EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: newtekuser on May 01, 2023, 12:59:32 am
-
I am still learning how to use an oscilloscope and at the same time trying to make sense of the measurements I take but have trouble discerning whether what I'm seeing is a result of a faulty design in my circuit, faulty oscilloscope, or probe.
As part of the troubleshooting I tried taking measurements without any probes attached and baffled about what I'm seeing:
i.e.:
* Readings with no probe attached:
5v/div: 800mVpp
2V/div: 400mVpp
So it's picking up some noise from somewhere. My coupling is set to DC, I don't see a difference.
* 500MHz P5050 probe attached to signal compensation hook:
5v/div: 1.8Vpp
2V/div: 1.44Vpp
1v/div: 1.2Vpp.
* 100MHz inexpensive probe from ebay attached to signal compensation hook:
5v/div: 800mVpp up to 1Vpp
2v/div: 400mVpp up to 480mVpp
In both cases, probe attenuation is set to 1:1.
Is this normal and so should I just subtract these values from the readings I get when measuring signals in my circuit?
LE:
The compensation setting on the cheap 100MHz probe is set to 10X so I set my horizontal to 500mV/div. It now measures 80mVpp up to 100mVpp.
However, the P5050 probe is set to 10X from factory and has no adjustment switch, but when I set the horizontal to 500mV/div it doesn't measure peak to peak, instead it displays an exclamation mark instead of the reading.
Do I have a bad P5050 probe then, or is there some other setting that needs adjusting?
-
The compensator output is intended to be used with the probe set to 10x. You're only looking for distortion in the square wave (undershoot and overshoot) and not the absolute value.
https://www.rohde-schwarz.com/us/products/test-and-measurement/essentials-test-equipment/digital-oscilloscopes/understanding-probe-compensation_254520.html (https://www.rohde-schwarz.com/us/products/test-and-measurement/essentials-test-equipment/digital-oscilloscopes/understanding-probe-compensation_254520.html)
Short the BNC at the scope input and see what you get. Everything else can be blamed on stray fields and such. Reconnect the probe and short the tip to the ground lead. What do you get?
-
Most scopes have an adjustment for 1X, 10X and possibly even 100X (and others). You should still set the V/div to 5V and let the scope worry about the fact that it is really jut 0.5V at 10X. The scopes are really good at doing this.
The Rigol DS1054Z has 16 ratios from 0.01X to 1000X in the usual 1,2,5 sequence. 0.01X, 0.02X, 0.05X, ... 1X, 2X ... 1000X.
One way to think about it: At 10X, the probe is dividing the signal by 10 and the scope is multiplying the V/div by 10 to compensate.
It might be handy to just use a 10X probe every chance you get. It presents less load to the device under test. To that end, I have a set of fixed 10x probes (Probemaster) and they don't even have a switch. I'm of a mind to glue the switch in the 10X position on the other probes. Testing high voltages assuming you are in 10X and actually being in 1X is a good way to get a new scope.
-
Most scopes have an adjustment for 1X, 10X and possibly even 100X (and others). You should still set the V/div to 5V and let the scope worry about the fact that it is really jut 0.5V at 10X. The scopes are really good at doing this.
The Rigol DS1054Z has 16 ratios from 0.01X to 1000X in the usual 1,2,5 sequence. 0.01X, 0.02X, 0.05X, ... 1X, 2X ... 1000X.
One way to think about it: At 10X, the probe is dividing the signal by 10 and the scope is multiplying the V/div by 10 to compensate.
It might be handy to just use a 10X probe every chance you get. It presents less load to the device under test. To that end, I have a set of fixed 10x probes (Probemaster) and they don't even have a switch. I'm of a mind to glue the switch in the 10X position on the other probes. Testing high voltages assuming you are in 10X and actually being in 1X is a good way to get a new scope.
The P5050 is a 10X probe, so I assume I don't have to make any adjustments in this regard. Since it is measuring 800mVpp without probing anything, do I need to subtract this value from the value when probing something?
-
ground the probe tip to the ground clip, what do you see? Should be a few millivolts at most.
Even grounded, depending your environmental factors, you might see some noise from local transmitters. I see signals around 97.3 and 101Mhz as there are two strong transmitters that reflect off the Bay but with the probes grounded, they are barely detectable. I can see these transmitters on my spectrum analyzers pretty much all the time.
If when grounded, you still see a signal of a few millivolts, then maybe your probe is defective with a broken ground. Try another channel and probe. The only scope I had at hand was a Siglent 1204x-e. With a 10x probe grounded it displays 5mv of noise and with it open it is all over the place.
-
Maybe I have a bad scope acquisition board? A second P5050 probe arrived in the mail today and seeing the following:
No probes connected to any of the channels (all active) - verticals set to 10 mV/div:
CH1: 2.4 mVpp
CH2: 2.4 mVpp
CH3: 2.4 mVpp
CH4: 2.4 mVpp
CH1 + P5050 Probe (shorted to its ground terminal) - fluctuates between 12 mVpp up to 17 mVpp
CH2 + P5050 Probe (shorted to its ground terminal) - fluctuates between 11 mVpp to 13 mVpp
(moved the first two probes to CH3, CH4:)
CH3 + P5050 Probe (shorted to its ground terminal) - fluctuates between 12 mVpp up to 15 mVpp
CH4 + P5050 Probe (shorted to its ground terminal) - fluctuates between 10 mVpp to 12 mVpp
However, scope passes all tests including long SPC calibration and probe calibration.
-
That sounds reasonable to me, 2.4 mVpp is an extremely small signal.
The TDS5000 scopes are nice instruments, they have TekProbe interfaces that utilize the readout pin on compatible Tek probes to automatically set the vertical scale depending on the attenuation factor of the probe.
-
I believe I was using the probe and scope incorrectly. When I click on auto-set, the scope automatically sets the vertical to 500mV and I do see the 1V signal produced from the probe compensation hook to be exactly 1V.
Also, when I measure peak to peak it reads about 1.12V which is exactly what I also read from the reticle.
Am I reading this correctly?