GW Instek FRA Software

[rigol52]

Anyone tested GW Instek DSO scope with AWG (like model MSO-2204E),
using this new GW Instek FRA (Frequency Response Analysis) software?:

http://www.gwinstek.com/en-global/products/Oscilloscopes/Digital_Storage_Oscilloscopes/MSO-2000_Series

https://www.google.si/url?sa=t&rct=j&q=&esrc=s&source=web&cd=10&ved=0ahUKEwjswuDe7N_YAhUGEiwKHVKgDkQQFgiAATAJ&url=http%3A%2F%2Fwww.gwinstek.com%2Fen-global%2FDownload%2FDownloadFile%2FDownloadFile%2Fdownload%2523_%252301_Oscilloscope%2523_%2523MDO-2000E%2523_%2523Manual%2523_%252320180108_UM_MDO-2000_FRAapp_E.pdf&usg=AOvVaw0glcBayhwuddQWUJmAqVp4

[rhb]

I plan to.  I've got an MSO-2204EA and a VNWA 3E.  As it happens, I suggested the idea to GW during some pre-purchase discussions.  I'm both surprised and pleased that they took up the idea.  It will be interesting to see how they did.  In principle, it should provide usable results over most of the HF spectrum, though with some limitations on dynamic range.

[rigol52]

I am looking forward to your test with great interest, thanks.

[rhb]

It will take a while.  I've not even plugged the VNWA in yet, nor my new GPSDO.  I *just* got most stuff  plugged back in, but the mat of power cords to the power strip is so thick I may need to redo it.  I've still got a bunch of stuff to plug in if I can find room.  It's a cramped space.

[rfdes]

I just downloaded the APP but see no instructions on how to install.  The technique appears different than a firmware upgrade.
Anyone installed the APP yet?  If so, please explain the steps.
Thanks
Jim

[nctnico]

I just downloaded the APP but see no instructions on how to install.  The technique appears different than a firmware upgrade.
Anyone installed the APP yet?  If so, please explain the steps.

It is just a matter of selecting the APP in the file browser and pressing the select button.

[rfdes]

Thanks for responding.

These 5 files are included in the APP download.  None of these will initiate an install when selected with the SELECT button.

FRA.txt
FRA.inf
FRA.lua
FRA.png
nLicense.png

I thought the FRA.txt file may be the instructions but I cannot read the file with a standard text editor.
The user manual also has no installation instructions.

[Kevin]

I have ever been installing Digital Filter App on GDS-1000B.  Try again, just select the .gz file(the original file).

[rigol52]

  Try again, just select the .gz file (the original file).

As rfdes say, after unpacking .gz file, we got this files:

FRA.txt, FRA.inf, FRA.lua, FRA.png, nLicense.png

FRA.txt probably contain Chinese characters only  (not readable by any editor available to me).

I am not GW Instek Scope owner, just preliminary interested due those FRA software feature.

[rfdes]

I sent a technical support request yesterday to GW.  Once I find the answer, I will post my findings.
take care - Jim

[rfdes]

I have ever been installing Digital Filter App on GDS-1000B.  Try again, just select the .gz file(the original file).

Kevin - You were absolutely right.  Not sure how I screwed this up but simply selecting the .gz file did the trick.
Thanks again.
Take care - Jim

[rigol52]

Does this mean, that final installation and run location of this FRA software is scope and not desktop PC?

That this, in fact, is about scope firmware upgrade?

[rhb]

Yes, it is a FW upgrade.  Requires V 1.34 and the FRA app.  Only works on the scopes with built in AWGs.

[rigol52]

This is even better FRA solution, than I thought at first (that it run on PC, processing signals, taken over USB cable).

[nctnico]

Perhaps somebody would care to post some screenshots?

[rhb]

Here are some screen shots.   Actually, I'd be embarrassed to even show it as a work in progress as it is not useful even as a pedagogical tool for the most basic demonstrations. It's pretty hard to find a more basic example than the series response of a crystal.

First shot is just 50 Ohm BNC cables with Tees and terminators at the scope input sweeping from 100 Hz to 25 MHz

The second is the response of a 40 MHz 3rd overtone xtal sweeping from 6-25 MHz

The 3rd is the parameters for the xtal pass.  Note it's all of 56 points.  The plots are all at the maximum resolution.

If you reduce the span it just gives you shorter traces on the same display.

It's so poorly done I'm not going to bother comparing it to the VNWA 3E.  I'm going to reinstall 1.32 as the SA demo was somewhat useful and they have removed or disabled that on the MSO line.

[rf-loop]

Here is Siglent plot.

And first, it is not at all perfect. Far away from perfect. Useability and visual ergonomy need lot of finishing. Colors, lay-out and some adjustments need some work for more sohistivated useability. But it is first FW where is BodePlot.
Also, SDS1000X-E series do not have internal signal generator. Signal generator is external and it is automatically controlled via USB and SDG do not need even touch at all, exept turn it on. Bode Plot function control it automatically. 
For this I did not use SDS1104X-E optional external AWG model. In this playing I use SDG1032X-h.
DUT was random  14.2MHz (damaged) filter where pass band was out of order and highly attenuated.

in other image there is also data table open (501 data points) and bit different center freq and span adjusment.
Other image without data table and span 25kHz and center 14195kHz

In this case scales are linear. They can be also log and also vertical can use dBm  with user adjustable DUT impedance.

It is strange that chinese GoodWill max resolution is around 50 (Siglent 500) and very strange is this display GW plot trace scaling. (Or is it user error?)
But also Keyshit BP resolution is total shame if it really is 10pts/decade max. Really - per decade - for what?
How many points is here if calculate points/decade. 14MHz center, span 25kHz and this 25kHz have 500 points.
How are these GW data points defined. Always bit over 50 points for used span and or for display width or what?

