Cable to connect function generator to oscilloscope

[fishandchips]

Hello, I need a cable to connect from the function generator to the oscilloscope. Any good recommendations? What is a commonly used length? Thanks

[alm]

Most of your question is answered in this thread:
https://www.eevblog.com/forum/testgear/50-ohm-bnc-cable-recommendation-for-scopefreq-gen/

As for length, it obviously depends on your bench setup. For a small bench 1m-2m should suffice. Possibly longer for a very wide bench or equipment on overhead shelves.

[fishandchips]

Thanks. Do signals degrade as the length increases? I suppose it should be fine for lengths between 1-2m. I probably get one at 1.5 m.

http://physics-astronomy-manuals.wwu.edu/Kenwood FG-273A Function Generator Manual.pdf

In the manual and on the function generator, there is no mention about the impedance of the output ports. Will the "RG-58 cable with 50 Ohm BNC", "RG316 cable with 50 Ohm BNC" and those "male BNC to some kind of grabber cable, male BNC to banana plugs, female BNC to grabbers that you put on the end of a regular BNC - BNC cable!" work? I also need a T-shape connector.

I plan to use it with the Micsig tBook mini TO1104 scope. Not sure if those four inputs to the scope are 1M Ohm input nctnico mentioned in the thread you cited.

[alm]

Your link does not work, working version:
http://physics-astronomy-manuals.wwu.edu/Kenwood FG-273A Function Generator Manual.pdf

Under 'output characteristics', it states that the output impedance is 50 Ohm ± 10%. I did not find the Micsig manual in a few seconds of searching, but I am pretty sure that it will state that the inputs will be 1 MOhm in parallel with some amount of capacitance. So the comments from the earlier thread and the linked video about termination applies.

Do connect the generator to the scope, you would only need a BNC-BNC cable and termination. But connecting the scope to the function generator is not that useful unless you are testing the generator or the scope. Usually you want to connect a DUT inbetween. In that case you might connect your DUT to the generator with some sort of BNC to whatever cable: the DUT end could be BNC, some kind of grabbers/clips, or other connectors like pin headers or SMA. The scope will usually be connected to the DUT through a scope probe.

Yes, signals degrade as the length increases, but you are not likely to notice this unless your cable is dozens of meter long or your frequencies are much higher than the 2 MHz your generator is capable of. I attached a table of the attenuation at various frequencies of an RG-58 cable. You'll see that at 1 MHz the attenuation per hundred meter is only 1 dB, so for a 1 meter cable it would be 0.01 dB, or 0.2% power loss. You would need some very sensitive equipment do measure this. Thinner cables may be slightly worse, but again, not something you will notice at these lengths and frequencies.

Edit: belated attachment.

[StillTrying]

http://physics-astronomy-manuals.wwu.edu/Kenwood%20%20FG-273A%20Function%20Generator%20Manual.pdf

[fishandchips]

Is the different between RG316 and RG58 is that the former is more flexible?

[alm]

RG-316 is more flexible and the performance is probably slightly worse (which is irrelevant for your application). Either should serve you perfectly fine.

[JLNY]

RG-58 is going to be thicker and therby less flexible. RG-316 is much thinner and more flexible, but higher loss at RF Frequencies. The attenuation through a 2m length of RG-316 @ 2MHz or less will be around 0.1dB or less, so I wouldn't worry about it and just go with a thinner cable.

My personal preference is to not have huge gangly wires everywhere on my bench, so if it were me, I would probably go on the shorter side at 1m.

[technogeeky]

As Dave has mentioned in his videos, you can also cut out the middleman and just use a regular oscilloscope probe with a BNC adapter tip.

These look like:

.

The downside to these is that they have to fit your oscilloscope probe tip rather exactly, so it may be a slight hassle finding one to match.

This gets rid of the impedance mismatch problem entirely: since you aren't using any 50 ohm transmission line, you don't need to 50 ohm terminate.

[David Hess]

RG-316 is the silver plated Teflon version of 0.1 inch diameter RG-174.  RG-58 is the common solid polyethylene 0.2 inch diameter cable.  Either will work fine for typical function generator applications.  RG-316 is more flexible but RG-58 is more rugged.

[fishandchips]

Which brand and material are lgood? Is gold better than copper? Are those from oscilloscope companies such as G. Instek better?

[jpanhalt]

I am surprised no one has given a link to Dave's discussion of connecting a signal generator to a scope yet.  Here it is:


Enjoy.

John

[fishandchips]

I am surprised no one has given a link to Dave's discussion of connecting a signal generator to a scope yet.  Here it is:


Enjoy.

John

So, either I get the gray high frequency x10 oscilloscope probe or get a 50 Ohm terminator for the coaxial cable like technogeeky mentioned?

I plan to use the Kenwood function generator with the Micsig tBook Mini TO1104.

Where do I get that gray high frequency x10 probe that can be connected to the function generator? Am I right that if I also feed in the signals from the function generator to the electronic circuit, I get a BNC T adapter first. Connect one end of the T adapter from the function generator to the scope via the the high frequency x10 probe. The the other end of the T adapter to the circuit via some kind of grabbers/clips, or other connectors like pin headers? Not sure what DUT mentioned by alm means.

[alm]

I am surprised no one has given a link to Dave's discussion of connecting a signal generator to a scope yet.  Here it is:

It was linked in the second post of the very similar thread I linked to.

