Author Topic: 50 ohm BNC cable recommendation, for Scope/Freq Gen  (Read 6944 times)

0 Members and 1 Guest are viewing this topic.

Offline JenniferG

  • Frequent Contributor
  • **
  • Posts: 255
  • Country: us
50 ohm BNC cable recommendation, for Scope/Freq Gen
« on: March 18, 2017, 09:41:56 am »
I need a cable to run from my frequency generator to oscilloscope, any recommendations? 

Also I need to run frequency generator to a circuit.. what is the the best cable and way to go about this?

I see 50 ohm and it's a BNC like 10Base2 ethernet.  If I recall correctly, I remember using 50ohm terminators on end of 10base2 cables -- back in mid 90's.   Do you ever need to put a T at the end and terminate with 50 ohm resistor when hooking up to scope or freq generator?   
Test Equip:  Instek 8251a w/ Probemaster 8028S, Commercial Electric MAD830B, Tektronix 2225
Power Supplies: HP 6236B, GPD-3303S (w/o overshoot problem)
Soldering Station:  Hakko 926, Backup: Velleman VTSS5
 

Offline Avacee

  • Supporter
  • ****
  • Posts: 298
  • Country: gb
Re: 50 ohm BNC cable recommendation, for Scope/Freq Gen
« Reply #1 on: March 18, 2017, 09:44:59 am »
Running a cable direct from function gen to the oscilliscope is not the optimal setup.
Dave explains what to not to do and demonstrates why and also what to do and why.



« Last Edit: March 18, 2017, 09:47:37 am by Avacee »
 
The following users thanked this post: JenniferG

Offline alm

  • Super Contributor
  • ***
  • Posts: 1258
  • Country: 00
Re: 50 ohm BNC cable recommendation, for Scope/Freq Gen
« Reply #2 on: March 18, 2017, 12:01:59 pm »
As for the actual cable, at the frequencies and lengths you are likely to use, I would suggest an RG-58 cable with 50 Ohm BNC connectors on both ends. RG-58 is probably the most common type of 50 Ohm coax cable out there (also commonly used for 10base2) and reasonably sturdy. It is not the most flexible cable, or the best shielded, or the lowest loss, but none of them should be terribly important for your application. You are not sending many Watts of power across a long cable to an antenna somewhere way up a mast.

Make sure you do not buy 75 Ohm cable (e.g. RG-59), as is commonly used for video applications.

Yes, if you want to connect your function generator (which will almost certainly have a 50 Ohm output impedance) to your scope, you either need to set the input of your scope to 50 Ohm instead of 1 MOhm (back in the Tek 2225 days this was mostly limited to higher bandwidth scopes, so I would guess the 2225 does not have that option) or use an external terminator.

The 50 Ohm feed-through terminator that Dave showed around the 7 minute mark is basically the same as the BNC tee and 50 Ohm terminator used in the 10base2 days. However, as you can plainly see, putting a tee creates a 'stub' from the tee to the scope input. As Dave discussed in his video, having an unterminated line results in reflections that can distort your signal. This stub would do the same at frequencies above 50 MHz or so. Hence the more compact feed-through terminator which is good up to about 250 MHz. Because feed-through terminators are quite niche (mostly used for this exact application), they tend to be quite expensive ($30-$50 or so last time I looked). For many applications the cheaper tee + terminator works just as well.

In addition to the cable and some kind of terminator, you might also want some way of connecting the function generator to your circuit, which may not have BNC inputs (e.g. a breadboard). There are various solutions, like a 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, or you can put one of these on the function gen and use banana leads. If you want to be cheap, you could even just grab a female banana jack and clip / solder some wires to it.

For low frequencies anything works fine. The higher the frequency (or the more sensitive you are to noise and interference), the more important it is to keep things in a coaxial environment for as long as possible, so in that case you would not run long banana leads from your function generator but convert to flying leads as late as possible. For a beginner I would probably just convert to binding posts for convenience. I like to use the short breakout leads with female BNC connector and clips or banana plugs on the end because they are modular.

There are other manufacturers than Pomona and many of them you can order from China for much cheaper, I just used Digikey for convenient links.
 
The following users thanked this post: CustomEngineerer, HoracioDos

Online nctnico

  • Super Contributor
  • ***
  • Posts: 18580
  • Country: nl
    • NCT Developments
Re: 50 ohm BNC cable recommendation, for Scope/Freq Gen
« Reply #3 on: March 18, 2017, 12:31:08 pm »
I have these for general purpose use. The cable is RG316 which is thin and relatively easy to work with.

This is the Ebay listing and the seller has other lengths as well.
http://www.ebay.com/itm/BNC-Male-plug-to-BNC-Male-Plug-RG316-Pigtail-RF-Jumper-Cable-50cm-/281487763038
These cables are cheap enough to cut up and solder one end to a board or something else. On cheap cables the connectors can be flaky but so far these cables have not shown that problem.

