Products > Test Equipment
Equipment to measure RC filter attenuation
<< < (4/5) > >>
mawyatt:
Agree with tautech about using the Hi Z modes, use directly with AWG BNC cable shown to DUT and 10X/1X scope probe for DSO CH1 (Input) and set to 1X scope probe for CH2 (Output). You want to place the highest possible signal on the DUT input assuming the filter input can handle the larger voltage swing (caution if polarized caps are present, then you'll need to add DC bias), this produces the highest DR within the Bode Plot routine since the input and outputs are dynamically scaled during the sweeps which produces a DR well beyond what an 8 bit ADC can provide. The built-in AWG with the SDS2000X+ has lower output capability and you would be better served with a separate AWG which can supply a higher voltage stimulus for the input.

What are the details of the LP filters you wish to measure?

We were/are impressed with the Bode capability built-in the SDS2000X+. The "Frequency Selective Measurement" method works very well to "pull out" the signal from the noise environment, and the dynamic scaling also works well to enhance the overall DR. This is not a typical hobby level feature you might expect on a lower end DSO, but in the hands of a capable user, a very powerful and useful feature and our only criticism is it's a little slow and forgets the IP address of AWG when using LAN (fixed with latest firmware update  :-+ )

We found this Bode feature so useful we built up a couple Isolation Transformers for use with Closed Loop Bode Measurements that the OP might find worthwhile.

https://www.eevblog.com/forum/testgear/diy-transformer-for-use-with-bode-plots/

Anyway, hope this helps.

Best,

tautech:
Thanks very much Mike and FYI slugrustle has the SDG1032X standalone AWG with similar capability to your SDG2000X and IIRC you have a SDG6000X too ?

And I still need find time to get onto this............   |O
rf-loop:

--- Quote from: tszaboo on September 14, 2022, 09:50:14 pm ---I'm, not surprised, 80dB is a lot. Scopes have 8 bit ADCs so very low dynamic range when it comes to these sort of measurements. You could try setting the AWG into white noise mode and measure the filter with your scope's FFT mode, based on your windows and your setting, you might be able to lower the noise floor. Higher dynamic range is usually done by instruments called dynamic signal analyzers. But  I only used one once, not sure how to set it up. DMM is useful because it can measure 10V and 100mV without too much error, but that's also going to have trouble above 80dB dynamic range.
How about specialized audio gear?

--- End quote ---

Siglent FRA have continuous AGC and also receiver channels are frequency selective.

FRA itself whole dynamic range is over 130dB  independent of if 8 bit ADC as in SDS2kX+ or 12bit ADC in SDS2kXHD and even roughly same with SDS1kX-E.

But there is no single simple truth. Now of course it need define what we mean with dynamic range. In bottom level we need define what noise level we accept - what level measurement uncertainty and so on, so on.

Siglent FRA (also known as "BodePlot") have continuous AGC (Automatic channel Gain Control) what can step full level range in every freq step(1). And more,  also receiver channels are frequency selective (center is moving during sweep).
(1) 50ohm 500uV/div to 1V/div and 1M 500uV/div to 10V/div.

FRA itself whole dynamic range is (nearly (2)) independent of if 8 bit ADC as in SDS2kX+ or 12bit ADC in SDS2kXHD and even roughly the same also with SDS1kX-E.
(2) If AGC is on, as is default. User can turn it also off (more fast sweep but reduced dynamic range)

But there is no single simple truth - as nowhere. Now of course it needs to define what we mean by dynamic range. In bottom level we need to define what noise level we accept - what level measurement uncertainty and so on, so on.

One limit is also maximum level what  depends about generator and other things..

If we think in some freq sweep DUT top level in some frequency point is  0dBm. Distance to FRA bottom noise level is over 100dB.
If DUT max out level is 10dBm then distance to FRA noise level is over 110dB... And so on.
If use oscilloscope internal 50ohm it set limit to maximum level. It isa then around 22dBm (because 1V/div max)
If use external 50ohm load and inputs 1M. Max is now 42dBm. Now FRA noise level is over 140dB down from this level.

One common way to tell oscilloscope FRA dynamic range is tell dynamic range from 0dBm level to down. (example Keysight use this, but without definition about bottom level quality etc)
Naturally if user turn FRA  AGC off then dynamic range is reduced.

If do not want talk dBm levels everyone can calc what are dBV or what ever other units.

Here is one tiny example about FRA  dynamic range down from 0dBm level. (SDS2504X HD)


C4 is just only for display FRA bottom noise level. (cascaded 2 cheap IF filter for more stop band attenuation)





Here other example, old quick test with SDS2kX Plus. And in this case I have extremely heavy problems to step this attenuator.
(0dBm signal go to attenuator as DUT in  and attenuator out is as DUT out.)
Most of sweeps go to garbage due to atten step problems  and finally I get this... and after then I throw this mad stepping attenuator "out from window". Steps levels was ok but stepping mostly fails ... arrghhh.., )


C3 trace is just for display FRA bottom noise level.
As can see -90dBm have some random noise and -100dBm more. It is up to user who can define what he accept. Still visually it is quite clear where is average.
But if there is narrow dip...and need measure its bottom level. After every sweep level is different, how accurate user need measure, again up to user needs.




In next image:  Vertical top 40dBm to bottom noise level is Siglent FRA dynamic range if we define it using noise level.
Note: Base noise level depends freq. (here limited to 60MHz, you can extrapolate with eyes how it continue to 120)
If want think dBV then top is 27dBV and 100dBm line is -113dBV
(As can see sweet pot is around 400kHz just nice for 450-500kHz IF filters)
This image, below,  is also with SDS2kXHD  but Xplus is roughly same in this matter.


Sad we do not have trace average function for reduce FRA displayed random noise.

---------------
For OP

I believe Siglent FRA (SDS2kXPlus) and SDG1000X  can do this job quite easy. Specially if you can drive circuit using 20Vpp and filter out in passband have not significant attenuation. (What level is not possible using SDS internal generator).
With 10x probe (DUT out) still  available dynamic range is quite lot. But then, this BodePlot is also very slow specially with low frequencies.
SDS2kPlus  FFT can also try.
mawyatt:

--- Quote from: tautech on September 15, 2022, 03:19:22 am ---Thanks very much Mike and FYI slugrustle has the SDG1032X standalone AWG with similar capability to your SDG2000X and IIRC you have a SDG6000X too ?

And I still need find time to get onto this............   |O

--- End quote ---

Yes we have both , and put both to good use often. The waveform quality/performance is outstanding for the $, so one can overlook the UI quirks ::)

For the OP, since the DR limiting factor with the Bode Mode as shown by rfloop is the DSO Output Channel noise floor. One wants to drive the DUT input with the largest practical input signal as we mentioned and use no attenuation on the DSO Output Channel, which gets the largest signal to the DSO Channel for the best Output sensing SNR.

Also the OP mentioned measuring filter responses with a quality DMM, this is certainly practical at lower frequencies but does not convey any phase information. Many filter applications the phase response is very important (signal processing for example), so here the DSO Bode Mode or a full fledged Network Analyzer is in order.

For analysis of power line/supply filters such as "T" Notch, LRC Low Pass and RC Low Pass the modern Bode Capable DSO is an excellent instrument of choice IMO. One can "see" the effects of loading, DC bias in somewhat real time (OK the SDS is a little slow) without having to resort to a PC and plotting.....this is why we always prefer a stand alone DSO rather than a PC enabled scope (we have the excellent Picosope 4262 but rarely used).

Best,
Someone:

--- Quote from: rf-loop on September 15, 2022, 08:25:37 am ---But there is no single simple truth - as nowhere. Now of course it needs to define what we mean by dynamic range. In bottom level we need to define what noise level we accept - what level measurement uncertainty and so on, so on.

One limit is also maximum level what  depends about generator and other things..

If we think in some freq sweep DUT top level in some frequency point is  0dBm. Distance to FRA bottom noise level is over 100dB.
If DUT max out level is 10dBm then distance to FRA noise level is over 110dB... And so on.
If use oscilloscope internal 50ohm it set limit to maximum level. It isa then around 22dBm (because 1V/div max)
If use external 50ohm load and inputs 1M. Max is now 42dBm. Now FRA noise level is over 140dB down from this level.
--- End quote ---
It is very complex with no single truth as you say. Were your example plots using a 20 or 40dBm injection signal? the channels didn't seem to be high enough scale for that. Other effects come to play with increasing the stimulus signal, crosstalk of the scope channels, and leakage/crosstalk in the DUT. So it is not always adding more stimulus amplitude = more dynamic range.

It would be good to show an actual configuration achieving more than 120dB dynamic range, rather than just moving the reference value and promising it might could achieve it.
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod