• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 51 subscribers
  • Views 432 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

AMC3301EVM: AMC3301EVM output oscillates

mhn
Prodigy 30 points
Part Number: AMC3301EVM
Other Parts Discussed in Thread: AMC3301

Hi

I ordered 3 pcs of AMC3301EVM evaluation boards in 08/2020.
I have found that the outputs of these boards oscillate. The problem is similar in all of these boards.

Here is an oscilloscope screen capture:

AMC3301EVM's INP terminal driven by signal generator with 0Vdc. INN terminal grounded. Power supply's minus terminal grounded.

Math 1: VDD - GND (power supply voltage)
Ch 3: Voltage between board INP and INN.
Ch 7: OUTP
Ch 8: OUTN
Math 2: OUTP - OUTN

As it can be seen, the outputs oscillate. There is no excitation at the input nor in power supply voltage at the same frequency.

The situation is similar when the input is shorted. The situation gets worse if the input is open:

Ch 7 frequency measurement "Meas 8" happened to be 1 MHz at the instant this above capture was taken. Most of time it was ~312 kHz.

Here are the package markings of ICs used in these boards:

03AH9ST
PAMC3301Q
G4

Here is the board used in these measurements:

Is there a design flaw in these ICs or what is going on?

  • 0
    Prodigy 30 points

    Here is the board used in these measurements:

  • 0 in reply to mhn
    Guru 140200 points

    Hi,

    output filtering is obligatory for these chips. See figure 8.2 of datasheet :-)

    Kai

  • 0 in reply to kai klaas69
    TI__Mastermind 47625 points

    Hi mhn,

    Thanks for bringing this to our attention. 

    How is your signal generator connected to the input? 

    If you are not dropping the signal generator output across a resistor, I suggest doing so. Oftentimes, signal generators struggle to source infinite amounts of current and require some sort of current limiting to operate effectively. 

    How were the inputs shorted? Bus bar between the two input terminals? If the short is made directly at the input pins does the issue persist? 

    Are you signal probes measuring differentially? VOUTP/N with respect to the low-side GND?

    As Kai mentioned, we do suggest filtering the output. The default output filter cutoff frequency is very high - the cutoff frequency can be modified to fit your applications requirements.  

  • 0 in reply to Alexander Smith
    Prodigy 30 points

    Hi Alexander.

    "How is your signal generator connected to the input?"

    The signal generator has a BNC connector, so I have a BNC to terminal block adapter from which the input is wired with two 0,75mm^2 wires. BNC connector outer rim is grounded and connected to INN, and center tap (signal source) is connected to INP.

    The signal generator was connected directly to input without any resistor. However, there are 10ohm series resistors on AMC3301EVM board at the input. Also, AMC3301 has a differential input impedance of 22 kohm.

    "How were the inputs shorted? Bus bar between the two input terminals? If the short is made directly at the input pins does the issue persist?"

    With a 2cm long 0.75mm^2 solid copper wire. Issue persists.

    "Are you signal probes measuring differentially? VOUTP/N with respect to the low-side GND?"

    I have both OUTP & OUTN wired directly to oscilloscope input. Then the low-side GND connected to earth ground. Also power supply minus terminal is earth grounded. This yields the lowest amount of disturbance in output. Still it is way too much. Math 2 is the differential signal: OUTP - OUTN. See below.

    Here is the test setup. Simple and not very pretty.

    "As Kai mentioned, we do suggest filtering the output. The default output filter cutoff frequency is very high - the cutoff frequency can be modified to fit your applications requirements."

    I can see your point here, but I would like to be able to evaluate the IC with an evaluation board. It should be usable in a real application straight away. On board filtering: input filter -3dB freq. ~ 970 kHz. Output filter -3dB freq. ~ 46 MHz. However, as is shown in datasheet :

    There is active low pass filter in the output. Also, "BWOUT Output bandwidth 334 kHz typical", which in my mind means that the active filter -3dB point is 334kHz and if the active filter gain is ~ 6.24dB (=2.05) it starts to attenuate quite quickly after that frequency. (4th order low pass filter -80dB/decade) So there shouldn't be > 1 gain at the frequency of oscillation (> 670 kHz). So, the output is low pass filtered to much lower frequency than what the on board output filter -3dB point is.

    As datasheet states: Output noise = 300μVRMS (INP = INN = HGND, fIN = 0 Hz, BW = 100 kHz). I have 6 mVrms differential.

    I also found out that when I cycle the power to board (without touching the board at all), the peak-to-peak voltage oscillation may be higher on OUTP on one time and then higher on OUTN on second, and sometimes even.

    So, I'll ask again. What is going on with these boards?

  • 0 in reply to mhn
    TI__Mastermind 47625 points

    Hi mhn,

    Previously, your screen shots showed 300kHz and 1Mhz, which I could not easily explain - hence all of my questions. 

    Your most recent screen shot shows ~625kHz, which is expected behavior due to a chopper. Additional filtering can be added to attenuate this noise, or it should be canceled out when doing a differential measurement. If you are indeed seeing 300kHz and 1MHz noise, there could be something in the environment causing the disturbance such a fluorescent lights or a faulty ground connection somewhere in the system.