Is it normal to be able to measure frequency and duty cycle with only one probe of a digital multimeter?

I found out accidentally that when I only had the red probe touching the anode of an LED I was using for PWM testing, I got the number I thought I was supposed to get for both duty cycle and frequency.

When using the black probe on the cathode the frequency stayed the same, but the duty cycle was changed. Would this be from the voltage drop over the LED? (I was expecting 10% duty cycle, which I got when using only the red probe. With the black probe connected I got about 12.8%)

Hi chuckles73,

Was the multimeter touching nothing at all? It's strange that the multimeter gave you any reasonable reading at all without a reference voltage to read from. Personally I would trust the reading when everything was connected correctly.

Humberto

chuckles73-

hevans is right. Any other result could be purely capacitive. What meter were you using, and how was it set?

BM

I'm using a TekPower TP4000ZC.

It was touching nothing. Set to Hz. Holding the black probe in midair, red probe touching the anode. I confirmed this multiple times this weekend when the red probe by itself seems to get correct readings.

I was surprised by the reading, as I didn't think it was possible. I'm glad to know that I understand what's supposed to be happening, even when it isn't what's actually happening. Maybe the meter is using it's internal 9v's neg. terminal as a ground, whether intentional or not. shrugs

Thanks for the responses.

Chuckles73,

Are you by any chance using the RS232 interface connection? If so the ground may be connected through your electrical ground. I think this may be the case if you power your Nerdkit using external power adapter and the earth/ground connection gets completed through external means.

-Raja

I'm not using the RS232 (ie nothing plugged in), but I think it's on by default unless I press the RS232 button on the DMM. Tonight I'll try hitting the button and seeing if my readings change.

Thanks for the explanation, the only question I have left is, what frequency are you getting... 60Hz?

BM

I was trying to hit 50Hz for servo control (I have no servo, I was just playing around, getting to know the MCU, DMM and various parts).

I was doing this with an LED to mess with PWM and dimming at the same time as proof-of-concepting my ability to do servo control, if I had a servo.

The LED was blinking noticeably (well, noticeable when you jerked it around so your eyes couldn't just blend the images together), so I upped it to 100Hz. Then I let fly with a non-prescaled 8-bit timer... Then decided I wanted phase correcting, so I switched to a 16-bit timer with a TOP of 255. I should have divided it in half, but I didn't think about that at the time.

When I was measuring I was doing this last one, I was getting 31.something KHz on the multimeter (once again measuring with only one probe). Doing some maths (16000000 / (255 * 2)) shows I should have been getting 31,372 cycles per second. In fact, measurements at all stages of this had the frequency correct. Also, again, duty cycle measurements with only the positive probe at all stages of this matched expected results.

Yeah, measuring with just the positive probe looks like it gets me accurate results. I'm just unsure as to why.

I didn't get a chance to test the RS232 tonight, I'll return to this post once I've tried flipping that off.

I'd love to see a screenshot of the meter while you're getting this, so we can see display, dial settings, etc.

BM