50 ohm terminator??

[UnkleBidii]

What is the purpose of a 50 ohm terminator? I've never used one and quite frankly in school I don't remember ever using one. I just connect one of the probes directly to the scope for my small projects.. I would assume that it would cut out any unwanted noise but where and when would that noise play a role?

Thanks guys. I'm new here as well.

[helius]

A 50 ohm terminator will terminate a signal into 50 ohms. It is not used with the kind of probes you have.

When your signal comes from a 50 ohm output over a 50 ohm coax instead of a probe, the end of the cable connected to the scope needs to be terminated (needs to see an impedance equal to the cable's characteristic impedance). https://en.wikipedia.org/wiki/Characteristic_impedance#Transmission_line_model
Most scopes have a switch that activates an internal 50 ohm termination, but it is very limited in power dissipation capability and must not be used with signals over 5V peak.

[UnkleBidii]

This would only apply to say a direct coax cable connection from function generator to scope?

[Zero999]

This would only apply to say a direct coax cable connection from function generator to scope?

Or a circuit which is designed to drive a 50 Ohm load, yes that's one application.

Another thing you might need a 50R load for is for a low impedance probe. An ordinary high impedance ×10 probe, may have a DC resistance of 10M but has a fairly high parasitic capacitance, about 10pF to 15pF, giving it a low impedance at high frequencies: 265R at 50MHz, assuming a capacitance of 12pF. At higher frequencies, it's better to replace the 10M probe with a low impedance probe, consisting of a piece of 50R co-axial cable with a 450R resistor at the probe end and the 'scope set to 50R input or a 50R terminator added, if it doesn't have it built-in. The new probe will have an impedance of 500R from DC, well into the UHF region and will load the circuit less than a 10M probe.

[nour]

This would only apply to say a direct coax cable connection from function generator to scope?

Or a circuit which is designed to drive a 50 Ohm load, yes that's one application.

Another thing you might need a 50R load for is for a low impedance probe. An ordinary high impedance ×10 probe, may have a DC resistance of 10M but has a fairly high parasitic capacitance, about 10pF to 15pF, giving it a low impedance at high frequencies: 265R at 50MHz, assuming a capacitance of 12pF. At higher frequencies, it's better to replace the 10M probe with a low impedance probe, consisting of a piece of 50R co-axial cable with a 450R resistor at the probe end and the 'scope set to 50R input or a 50R terminator added, if it doesn't have it built-in. The new probe will have an impedance of 500R from DC, well into the UHF region and will load the circuit less than a 10M probe.

If my scope doesn't have a 50R impedance, only High impedance (Rigol ds1054z) how I can properly terminate it to have 50R impedance? can i just add 50R to the cable and that's it?

[Zero999]

This would only apply to say a direct coax cable connection from function generator to scope?

Or a circuit which is designed to drive a 50 Ohm load, yes that's one application.

Another thing you might need a 50R load for is for a low impedance probe. An ordinary high impedance ×10 probe, may have a DC resistance of 10M but has a fairly high parasitic capacitance, about 10pF to 15pF, giving it a low impedance at high frequencies: 265R at 50MHz, assuming a capacitance of 12pF. At higher frequencies, it's better to replace the 10M probe with a low impedance probe, consisting of a piece of 50R co-axial cable with a 450R resistor at the probe end and the 'scope set to 50R input or a 50R terminator added, if it doesn't have it built-in. The new probe will have an impedance of 500R from DC, well into the UHF region and will load the circuit less than a 10M probe.

If my scope doesn't have a 50R impedance, only High impedance (Rigol ds1054z) how I can properly terminate it to have 50R impedance? can i just add 50R to the cable and that's it?

The 50R terminator needs to be as near to the input of the 'scope as possible. You need to add a 'T' piece BNC to the connector on the 'scope with the terminator on one side and co-axial cable on the other.

[eb4fbz]

If my scope doesn't have a 50R impedance, only High impedance (Rigol ds1054z) how I can properly terminate it to have 50R impedance? can i just add 50R to the cable and that's it?

Get an external termination http://www.aliexpress.com/item/P7001-2PCS-High-quality-BNC-to-BNC-Female-seat-50KY-50-ohm-matching-device-Q9-adapter/32661228102.html

[Zero999]

If my scope doesn't have a 50R impedance, only High impedance (Rigol ds1054z) how I can properly terminate it to have 50R impedance? can i just add 50R to the cable and that's it?

Get an external termination http://www.aliexpress.com/item/P7001-2PCS-High-quality-BNC-to-BNC-Female-seat-50KY-50-ohm-matching-device-Q9-adapter/32661228102.html

That's much better, although the maximum power dissipation might not be as high as the OP's terminator.

[vasco]

BNC 50 ohm terminators were used at the end points of old thin ethernet networks in order to kill reflections. As I don't believe there's still a market for thin ethernet I'm sure that now a days they have another use.
I know that 75 ohm terminators are used in video application (and were used when connecting synchronous terminals to mainframes).
I once found a 92 ohm terminator connected to an ICL data-entry machine, possibly related in some proprietary data link.

[w2aew]

What is the purpose of a 50 ohm terminator? I've never used one and quite frankly in school I don't remember ever using one. I just connect one of the probes directly to the scope for my small projects.. I would assume that it would cut out any unwanted noise but where and when would that noise play a role?

Thanks guys. I'm new here as well.

You might enjoy my video tutorial on terminations:

[ed_reardon]

I was about to link your video sir!

Great series of videos, thank you for taking the time to share them with us, the coax-length one really did give me a way of thinking I hadn't thunk before!

[csar]

Hi guys,

I'm confused about termination. I have a function generator connected to my breadboard through a 50 Ohms coax cable.  My circuit is a simple RLC serie circuit. To match impedance, I used a 50 Ohms resistor. My question is: does the RLC circuit connected directly to the 50 Ohms resistor change the termination? If so, what is the solution?

Thanks.

[basinstreetdesign]

csar:  Yes your RLC circuit connected to the terminated coax cable does indeed change the termination impedance.  It will do so to the extent that the RLC impedance approaches 50 Ohms near the resonant frequency.  If your cable is shorter than about 2m then you may not notice anything besides a frequency-dependent load on the cable.  If it is significantly longer then reflections from the load end will also make the wave-forms at both ends messier.  The longer the cable, the more travel time  the reflections take and the messier it gets.  Chances are, your function generator is source-terminated in 50 Ohms so reflections shouldn't happen there as well.

In order to fix this, your circuit at the load end needs to do one of two things:
a - present 50 Ohms at all frequencies so you can remove the 50 Ohm resistor, or
b - keep the 50 Ohm resistor as terminator and buffer the received signal with an active (powered) circuit.  Then you can build whatever you want downstream of the buffer to deal with the signal.

[bson]

Any filter is generally designed to match the source impedance, so it doesn't need termination - it is the termination.  Consider a 1k source impedance in the form of a series resistor vs a 50ohm one charging up a capacitor; clearly the 50ohm source will charge the capacitor much faster, and to get the same time to charge it you need a much bigger capacitance value; 20 times bigger, in fact.  So clearly the filter component values need to be selected to match the source impedance.  A capacitor in a pi filter facing a very low impedance like a 10mohm driver is utterly pointless and it will do absolutely nothing other than waste power (unless it's humongous).  The same on the other end - the filter's output side source impedance is matched to the load.  For just experimenting stick a 1k resistor in series with the signal generator on the breadboard or whatever you use and design filters for 1050ohm.  If the load following the filter is practically infinite (such as a non-inverting op amp) add a known termination such as 47k to design against.  Larger values lets you reduce filter component values to where you can use off-the-shelf linear film capacitors and simple leaded inductors.

You want as much source and load resistance as is feasible for the signal levels and frequencies.  A 50ohm source is a good starting point for a function generator - you can always add more to suit!

[csar]

Thanks basinstreetdesign and bson 

basinstreetdesign: Is the use of buffer feasible even at high frequencies?

bson: to avoid reflections, the termination must be 50 Ohms. But, if the filter itself is the termination and its impedance changes with frequency, how the reflections are avoided?

[bson]

if the filter itself is the termination and its impedance changes with frequency, how the reflections are avoided?

They're not; a run of the mill passive LCR filter will reflect in the stop band.  It's possible to design low-reflection and reflectionless filters.

[tautech]

Some more understanding on why 50 termination is sometimes necessary to combat reflections and ringing:
https://www.eevblog.com/forum/testgear/show-us-your-square-wave/

[UnkleBidii]

Hey @w2aew! Yes, I do watch your videos...slowly making my way through. I'm Lyman on there...Thanks!