oscilloscope probes VS BNC cable
john
Is BNC cable is appropiate cable to make oscilloscope readings or
oscilloscope probes? My co-worker is insisting that BNC Cable is more
shielded than the probes? I suspect that its not all about shielding
only. There is a reason that probes exsists.
We are trying to measure from an amplifer ( multiclamp700B). Please
advice
Thanks
John
Phil Allison
"john"
>
> Is BNC cable is appropiate cable to make oscilloscope readings or
> oscilloscope probes?
** Yes - it is done all the time where a piece of equipment provides the
signal for the scope.
> My co-worker is insisting that BNC Cable is more
> shielded than the probes?
** Might be - bit it hardly ever matters.
> I suspect that its not all about shielding
> only. There is a reason that probes exsists.
** Yes - they exist to allow you to " probe " inside equipment & and
pick up signals from wherever you like.
Most scope probes are made with thin, low capacitance per foot cable ( using
a very thin inner conductor) and are fitted with an adjustable head
incorporating a 1:1 or 10:1 switch. The 10:1 position giving a 10 megohm
load in parallel with a few pF of capacitance which is more suitable for
probing sensitive circuit locations.
Also scope probes are *designed* to work with standard 1 megohm in parallel
with 20 pF input circuits of typical scopes.
> We are trying to measure from an amplifer ( multiclamp700B).
** If it has a BNC output - then use a BNC to BNC lead.
...... Phil
rebel
First up, "BNC " is a family of connectors, not a cable type.
But assuming you mean coaxial cable such as 50ohm transmission line, the use of
a proper CRO probe system provides a lower capacitance cable. Part of the
objective is to not load the device under test.
The shielding on the proper probe cable is indeed typically not as complete as
even something as mundane as RG-58C/U, but that really IS secondary.
D from BC
wrote:
>Hi,
Scope probes have a capacitance adjustment.
This is to compensate for cable capacitance and scope input
capacitance.
It can make square waves look more square.
But I guess the same circuit can be added to some qualified coax + a
BNC connector. (Example: Maybe cable tv coax could be ok?)
D from BC
Rich Grise
It's not necessarily "better shielded". Most scope probes have a
very tiny center conductor in their cable, yielding a high-impedance,
low-capacitance cable. As D said, the cable comprises part of a
capacitive divider.
I have a salvaged piece of RG-58/U with a BNC on one end and bare
wire at the other. I've soldered a 1K resistor in series with the
center conductor, and it's "good enough", at DC and audio frequencies.
But I was thinking of making a capacitive divider with it - I've got
a 3' piece of coax, which is listed at, I think, 27 pF per foot. So,
with 3 ft, that'd be 81 pF. Which, if I remember my old formulas, I
would need to stick an 8.1 pF cap in series with my center conductor
to yield a 10:1 probe.
I haven't checked my numbers, or tried anything like this, primarily
because I don't have an 8.1 pF cap lying around. And, of course, a
trimmer would be better, to trim out the tolerances and square up those
edges! :-)
But, the coax should be good enough for the audio range, but I
recommend an isolation resistor between it and the DUT.
Hope This Helps!
Rich
Phil Allison
"D from BC"
> Scope probes have a capacitance adjustment.
** ONLY divider probes have that.
> This is to compensate for cable capacitance and scope input
> capacitance.
** ONLY against the effect it has on the divided down signal.
> It can make square waves look more square.
** ONLY divided down ones.
> But I guess the same circuit can be added to some qualified coax + a
> BNC connector.
** NO.
Cos it ain't a divider.
....... Phil
Martin Riddle
> It can make square waves look more square.
>
You didn't really mean that, did you?
Cheers
D from BC
Oops... That's right.. No comp on 1x probes (direct to scope).
I think I just assumed everybody uses 1x10x divider probes..
Divider probes and other reactances (without googling) is:
------probe---+---\/\R1\/\---+
| |
|------| |---- +------+---------+
Ccomp | | |
\ = Ccable = CinScope
/ | |
R2 | |
\ | |
/ | |
Gnd Gnd Gnd
What I meant was R1,R2 and Ccomp could be put on a qualified coax to
make a divider probe.
D from BC
D from BC
Noooo... I don't mean it's going to make 45% duty look like 50% duty.
:P
1x probe cable capacitance and scope input capacitance can have an RC
effect on some circuits with fast edges. The scope can present a
slower than actual edge times.
With a 10x probe the circuit sees less applied capacitance. The
circuit sees a capacitor divider composed of Ccomp in series with
Ccable + Cscope.
Therefore a closer to actual edge time can be viewed.
Hope I got that right :|...
D from BC
David L. Jones
> Hi,
>
> Is BNC cable is appropiate cable to make oscilloscope readings or
> oscilloscope probes?
It can be if the device you are "probing" provides a low impedance
signal output designed for driving a coax cable. For example, if you
are measuring the output from a signal generator it's common to use a
standard BNC cable rather than a probe because it's a more convenient
and provides a better quality connection. Also, a standard BNC cable
can give you more bandwidth for your $$, as quality high speed probes
are expensive.
> My co-worker is insisting that BNC Cable is more
> shielded than the probes? I suspect that its not all about shielding
> only. There is a reason that probes exsists.
Yes, probes exist to allow you to more conveniently "probe" point of
interest in a circuit with minimum loading (they have lower
capacitance). Also, a 10:1 probe allows you to measure much higher
bandwidth (faster) signals than just a standard coax cable can.
> We are trying to measure from an amplifer ( multiclamp700B). Please
> advice
In that case you would use a BNC cable, with a BNC T piece if needed.
Dave.
Rich Grise
> On Sat, 15 Sep 2007 09:06:41 +1000, "Phil Allison"
>>> Scope probes have a capacitance adjustment.
>>** ONLY divider probes have that.
> Oops... That's right.. No comp on 1x probes (direct to scope).
> I think I just assumed everybody uses 1x10x divider probes..
> Divider probes and other reactances (without googling) is:
>
> ------probe---+---\/\R1\/\---+
> | |
> |------| |---- +------+---------+
> Ccomp | | |
> \ = Ccable = CinScope
> / | |
> R2 | |
> \ | |
> / | |
> Gnd Gnd Gnd
>
> What I meant was R1,R2 and Ccomp could be put on a qualified coax to
> make a divider probe.
What if my scope has 1M, 25 pF at its input, such that R2, 1M, is to the
right of Ccable; Ccable is approx, 27 pF per foot, and it's just under
3' long - would R1 be 10M and Ccomp about 9.1 to 10.0 pF?
Is there a formula for gimmicks somewhere? :-) Do I have to buy a 10M
resistor to make a proper probe?
Thanks,
Rich
John Popelish
The resistor divider has to drop the voltage to 1/10th, so
the resistors have to have a ratio of 9 to 1, since the
voltage is proportional to R2/(R1+R2). So a 9.1 Meg
resistor is pretty good. Likewise, the cap that parallels
the 9.1 meg resistor has to be about 1/9th of the total
input capacitance (cable capacitance plus scope input
capacitance. Now you see why they go to so much trouble to
build special low capacitance cable for scope probes. I
wonder if you can pull out the center conductor and pull in
a very slinky high E guitar string (tinned with silver
solder at each end for electrical connections).
D from BC
wrote:
E 0.20mm
http://www.mohomods.com/strings.htm
RG59 cable core wire AWG 22 = 0.64516mm
http://rocky.digikey.com/WebLib/Alpha%20Wire/Web%20Info/9055,58,59%209803,30,48,58%20p279.pdf
Diameter reduction figure: 0.65/0.2 = 3.25
100Mhz Scope probe core wire diameter ???
Not here http://en.wikipedia.org/wiki/Test_probe
Not here
http://www.tek.com/products/accessories/oscilloscope_probes/passive.html
Nope... but it's a page on making a DIY scope probe...
http://www.cromwell-intl.com/radio/probes.html
Fk it... I'll guess 33 AWG which ~ the E string
Anyways...
Getting an E string preferrably centered in 3ft of RG59 (or other)
might be tough.
Pulling out the coax core wire might be difficult.
Some oven heating might help. :P
RG6 replaces RG59.
http://en.wikipedia.org/wiki/RG-6
It has a fatter 18 AWG wire.
The guitar string idea gets better stuck in RG6.
RG6 5.00 dB attenuation per 100' at 750 MHz
http://www.atechfabrication.com/products/coaxial_cable.htm
But I think that's with video source and load impedances.
D from BC
John Popelish
I would look for low capacitance (high impedance) teflon
insulated coax. Besides, it is slippery when you try to
push the E string through it. E.g.:
http://glacier.lbl.gov/~gtp/DOM/delay/Belden_RG-180.html
D from BC
wrote:
>D from BC wrote:
I'm getting a bad vibe about that RG-180...I think it's going to be
pricey, min quality'd and difficult to find a source..
RG6 can probably be found a Radio Shack, London Drugs, Home Depot or a
surplus electronics store.
Or perhaps stop the local cable guy and give'm 5 bucks...
D from BC
John Popelish
> I'm getting a bad vibe about that RG-180...I think it's going to be
> pricey, min quality'd and difficult to find a source..
> RG6 can probably be found a Radio Shack, London Drugs, Home Depot or a
> surplus electronics store.
> Or perhaps stop the local cable guy and give'm 5 bucks...
Does D from BC mean you are in British Columbia? I have
some RG-180 I got on eBay. It is sort of stiff, but not
much worse than most coax that is not made to be an o-scope
probe. If governments don't get in the way too badly,
perhaps I could send you 10 feet, or so, in an envelope, to
play with.
D from BC
wrote:
>D from BC wrote:
Yup..I'm in British Columbia.
Land of really bad traffic, great skiing, marijuana and Chinese food.
:P
Surrey BC might still be car theft capital of North America.
Weather is like 50% continuous sun and 50% continuous rain.
And a V6 is must due to some roller coaster type roads.
Also, the women are in great shape because they don't make enough
money to eat alot :P
About the wire..
Thanks for the offer..but I have cheapo 100Mhz probes I got off ebay.
I have an interest in DIY probes but stay away from making one :)
D from BC
John Devereux
There are some situations where a piece of cable can give much better
results than a probe, where the circuit impedance is low (like your
amplifier output perhaps).
A common situation for me is measuring mV ripple on the output of a
switch mode power suppy. A normal x10 probe would attenuate the signal
unnecessarily, and the ground lead would pick up a *lot* of magnetic
noise from the switching inductor. Simply connecting a "BNC" (RG58)
cable gives much cleaner results.
Also see e.g. Jim Williams' classic AN47.
<http://www.google.com/search?q=ltc+an47>
--
John Devereux
Rich Grise
> Rich Grise wrote:
>>
>> What if my scope has 1M, 25 pF at its input, such that R2, 1M, is to the
>> right of Ccable; Ccable is approx, 27 pF per foot, and it's just under
>> 3' long - would R1 be 10M and Ccomp about 9.1 to 10.0 pF?
>>
>> Is there a formula for gimmicks somewhere? :-) Do I have to buy a 10M
>> resistor to make a proper probe?
> capacitance. Now you see why they go to so much trouble to
> build special low capacitance cable for scope probes. I
> wonder if you can pull out the center conductor and pull in
> a very slinky high E guitar string (tinned with silver
> solder at each end for electrical connections).
Well, I _could_, but it's the only BNC I have and it came already
attached to the coax. I guess I'll just save up my pennies and buy
a proper probe when I can afford it. :-)
Thanks!
Rich
Helmut Sennewald
news:1189779039.1...@g4g2000hsf.googlegroups.com...
> Hi,
>
> Is BNC cable is appropiate cable to make oscilloscope readings or
> oscilloscope probes?
Hello John,
It depends on your requirement. It would be ok for signals up to 100kHz.
The disadvantage would be the higher capacitance of the normal coaxial
cable. Probe cables have only half the capacitance of 50Ohm coaxial
cables.
> My co-worker is insisting that BNC Cable is more
> shielded than the probes?
I don't agree.
> I suspect that its not all about shielding
> only. There is a reason that probes exsists.
The most important thing is the distributed series resistance of the
inner conductor. A value of 100 to 200 Ohm per meter gives the
best pulse performance. If you would make a 10:1 probe using
a normal coaxial cable, you would have nearly 100% overshoot
for rise times below ten nanoseconds.
Best regards,
Helmut
whit3rd
>
> Is BNC cable is appropiate cable to make oscilloscope readings or
> oscilloscope probes?
BNC connectors are commonly affixed to 50 ohm, 75 ohm, sometimes 93
ohm
transmission-line coaxial cables. If you have a suitable source, and
use
the appropriate terminator at the oscilloscope, those coaxial cables
are
very good input wiring.
An audio amplifier with 6 ohm output impedance will best be measured
with
50 ohm cable fitted with 43 ohms series resistance to the amplifier
output.
With a 50 ohm terminator at the 'scope, expect 2:1 attenuation, of
course.
If, on the other hand, your source has high impedance, it will
possibly
fail to drive the termination resistor, or will be loaded by the cable
capacitance.
The termination resistor will attenuate the signal, the cable
capacitance will
cause additional high-frequency attenuation. If your source has low
impedance
(lower than the terminator) the cable inductance can cause some
overshoot (high
frequency boost).
And if you omit the terminator resistance, and the source is not
series-terminated,
one would expect high frequency distortions from standing-waves in
that
transmission line. For short cables (10 feet) and low frequencies
(under 10 MHz)
it isn't a big problem. Not usually.
Phil Allison
"whit3rd"
>
> BNC connectors are commonly affixed to 50 ohm, 75 ohm, sometimes 93
> ohm transmission-line coaxial cables. If you have a suitable source, and
> use the appropriate terminator at the oscilloscope, those coaxial cables
> are very good input wiring.
** Works just fine even without those restrictions.
> An audio amplifier with 6 ohm output impedance will best be measured
> with 50 ohm cable fitted with 43 ohms series resistance to the amplifier
> output.
** Complete bollocks.
> If, on the other hand, your source has high impedance, it will
> possibly fail to drive the termination resistor,
** So leave it out.
> or will be loaded by the cable capacitance.
** All (passive) probes have cable capacitance.
> If your source has low impedance
> (lower than the terminator) the cable inductance can cause
> some overshoot (high frequency boost).
** Total bollocks.
A correctly terminated BNC lead is a resistive load.
....... Phil
freda
frequency models that also include 50-ohm inputs.
Use a 50-Ohm Coax to drive a 50-Ohm terminated scope input direcly.
To make a 10X probe for 50-Ohm system, use a 450 Ohm resistor at the
circuit end of a 50-ohm coax, again into the 50-ohm input of a
scope....you'll need to factor this attenuator into the scopes readout
as it won't know about the 10X unless you can force it somehow. This
forms a high quality probe for cheap, but only for 50-ohm systems.
Keep the braid over the resistor as best a possible to minimize
impeadance discontinuities.
Watch how much power you dump into the attenuator. Most won't take
more then 5W. Some scopes don't have any over temp protection on the
input so you can fry them pretty quickly if you are not careful.
If you aren't in a 50-ohm system OR if your scope only has 1 Mohm
inputs, use a probe designed to match it's input capacitance.
You NEED to compensate the probe to match the input - this is a LF
comp and is usually done by using the calibrator and tweaking until
the square-wave input has flat tops and bottoms.