• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 20 subscribers
  • Views 1503 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.

TPL5110: Timer interval accuracy vs. datasheet

Avihai O
Prodigy 170 points
Part Number: TPL5110


Hello,
I'm testing the TPL5110 accuracy in order to decide to use it in my projects.
In the datasheet, the timer accuracy is reported to be 0.6%, but in my tests I get errors of 3-10% in the interval time in "EN" mode (timer).
I'm calculating the t_ip according to the Rext resistance measure with a multimeter.
I used several resistors and all gave me errors. The only case when I got a smaller error of 0.5-1% is with Rext=500ohm (for t_ip=100ms).

Questions concerning the accuracy in section 7.5 in the datasheet:
1. "Oscillator accuracy over supply voltage" - what is the base voltage to calculate from, 0V or 2.5V?

2. same question as #1 for "Time interval Setting Accuracy over supply voltage"

3. "Time interval Setting Accuracy" - does it already include the accuracy of #1 and #2? which errors does it include?

4. Concerning the quantization error - Equation 3 seems to give very big numbers. for example:

for Rext=8.215kohm (Tdesired=3.85sec), Tadc=INT[5.5]=6sec. Equation 3's result is 56% which is a strange big number.

Can you clarify how to use this equation to check my expected error?

Is it taken into account in the timer accuracy?


Thanks,

Avihai

  • 0
    TI__Genius 16845 points
    Hi Avihai,

    Q1. "Oscillator accuracy over supply voltage" - what is the base voltage to calculate from, 0V or 2.5V?
    A1. The product of your supply voltage and this spec give your accuracy. For the calculation, use the VDD in your application.

    Q2. same question as #1 for "Time interval Setting Accuracy over supply voltage"
    A2. Again, multiply this spec by VDD for your accuracy.

    Q3. "Time interval Setting Accuracy" - does it already include the accuracy of #1 and #2? which errors does it include?
    A3. No, this does not include inaccuracies due to supply voltage, temperature, and so on. This specification only accounts for the accuracy of the time interval at VDD = 2.5V and 25°C ambient.

    Q4.Concerning the quantization error - Equation 3 seems to give very big numbers. for example:
    for Rext=8.215kohm (Tdesired=3.85sec), Tadc=INT[5.5]=6sec. Equation 3's result is 56% which is a strange big number.
    Can you clarify how to use this equation to check my expected error?
    Is it taken into account in the timer accuracy?
    A4. The quantizer uses Rext to select between 1650 different possible time intervals. This means that although you can choose any value for Rext, there is a fixed set of time intervals that can be programmed. For a desired time interval of 3.85 sec, the actual time interval the device will generate is Tadc=INT[5.5]= 5 sec.

    I hope this helps.

    Kind regards,
    Lane Boyd
  • 0 in reply to Lane Boyd
    Prodigy 170 points

    Lane, thank you very much for you answers.

    I think that there is a mistake in the equation in the datasheet. Rd should be equal to Rext.


    Best regards,
    Avihai

  • 0 in reply to Lane Boyd
    Prodigy 170 points

    Other questions please:

    1. taking into account all those errors doesn't give me an error close to the actual number which I get in measurements:

    for example, Vdd=4V, Rext=9.226kohm at 25 Celcius degrees gives 6.112s instead of 5.901sec (error of 3.6%)

    what could also contribute to the error while it's a completely new chip and I get similar results in 4 other chips?

    2.When using the timer with Vdd=1.8V, the actual minimal timer period is 200ms (instead of 100ms).
    Is this a known or expected issue?
    Also, at this voltage I don't get a square wave, but something more like a toothsaw.

    Thanks again for your help.

  • 0 in reply to Avihai O
    TI__Genius 16845 points
    Hi Avihai,

    Q1. taking into account all those errors doesn't give me an error close to the actual number which I get in measurements:
    for example, Vdd=4V, Rext=9.226kohm at 25 Celcius degrees gives 6.112s instead of 5.901sec (error of 3.6%)
    what could also contribute to the error while it's a completely new chip and I get similar results in 4 other chips?

    A1. I tried to replicate your measurement and got similar results. For VDD = 4V, Rext = 9.23kohm at room temp, I measured t_DRV = 6.07 sec. The timer interval period for this Rext is 6 sec, so there is an error of 1.17%. This is within my expectation as the datasheet typical error given of 1%.

    Q2. When using the timer with Vdd=1.8V, the actual minimal timer period is 200ms (instead of 100ms).
    Is this a known or expected issue?
    Also, at this voltage I don't get a square wave, but something more like a toothsaw.

    A2. I wasn't able to replicate this issue; I observed a square-wave output with 100 ms period on DRV in timer mode. One thing to try is reducing the capacitance on DRV to increase the edge rate.

    Thanks for pointing out the datasheet mistake, I will make sure this is corrected in the next revision.

    Kind regards,
    Lane Boyd
  • 0 in reply to Lane Boyd
    Prodigy 170 points

    Thanks Lane.