[nctnico]

According the GW Instek's info the FRA function is there to compete with Keysight's DSOX1000 so I'm not surprised that it isn't that sophisticated. Using it on an RC network or an amplifier will probably yield more useful results.

[rf-loop]

According the GW Instek's info the FRA function is there to compete with Keysight's DSOX1000.

Only for this? Not for use as tool?

[simone.pignatti]

According the GW Instek's info the FRA function is there to compete with Keysight's DSOX1000.

Only for this? Not for use as tool?

ahah good one!

[nctnico]

According the GW Instek's info the FRA function is there to compete with Keysight's DSOX1000.

Only for this? Not for use as tool?

It depends on what you want to use it for. If the goal is to look at amplifiers and RC networks then it will work just fine. For things like looking at resonances you'll need more dynamic range anyway. You have to be careful not to want too much from an oscilloscope which is inherently limited by a small dynamic range compared to a spectrum analyser and/or network analyser.

[JPortici]

According the GW Instek's info the FRA function is there to compete with Keysight's DSOX1000.

Only for this? Not for use as tool?

[rhb]

The problem is the FRA interface which insists on using a logarithmic X axis and does not match the window to  the length of the data. This where having the ability to program and modify apps for the instrument would make a huge difference whether Instek made their version open source or not.  Though, if it were open source, I'd have fixed it  within  day or two at at most.

The AWG has the ability to generate signals with offsets.  So judicious use of DSO and AWG settings can substantially improve the sampling quality beyond the 8 bits ADC.  All of this should be scriptable using SCPI commands.  So doing it on the scope should be a standard feature.

As a standard user interface feature, the user should be able to select a SCPI script from a USB drive or from internal flash and run it.  It would be preferable to have the Lua framework that Instek uses as an official product feature, but once one has the ability to program the instrument, lots of things are possible.

[maex]

Here is a Bode Plot from my Micsig DP10013:

the first with x50 the second with x500

[rigol52]

Here is a Bode Plot from my Micsig DP10013:

Hi maex

Nice plot.

Can you explain your testing configuration (scope model , AWG)?

Micsig DP10013 is only Differential Probe and you probably didn't upgrade
GW Instek Software/Firmware to Micsig scope firmware? Thanks.

[rhb]

Insert a filter or crystal in the path and see what you get.  I could not get a usable xtal response with anything I tried.  The problem is lack of user control of the sampling.  Trivial to fix if we had the source code.

FWIW I've had the 2072E on the bench identifying chips and  function. It looks very much like an FPGA dev board in a nice package with lots of inputs and a good analog front end.  One thing that struck me looking at the board was how easy it would be to generate basic service data for.  And the SMD parts make replacement pretty simple with everything on a couple of planes.

[nctnico]

Another problem could be impedance mismatch so you can't get enough dynamic range and/or the crystal is being loaded too much so it can't resonate.

[rhb]

It's a programmer error.

[rigol52]

Does anyone have live contacts with Siglent staff, to ask about their possible plans

on FRA (Bode plot) solutions with their low range oscilloscopes like SDS1202X-E ?

I wrote to distributer but no answer.

[Performa01]

Does anyone have live contacts with Siglent staff, to ask about their possible plans

on FRA (Bode plot) solutions with their low range oscilloscopes like SDS1202X-E ?

I wrote to distributer but no answer.

As far as I can tell there are no plans for the SDS1202X-E.

For Bode plotter to work, the scope requires either a built-in signal generator or the ability to control an external one. Both options are missing on the SDS1202X-E. It also would be a bit limiting as it is a 2 channel scope and could do just single channel frequency/phase analysis. But the SDS1104X-E and SDS1204X-E come with a proper 3-channel bode plotter that is meant for real work and not just another me-too feature.

Anyway, this is the wrong thread to discuss this topic. If you're interested, watch out for a review (coming soon) here:

https://www.eevblog.com/forum/testgear/siglent-sds1104x-e-in-depth-review/

[rigol52]

Hi Performa01, thanks for info. Exactly what I am looking for.

[SWR]

My apologies if these are obvious points you've already tried with the crystal series resonance:
- Increase Points/Decade from 10 to 90.
- Try different AWG amplitudes.
- Try both 50ohm and High Z AWG load setting.

I agree that it would be a killer feature to make these apps open source.
They seemingly already implemented a script interpreter, so why not make it public?

It's strange that they didn't implement the resolution parameter as Points/Division instead of Points/Decade?
That ought to be a small script change and would make it useful in a wider range of applications IMHO.

I'm still waiting for my MDO-2204EX to arrive, but I'm looking forward to playing with this.

Best regards
Soren

[rhb]

I tried all of those which is why I was so disappointed.   For the xtal sweep range I only had 56 points which is why the crap result.   They are using Lua for the scripting language, so there is the decompiling byte code problem.  Not impossible, but a nuisance.

I'm almost ready to start real work.  The heatsink for the BeagleBoard X15 ARM chip arrived today and the glue is drying. I didn't realize I needed to ground Power-On# and cut it off and heatshrinked the end.  So I've got to take it apart and fix that.  I also need an assortment of bulkhead mounted connectors.  But the MicroZed is ready to go.  I just need a block of time without any interruptions to start on the FPGA  tutorials for the MicroZed.

GW will not allow the SA app to run on my MSO-2204EA.  The marketing guy was clearly unhappy that I had managed to subvert his new product. He "fixed" the mistake in 1.34. So I reverted to 1.32 as the SA app is usable whereas the FRA is not.