Author Topic: Frequency response of your DSO  (Read 34743 times)

0 Members and 1 Guest are viewing this topic.

Offline tinhead

  • Super Contributor
  • ***
  • Posts: 1801
  • Country: 00
    • If you like my hacks, send me a donation
Frequency response of your DSO
« on: January 30, 2013, 10:53:16 am »
while making some measurments for another one thread

http://www.eevblog.com/forum/projects/transmission-line-avalanche-pulse-generator/

i thought it could be nice to see the frequency response of other DSOs on the market.
I know, here and there other people posted already some pictures/results, but to have it in one thread could be more transparent.

So if you, you or you have a (proper) equipment to run some measurments on your DSOs, please do it so and add the information here. You can use as well the attached excel file to add/compare with my result.

Note: please don't use your DSO probes, unless you have some nice active probes and know what you doing. A RG-xxy cable from Signal Generator, terminated properly on the DSO end (this can be internal 50R termination, but as well t-connector with 50R at one end) should be the best option. Signal Gen should be something 1-200MHz at least, sine of course, when possible not one-hung-low type.

EDIT: proper means at least known signal source with max. deviation < ±1dBm over the measure range. This should be good enough to have an compare and overview

 
« Last Edit: January 31, 2013, 07:49:46 am by tinhead »
I don't want to be human! I want to see gamma rays, I want to hear X-rays, and I want to smell dark matter ...
I want to reach out with something other than these prehensile paws and feel the solar wind of a supernova flowing over me.
 

alm

  • Guest
Re: Frequency response of your DSO
« Reply #1 on: January 30, 2013, 11:27:27 am »
A few years ago, I posted the frequency response of an old Tek TDS-220 100 MHz lunch box DSO (though it lacks space for your lunch, unlike some cheap bench meters). I also compared the various vertical attenuator settings, this made a surprising amount of difference. I'll see if I can dig up the data later.

I remember several people with access to decent RF equipment (eg. rf-loop) quoting -3 dB numbers for the Rigol DS1052/1102E, but I don't think they ever posted the actual frequency response.
 

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 8243
  • Country: my
  • reassessing directives...
Re: Frequency response of your DSO
« Reply #2 on: January 30, 2013, 11:34:31 am »
OT: since iirc i've raised this issue here and there without satisfactory reply... i have no proper equipment to test, so i had to find a workaround, first using diode and cap and then others and fancy diy software etc etc, now i have a way that makes the DSO freq respond is 'almost' irrelevant. i can plot my DUT freq responds without having to worry too much of this matter.
if something can select, how cant it be intelligent? if something is intelligent, how cant it exist?
 

Online Jay_Diddy_B

  • Super Contributor
  • ***
  • Posts: 1668
  • Country: ca
Re: Frequency response of your DSO
« Reply #3 on: January 30, 2013, 12:51:50 pm »
Hi Group,
Here is the frequency response of my Tektronix TDS 754A 500 MHz scope.
A HP8657B RF generator was used. The cable was a very quality Goreflex.




I have also attached a zip file of the raw data if anybody wants you manipulate the data.

Jay_Diddy_B
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: Frequency response of your DSO
« Reply #4 on: January 30, 2013, 01:46:22 pm »
OT: since iirc i've raised this issue here and there without satisfactory reply... i have no proper equipment to test, so i had to find a workaround, first using diode and cap and then others and fancy diy software etc etc, now i have a way that makes the DSO freq respond is 'almost' irrelevant. i can plot my DUT freq responds without having to worry too much of this matter.

Please tell us the details. :)

Online Jay_Diddy_B

  • Super Contributor
  • ***
  • Posts: 1668
  • Country: ca
Re: Frequency response of your DSO
« Reply #5 on: January 30, 2013, 02:20:54 pm »
Hi,
 This time I measured a Tektronix 11A71 in a DSA602A mainframe. I could not get a stable waveform beyond 1.4 GHz

The setup was a HP 8657B signal generator and a Goreflex cable.



I have attached a zip file with the measurements for anybody who wants manipulate the data.

Jay_Diddy_B
 

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 8243
  • Country: my
  • reassessing directives...
Re: Frequency response of your DSO
« Reply #6 on: January 30, 2013, 02:36:52 pm »
Please tell us the details. :)
original message transferred to... http://www.eevblog.com/forum/projects/poor-man's-frequency-respond-cir-kit
i dont want to hijack this thread. Cheers.
« Last Edit: January 30, 2013, 03:14:40 pm by Mechatrommer »
if something can select, how cant it be intelligent? if something is intelligent, how cant it exist?
 

Online ve7xen

  • Frequent Contributor
  • **
  • Posts: 644
  • Country: ca
    • VE7XEN Blog
Re: Frequency response of your DSO
« Reply #7 on: February 01, 2013, 06:31:13 pm »
I have collected data using a Rigol DS2072 (70MHz) and HP 8657A signal generator (set for 1Vrms), connected with 1m of crappy RG-58 coax and a T with 50R termination. This generator is spec'd with flatness of +/- 0.5dB, but only at 0dBm, I'm running it here close to its max output of 17dBm. It's also probably not been calibrated in a long time, though I tuned it 6 months or so. I've somewhat 'guessed' at the RMS value for some of the readings, as the Rigol's RMS measurement can change 20-30mV depending on the horizontal scale. Not sure what that's about, but tried to get the most consistent values. All that to say this is not a very scientific survey.

Don't use Excel, so I created my own sheet in Google Docs, which is shared here if you're interested in the raw data or prettier graphs.

Edit: I was wondering if the dip at 30MHz was my generator/cabling or the scope itself, so I tried to measure it visually with my Tek 2230 analogue scope. By eye, it looks to be about 0.5dB down at 30MHz too, so I think this might be a measurement artifact.
« Last Edit: February 01, 2013, 06:56:05 pm by ve7xen »
73 de VE7XEN
 

Offline tinhead

  • Super Contributor
  • ***
  • Posts: 1801
  • Country: 00
    • If you like my hacks, send me a donation
Re: Frequency response of your DSO
« Reply #8 on: February 02, 2013, 05:02:14 am »
Other ReMark: the impedance of the input of the Rigol is 1 mOhm at 16 pf,

typical 16p, and thats good for 160-250MHz bw. The LPF in LMH6518 is doing the job
as well - and the software.

so how do you want to measure.....???
i don't want to measure probes nor the input stage itself (because in todays designs with FIR filters the
"input stage" ends up in DSP/FPGA), what the DSO is doing with the signal is also more complex
and i wish to see the results "on screen" (read back with eyes/usb/gpib).

However i don't want to ignore impendance match completly. Therefore the test signal should be
terminated in the DSO with 50R, and in case of Rigol or other DSOs without internal 50R then simply externaly
with 50R at the end of t-connector or feed through 50R temintor <- whatever works better for given DSO.
I don't want to be human! I want to see gamma rays, I want to hear X-rays, and I want to smell dark matter ...
I want to reach out with something other than these prehensile paws and feel the solar wind of a supernova flowing over me.
 

Offline EV

  • Frequent Contributor
  • **
  • Posts: 471
  • Country: fi
Re: Frequency response of your DSO
« Reply #9 on: February 02, 2013, 05:25:07 am »
What reading shall we use.., the reading of the signalgenerator or the real measurement  direct on the input.
Or just ingmore, but the this in the advantage op the DSO's with 50 ohm input.

Why not use the reading on the ascilloscope display, if we are testing scope?
 

Offline tinhead

  • Super Contributor
  • ***
  • Posts: 1801
  • Country: 00
    • If you like my hacks, send me a donation
Re: Frequency response of your DSO
« Reply #10 on: February 02, 2013, 05:45:25 am »
What reading shall we use.., the reading of the signalgenerator or the real measurement  direct on the input.
Or just ingmore, but the this in the advantage op the DSO's with 50 ohm input.

Why not use the reading on the ascilloscope display, if we are testing scope?

this is exactly what i'm trying to achieve, to read values from display - manuall or usb/gpib/net/...

A typical frontend frequency response would be not measured like that, but this is what i don't care about
knowing that todays DSOs manipulating the data anyway.

Example - my Tekway ST1202B is having Butterworth and not Gaussian respone (on display), where i know that
EXACT the same frontend is used by ATTEN/Siglent/Rigol - and from what i remember Rigol DS1k is having Gaussian response.
I know additionally that the frontend itselt have no issue with up to 500MHz, and flat enough for 300-350MHz.
I know however as well that due parallel connected ADCs (their capacity) the maximum frequency is cut, which is
why Rigol never released DS1202E (10ADCs) but Tekway did (8ADCs).

This thread should therefor show us what all these DSOs are having on display, to be able to compare implementations
and not pure hardware.
I don't want to be human! I want to see gamma rays, I want to hear X-rays, and I want to smell dark matter ...
I want to reach out with something other than these prehensile paws and feel the solar wind of a supernova flowing over me.
 

Online Marco

  • Super Contributor
  • ***
  • Posts: 3725
  • Country: nl
Re: Frequency response of your DSO
« Reply #11 on: February 02, 2013, 07:34:12 am »
Shouldn't you use a low impedance source and short cabling to test this? So impedance mismatch doesn't completely throw off your measurements? In fact a low impedance source and the original x10 probes seem a far better test than a 50 Ohm impedance source and a meter of coax to me.

With 50 Ohm termination on the input my Owon will only have around 30 Ohm input impedance at 200 MHz...
 

alm

  • Guest
Re: Frequency response of your DSO
« Reply #12 on: February 02, 2013, 08:04:53 am »
Good luck finding a << 50 ohm leveled sine wave generator for 200 MHz. At 50 ohm source impedance, the voltage over the probe input will be 100% of the generator output at DC (1 MOhm impedance), and 50% of the generator output at 200 MHz (50 ohm impedance for a probe with 16 pF of capacitance). This will introduce a much larger error than the change from 50 to 30 Ohms.
 

Online Marco

  • Super Contributor
  • ***
  • Posts: 3725
  • Country: nl
Re: Frequency response of your DSO
« Reply #13 on: February 02, 2013, 08:37:29 am »
So use a buffer (active probe will do) or at least use a 10x or 100x compensated attenuator with nominally 50 Ohm input impedance instead of simple termination.

Trying to gauge the quality of scopes not designed to have an accurate 50 Ohm input impedance over it's range with a 50 Ohm impedance source seems foolhardy to me ... is fine for the high end scopes with real 50 Ohm inputs, but not for the cheap DSOs. However unrealistic it is, they are designed for low impedance sources at high frequencies (of course even expensive passive probe scopes are the same in that respect).
« Last Edit: February 02, 2013, 09:08:00 am by Marco »
 

Offline tinhead

  • Super Contributor
  • ***
  • Posts: 1801
  • Country: 00
    • If you like my hacks, send me a donation
Re: Frequency response of your DSO
« Reply #14 on: February 02, 2013, 11:38:46 am »
So use a buffer (active probe will do)
i did, and this didn't changed anything (as the probe need to be anyway terminated to 50R).
I've tested some attenuators as well, no change.

The only bad influence is when usind DSO own attenuator, everything without attanuator (max 200mV/DIV by x1) is fine.
But that's fine, when using active probe with x10 2V/DIV are sufficient.

is fine for the high end scopes with real 50 Ohm inputs, but not for the cheap DSOs.

don't be so pessimistic, in china they using the same parts and same simulations, i don't think they stupid.
Some trying to design own frontends (wasn't Owon SDS having already 3 different frontends?), others simply
copying proven design (Tekway/Rigol CA,E,D series/Agilent DSO1kB/ATTEN/Siglent/LeCroy WaveAce/...).

But hey, you can run your own simulation of that China-most-used-frontend, schematic is here

http://www.eevblog.com/forum/chat/hantek-tekway-dso-hack-get-200mhz-bw-for-free/?action=dlattach;attach=12233

I ran my and i think it's not bad at all. I've added as well relais (worst case) parasitic capacitance (3p instead of 1p a in datasheet).

« Last Edit: February 02, 2013, 11:54:24 am by tinhead »
I don't want to be human! I want to see gamma rays, I want to hear X-rays, and I want to smell dark matter ...
I want to reach out with something other than these prehensile paws and feel the solar wind of a supernova flowing over me.
 

Online Marco

  • Super Contributor
  • ***
  • Posts: 3725
  • Country: nl
Re: Frequency response of your DSO
« Reply #15 on: February 02, 2013, 12:49:19 pm »
i did, and this didn't changed anything (as the probe need to be anyway terminated to 50R).
With a buffer I mean either a buffer at the scope connector OR a buffer at the signal generator so you can just use the scope's own compensated 10x probe.
Quote
I've tested some attenuators as well, no change.
An attenuator in and of itself does little, but attenuation gives you room for compensation. I agree though, active probes won't help ... I thought for sure they'd be better designed, but looking at something like the P6012 all it does is add inline termination for high impedance scopes ... bleh.
Quote
don't be so pessimistic
I'm not pessimistic, I'm just cognizant of the fact that they are designed for very low impedance sources at high frequencies and a compensated probe to get the correct frequency response. A cheap scope simply can't be tested as a separate system from it's probe unless you are willing to build specialized circuitry, inline termination is not the same as having an input designed to have 50 Ohm impedance over the device's usable frequency range ... the probe and it's compensation are integral parts of what makes up the frequency response.
 

Offline tinhead

  • Super Contributor
  • ***
  • Posts: 1801
  • Country: 00
    • If you like my hacks, send me a donation
Re: Frequency response of your DSO
« Reply #16 on: February 02, 2013, 01:27:44 pm »
input designed to have 50 Ohm impedance over the device's usable frequency range

what's about this one? LeCroy, 1M/50R input, 300MHz BW

http://teledynelecroy.com/oscilloscope/oscilloscopemodel.aspx?modelid=2122#spec


I don't want to be human! I want to see gamma rays, I want to hear X-rays, and I want to smell dark matter ...
I want to reach out with something other than these prehensile paws and feel the solar wind of a supernova flowing over me.
 

Online Marco

  • Super Contributor
  • ***
  • Posts: 3725
  • Country: nl
Re: Frequency response of your DSO
« Reply #17 on: February 02, 2013, 01:59:56 pm »
What about it? It has internal termination, so you don't need inline termination ... as for whether it's designed to be 50 Ohm across it's rated bandwidth I have no idea,  you'd hope so for a LeCroy (the limited voltage range for 50 Ohm does suggest it).
« Last Edit: February 02, 2013, 02:15:34 pm by Marco »
 

Offline tinhead

  • Super Contributor
  • ***
  • Posts: 1801
  • Country: 00
    • If you like my hacks, send me a donation
Re: Frequency response of your DSO
« Reply #18 on: February 02, 2013, 02:20:35 pm »
What about it? It has internal termination, so you don't need inline termination ... as for whether it's designed to be 50 Ohm across it's rated bandwidth I have no idea,  you'd hope so for a LeCroy.

Marco,

and that's exact the point, this scope is using exact the same frontend as Tekway/Rigol CA,E,D/Agilent DSO1kB/ATTEN/Siglent,
an frontend designed to be used as 1M/50R. The only diff is that models with internal termination are using 49.9R resistor direct
near BNC, which is from signal path perspective equal to feed through terminator.

Therefor statement like

Trying to gauge the quality of scopes not designed to have an accurate 50 Ohm input impedance over it's range with a 50 Ohm impedance source seems foolhardy to me ... is fine for the high end scopes with real 50 Ohm inputs, but not for the cheap DSOs.

is simply useless, you can't judge about things you don't know.
I don't want to be human! I want to see gamma rays, I want to hear X-rays, and I want to smell dark matter ...
I want to reach out with something other than these prehensile paws and feel the solar wind of a supernova flowing over me.
 

Online Marco

  • Super Contributor
  • ***
  • Posts: 3725
  • Country: nl
Re: Frequency response of your DSO
« Reply #19 on: February 02, 2013, 02:40:17 pm »
The only diff is that models with internal termination are using 49.9R resistor direct near BNC, which is from signal path perspective equal to feed through terminator.
Judging a scope which has a specified 50 Ohm input on the quality of that input is meaningful ... if it's not actually ~50 Ohm across it's frequency range that's a black mark on them.

Judging a scope which doesn't have a 50 Ohm input on it's frequency response to a 50 Ohm source with an inline terminator for which it was never designed is not meaningful ... if you pretend that's doing anything more than performing a roundabout measurement of input capacitance that's on you.
« Last Edit: February 02, 2013, 11:07:59 pm by Marco »
 

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 8243
  • Country: my
  • reassessing directives...
Re: Frequency response of your DSO
« Reply #20 on: February 02, 2013, 03:07:04 pm »
imho. the Jim Williams' quote comes to mind. "the best probe is no probe". if you somehow can bring the generator as close as possible to the point less than reflection length, you can connect them without termination on the dso input end. but whats the point? it may not be feasible bringing tonnes of equipment close together. a quality 50ohm coax and a quality inline terminator is all you need to get the "exact half" of the generator's signal appearing at the scope's input. as someone (Janne) proved, good (flat) to few GHz range (i believe it should be here http://www.eevblog.com/forum/beginners/how-to-better-wire-the-bnc-connector-and-ghz-issue/msg37963/#msg37963 but the proof has been erased)
if something can select, how cant it be intelligent? if something is intelligent, how cant it exist?
 

Online Marco

  • Super Contributor
  • ***
  • Posts: 3725
  • Country: nl
Re: Frequency response of your DSO
« Reply #21 on: February 02, 2013, 03:21:18 pm »
A quality inline terminator with a parallel 10 pF capacitor is no longer a quality inline terminator.
 

Offline tinhead

  • Super Contributor
  • ***
  • Posts: 1801
  • Country: 00
    • If you like my hacks, send me a donation
Re: Frequency response of your DSO
« Reply #22 on: February 03, 2013, 12:26:32 am »
Marco,

i see you still didn't got what the reason for this thread is. Ok, i will try once again. It does not matter WHAT an unknown
ideal scope, with calibrated 50R internal input is showing at rated XXX frequency - because we know this anyway from
manufacturer datasheet. The same for unknown DSO with provided probes, they almost always too cheap, they do
have almost always unknow frequency response, they didn't always matching the DSO, not always properly compensated.

And here is the point, having signal source with known level at specific Z we can measure what a DSO, with no calibrated
internal 50R temrination, is doing with this signal. For why? Well, this is the reference level (and this is exactly what you using,
not an imaginary or real value on input but what on display), a 50R teminator will be always 50R terminator, coax cable still a
coax cable (and yes there are diff, but they didn't matter on "almost DC").

And this is reference level for all DSOs then, no matter if they calibrated to 50R initernal termination or not, e.g. product A (Rigol DS1052E) will have on paper (from hardware point of view) exact the same response as product B (Agilent DS1052B).
Why ? Becuse they using exact the same hardware. The questions is, due the fact that the software might be different, are the result on screen equal as well? And same apply to majority of chinese (and as well LeCroy, Agilent low end) products.

For you, as owner of Owon, the kind of measurment is still important. You was looking for active probe, which would be then
anyway terminated external with 50R. So for you as well would be interessting to see what such probe, giving on product A such
respone (with external termination) or product B (with internal calibrated termiantion) another response, is doing on your DSO.
Btw, consider these active probe, i'm using two of them and the performance (up to ~500MHz) as as good as P6243
(i'm cheating my TEK with own eeprom):

http://welecw2000a.sourceforge.net/docs/Hardware/Aktiver_Tastkopf_mit_OPA659.pdf

Another aspect of this thread was to show kind of response, is it flat or gaussian? And here does not matter if you at DSO
rated frequency off by 1.8dB (as my test DSO), the kind of response is still the same:



« Last Edit: February 03, 2013, 12:28:22 am by tinhead »
I don't want to be human! I want to see gamma rays, I want to hear X-rays, and I want to smell dark matter ...
I want to reach out with something other than these prehensile paws and feel the solar wind of a supernova flowing over me.
 

Offline Wim13

  • Regular Contributor
  • *
  • Posts: 225
  • Country: nl
Re: Frequency response of your DSO
« Reply #23 on: February 03, 2013, 12:39:25 am »
Back to the topic, to compare the graphs of different DSO's,

If you want to compare you have to use everytime the same setup,
and thats the point i want to make, otherwise its compare has no meaning.

If a use a constant voltage on my DSO ( Rigol 2072 ) of 0 dbM ( 0,223 V )
on the BNC connector, 1 Mohm 16 pf, i have a straight line from 70 Mhz 0 dB to 200 Mhz -3 dB.

if a use a quality signal generator, 50 ohm, 1 meter of coax, T connector on the DSO with 50 ohm
terminator with a kown specification i get a straight line form 70 Mhz 0 db to 200 Mhz -6 dB.
Thats is making the difference.

The uncertainty of this whole setup is within 1 db.

If everyboy is using this samen setup, and also specify the uncertainy, then we can use all the info
 
 

Online Marco

  • Super Contributor
  • ***
  • Posts: 3725
  • Country: nl
Re: Frequency response of your DSO
« Reply #24 on: February 03, 2013, 12:50:12 am »
For you, as owner of Owon, the kind of measurment is still important. You was looking for active probe, which would be then  anyway terminated external with 50R.
I kinda assumed an active probe designed for high impedance scopes either put a buffer on the scope side connection or put a 10x compensated attenuator at the scope side which should get you close to a much more bearable 1 pF in parallel with the terminator, that's how I would do it.
« Last Edit: February 03, 2013, 12:53:42 am by Marco »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf