

    Thanks and Best Regards,
    Andrew

  • We have tested a new lot with about 100 qty till now, further 200 qty is in assembly. In all of the new lots, we have same observation.

  • Hi Mechtex,

    Thanks for the update - 

    Would it be possible to send a few units from each lot (new lot and old lot) to me, so that our team can analyze them further? 

    The address to direct them to would be: 

    12500 TI Blvd, Dallas, TX 75243
    addressed to Andrew Liu, office coordinates J2-1635

    Thanks and Best Regards,

    Andrew 

  • Hi Andrew,

    Do you want just the chips or do you want us to send the whole PCB and motor? We already have a shortage of parts and no new chips are available and this is exacerbating the issue.

    Regards,

    Mechtex Team.

  • Hi Mechtex,

    Understood on your concern, and we will provide support accordingly 
    I am checking with the team to confirm our local bench capabilities. Will let you know if the PCB and/or motor are needed in the shipment.

    Before we initiate the shipment, I would like to ask for a clarification of the problem - 

    The phase output in our Old lot is UVW, however, in a new lot, it is VWU, remaining all setup and connections are the same.

    The 1xPWM commutation logic table is in page13 of the datasheet.

    • Link: https://www.ti.com/lit/ds/symlink/drv8306.pdf
    • Question: Can you help confirm that you are observing behavior that doesn't match this table?   
      • some waveform scope captures may help confirm this issue for our team to investigate root causes 
      • this would be something like: 
        • scope capture of Hall A,B,C on same time domain as GHA/GLA (Phase A) 
        • scope capture of Hall A,B,C on same time domain as GHB/GLB (Phase B) 
        • scope capture of Hall A,B,C on same time domain as GHC/GLC (Phase C) 
        • scope capture of Phase A, B, C outputs on same time domain: SHA, SHB, SHC

    Best Regards,
    Andrew 

  • Hi Andrew,

    We have a 4 channel scope so we can do the following:

    • scope capture of Hall A,B,C on same time domain as GHA (Phase A) 
    • scope capture of Hall A,B,C on same time domain as GLA (Phase A) 
    • scope capture of Hall A,B,C on same time domain as GHB (Phase B) 
    • scope capture of Hall A,B,C on same time domain as GLB (Phase B) 
    • scope capture of Hall A,B,C on same time domain as GHC (Phase C) 
    • scope capture of Hall A,B,C on same time domain as GLC (Phase C)
    • scope capture of Phase A, B outputs on same time domain: SHA, SHB
    • scope capture of Phase B, C outputs on same time domain: SHB, SHC

    DIR and PWM will be pulled high to VREF voltage for all of the above experiments. This will be performed for 1 PCB made from the old lot and 1 PCB made from the new lot.

    Will that work?

    Also, we are based in Mumbai, India, so shipping to anywhere in India will be better (time a paper work-wise). Not to mention faster communication (as we are in the same time zone). 

  • Hi Mechtex,

    The itemized list of scope captures looks good - acknowledged on the 4-channel limitation of the scope. 

    The first six will allow us to compare the HALL input signals and resulting GHx/GLx outputs to the commutation logic table (posted earlier above).

    For the last two relating to SHA/SHB/SHC, wanted to ask if it's possible to get all 3 on the same scope capture, using 3 of the channel voltage probes?

    if not, it may still be fine - we are requesting that one just to double-confirm the order of U->V->W 

    For the remaining info on test conditions, 

    • ok to have DIR consistent, as we want the test cases to be all in one direction of motor rotation. 
    • one test per PCB (new lot vs old lot) should be ok. Is there any difference in the PCBs? 

    For the comment regarding India shipment logistics and time-zone communication, I'll look into what can be done. 
    Some options could be to have a late-night or early meeting live to resolve the communication cycletime. However, I will have to ask my team if there are any applications personnel in our India office that can help evaluate the physical devices. 

    Thanks and Best Regards,

    Andrew 

  • For the last two relating to SHA/SHB/SHC, wanted to ask if it's possible to get all 3 on the same scope capture, using 3 of the channel voltage probes?

    So you just want us to probe the 3 motor phases, right? That can be done instead of the other two readings. 

    • one test per PCB (new lot vs old lot) should be ok. Is there any difference in the PCBs? 

    No, the PCBs are the same. They are from the same manufacturing lot (For the PCBs). The only difference is the old and new lot driver ICs.

    For the comment regarding India shipment logistics and time-zone communication, I'll look into what can be done. 
    Some options could be to have a late-night or early meeting live to resolve the communication cycletime. However, I will have to ask my team if there are any applications personnel in our India office that can help evaluate the physical devices. 

    Sure we can see what time suits both of the teams (lets discuss about this on mail). Please check if there is any local support available as that would be preferable. We will be sending you other data via email soon.

  • Hi Mechtex,

    1. For the SHA/SHB/SHC, correct - probing the 3 motor phases should be good
    2. Thanks for the information on the PCBs. Looking forward to reviewing the results 
    3. Understood - I will align with my manager today on the possibility of local support in India
      1. acknowledged on the availability of a future meeting invite. Will anticipate details over email 

    Thanks and Best Regards,

    Andrew

  • Hi Aadit, Mechtex Team,

    Following up on some of our other recent discussion items,

    1. India local support
      1. talked to my manager about this, and got feedback that we don't have the relevant diagnostic labs or bench equipment in our India-location office. Therefore, we will have to rely on our Dallas location for the analysis of physical samples 
    2. Samples shipment
      1. While Avinash and team are collecting the waveform images, would it be possible to initialize the shipment of some samples to our Dallas location in the U.S.? I have included the address below again for convenience.
      2. hoping to work on the physical samples analysis in parallel. Something like 3-5 samples from each lot, labeled, might be sufficient for the shipment. 

    Samples shipment destination: 
    12500 TI Blvd, Dallas, TX 75243
    addressed to Andrew Liu, office coordinates J2-1635

    Best Regards,
    Andrew

  • Hi Andrew,

    talked to my manager about this, and got feedback that we don't have the relevant diagnostic labs or bench equipment in our India-location office. Therefore, we will have to rely on our Dallas location for the analysis of physical samples 

    Alright, we coordinate with our logistics and sales teams to arrange shipments for you guys.

    While Avinash and team are collecting the waveform images, would it be possible to initialize the shipment of some samples to our Dallas location in the U.S.? I have included the address below again for convenience.

    We have sent you the waveforms as requested by email. Please have a look and let us know. If the same course of action needs to be followed.

    hoping to work on the physical samples analysis in parallel. Something like 3-5 samples from each lot, labeled, might be sufficient for the shipment. 

    What kind of samples do you want? do you want our PCB and Motor? or just the ICs. I can send you more samples from the new lot. However, we only have one sample from the old lot, as all the rest were sent to our customers. Please let me know.

    Regards,

    Aadit.

  • Hi Aadit,

    Thanks for sending over the waveforms and data via email -

    I was looking at them just now to review the info, and will bring it up to our team members today to discuss possible root causes & share feedback accordingly. 

    For the samples shipment, we would like just the ICs. No motor/PCB necessary, based on the info provided about both lots' PCBs being the same. 

    For the new lot, please help ship 5-10 samples or so to us at the address below.
    For the old lot, please help ship the one remaining sample 

    Thanks and Best Regards,
    Andrew 

  • Hello Aadit,

    I think we have been able to resolve our questions directly. May I close this thread?

    Thanks,

    Matt