How do I measure signals in the GHz range?

S

Saul Bernstein

Guest
Hello,

I've got a storage oscilloscope here with a bandwith of 6 GHz. Unfortunately
only the default probes with 500 MHz bandwidth are included.

Now I would like to measure signals in the range of 2.5 gigabit per second.
Thus, I am looking for suitable probes with sufficient high bandwidth. But I
think bandwidth is not all there is to consider. There is also the question
whether to use active or passive probes and probably impedance and
capacitance will also play a mayor role.

In addition, the signals I would like to measure on the board are routed
differentially (selfevident at the frequencies!), so that differential probe
would surely be the suitable choice here. But wouldn't it be possible to
measure asymmetrically to ground potential, anyway?

Guys, how do I find the right solution for my task? Unfortunately I'm an
absolute newbie in measurement technique so pardon me for bothering you. I
already hit some books but I couldn't find out, anyhow.

Thanks for support. Saul
 
"Saul Bernstein" <jiffylube@freenet.de> wrote in message
news:48e4a015$0$6607$9b4e6d93@newsspool3.arcor-online.net...
Hello,

I've got a storage oscilloscope here with a bandwith of 6 GHz.
Unfortunately only the default probes with 500 MHz bandwidth are included.

Now I would like to measure signals in the range of 2.5 gigabit per
second. Thus, I am looking for suitable probes with sufficient high
bandwidth. But I think bandwidth is not all there is to consider. There is
also the question whether to use active or passive probes and probably
impedance and capacitance will also play a mayor role.

In addition, the signals I would like to measure on the board are routed
differentially (selfevident at the frequencies!), so that differential
probe would surely be the suitable choice here. But wouldn't it be
possible to measure asymmetrically to ground potential, anyway?

Guys, how do I find the right solution for my task? Unfortunately I'm an
absolute newbie in measurement technique so pardon me for bothering you. I
already hit some books but I couldn't find out, anyhow.

Thanks for support. Saul
Click to expand...
Talk to whoever made your scope - (Tek, Agilent or LeCroy ?).
The probes are going to cost you well over Ł1000 each so you might be able
to arrange a demo and get some free instruction that way.
The big 3 manufacturers all have web sites with a reasonable amount of
information about measuring technique.

Michael Kellett
 
On Thu, 2 Oct 2008 12:19:37 +0200, "Saul Bernstein"
<jiffylube@freenet.de> wrote:

Hello,

I've got a storage oscilloscope here with a bandwith of 6 GHz. Unfortunately
only the default probes with 500 MHz bandwidth are included.

Now I would like to measure signals in the range of 2.5 gigabit per second.
Thus, I am looking for suitable probes with sufficient high bandwidth. But I
think bandwidth is not all there is to consider. There is also the question
whether to use active or passive probes and probably impedance and
capacitance will also play a mayor role.

In addition, the signals I would like to measure on the board are routed
differentially (selfevident at the frequencies!), so that differential probe
would surely be the suitable choice here. But wouldn't it be possible to
measure asymmetrically to ground potential, anyway?

Guys, how do I find the right solution for my task? Unfortunately I'm an
absolute newbie in measurement technique so pardon me for bothering you. I
already hit some books but I couldn't find out, anyhow.

Thanks for support. Saul

Though I never tried GHz range stuff, I did use the resistive probe as
Click to expand...
described:

http://www.signalintegrity.com/Pubs/straight/probes.htm

And cheap too :)


--
- Blarp the Enigmatic
 
On Thu, 2 Oct 2008 12:19:37 +0200, "Saul Bernstein"
<jiffylube@freenet.de> wrote:

Hello,

I've got a storage oscilloscope here with a bandwith of 6 GHz. Unfortunately
only the default probes with 500 MHz bandwidth are included.

Now I would like to measure signals in the range of 2.5 gigabit per second.
Thus, I am looking for suitable probes with sufficient high bandwidth. But I
think bandwidth is not all there is to consider. There is also the question
whether to use active or passive probes and probably impedance and
capacitance will also play a mayor role.

In addition, the signals I would like to measure on the board are routed
differentially (selfevident at the frequencies!), so that differential probe
would surely be the suitable choice here. But wouldn't it be possible to
measure asymmetrically to ground potential, anyway?

Guys, how do I find the right solution for my task? Unfortunately I'm an
absolute newbie in measurement technique so pardon me for bothering you. I
already hit some books but I couldn't find out, anyhow.

Thanks for support. Saul
Click to expand...
What kind of scope is it?

John
 
On 10/2/08 11:18 AM, in article
ca183c07-1d51-459c-b324-a8ce537cadc2@25g2000hsx.googlegroups.com,
"a7yvm109gf5d1@netzero.com" <a7yvm109gf5d1@netzero.com> wrote:

On Oct 2, 6:19 am, "Saul Bernstein" <jiffyl...@freenet.de> wrote:

In addition, the signals I would like to measure on the board are routed
differentially (selfevident at the frequencies!), so that differential probe

What's so "self-evident" about it? Your sat TV receiver works at 12GHz
and is 100% unbalanced.

You don't say what the measurements are supposed to be for. When
working with data at these rates, the correct tool these days is a
BERT, or J-BERT, not a scope.
Click to expand...
The initial test on multi-GB systems I know of, is done by looking at the
recovered signal with a scope. BERT is only good on a path that was once
known "good" but is now showing system errors. IMO a BERT test set is
something you use to prove there really is (or isn't) something wrong, just
before you grab your scope.



Unless your "scope" is a BERT with a missing personality module...
Click to expand...
 
On Oct 2, 6:19 am, "Saul Bernstein" <jiffyl...@freenet.de> wrote:
In addition, the signals I would like to measure on the board are routed
differentially (selfevident at the frequencies!), so that differential probe
Click to expand...
What's so "self-evident" about it? Your sat TV receiver works at 12GHz
and is 100% unbalanced.

You don't say what the measurements are supposed to be for. When
working with data at these rates, the correct tool these days is a
BERT, or J-BERT, not a scope.

Unless your "scope" is a BERT with a missing personality module...
 
On Thu, 2 Oct 2008 11:18:22 -0700 (PDT), a7yvm109gf5d1@netzero.com
wrote:

On Oct 2, 6:19 am, "Saul Bernstein" <jiffyl...@freenet.de> wrote:

In addition, the signals I would like to measure on the board are routed
differentially (selfevident at the frequencies!), so that differential probe

What's so "self-evident" about it? Your sat TV receiver works at 12GHz
and is 100% unbalanced.

You don't say what the measurements are supposed to be for. When
working with data at these rates, the correct tool these days is a
BERT, or J-BERT, not a scope.

Unless your "scope" is a BERT with a missing personality module...
Click to expand...
Scopes do the basic electrical signal integrity measurements, levels
and timing and jitter and eye diagrams. Packets and protocols are
higher levels from there.

John
 
On Oct 2, 6:19 am, "Saul Bernstein" <jiffyl...@freenet.de> wrote:
Hello,

I've got a storage oscilloscope here with a bandwith of 6 GHz. Unfortunately
only the default probes with 500 MHz bandwidth are included.
Click to expand...
Heterodyne the 2.5GHz signals down to 200MHz or so ;)
 
[F'up2 was missing --- fixed]

a7yvm109gf5d1@netzero.com wrote:
On Oct 2, 6:19 am, "Saul Bernstein" <jiffyl...@freenet.de> wrote:
In addition, the signals I would like to measure on the board are routed
differentially (selfevident at the frequencies!), so that differential probe

What's so "self-evident" about it? Your sat TV receiver works at 12GHz
and is 100% unbalanced.
Click to expand...
Pardon the French, but that's nonsense. Nothing in a sat receiver works
at 12 GHz. That's self-evident if you take just a quick look at the
antenna cable --- no way a 12 GHz signal can survive 10+ meters of that
coax cable and connectors.

The only place 12 GHz take place in a sat TV setup is in the LNB (for
those who don't know: that's the gadget at the end of that arm, near the
center of the dish). From there onwards the signal is in the 1..2 Ghz
range, at most.
 
On Fri, 3 Oct 2008 09:31:49 -0700 (PDT), Alex
<christiansen_alex@hotmail.com> wrote:

On Oct 2, 12:19 pm, "Saul Bernstein" <jiffyl...@freenet.de> wrote:
Hello,

I've got a storage oscilloscope here with a bandwith of 6 GHz. Unfortunately
only the default probes with 500 MHz bandwidth are included.

Now I would like to measure signals in the range of 2.5 gigabit per second.
Thus, I am looking for suitable probes with sufficient high bandwidth. But I
think bandwidth is not all there is to consider. There is also the question
whether to use active or passive probes and probably impedance and
capacitance will also play a mayor role.

In addition, the signals I would like to measure on the board are routed
differentially (selfevident at the frequencies!), so that differential probe
would surely be the suitable choice here. But wouldn't it be possible to
measure asymmetrically to ground potential, anyway?

Guys, how do I find the right solution for my task? Unfortunately I'm an
absolute newbie in measurement technique so pardon me for bothering you. I
already hit some books but I couldn't find out, anyhow.

Thanks for support. Saul

Allthoug I dont work with HF ... I think there will be a problem with
measuring a 2.5 G bit signal with a 6 Ghz scope
unless the signal is a sine. ..if its a square signal ,the third
harmonic have to be shown too ,to get a usable picture ,and
that is allready 7.5 Ghz. So if you could get a picture of the signal,
you would not know if it was true.
alex
Click to expand...
6G's should be fine, if it can be probed properly. 2.5 GBPS is
equivalent to a 1.25 GHz square wave, 400 ps per data element, and the
rise time of a 6GHz scope will be around 60 ps.

John
 
On 10/3/08 9:31 AM, in article
542c1645-74fa-44a3-8b0a-875051d14694@e17g2000hsg.googlegroups.com, "Alex"
<christiansen_alex@hotmail.com> wrote:

On Oct 2, 12:19 pm, "Saul Bernstein" <jiffyl...@freenet.de> wrote:
Hello,

I've got a storage oscilloscope here with a bandwith of 6 GHz. Unfortunately
only the default probes with 500 MHz bandwidth are included.

Now I would like to measure signals in the range of 2.5 gigabit per second.
Thus, I am looking for suitable probes with sufficient high bandwidth. But I
think bandwidth is not all there is to consider. There is also the question
whether to use active or passive probes and probably impedance and
capacitance will also play a mayor role.

In addition, the signals I would like to measure on the board are routed
differentially (selfevident at the frequencies!), so that differential probe
would surely be the suitable choice here. But wouldn't it be possible to
measure asymmetrically to ground potential, anyway?

Guys, how do I find the right solution for my task? Unfortunately I'm an
absolute newbie in measurement technique so pardon me for bothering you. I
already hit some books but I couldn't find out, anyhow.

Thanks for support. Saul

Allthoug I dont work with HF ... I think there will be a problem with
measuring a 2.5 G bit signal with a 6 Ghz scope
unless the signal is a sine. ..if its a square signal ,the third
harmonic have to be shown too ,to get a usable picture ,and
that is allready 7.5 Ghz. So if you could get a picture of the signal,
you would not know if it was true.
alex
Click to expand...
That all depends on what you are looking for. Time relationships, the
relation of a pulse position relative to others, is valid.
 
On Fri, 03 Oct 2008 12:42:37 +0100, Hans-Bernhard Bröker <HBBroeker@t-online.de> wrote:

F'up abuse detected.

[plonk]

a7yvm109gf5d1@netzero.com wrote:
On Oct 2, 6:19 am, "Saul Bernstein" <jiffyl...@freenet.de> wrote:
In addition, the signals I would like to measure on the board are routed
differentially (selfevident at the frequencies!), so that differential probe

What's so "self-evident" about it? Your sat TV receiver works at 12GHz
and is 100% unbalanced.

Pardon the French, but that's nonsense. Nothing in a sat receiver works
at 12 GHz. That's self-evident if you take just a quick look at the
antenna cable --- no way a 12 GHz signal can survive 10+ meters of that
coax cable and connectors.

The only place 12 GHz take place in a sat TV setup is in the LNB (for
those who don't know: that's the gadget at the end of that arm, near the
center of the dish). From there onwards the signal is in the 1..2 Ghz
range, at most.
Click to expand...


--
http://www.petersparrots.com http://www.insanevideoclips.com http://www.petersphotos.com

What lives in the sea, and goes dah di dah dit, dah dah dit dah?
A morse cod.
 
On 10/3/08 10:27 AM, in article op.uiglf5dx4buhsv@fx62.mshome.net, "Peter
Hucker" <none@spam.com> wrote:

On Fri, 03 Oct 2008 12:42:37 +0100, Hans-Bernhard Bröker
HBBroeker@t-online.de> wrote:

[F'up2 was missing --- fixed]

F'up abuse detected.

[plonk]

a7yvm109gf5d1@netzero.com wrote:
On Oct 2, 6:19 am, "Saul Bernstein" <jiffyl...@freenet.de> wrote:
In addition, the signals I would like to measure on the board are routed
differentially (selfevident at the frequencies!), so that differential
probe

What's so "self-evident" about it? Your sat TV receiver works at 12GHz
and is 100% unbalanced.

Pardon the French, but that's nonsense. Nothing in a sat receiver works
at 12 GHz. That's self-evident if you take just a quick look at the
antenna cable --- no way a 12 GHz signal can survive 10+ meters of that
coax cable and connectors.

The only place 12 GHz take place in a sat TV setup is in the LNB (for
those who don't know: that's the gadget at the end of that arm, near the
center of the dish). From there onwards the signal is in the 1..2 Ghz
range, at most.
Click to expand...
Heard 599 here.
 
On Oct 2, 12:19 pm, "Saul Bernstein" <jiffyl...@freenet.de> wrote:
Hello,

I've got a storage oscilloscope here with a bandwith of 6 GHz. Unfortunately
only the default probes with 500 MHz bandwidth are included.

Now I would like to measure signals in the range of 2.5 gigabit per second.
Thus, I am looking for suitable probes with sufficient high bandwidth. But I
think bandwidth is not all there is to consider. There is also the question
whether to use active or passive probes and probably impedance and
capacitance will also play a mayor role.

In addition, the signals I would like to measure on the board are routed
differentially (selfevident at the frequencies!), so that differential probe
would surely be the suitable choice here. But wouldn't it be possible to
measure asymmetrically to ground potential, anyway?

Guys, how do I find the right solution for my task? Unfortunately I'm an
absolute newbie in measurement technique so pardon me for bothering you. I
already hit some books but I couldn't find out, anyhow.

Thanks for support. Saul
Click to expand...
Allthoug I dont work with HF ... I think there will be a problem with
measuring a 2.5 G bit signal with a 6 Ghz scope
unless the signal is a sine. ..if its a square signal ,the third
harmonic have to be shown too ,to get a usable picture ,and
that is allready 7.5 Ghz. So if you could get a picture of the signal,
you would not know if it was true.
alex
 
On Oct 3, 7:16 pm, Don Bowey <dbo...@comcast.net> wrote:
On 10/3/08 9:31 AM, in article
542c1645-74fa-44a3-8b0a-875051d14...@e17g2000hsg.googlegroups.com, "Alex"





christiansen_a...@hotmail.com> wrote:
On Oct 2, 12:19 pm, "Saul Bernstein" <jiffyl...@freenet.de> wrote:
Hello,

I've got a storage oscilloscope here with a bandwith of 6 GHz. Unfortunately
only the default probes with 500 MHz bandwidth are included.

Now I would like to measure signals in the range of 2.5 gigabit per second.
Thus, I am looking for suitable probes with sufficient high bandwidth. But I
think bandwidth is not all there is to consider. There is also the question
whether to use active or passive probes and probably impedance and
capacitance will also play a mayor role.

In addition, the signals I would like to measure on the board are routed
differentially (selfevident at the frequencies!), so that differential probe
would surely be the suitable choice here. But wouldn't it be possible to
measure asymmetrically to ground potential, anyway?

Guys, how do I find the right solution for my task? Unfortunately I'm an
absolute newbie in measurement technique so pardon me for bothering you. I
already hit some books but I couldn't find out, anyhow.

Thanks for support. Saul

Allthoug I dont work with HF ... I think there will be a problem with
measuring a 2.5 G bit  signal with a 6 Ghz scope
unless the signal is a sine. ..if its a square signal ,the third
harmonic have to be shown too ,to get a usable picture ,and
that is allready 7.5 Ghz. So if you could get a picture of the signal,
you would not know if it was true.
alex

That all depends on what you are looking for.  Time relationships, the
relation of a pulse position relative to others, is valid.- Hide quoted text -

- Show quoted text -
Click to expand...
oh yes ..your right John
it is , only 1.25 G hz, and
Also ,I agree , Don.
Alex
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top