TMAG6181-Q1: Sensors forum

Ivo Hofman

Part Number: TMAG6181-Q1

I need to read turns counter from the TMAG6181-Q1 exactly, but the output signal has the fluctuations around 200ns, what is more than last digit.

So I do not know the quadrant.

Power is blocked by electrolyt and the ceramic, nothing helps.

Can you help me?

over 1 year ago

0 Scott Bryson over 1 year ago

TI__Mastermind 21020 points

Ivo,

Does the fluctuation occur at all angles or only near the switch points? Can you read the marking on the top of the device?

Thanks,

Scott

0 Ivo Hofman over 1 year ago in reply to Scott Bryson

Prodigy 10 points

Dear Scott,

the fluctuatios occur at all angles. The top marking is 41 T1 6181.

Thank you,

Ivo

0 Scott Bryson over 1 year ago in reply to Ivo Hofman

TI__Mastermind 21020 points

Ivo,

thank you for providing this. I will try to get a bit more information what might be going on. I plan to have a better idea by tomorrow.

How often are the fluctuations? Since the fluctuations are only about 1 step wide, a possible spurious increment should be able to be ignored since it requires 2 code steps to move a full 180 degrees. If you are reading the sine and cosine outputs, then you should be able to distinguish the false step. This is not a preferred option, but might be a step to consider while I look to get you more details what might be happening.

Are you able to capture any output plots of the PWM?

Thanks,

Scott

0 Ivo Hofman over 1 year ago in reply to Scott Bryson

Prodigy 10 points

Scott,

I sended video on the www.uschovna.cz/zasilka/PEK2F8IAMH82MGRN-VS8. I used osciloscope and it is vissible from it, that

the fluctuations of the falling edge are so big, that it is not possible to have correct turns from it. It is connected to the development

board with the PIC micro. I tryed also another supply, but it had no effect to fluctuations.

Thank you,

Ivo

0 Scott Bryson over 1 year ago in reply to Ivo Hofman

TI__Mastermind 21020 points

Ivo,

Thanks for the video. I will review this with our validation team. Do you have the lot trace code from the packaging for your order?

Scott

0 Ivo Hofman over 1 year ago in reply to Scott Bryson

Prodigy 10 points

Hi Scott,

the photo of the packaging is on www.uschovna.cz/zasilka/PEKC8C79PL6YSF64-WK3

Ivo

0 Scott Bryson over 1 year ago in reply to Ivo Hofman

TI__Mastermind 21020 points

Ivo,

Thank you for confirming, I am following up with the validation team and we are looking to see if we can reproduce this.

Thank you,

Scott

0 Scott Bryson over 1 year ago in reply to Scott Bryson

TI__Mastermind 21020 points

Ivo,

I setup on the scope in the lab, and did see a similar event when I zoomed in on just the falling edge. The one challenge posed here is this is only half of the duty cycle measurement. I also see the same motion on the subsequent rising edge. This means that the period length has some deviation to it, and is the cause of the deviation in the falling edge timing. The timing of both rising and falling edge events are required to accurately measure the full duty cycle of the PWM.

Unit to unit may see different oscillator frequencies, and we see here that it is also possible to have small deviations in the period length, and so you cannot accurately determine the duty cycle unless the full period is measured.

Thanks,

Scott