Just use T pieces and 50 Ohm terminator to feed the signal into a 1M Ohm scope input. The 1M Ohm input's capacitance will screw a signal up way before the stub added by the T pieces becomes a problem.
« Last Edit: March 18, 2017, 12:50:59 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 
The following users thanked this post: _Wim_

Offline HoracioDos

  • Frequent Contributor
  • **
  • Posts: 335
  • Country: ar
  • Just an IT monkey with a DSO
Re: 50 ohm BNC cable recommendation, for Scope/Freq Gen
« Reply #4 on: March 18, 2017, 01:01:20 pm »
Thanks for asking this and for Alm's answer. This will save me a lot of frustration.

I've bought this cable last week to connect a Feeltech FY3200S to a Rigol  DS1054Z.  It's an allready made RG58 cable with connectors on both ends.

http://www.ebay.com/itm/161769013895?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT

I built a 1Khz oscillator and wanted to check the sine wave against a known reference to see if they match.
 

Offline HoracioDos

  • Frequent Contributor
  • **
  • Posts: 335
  • Country: ar
  • Just an IT monkey with a DSO
Re: 50 ohm BNC cable recommendation, for Scope/Freq Gen
« Reply #5 on: March 18, 2017, 02:50:34 pm »
....they tend to be quite expensive ($30-$50 or so last time I looked). For many applications the cheaper tee + terminator works just as well.

Pomona BNC Feed-Thru adapter is quite expensive ($45) and the max frequency is up to 500MHz.
http://www.digikey.com/product-detail/en/pomona-electronics/4119-50/501-1036-ND/603345
I've found this chinese adapter (2pcs for $9.99) but It doesn't say the max freq.
http://www.ebay.com/itm/162328795819 
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 10486
  • Country: us
  • DavidH
Re: 50 ohm BNC cable recommendation, for Scope/Freq Gen
« Reply #6 on: March 18, 2017, 04:09:23 pm »
I like using RG-316 because of its flexibility or RG-400 for its better shielding but RG-174 and RG-58 are almost always sufficient.  Teflon insulated coaxial cable is easier to use because they dielectric does not melt when soldering.

If you want to run the function generator output to the oscilloscope *and* the circuit under test without a splitter, then an easy way to do this is to use a BNC-T at the oscilloscope's high impedance input and a termination at the circuit; so the cable goes from the generator to the oscilloscope and then to the circuit.  At lower frequencies though where most function generators operate, a BNC-T can be used as a splitter instead of inline without terminations and even 75 ohm coaxial cable does not matter.

On my 10 MHz Tektronix FG502, termination or no termination makes hardly any difference with a 3 foot cable and it is faster than most function generators.
 

Offline djnz

  • Regular Contributor
  • *
  • Posts: 169
  • Country: 00
Re: 50 ohm BNC cable recommendation, for Scope/Freq Gen
« Reply #7 on: March 18, 2017, 06:05:27 pm »
Same feed-through is cheaper on AliExpress:

https://www.aliexpress.com/store/product/P7001-2PCS-High-quality-BNC-to-BNC-Female-seat-50KY-50-ohm-matching-device-Q9-adapter/1810103_32661228102.html

I bought these a long time ago. At first, I received one good one (50 ohm) and one bad one (100 ohm instead of 50). I brought this to the seller's notice and they shipped me two more good ones. They are not bad, but the fact that one of them was 100 ohms probably means that internally, they are not using a uniform disk resistor but rather are probably using other higher value regular resistors in parallel. Probably no good for GHz use, but fine for non-critical low-hundreds of MHz. VSWR won't be stellar, but very adequate for most purposes I think.

Also, buy some 50 ohm cable(s) that are BNC on one end and alligator grabber / e-z-hook on the other. Quite useful, even with their impedance match limitations.



« Last Edit: March 18, 2017, 07:24:22 pm by djnz »
 

Offline TheSteve

  • Supporter
  • ****
  • Posts: 3076
  • Country: ca
  • GHz
Re: 50 ohm BNC cable recommendation, for Scope/Freq Gen
« Reply #8 on: March 18, 2017, 06:43:29 pm »
I bought 2 of the 50 ohm feedthrough terminators last year from banggood but they never arrived. I had planned to test them to see what frequency range they were good to.
VE7FM
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 15318
  • Country: za
Re: 50 ohm BNC cable recommendation, for Scope/Freq Gen
« Reply #9 on: March 18, 2017, 06:54:45 pm »
Get the ones with probe grabber tips on them, much more useful over the croc clips. these are much more useful on a PCB or on leads to do a quick and dirty probe, and if they are used eg across a current sense resistor in the common supply you can use the other probe lead as a single ended probe, just remember that you will have poor bandwith and HF ringing, but in many cases you can use this to find PSU ripple on a secondary side with ease, plus you can trigger off the current waveform as well with ease.

Did this earlier this month to make a quick and dirty octopus tester using a bag of random resistors ( SOT for the measuring resistor and the current limiting one) and a small isolated 12VAC variac to check some small diodes and do a little demo ( Thanks Ian for the scopes there as well) at Ham meeting.
 

Offline alm

  • Super Contributor
  • ***
  • Posts: 1258
  • Country: 00
Re: 50 ohm BNC cable recommendation, for Scope/Freq Gen
« Reply #10 on: March 18, 2017, 07:08:48 pm »
I agree with the grabber/hook type clips over alligator clips. Alligator clips only seem useful for wires or terminals too large for a retractable hook, but it is rare for me to want to connect a function generator to 10 awg wires or blade terminals. Even 0.25 W resistor legs are a challenge with alligator clips.
 
The following users thanked this post: SeanB

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 10486
  • Country: us
  • DavidH
Re: 50 ohm BNC cable recommendation, for Scope/Freq Gen
« Reply #11 on: March 18, 2017, 07:21:31 pm »
The stub and input capacitance of the oscilloscope cause more problems then non-ideal performance of the feedthrough termination and high impedance oscilloscope input only go to 500 MHz.  At 100 MHz you can get away with a BNC-T and reasonable 50 ohm dummy load and how many function generators even go that high?

It does have an effect on the sync output though since it has a much faster edge rate than a 10 MHz function generator will support but does pulse fidelity manner?

Also a feedthrough termination made with a pair of surface mount 100 ohm resistors mounted on a coplanar waveguide is not bad.

They make special feedthrough terminations for testing oscilloscope probes where a disc resistor is used right at the coaxial probe tip connection.
 

Offline djnz

  • Regular Contributor
  • *
  • Posts: 169
  • Country: 00
Re: 50 ohm BNC cable recommendation, for Scope/Freq Gen
« Reply #12 on: March 18, 2017, 07:28:19 pm »
Yup - grabber / e-z-hook are better than alligator.
 
The following users thanked this post: SeanB

Offline alm

  • Super Contributor
  • ***
  • Posts: 1258
  • Country: 00
Re: 50 ohm BNC cable recommendation, for Scope/Freq Gen
« Reply #13 on: March 18, 2017, 07:40:28 pm »
The stub and input capacitance of the oscilloscope cause more problems then non-ideal performance of the feedthrough termination and high impedance oscilloscope input only go to 500 MHz.  At 100 MHz you can get away with a BNC-T and reasonable 50 ohm dummy load and how many function generators even go that high?
Obviously for this particular application pretty much any termination will do. Even a crappy termination (axial resistor soldered to a BNC bulkhead connector) is still infinitely better than an unterminated piece of coax, even at relatively low frequencies.

Note that the non-ideal reactance of the high-impedance input is still mostly present with scopes that have an internal 50 Ohm termination. On the ones I have seen they just switch a 50 Ohm resistor in parallel with the 1 MOhm input, so the only thing you win relative to an external terminator is the slightly shorter stub. Only the (expensive) 50 Ohm-only scopes are better, and even those are not all great. There is a reason why scope manufacturers will often put a bunch of attenuators on their signal generators when testing scopes: to get rid of all those reflections from the '50 Ohm' scope front-ends.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 10486
  • Country: us
  • DavidH
Re: 50 ohm BNC cable recommendation, for Scope/Freq Gen
« Reply #14 on: March 18, 2017, 07:54:43 pm »
Note that the non-ideal reactance of the high-impedance input is still mostly present with scopes that have an internal 50 Ohm termination. On the ones I have seen they just switch a 50 Ohm resistor in parallel with the 1 MOhm input, so the only thing you win relative to an external terminator is the slightly shorter stub. Only the (expensive) 50 Ohm-only scopes are better, and even those are not all great. There is a reason why scope manufacturers will often put a bunch of attenuators on their signal generators when testing scopes: to get rid of all those reflections from the '50 Ohm' scope front-ends.

The old 350 MHz Tektronix 485 is an example where the inputs are actually 50 ohms in 50 ohm mode.  The high input impedance attenuators and buffer are switched out but I suspect it was for greater bandwidth.  In high impedance mode, it has the same 250 MHz bandwidth as the 475A.

Tektronix had problems with their 1 GHz 7A29 vertical amplifier which they tracked down to the "50 ohm" BNCs not actually being 50 ohms.  So they made their own custom 50 ohm BNCs.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 17074
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Avid Rabid Hobbyist
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf