Bandwidth check of my oscilloscope with simple means ?

[tggzzz]

The best way to measure just the scope itself is to get one of leo's pulse gens you can directly attach and once you have a reference you can then attach your probes and see what they are rated for.

Direct attachment also avoids the VSWR issues associated with the mismatch to 50ohms//15pF. There's a reason probe cable is very lossy

Such step generator tools are (arguably) more valuable in that they can show you the consequences of poor probing technique.

You can also check things like bnc cables... half the time people dont know just how horrible they are.  On the other hand most of this is moot if you are dealing with 100MHz and lower

Coax (and other) transmission lines have a well-defined attenuation-vs-frequency characteristic, and the attenuation rises quites steeply with frequency. Pity the engineers that had to span the Atlantic ocean with coax cable; the repeaters had a lot of high frequency pre-emphasis. (Plus the TAT7 cable's impedance was unavoidably 61.8ohms, IIRC)

[vk6zgo]

This is second topic with exactly same question from OP.

This one also devolved into discussion about creating fast pulses, like most of them..

Again, answer to OP question, no, there is no simple improvised way to measure -3dB point of scope front end without leveled signal generator.
If you had 200-500 signal generator with decent amplitude accuracy across the frequency (like SDG6000X or similar) you could fairly well estimate what the curve is.

OP never explained what he wants to accomplish or why he asked the question.  Without that it is hard to help him.

If you have a decent VTVM with an RF probe, you can look at spot frequencies across the band of interest, adjusting the signal generator to the same level for each spot, which is how we did things back in the day, when absolute accuracy of sig gens was more the exception than the rule.
The beauty of it is that the device you use to check the level doesn't need to have high absolute accuracy, just good frequency response accuracy.

Of course, if you had an RF probe for use with a DMM, with a known accuracy, you could do the same.

PS:-You could also use the precision detector Berni referred to in the edit to his post 18.

[radiolistener]

Here's the results from a coin cell battery held like this:

The capacitor was 100% necessary to get a good time. Without it I was getting results in the uS range.

Here is my result. I connected lab PSU with 12.9 V to oscilloscope BNC with 1 meter wires with banana jacks.
Lab PSU has some uF range capacitor on the output.

I think this happens due to wire inductance.

[radiolistener]

For going faster i would recommend just buying Leo Bodnars fast pulse generator:
http://www.leobodnar.com/shop/index.php?main_page=product_info&products_id=295

Nice device, I want it, but very expensive.

[tggzzz]

Without it I was getting results in the uS range.

How do you use your scope to measure conductance?

(That's a pet peeve of mine)

[tggzzz]

This is second topic with exactly same question from OP.

This one also devolved into discussion about creating fast pulses, like most of them..

Again, answer to OP question, no, there is no simple improvised way to measure -3dB point of scope front end without leveled signal generator.
If you had 200-500 signal generator with decent amplitude accuracy across the frequency (like SDG6000X or similar) you could fairly well estimate what the curve is.

OP never explained what he wants to accomplish or why he asked the question.  Without that it is hard to help him.

Yes, as I asked the OP in reply #2.

It has only been a day; let's hope he will return with some comments.

[nctnico]

The best way to measure just the scope itself is to get one of leo's pulse gens you can directly attach and once you have a reference you can then attach your probes and see what they are rated for.

Big fat NO
See my earlier reply.

[vk6zgo]

Without it I was getting results in the uS range.

How do you use your scope to measure conductance?

(That's a pet peeve of mine)

 1 uSiemens would be about 1M , so you could use the 'scope as the null detector for a bridge.

[2N3055]

Someone needs to stay out of the special candies...

Case in point to not always go by what the mfg says... Siglent SDS2100x+.. can ONLY be legitimately raised to 350MHz but actually has a front end on all units that goes to 500MHz+, typically -3db of around 580-620MHz.

The best way to measure just the scope itself is to get one of leo's pulse gens you can directly attach and once you have a reference you can then attach your probes and see what they are rated for.  You can also check things like bnc cables... half the time people dont know just how horrible they are.  On the other hand most of this is moot if you are dealing with 100MHz and lower

That is absolutely wrong.  You can officially buy 500MHz bandwidth, but officially you can upgrade only from 350Mhz:

Siglent SDS2000XP-4BW50 Upgrade 500 MHz 4 channel
Bandwidth Upgrade from 350 MHz to 500 MHz for the SDS2354X Plus. Only in interleaving mode For this upgrade the latest firmware is required.
Nr: SDS2000XP-4BW50
EAN: SDS2000XP-4BW50

Also, by using Leo's pulser, you can absolutely make a great verification of scope's  risetime.  Bandwidth is measured by sweep with leveled gen. Period.

If you upgrade the scope, you can use fast pulser to check whether upgrade went through, because you will see faster risetime. And that is all.

Sometimes, some people don't really need to know bandwidth, they only need to check how fast is their scope in showing fast digital edges. That is what pulser is born for.

[tggzzz]

Sometimes, some people don't really need to know bandwidth, they only need to check how fast is their scope in showing fast digital edges. That is what pulser is born for.

Usually if you are interested in bandwidth/frequency, then it is better to use a frequency domain tool such as a spectrum analyser.

(Or, say it quietly, a scope with an FFT)

[Elasia]

Someone needs to stay out of the special candies...

Case in point to not always go by what the mfg says... Siglent SDS2100x+.. can ONLY be legitimately raised to 350MHz but actually has a front end on all units that goes to 500MHz+, typically -3db of around 580-620MHz.

The best way to measure just the scope itself is to get one of leo's pulse gens you can directly attach and once you have a reference you can then attach your probes and see what they are rated for.  You can also check things like bnc cables... half the time people dont know just how horrible they are.  On the other hand most of this is moot if you are dealing with 100MHz and lower

That is absolutely wrong.  You can officially buy 500MHz bandwidth, but officially you can upgrade only from 350Mhz:

Siglent SDS2000XP-4BW50 Upgrade 500 MHz 4 channel
Bandwidth Upgrade from 350 MHz to 500 MHz for the SDS2354X Plus. Only in interleaving mode For this upgrade the latest firmware is required.
Nr: SDS2000XP-4BW50
EAN: SDS2000XP-4BW50

Also, by using Leo's pulser, you can absolutely make a great verification of scope's  risetime.  Bandwidth is measured by sweep with leveled gen. Period.

If you upgrade the scope, you can use fast pulser to check whether upgrade went through, because you will see faster risetime. And that is all.

Sometimes, some people don't really need to know bandwidth, they only need to check how fast is their scope in showing fast digital edges. That is what pulser is born for.

[2N3055]

Sometimes, some people don't really need to know bandwidth, they only need to check how fast is their scope in showing fast digital edges. That is what pulser is born for.

Usually if you are interested in bandwidth/frequency, then it is better to use a frequency domain tool such as a spectrum analyser.

(Or, say it quietly, a scope with an FFT)

Sometimes amplitude is important...
But for strict frequency domain measurements SA is primary instrument... That is not questionable.

Although, i managed to some nice quick checks with a scope on occasion, at 432 MHz and 868 MHz telemetry transceivers.

[2N3055]

Someone needs to stay out of the special candies...

Case in point to not always go by what the mfg says... Siglent SDS2100x+.. can ONLY be legitimately raised to 350MHz but actually has a front end on all units that goes to 500MHz+, typically -3db of around 580-620MHz.

The best way to measure just the scope itself is to get one of leo's pulse gens you can directly attach and once you have a reference you can then attach your probes and see what they are rated for.  You can also check things like bnc cables... half the time people dont know just how horrible they are.  On the other hand most of this is moot if you are dealing with 100MHz and lower

That is absolutely wrong.  You can officially buy 500MHz bandwidth, but officially you can upgrade only from 350Mhz:

Siglent SDS2000XP-4BW50 Upgrade 500 MHz 4 channel
Bandwidth Upgrade from 350 MHz to 500 MHz for the SDS2354X Plus. Only in interleaving mode For this upgrade the latest firmware is required.
Nr: SDS2000XP-4BW50
EAN: SDS2000XP-4BW50

Also, by using Leo's pulser, you can absolutely make a great verification of scope's  risetime.  Bandwidth is measured by sweep with leveled gen. Period.

If you upgrade the scope, you can use fast pulser to check whether upgrade went through, because you will see faster risetime. And that is all.

Sometimes, some people don't really need to know bandwidth, they only need to check how fast is their scope in showing fast digital edges. That is what pulser is born for.

OK so I didn't understand what was your point here..  Sorry, still don't, but I see that I responded to something else. Sorry for that.

[Bud]

There is a neat method how to measure your oscilloscope's bandwidth that Keysight published in a couple of their Blogs. What you do is connect a fast pulser to input, apply Math "differentiate" function, then apply FFT to the result. This will produce a plot in frequency domain. Now use cursors to read the frequency point where level on the FFT plot drops to -3dB from the top.

https://community.keysight.com/community/keysight-blogs/oscilloscopes/blog/2016/09/01/how-to-measure-your-oscilloscope-and-probe-s-bandwidth-yourself

[JDubU]

I have a Rigol DS2000 scope that has a trigger output signal with a surprisingly fast risetime (faster than the scope's bandwidth).
To measure the risetime of the scope, I set the scope to auto trigger on one channel and connect the trigger out to the input of the other channel with a BNC cable.  Since it's auto triggering, the scope will free run and the trigger out pulse shows up, in sync, on the second channel.  Set the second channel to measure risetime and calculate the bandwidth from it (BW= 0.35/risetime).

Don't know if the RTB2000 has a fast trigger output risetime.

[Fungus]

FWIW I just did another test using the little probe spring instead of the clip-on lead.

As expected, the ringing was much reduced.

(Note for newbies: The "ringing" on the display probably isn't in the signal, it's coming from the probe clip. Use the spring whenever possible! BNC is even better!)



The measured rise times also seemed a teeny bit smaller, but there's not much in it.



PS: I just noticed that the crocodile clips on these Micsig probes aren't detachable. That's annoying. 

Edit: Re-did the screen captures after I remembered the 'scope can measure overshoot. 

nb. For those people just joining the thread, the capacitor is just shoved into the GND and +5V on the Arduino. I'm just using it as a 5V voltage source.

[tggzzz]

Sometimes, some people don't really need to know bandwidth, they only need to check how fast is their scope in showing fast digital edges. That is what pulser is born for.

Usually if you are interested in bandwidth/frequency, then it is better to use a frequency domain tool such as a spectrum analyser.

(Or, say it quietly, a scope with an FFT)

Sometimes amplitude is important...
But for strict frequency domain measurements SA is primary instrument... That is not questionable.

Although, i managed to some nice quick checks with a scope on occasion, at 432 MHz and 868 MHz telemetry transceivers.

My 350MHz Tek 485 triggers nicely on a 1GHz signal. The beam deflection leaves something to be desired, though

You can also estimate a signal generator's risetime using a spectrum analyser: https://entertaininghacks.wordpress.com/2015/08/11/measuring-digital-signal-edge-rates-without-an-oscilloscope/

[tggzzz]

(Note for newbies: The "ringing" on the display probably isn't in the signal, it's coming from the probe clip. Use the spring whenever possible! BNC is even better!)

The ringing is the consequence of the interaction between the scope probe's tip capacitance and a ground lead's inductance.

You can shift the frequency higher by reducing either the capacitance or the inductance.
https://entertaininghacks.wordpress.com/2015/04/23/scope-probe-accessory-improves-signal-fidelity/
https://entertaininghacks.wordpress.com/2016/09/17/scope-probe-accessory-higher-frequency-results/

BNC leads have a mis-match between the 50ohm cable impedance and the 50ohm//15pF scope input. That will form a comb filter at frequencies determined by the cable's length.

[Berni]

There is a neat method how to measure your oscilloscope's bandwidth that Keysight published in a couple of their Blogs. What you do is connect a fast pulser to input, apply Math "differentiate" function, then apply FFT to the result. This will produce a plot in frequency domain. Now use cursors to read the frequency point where level on the FFT plot drops to -3dB from the top.

https://community.keysight.com/community/keysight-blogs/oscilloscopes/blog/2016/09/01/how-to-measure-your-oscilloscope-and-probe-s-bandwidth-yourself

Yep that method works too. Here it is done using Leos fast pulser on a MSO9204H



First trace is the Ch 1 input, next trace is the differential of that signal (Showing a peak slewrate of about 4 GV/s) and then the FFT of that showing the bandwith.
And yes i know the cutoff does not look all that nice, but it is only a 2GHz scope so getting about 3.2GHz out of it at -3dB is not bad.

[tggzzz]

And yes i know the cutoff does not look all that nice, but it is only a 2GHz scope so getting about 3.2GHz out of it at -3dB is not bad.

It would look even better without that ringing/overshoot

But yes, it is fun to be able to visibly apply the theory of time-frequency duality, to practically measure things indirectly.

How about not differentiating, and comparing the FFT to that shown here: https://entertaininghacks.wordpress.com/2015/08/11/measuring-digital-signal-edge-rates-without-an-oscilloscope/
especially:

[Frank_MV]

Thank you for all your contributions, I will read everything at my leisure.

Background: I have upgraded my RTB2004 from 70MHz to 100MHz and would like to check if it was successful.

Br,
Frank

[Bud]

it is only a 2GHz scope

  

[David Hess]

As mentioned earlier:

If you use a probe, which is not a bad idea since what really usually matters is the bandwidth at the probe tip, then the connection should be made coaxially to the probe tip without the ground lead.

Leo Bodnar's pulse generator is an inexpensive source for suitable fast edges:

http://www.leobodnar.com/shop/index.php?main_page=product_info&products_id=295

Also:

Another source for clean and fast edges is the horizontal and vertical sync outputs on pins 13 and 14 on a VGA connector.

[Berni]

One thing to keep in mind when testing probes like this is that at high frequency the >10pF of probe capacitance starts to really matter and can drag down a 50 Ohm source, but up to 100MHz it's still pretty reasonable.

it is only a 2GHz scope

 

Well compared to other Infiniium family scopes it is pretty low.

But mostly was saying in terms of the 3.2GHz measured bandwidth compared to the 2GHz rated bandwidth. If you turn on filtering to get actual 2GHz of bandwidth out of it the ringing goes away.

[tggzzz]

One thing to keep in mind when testing probes like this is that at high frequency the >10pF of probe capacitance starts to really matter and can drag down a 50 Ohm source, but up to 100MHz it's still pretty reasonable.

15pF at 100MHz is ~100ohms. So much for such a probe being "high impedance 10Mohm"!

A 10:1 " low impedance" resistive divider Z0 probe will be 500ohms input impedance, and have a multi GHz bandwidth.