[Boatanchors] Fw: Probe Refresher

[Doug Hensley]

Forwarding per Alex's request.  He can usually answer any question you might have on the Oscilloscope.


Cheers,


Doug


________________________________
From: Oscilloscopes at aol.com <Oscilloscopes at aol.com>
Sent: Sunday, April 30, 2017 11:30 AM
Subject: probe refresher

Hello again Doug,

(Please broadcast this kind of notice to all scope users),

 I'm sure you realize that scope probes (unlike like DMM probes) use shielded cable. The greater need to shield the probe from stray pickup is in large part due to the far greater bandwidth a scope input is sensitive to as compared to a voltmeter.

The shielded direct (1:1) probe is usually limited in useful bandwidth to about 10MHz. A direct probe is not the same as a 50 ohm coax line connecting a 50 ohm source to a 50 ohm load, which usually has a much greater useful bandwidth. The biggest difference is that the center conductor of a probe is a resistive wire. This helps reduce internal reflections as usually neither the source, nor the load impedance match each other, or the probe's coax cable. The resistance value of a probe's center conductor is on the order of a hundred ohms, which in series with the scope's 1 Mohm input still leaves the overall load as 1 M ohms, and the gain as unity. A scope input typically also offers about 20pF of shunt capacitance, and a direct probe adds about 100pF, for a total of about 120pF. It is this 120pF burden at test points that limits upper bandwidth to about 10MHz.

A divide-by-ten probe (aka 10:1) uses a 9M ohm resistor at the tip which will generally show the test points only about a 10pF & 10Mohm burden. This reduced burden enables such 10:1 probes to offer much greater bandwidth than direct probes, at the expense of tenfold reduced sensitivity.

For a 400MHz probe to pass 400MHz with high fidelity, one more condition needs to be satisfied. The usual 6 inch (15cm) alligator ground lead that sticks out the side of the probe needs to be made/kept shorter. Those 15cm are a stray inductance with broad resonance around 150MHz, well within the bandwidth capabilities of scopes I sell. By shortening the grounding lead to less than a third (5cm or less), we move that resonance to beyond what 400MHz scopes care about. Since this seems inconvenient to do all the time, you might just do it when you plan to probe at over 50 or 100 MHz.

Lastly, if a scope can tolerate 400V at its input (as in case of 2465B), hooking up a 10:1 divider probe does not mean that now you can push 4000V before causing damage. This is because most divider probes are also only rated for 400V or so.


Thank you,



Alexander Schonfeld
OSCILLOSCOPES, ETC.
4828 Cache Peak Dr
Antioch CA 94531 USA
mobile 1 408 499 2552

PS: 10;1 probes can be identified by having an adjustment called "compensation". If they are to be 10:1 dividers, they not only have to be 10:1 resistive dividers, but capacitive, and inductive as well, to be frequency flat