Voltage divider for PWM Signal NI9220

[Martin88]

Hi Froum,

i uses a NI9220 (+-10V input) with cDAQ i would like to measure a PWM Signal that varias from +13V to -13V with different states from +9, +6 and +3V and the flowing current is in few mA range. 
I am interessted to measure the quality of the PWM signal on all states (rising time and falling time, frequency pulse width) so it should be accuarate. I was thinking to use a simple 10:1 voltage divider with accuarate 10kOhm and 1kOhm resistors. Does anyone has experience if this could work or how accuarte the signal will be?

Thanks 

Martin

[johnsold]

Martin,

The combination of 10 kohm and 1 kohm resistors does not give you a 10:1 voltage divider. It is 11:1. Aside from that  a votage divider should be fine. That divider will draw slightly more than 1 mA with 13 V input. If the source is current limited, you may need to go to higher value resistors.

The stray capacitance of the wiring to the voltage divider will limit the bandwidth. If your PWM frequency is less than ~100 kHz, this should not be a problem. If the frequency goes into the megahertz range, you may need carefully evaluate the frequency response.

I do not uderstand what you mean by "...states from +9, +6 and +3V..." Are these the desired output voltages? Or do you have a Pulse Amplitude Modulated (PAM) system?

Lynn

[Martin88]

Hi Lynn,

thanks for the fast reply. 

Yes the PWM frequency is about 1000Hz so I should not be limited by the badwidth. 

Its also a PAM system where it shows the state with amplitude change. (see attached Picture)

[johnsold]

I am curious. What is the purpose of a signal like that which changes both amplitude and pulse width?

Lynn

[Martin88]

With the amplitude high i try to detect different error stages in the device. The PWM then gives me informations about how much current can flow.