开云体育

The usual way to build such a probe is simple and excellent for measuring very fast sub-nanosecond pulses. It consists of a length of 50 ohm coax that's terminated in 50 ohms at the scope, either with a 50 ohm input on the scope (usual on high bandwidth scopes) or with a separate? termination. At the input end a small, low inductance resistor is soldered to the center conductor and the shield twisted into a small tail for the ground connection. The lead length is kept as small as possible and the leads often soldered directly to the test points. With a 450 ohm input resistor the probe divides by ten, a 4950 ohm resistor divides by 100. The input resistance is 500 and 5000 ohm respectively. This seems very small, however at high frequencies any capacitive loading is severe, the reactance of 1 pf at 1 GHz is approx. only 160 ohms. These probes are quick and cheap and the short leads are essential for reliable signal coupling; an inch of lead has significant inductance at high frequencies. Even the most professional engineers with access to the latest equipment make them up since they are so handy. Incidentally, the multi thousand dollar many GHz probes are often soldered directly into the test points to minimize lead inductance. I've attached the instructions for making a probe, they are often simpler if a commercial termination is used at the scope end and doesn't have to be built into the cable. The resistor can be tinier too. It's a bit of art to make them with a really flat response, suitable test equipment to check response is valuable, but if made carefully they are capable of useful performance as is. There was a link to a you-tube discussion forum a few weeks back on probes that was very good, if anyone has the link to it, it's worth watching. Tektronix sold some coax probes like this, I don't know if they still do so. I have a 100:1 probe that has a switchable 50 ohm termination at the scope end. For really high frequencies the simple minimalist home made ones are probably better since the lead length is small. The ground lead in particular is important and is often a short spike to touch a ground point.
On the other hand, if you want to look at mid frequencies, use a standard high impedance 10:1 probe. The low impedance of a coax cable will load the circuit. An unterminated coax with clip leads is OK for audio frequencies but has too much capacitance for much higher than that.

Don Black.


Don Black.

On 20-Feb-13 3:05 PM, Cliff White wrote: