Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Coaxial scope leads without the point and ground clip for noise-free measurements?

Status
Not open for further replies.
G

grizedale

Advanced Member level 3
Joined
Jun 13, 2011
Messages
838
Helped
17
Reputation
34
Reaction score
17
Trophy points
1,298
Activity points
8,804
Hello,

I recently had to probe a microcontroller pin which was on the same pcb as an smps.......the voltage there was full of noise, and my boss wasn't impressed, so i think i need a scope probe which is just a coaxial cable with a BNC connector at one end, then atthe other end, i will simply solder ground shield and centre conductor to the voltage and ground on the PCB.

But do you know where i can buy:-

~1 metre length of coaxial cable.
a BNC connector.
a tool for stripping the coaxial cable so that i can attach it to the BNC connector.
the tool needed to attach the BNC connector to the coaxial cable.

I cannot find these things anywhere on RS Farnell or digikey
 

 
Last edited:
Probe tip accessories like the one shown below are very effective at reducing induced signals from SMPS:
 
Thankyou very much...mtwieg ill bear yours in mind though it looks like its only for a specific point on a board.

Mister-rf i will buy that kit, it looks great.

also since its so devilishly fiddlesome to solder the end of the copaxial cable (sheath and centre conductor) to the pcb, is there anything we can buy to put on the end of the coax so as to make it easier to solder to the PCB?
 

grizedale, you can also achieve the same thing as shown in mtwieg's post using standard o'scope probes without the probe tip accessory by:
1.) Pull the clippy tip off of your oscilloscope probe (exposing the pointed tip and the ground connector)
2.) Angle tweezers or a small screwdriver to contact the probe ground to the board ground.
It's a little finicky, so you can also replace step (2) with:
2.) Grab some 22AWG solid wire (commonly used for breadboard prototyping), strip it, wind it around the probe's ground contact, and solder the other end to the PCB.
In this case, it's colloquially known as a "pigtail."
 
grizedale said:
Thankyou very much...mtwieg ill bear yours in mind though it looks like its only for a specific point on a board.
Click to expand...
Well they're generally pretty flexible, though mine is pretty bent up after stretching it too many times. But still much easier to use than stripped coax.

ZekeR said:
2.) Angle tweezers or a small screwdriver to contact the probe ground to the board ground.
Click to expand...
I've tried this and it always seems to create more problems than it fixes, especially if you're directly touching the conductive part of the tweezers.
2.) Grab some 22AWG solid wire (commonly used for breadboard prototyping), strip it, wind it around the probe's ground contact, and solder the other end to the PCB.
In this case, it's colloquially known as a "pigtail."
Click to expand...
I knew it had a name!
 
I always use pigtails, because I end up losing those neat little springy contacts. Maybe I should clean my bench... =)

- - - Updated - - -

I always use pigtails, because I end up losing those neat little springy contacts. Maybe I should clean my bench... =)
 
I've seen differential probe tips that have two prongs (signal
and return) right next to each other for minimum inductance.
Forget where; probably on some spam from Agilent or Tek.
 
zekeR, Thanks, ive tried that many times, but i find that because the ground cylinder at the end of the probe is so polished and smooth, the wire simply does not grip onto it well enough, and it keeps falling off.
 

grizedale, you must wrap the wire around the cylinder multiple times (my standard is 3 times). Pigtails will hold the probe in place quite securely. You can bend the pigtail so that the probe stands up vertically and the tip makes contact through gravity, or you can create another miniature pigtail for the probe's tip and then rest the probe horizontally (so it's suspended between two pigtails, one for the ground and one for the tip). Which one you choose generally depends on how much you intend to be probing that point.
 
If you don't manage to connect a ground spring or similar low inductance ground option, you should think about probe sockets soldered to the circuit. They are available from major instrumentation manufacturers, like this one **broken link removed**
 
Thankyou , i appreciate all points now.
-FvM, its great for some of our measurements but expensive to put in our circuits.
ZekeR i take your point.
mister-RF, i now doubt your solution since the coax leads that you kindly suggest have very very thin, fine outer conductor shrouds, and soldering them would be difficult, -i really need a very flexible coax with a thickish outer braid which i can pull back and twist to a point for soldering to the PCB.

-Another poin is that this will be used in the environmental chamber so a probe with pigtail is not that useful, i really need a decent 1200cm piece of coax with thick outer conductor.

When i worked at a telecoms company we used to have reels of it (Huber & Suhner) but i cannot find it in small lengths.

-though i must admit i dread the cost of the tool needed to connect the BNC to the end.
 

The main problem seems to be that you want a short piece of coax, but most online stores only sell reels. Perhaps you would have better luck with walk-in stores rather than online stores.

Have you asked at your local hardware store yet?

The sort of places that sell stereos, car sound equipment or TVs also tend to sell assorted cables. Perhaps you could find something usable there.

Co-ax is also used for TV and FM antennas. Perhaps you could buy a cheap "bunny ears" TV antenna, cut off the antenna, and just use the cable.

Similarly, you could go to an auto-spares place and see if they have any cheap FM whip antennas or extension cables for antennas.
 
If you really care about the fidelity of the signal, stick to 50 Ohm cable and set the o'scope's input termination to 50 ohms. This will prevent transmission line funnies. To avoid loading down your circuit with a 50 ohm termination, you can use a 950 ohm resistor (non-inductive!) in series with the signal cable (and set the o'scope to register a 20:1 division ratio).

If nothing else, you can just grab a BNC cable sitting around in your lab and cut the end off.
 
ZekeR said:
If you really care about the fidelity of the signal, stick to 50 Ohm cable and set the o'scope's input termination to 50 ohms. This will prevent transmission line funnies. To avoid loading down your circuit with a 50 ohm termination, you can use a 950 ohm resistor (non-inductive!) in series with the signal cable (and set the o'scope to register a 20:1 division ratio).

If nothing else, you can just grab a BNC cable sitting around in your lab and cut the end off.
Click to expand...


Yes, place a series R in the line otherwise the poor output pin and the poor 50 Ohm termination in the scope both may suffer.


Perhaps it is possible to use a probe rated to 2 or 3 GHz and presenting only 500 Ohm load to the circuit of interest?

. https://www.howardelectronics.com/auburn/rfprobe.html

Note: This probe is DC blocked, so one may not observe what one thinks to be the proper signal, but the rise and fall times will be accurate.


RF_Jim

- - - Updated - - -

mtwieg said:
Probe tip accessories like the one shown below are very effective at reducing induced signals from SMPS:
Click to expand...

I have to concur with mtwieg; the ground clip next to the scope probe is is VERY effective at keeping noise down when in the proximity of strong/noisy SMPSs ... had I not seen it myself I might not have believed how effective those probe tip grounds are.


RF_Jim
 
The main problem seems to be that you want a short piece of coax, but most online stores only sell reels. Perhaps you would have better luck with walk-in stores rather than online stores.
Click to expand...
I have no problem to get most coaxial cable types (e.g. RG 178) in single meter quantities from some professional and several DIY online stores.

I also agree to the comments about the advantage of passive resistive (500 or 1000 ohm) probes for real wideband measurements. Many circuits don't tolerate a 50 ohm respectively > 100 pF capacitive load. We would need to look into the application details to decide about "best probe". Active probes are perfectly suited for most measurement applications by the way, but rather expensive, unfortunately.
 
Need to consider also this:
grizedale said:
Thankyou , i appreciate all points now.
-i really need a very flexible coax with a thickish outer braid which i can pull back and twist to a point for soldering to the PCB.-Another poin is that this will be used in the environmental chamber so a probe with pigtail is not that useful, i really need a decent 120.0cm piece of coax with thick outer conductor.
.
Click to expand...
In my experience most of the time you can’t use for that purpose a normal antenna cable… at least the antenna cables they sold in shops. Now you can see how 200cm length RG178B/U cable fits easily in one standard envelope. :-D
 

Attachments

  • 150620121433.jpg
    150620121433.jpg
    297.6 KB · Views: 118
Last edited:
Thanks,

FvM and godfryl: I will ask at hardware store to see about short lengths of coax, hopefully with a BNC connector at one end.

If using coax as described than i would simply use the normal 1MOhm termination in the scope......I am not at all interested in wideband RF bandwidth measurements, i am simply concerned with reducing noise picked up into the measurement cable.

So anyway, i am not interested in RF matching, our products are simple LED lights with smps drivers.
 

It's fine to use a 1MOhm termination and connect the 50 ohm cable directly to the signal, as long as you're aware that the signal will also be loaded with ~100pF, which means rise/fall times will slow down. A pulse will see 50 ohms, gradually growing to 1MOhm as it propagates through the cable several times. Depending on the impedance of the node, it can slow down a lot, or it can cause the circuit not to work properly at all. I wouldn't trust any rise times faster than 1us.
 
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top