Where do I get that gray high frequency x10 probe that can be connected to the function generator?

It has been a while since I watched that video, but I am pretty sure it is just a basic x10 scope probe (could as well use a switchable 1x/10x probe set to 10x). Many probes come with an adapter that lets the tips connect to a BNC socket. I do not think these adapters are particularly generic, so I would not assume an adapter from a random other probe to necessarily fit your probes. If your probe did not come with a BNC adapter, it might come with one of the springs that fit around the ground sleeve and provides a very short (low inductance) ground path. This is easy to fabricate from a piece of wire (wind it around something with a slightly smaller diameter than the grounding sleeve). But that would mean having to hold the probe.

Am I right that if I also feed in the signals from the function generator to the electronic circuit, I get a BNC T adapter first. Connect one end of the T adapter from the function generator to the scope via the the high frequency x10 probe. The the other end of the T adapter to the circuit via some kind of grabbers/clips, or other connectors like pin headers? Not sure what DUT mentioned by alm means.

DUT = device under test, i.e. whatever device you are measuring. Might be a circuit on a breadboard, or a piece of equipment you are repairing, or a single component, whatever.

For the typical application of using the generator to provide a signal to the DUT, for example testing the gain of an amplifier, you would typically connect the generator directly with a cable to the DUT (this is where something like grabbers or hooks might come in handy), and use a scope probe to probe whatever point of the circuit (could be for example the input and output of an amplifier). Whether you would need to terminate your generator at the DUT depends on the input impedance of your DUT and the frequencies involved (for audio you typically would not bother).

You would not usually connect the scope directly to the generator unless you were testing either, or as some sort of baseline. But in my opinion things like BNC-BNC cables, tees and terminators are basic lab equipment, so it is not bad to have them.

[fishandchips]

Thanks alm. As far as I understand, the issue mentioned by Dave's video only happen when the frequency is very high. Am I right?

I recall that many years ago, I plugged in a T connector to the function generator. Then, one end of the T connector to the oscilloscope while the other end to an op-amp circuit. I adjusted the gain of the circuit to see how the input and output signals looked like on the scope.

[alm]

It happens when the required bandwidth is high compared to the length of the cable. For sine waves the frequency is identical to the bandwidth, but a square wave may require a bandwidth that is far beyond the frequency. This is why Dave was using square wave signals in his example. With a 2 MHz sine wave and a 2m cable, I doubt you would see any difference at all between a terminated and unterminated signal apart from the termination halving the amplitude. At those frequencies you can safely use a tee and a straight BNC cable to your scope. If the signal needs more bandwidth, then you may have to consider termination and stub length.

[David Hess]

Most function generators including your Kenwood FG-273 are not fast enough for termination to be required with cable lengths which are reasonable for a workbench.  Even the TTL sync output is not fast enough for any cable shorter than about 10 feet.  The important characteristic here is transition time which is 100 nanoseconds for the FG-273 function output and 25 nanoseconds for the TTL sync output.

As far as lengths, I usually make my own cables with lengths based on a propagation delays of 2 or 5 nanoseconds which comes out to 16.4 and 41 inches for Teflon cables like RG-400 (my favorite) and RG-316.  For the more common RG-58 and RG-174, this would be 15.6 (2 nanoseconds) and 38.9 (5 nanoseconds) inches but unless you are working at frequencies much higher than 100 MHz, having matched and accurate lengths is unimportant.  Also consider RG-8X cable which is lower loss and at least in my experience, is more flexible than RG-58.

There is a better way to use a BNC-T if you can accept a little bit of delay.  Place the BNC-T at the high impedance oscilloscope input and run the function generator output through the BNC-T and then to the terminated device under test.

[fishandchips]

Could anybody please check if the following cables and T adapters are OK? Thanks

http://www.active123.com/200303-BNC-Cable-RG58-Male-To-Male-3Ft-Prodview.html
http://www.active123.com/TL310-BNC-Plug-To-Seizer-Probe-Tips-36-Cord-Prodview.html
http://www.active123.com/TL305-BNC-Plug-To-Alligator-Clips-39-Cord-Prodview.html
http://www.active123.com/213651-BNC-T-Plug2-Jacks-Prodview.html

https://www.amazon.com/gp/product/B0002JJU4G/ref=ox_sc_act_title_1?smid=A33V2A8QM8RVCQ&psc=1
https://www.amazon.com/gp/product/B0002JJU14/ref=ox_sc_act_title_2?smid=ATVPDKIKX0DER&psc=1
https://www.amazon.com/gp/product/B0002JJU2S/ref=ox_sc_act_title_3?smid=ATVPDKIKX0DER&psc=1
https://www.amazon.ca/gp/product/B00MFRYGG0/ref=ox_sc_act_title_1?smid=A2D9L39QUE9EAD&psc=1

https://www.amazon.ca/Elenco-Electronics-TL-3-Minigrabber-Test/dp/B0002JJU14/ref=pd_bxgy_60_img_2?_encoding=UTF8&psc=1&refRID=5QEPNTVPET3DKRYTNY3T
https://www.amazon.ca/gp/product/B0002JJU4G/ref=ox_sc_sfl_title_1?ie=UTF8&psc=1&smid=A3DWYIK6Y9EEQB