Behavior after 50kHz max. frequency calibration


#1

Dear all,

I’ve found strange behavior of the system in configuration setting of 50kHz maximum waveform frequency. Attached worksheet is the log of my calibration experience.

After the “Device reference calibration” with 50KHz max. freq., warning message “Failed to achieve the specified voltage” was popped up at each switching action if “Auto_adjust_amplifier_gain” check was OFF.

Measured amplifier output voltage by DropBot is lower than actual.

On the other hand with “Auto_adjust_amplifier_gain” check was ON, higher output voltage than setting was found. It seems to be caused from miss reading of output voltage by DropBot.

These behavior isn’t found after 20kHz max. frequency calibration. So as a work around, I’m using the system calibrated with 20kHz maximum calibration even in the higher frequency operation.

I’d like to have your advise. Is there any miss understanding in my calibration steps?

Aki @ Univ. of Tokyo

calib_log.xlsx (11.5 KB)


#2

Hello Aki,

If your 20kHz calibrations give the proper voltage readings then I recommend just using that calibration. In the wheeler lab we have our DropBots calibrate using the default settings and recommend them for others as well. This calibration is used to determine the exact resistance and any parasitic capacitance in the reference resistors that the DropBot uses to calculate the voltage and these values can be used for the higher frequencies too. The ideal frequencies for normal use (moving, mixing, splitting) is around 10-20 kHz anyways so unless you are needing high frequencies for some special purpose default calibration settings cover the useful range.

When you did the calibration at 50 kHz did you change the number of steps or only the end frequency? If you didn’t then you will only have data points for 10 frequencies from 100 Hz to 50kHz when in the default settings you get the same amount of data points over a much smaller range. This means you might not get enough data points for each resistor which could lead to the improper voltage readings.

Hope that helps!

Josh


#3

Hi Josh,

Thanks for your reply.

Your recommendation is the same to our current setting as a work around. I understand this is more suitable to get stable operations and set it to the regular procedure.

Thanks,

Aki@Univ. of Tokyo


#4

Hi Aki,

I should warn you that if you are using the Trek amplifier, it becomes increasingly unstable at higher frequencies. Also, the Arduino can only sample at ~100kHz, and may not give accurate readings at higher frequency.

These two factors combined mean that things may not behave well at frequencies >20kHz. We tend to operate with an upper frequency limit of 10kHz for best performance.

-Ryan

···

On Tue, Jan 30, 2018 at 1:22 AM, ytoyo...@nojilab.t.u-tokyo.ac.jp wrote:

Hi Josh,

Thanks for your reply.

Your recommendation is the same to our current setting as a work around. I understand this is more suitable to get stable operations and set it to the regular procedure.

Thanks,

Aki@Univ. of Tokyo

You received this message because you are subscribed to the Google Groups “dropbot-dev” group.

To unsubscribe from this group and stop receiving emails from it, send an email to dropbot-dev+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.


#5

Ryan,

I’ve got your kind advise.

Basically, we are using 10kHz in our experiment. We need to set higher frequency for some of the case.

I will carefully monitor the output in such cases.

Thanks,

Aki@U-Tokyo

···