Can I use a scope probe to inject RF from my sig gen??

[Chris Wilson]

Can I use a normal X1 / X10 oscilloscope probe on the 50 Ohm output of my HP sig gen to inject RF into a dead transceiver for fault tracing? The service manual says to inject a 14MHz signal at about -110dBm . If the fault is after the IF the IF is circa 70Mhz (Kenwood TS-590 with dual IF paths). If not what can I do for a suitable probe on the end of the co-ax from the sig gen at these frequencies? Thanks.

[Ian.M]

Not in any predictable fashion - the scope probe coax is deliberately lossy, so even on x1 you'll be guessing the output level at the probe tip, and on X10 it will be a highly frequency dependent unknown attenuator. 

Also -110dBm is 10pW (picoWatts), so even if your generator output level is only 7mV RMS, you'd have to stack four 50R matched 20dB attenuators to get its output down to an appropriate level. 

[TheDane]

What kind of probe do you have?


If a normal probe to you is an active type, then the answer is no - as it has most likely an amp inside it.
 
If it is a passive 100MHz x1 probe with a 14MHz signal, then most likely yes - is my experience. Make sure there is no attenuation (it is an active short from input to output), unless you want/need it. -110dBm is not for beginners.

I recommend reading about some myth regarding probes, it's quite enlightening, especially the 3dB part in 1):
http://www.electronicdesign.com/test-measurement/11-myths-about-oscilloscope-probes

quote: "A widely accepted rule of thumb is to use a probe with 3X to 5X the clock frequency or the fastest toggle rate in a digital system. This makes it possible to capture the third or fifth harmonic of the fundamental frequency of the clock or digital signal, resulting in a signal on the screen of the oscilloscope that more accurately represents the true signal with square-like edges. Another useful rule of thumb is that BW × Tr = 0.35 (for 10-90 Tr). Using this rule of thumb, one can either determine the needed bandwidth to measure a given rise time or the fastest edge that can be measured for a probe of a certain bandwidth."

[james_s]

I seriously doubt it's an active probe, those are specialty hardware that one is not typically going to have unless they know what it is and why they need it.

I see no reason why a scope probe can't be used to inject RF. It may not be ideal, it may not give easily quantifiable results, but RF will come out the tip of the probe and it won't hurt anything.

[vk6zgo]

If all you want to do is poke some RF into the input & see if it comes out somewhere else, why not?
If you wanted to input a known signal level, youwould not use a 'scope probe.