50 Ohm Feedthrough

[TheDriver]

For the first time I connected my new function generator, Wavetek 288, to my scope, tek 2235, and found the amplitude to be doubled. At first I thought something was wrong with the generator but come to find out it was because of the impedance mismatching.

Is the only way to fix this problem a 50 Ohm feedthrough?

Any recommendations on what make, models, or specs to be looking for?

Thanks,
Matt

[Frank Berry]

If the function generator has a BNC output connector, use a BNC Tee on the input. Connect the cable to one port and a 50 ohm termination on the other port.

[TheDriver]

Could I use the T or feedthrough on the output of the generator instead? Or do I have to put it on the input of the scope?

Thanks,
Matt

[Frank Berry]

Input of the scope would be best. That way, you won't have "echos" due to cable impedance mismatch.

[frank1]

The generators output voltage is specified when terminated in 50 ohms. If a generator with 50 ohms output impedance is un-terminated, its output voltage will be +6dB (double).

Do a search for "BNC through termination".
You can get either 50 or 75 ohm ones to suit.

Frank

[Conrad Hoffman]

Be careful with the term impedance mismatch. At audio frequencies it's pretty much irrelevant and the output amplitude of the generator simply changes due to resistive loading. It's certainly due to impedance, but nothing is mismatched in the RF sense, it just is what it is. OTOH, if you've got the sig gen cranked up to a couple MHz, you will definitely see reflections and ringing if you have significant cable lengths and a mismatch.

[Frank Berry]

Originally posted by Conrad Hoffman
Be careful with the term impedance mismatch. At audio frequencies it's pretty much irrelevant and the output amplitude of the generator simply changes due to resistive loading. It's certainly due to impedance, but nothing is mismatched in the RF sense, it just is what it is. OTOH, if you've got the sig gen cranked up to a couple MHz, you will definitely see reflections and ringing if you have significant cable lengths and a mismatch.

Absolutely correct. The longer the cable length, the lower the frequency at which you'll begin to see ringing.

[TheDriver]

How does a T make the gen readout correct?

A feedthrough has a resistor to ground, correct? Could I measure the gen output across a 50 ohm resistor and get the gen read out to match the scope?

[Conrad Hoffman]

The usual way to solve this is with a 50 ohm terminator. These are usually a small block with a male BNC on one end and a female BNC on the other. Common from HP/Agilent and Tek. It's nothing more than a 50 ohm resistor from the center pin to the shield. If you put one at the end of your 50 ohm cable it will give you maximum signal fidelity on square waves. Some scopes have the same thing built in. Be sure not to end up with 25 ohms from using both at the same time. A feedthru usually refers to a feedthru capacitor, used for bringing power through the side of a chassis wall. It will have a capacitance to ground, but no resistor. I don't know what you have as a readout, but when I do audio gain or other measurements, I never trust anything except a measurement of both input and output levels, be it on a scope or with a high bandwidth voltmeter. It might be more common to have a meter on the generator in the RF world, where precision is much harder to achieve. In the RF world, the working assumption is that everything is always correctly terminated, as the problems caused by mis-termination are more serious.

[coloradosound]

see this:
http://cp.literature.agilent.com/lit...989-1456EN.pdf

Hint 2 page 3.
Might be useful to use the formulas to scale for other loads.
I don't think that at audio freqs. there would be too much
concern for reflection.