• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 559 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.

ADS131M04: Oscillator temperature rise issue

Amaury LAINE
Prodigy 50 points
Part Number: ADS131M04

Hello,

I made the exact same circuit that the ADS131M04EVM, but my oscillator SIT8924BA-22-33E-8.192000G has temperature rise of 10c comparing the other IC arround. It is not big but this is not normal for an oscillator. My footprints are good and all looks good. I don't understand why the oscillator has a temperature rise on my board and not on the dev board. I can see my clock too.

I

  • 0
    TI__Mastermind 18900 points

    Hello Amaury,

    Assuming you the EVM, have you compared against the actual EVM? 10C sounds like a lot but it may be the same as our EVM, I've never checked.

    Thermal issues are a principle of layout. Generally, current through metal is the source of an area getting hot. Less metal and more current results in more heat with radiates radially from the IC and path of current.

    Reviewing your layout, I notice that you have poured 3V3 on the top and bottom layers and one of your inner layers has a GND pour. we poured GND on the top and bottom layer of the PCB and within an inner layer.  So, 3.3V might be responsible for current into the oscillator, but GND (or GND_ISO in your design) is what holds the return current. It looks like all of the return current goes through 1 via and a relatively skinny trace (maybe no more than 15mil). 

    As a result, I'd either put more metal around the GND connection to the device and add more vias to connect down to the GND layer and dissipate heat in a wider area or just pour GND on the top layer.

    In addition, sometimes the current might not be flowing through the device, but there might be a lot of current flowing nearby which causes the device to heat up. I'd check your circuitry nearby and make sure nothing is shorted and pulling a lot of current which might flow under the device and heat it up.

    Alternatively, sometimes heat issues occur with manufacturing if the connection between the device pin and pad is not good, which is the equivalent of not putting a lot of metal in the path of current for the device. You might want to remove the device and put it back on to see if anything improves.

    Best,

    -Cole

  • 0 in reply to Cole Macias
    Prodigy 50 points

    Thanks Cole for the answer. I based my design on the dev board and it look like i have the same kind of trace width and via config. I have checked the temperature rise on the dev board and nothing to see, it is working properly. 

    I don't know why our oacillator is getting hot and the linear DCDC 5V to 3.3V too.

    I didn't played with SPI register yet and i was wondering if maybe some configuration might impact the oscillator consumption

  • 0 in reply to Amaury LAINE
    TI__Mastermind 18900 points

    Hello Amuary,

    Agreed the EVM looks similar. There's still more area dedicated to GND so more heat should be dissipated through the metal which lowers the temperature. You'll still need to double check if current from any other source might be running near the device. A good comparison is P=IV if you know the oscillator's quiescent or operating current. Oscillators are not supposed to any more than a couple of mA so something tells me that something is wrong, and your baseline test with the EVM kind of shows that.

    As a result, I would definitely look into manufacturing and remove the oscillator and put it back on--or even better, replace it with the one on the EVM and see if the thermal issue follows the oscillator or the board.

    Best,

    -Cole

    Edit: if you've never seen our layout recommendations for how to layout ADCs, I suggest you check out this post: https://e2e.ti.com/support/data-converters-group/data-converters/f/data-converters-forum/755516/faq-pcb-layout-guidelines-and-grounding-recommendations-for-high-resolution-adcs