DIY 10:1 oscilloscope probe

[peteb]

I am trying to make a 10:1 oscilloscope probe from a 1:1 probe that is just a straight wire. The guide I am following shows the 1:1 probe having its own resistance and capacitance. Mine does not.

the idea is that the impedance of the scope is 1 Meg and adding a 9 meg resistor on the end cuts 90% of the signal.

here is what happened using the square test signal from the scope itself. The square signal could be used for calibration purposes.


test signal thru A and B, both with 1:1 probes

Now the lower signal is thru the 10:1 extension. It is the same probe, I can just takes the signal in at two different places. The wave is reduced by approximately 10:1 but it no longer is square.

Now the 10:1 signal is multiplied by 10 by the oscilloscope. It actually does look lIke the incomplete DIY 10:1 probe is successfully reducing the signal by 10 times, and now the signal has a clear shape, but it clearly is a different shape from the original signal.

solution? Try adding the prescribed cap.

it is interesting that yesterday I went to the electronics store for parts for two problems. Both problems needed a single resistor with a bypassing cap.


in both cases I tried the resistor by its self, and in both cases the results were mixed. Now is the time to try the caps. I did both resistors with one warmup of the soldering iron, and now I will do the same with the caps. It is so common in electronics to see caps used with resistors. It is like the developers had a typical process of using resistors and then adding caps to make the resistor do what they had in mind in the first place.

 

[Digital Larry]

An exponential rise on the front edge of something that is supposed to be a square wave means you are losing high frequencies. This is often due to parasitic capacitance which is unavoidable and whose impact is worse in a high impedance circuit.

The cap with a series resistor in parallel with another resistor acts as a high frequency shelving boost, whose frequency response can be estimated by math. OMG MATH RUN AWAY! I think you are on the right track as regards trying to match the square wave. Be interested to know what you finally settle on.

 

[peteb]

The cap with a series resistor in parallel with another resistor acts as a high frequency shelving boost, whose frequency response can be estimated by math.

Thanks Larry.

what you describe is what they show inside the probe. I will give it a try. Would you put that in series with what I have, like their diagram shows?

An exponential rise on the front edge of something that is supposed to be a square wave means you are losing high frequencies. This is often due to parasitic capacitance which is unavoidable and whose impact is worse in a high impedance circuit.

this is what I would like to do. To understand the relationship between the resistors and the capacitors and the shape of the signal.

you say it is unavoidable. It does make sense that my RC probe end has a very high cut off frequency. What I have looks like a high pass filter at a high cut over frequency. However, is it not allowing the signal to return back to its own ground, kind of like a bypass filter cap allows the AC that developed across the cathode to return back to its ground, thru the bypass cap?


I did the cap like they show on the probe tip:

these two signals are at the same magnification on the O scope. The tip no longer cuts the signal by ten Times.

the tip now allows the scope to keep up with the transition, but it does not maintain the level.

it is fun to work on a scope without the hassle of the signal generator or the amp and the dummy load.

 

[peteb]

 

[peteb]

This is my new way to store two probes in one bag. Seperate them with a piece of paper so they don’t get tangled together.

Now I can take one probe out without taking both probes out.

 

[Digital Larry]

this is what I would like to do. To understand the relationship between the resistors and the capacitors and the shape of the signal.

you say it is unavoidable. It does make sense that my RC probe end has a very high cut off frequency. What I have looks like a high pass filter at a high cut over frequency. However, is it not allowing the signal to return back to its own ground, kind of like a bypass filter cap allows the AC that developed across the cathode to return back to its ground, thru the bypass cap?


I did the cap like they show on the probe tip:

these two signals are at the same magnification on the O scope. The tip no longer cuts the signal by ten Times.

Ok the new waveform shows that the high frequencies are "too much" relative to the low frequencies. Getting all this to balance out with the voltage ratio etc. is probably past even me at this point without resorting to SPICE or something.

Here are my rules of thumb for RC circuits.

If you have an RC circuit, the -3 dB point (typically called the cutoff frequency) will be the frequency at which the capacitive reactance equals the resistance it "sees" - which gets a bit complicated.

At low frequencies, capacitors are "not there"... meaning their capacitive reactance is so high as to not matter.

At high frequencies, capacitors are a "short circuit" - yes there is phase shift involved but this can be ignored in many simple cases.

Capacitive Reactance - an overview | ScienceDirect Topics

www.sciencedirect.com

Even if you are not mathy, look through the diagrams here and see how they correlate to what you're seeing.

https://web.pa.msu.edu/courses/2014spring/PHY252/Lab4.pdf

 

[peteb]

Thank you Larry.

it looks like I should just buy a 10:1 probe.

it has been interesting.

 

[LightningPhil]

What bandwidth are you going for? A straight restrictive divider works quite well to a point. The lower the impedance, generally the better the bandwidth.

There's a 200 kV divider kicking about somewhere that's good to about 100 kHz. No caps at all. Not as good as the Tektronix 25 kV high speed probes. But they're full of correction caps. Also the same price as a custom guitar mind.

 

[peteb]

Thanks Phil,

I would be interested in seeing only guitar related frequencies.

1 Hz - 2k Hz.


I feel I have lost my way. The resistor alone did what it was supposed to do, the cap undid what the resistor did.

I see no clear path forward other than to buy a 10:1 probe online for $12-16.


the only clear path I see is to retain the resistor. Make a high pass filter with a cut off around 1 Hz, and a low pass filter with a cutoff around 5k or something. I suspect however that this setup would reproduce the original signal without reduction.

might as well just buy one.