lnDSO150 icon indicating copy to clipboard operation
lnDSO150 copied to clipboard

calibrate voltage / wrong voltage

Open konwin opened this issue 3 years ago • 4 comments

Hello, I successfully flashed my old original DSO150 with STM32F103CB. The only thing I observed is that the measured voltage is around 25% to low. The test signal is shown as 2.5V and an AA battery I checked with my multimeter at 1.18V is shown at 0.9V. Is this adjustable? Or have I done the calibration wrong (connect probe clamp to the ground clamp and leave connected while pressing ok and switching to AC/DC as stated)? PXL_20221213_232847394~2

konwin avatar Dec 13 '22 23:12 konwin

Hi The problem is depending on your hw revision, the internal gain/ resistors are different, so it will always be wrong on some hw. The "basic calibration" only calibrates the zero, i.e. dc offset. So if you see zero when you ground the signal, you did it correctly.

On my todolist i have the "advanced" calibration to calibrate YOUR device properly. I did that in the past, it works fine, but the code is missing at the moment.

Short term, i can suggest either : 1- If you can, change capture/dso_adc_gain.cpp. it contains the gain for the different stages. 2- I can blindly change it, but i dont know your hw, might be better , might be worse.

BTW, the main goal of the "advanced" calibration was to be much more accurate. In the past, when it's done, the value are not bad at all.

mean00 avatar Dec 14 '22 06:12 mean00

The explanation is here : http://goingbacktoelectronic.blogspot.com/2019/02/dso150-input-gain.html

mean00 avatar Dec 14 '22 06:12 mean00

I'm manually splitting the voltage range issue into Issue #9 https://github.com/mean00/lnDSO150/issues/9

mean00 avatar Jun 18 '23 15:06 mean00

@konwin There is now in master a 5v calibration method to lower the error due to internal Vref variation It is assumed to be 1.2 but i've measured it up to be up to 10% wrong One calibrated, the absolute error goes down A LOT.

mean00 avatar Jun 18 '23 15:06 mean00