Why are freq ref inputs not always 50 ohm?

[magnehh]

Hi all!

My question is: Why are frequency reference inputs (e.g. 10MHz) not always 50 ohms?

I've been wondering about this for ages, but have never found an answer. The only reasonable explanation I can come up with is to be able to daisy-chain devices, with a terminator at the last device in the chain.

I just tested this with my beautiful, old Agilent 53131A, and it's not even close to 50 ohms. I did this by hooking a sig. gen. to a tee at the freq ref input of the 53131, then to a scope with another tee. I then tested with and without a 50 ohm terminator at the scope. Instead of halving, it only shrunk about 10% or so.

Any help would be appreciated!

[TheSteve]

I don't think your testing method will give very good results. Assuming you had the 50 ohm termination enabled on the counter and it isn't damaged it will be very close to 50 ohms. You can measure its resistance at DC with a multimeter which will give you a good idea if it is enabled properly and functioning. To really know its performance though you'd need something that can measure its return loss or SWR up to the rated input frequency. A network analyzer or a spectrum analyzer that can make scalar measurements using a tracking generator and a return loss bridge or directional coupler.

[magnehh]

Thanks for your input, TheSteve. Correct me if I'm wrong, but I think you may be thinking about the counter inputs on the front, which are indeed selectable 1Mohm or 50 ohm. I'm wondering about the input on the back, which provides the reference clock (Ref in).

Sometimes, these inputs are transformer coupled, hence my method of injecting a 10MHz signal to measure the loading.

[Vgkid]

The input impedance on the 5335A is 1K , I have seen values usually around a couple hundred ohms(5-600) .

[TheSteve]

Thanks for your input, TheSteve. Correct me if I'm wrong, but I think you may be thinking about the counter inputs on the front, which are indeed selectable 1Mohm or 50 ohm. I'm wondering about the input on the back, which provides the reference clock (Ref in).

Sometimes, these inputs are transformer coupled, hence my method of injecting a 10MHz signal to measure the loading.

Ahh, sorry, yes I was thinking of the front inputs. I don't think I've seen any 50 ohm ref inputs.

[maxwell3e10]

Having higher impedance allows one to connect more devices to one 10 MHz source. Even though the proper thing to do is to use a 10 MHz distribution amplifier, in practice it usually works fine with several devices on one cable. If some standing waves develop due to impedance mismatch its not a big problem as far as the frequency is concerned.