Why aren't there reflections in oscilloscope probes?

Go To Last Post
6 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

So - I'm assuming that oscilloscope cables are 50 ohms. So why aren't there huge reflections on them? I mean, on a 1X cable you have about a 1M ohm source impedance driving the 50 ohm cable, and the load at the other end of the cable is something like 1M as well. Shouldn't this cause major reflections due to the impedance discontinuity? What am I missing?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

'scope probes, e.g., the common 10x probe, are very high impedance so as to not affect most common circuits. It does not become part of the transmission line under observation.

You will see "ringing" on the fast transitions of pulses. These are due to the probe's ground lead being imperfect. Want big ringing? Look at a logic signal with a bare wire pair connected to the 'scope, without the high impedance 10x probe.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Not quite.

Well made scope probes use resistive center conductor in the coax between the probe and the scope connector. It is carefully tuned to just dampen the reflection. there are major tradeoffs between bandwidth/risetime, cable length, probe input C, and a few other factors.

The reflection mechanism is as follows:

1) pulse launched onto cable (low-Z compared to "termination") reflects from the high-Z scope input.

2) Reflection travels back to the nose and reflects from the Hi-Z of the 9Meg resistor in the nose.

3) Reflection travels back toward the scope.

4) Oscilloscope sees the reflections, back and forth, as ringing.

5) Distributed resistance in the center conductor provides enough loss at the ring frequency (which depends on the cable length) so that it is not significant.

The attenuator is not put at the scope input on probes with a bandwidth more than a few MHz because of capacitance. If you put the 9Meg resistor at the scope input, then the cable C appears as a shunt at the probe input. But, if you put that resistor out at the nose, the cable C is reduced by attenuation factor of the probe (as seen at the probe nose).

Hope this helps
Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Very careful design by somebody who knows what he's doing.

I've always wondered what the "probe compensation" adjustment actually was doing. I know how to use it, but never actually figured out what it did.

 

277,232,917 -1 The largest known Mersenne Prime

Measure twice, cry, go back to the hardware store

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Probe compensation works this way.

1) The scope input consists of parallel R & C. The R is typically 1 megohm for low frequency scopes. High frequency scopes often have a switch to change between 1 meg and 50 ohms.

2) If you simply add a 9 meg series resistor to this, you have a circuit that rolls off at high frequencies.

3) Now, add some C in parallel with the 9 meg. Lets call the scope input Rin & Cin. Lets call the attenuator components Ratten and Catten. Now, if Rin*Cin = Ratten * Catten, there is no rolloff (from this source).

4) Something in this system requires the ability to adjust because you cannot make any of the parts perfectly accurate in value. Early scope probes made Catten adjustable by making it coaxially shaped and the resistor run down the center. Twisting the probe nose would turn screw thread that would slide one part in or out with respect to the other. More modern construction add a "box" at the scope end of the cable with an adjustable cap, there.

5) High bandwidth probes often add compensation networks to try to correct for skin effect losses and other factors that make probes work less well above 50MHz.

6) The issue of ground lead inductance at the probe nose has bedeviled scope users ever since scopes did better than 25MHz (we are talking 1950's, here). There have been a variety of somewhat successful ways of dealing with this.

7) The "scope probe" that consists of a piece of coax with clip leads on the end is basically unusable for anything more than crude measurements above 100KHz. Remember, its not the repetition frequency that is important, but the rise/fall times.

Jim - Former Tek engineer (all this was figured out by engineers much smarter than me)

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I know that this is dragging up an old post, but this is entirely relevant and interesting.

"The Secret World of Oscilloscope Probes" is a good article written by Doug Ford for an Australian magazine. It explains exactly the issues discussed above.

http://www.dfad.com.au/links/THE...

-Tim