Author Topic: Use of SDS1204XE as Curve Tracer (Tauteg??)  (Read 1511 times)

0 Members and 1 Guest are viewing this topic.

Offline watchmakerTopic starter

  • Frequent Contributor
  • **
  • Posts: 612
  • Country: us
  • Self Study in EE
    • Precision Timepiece Restoration and Service
Use of SDS1204XE as Curve Tracer (Tauteg??)
« on: October 09, 2023, 12:17:20 pm »
I have a BK 501A curve tracer and 3 scopes, BK 1520, Agilent 54621A, and the SDS 1204XE.

The 501 A works best with the BK 1520.  Clean trace lines and the graticule means something (ie I can get results that match other TTs).The downside is high flicker rate which I guess is due to retrace.  V=.2mV and H=1.0V.

Using the same settings on the Agilent, I get the same good results, with less, but still some, flicker.

When I use the Siglent with the same settings, I cannot use the graticule. The divisions make no sense. Also, the "traces" are very broad vertically making measurement a guessing game.   I measure from the bottom of one trace to the bottom of the next. Very stable display and if I drop H sens I can reduce the digital step effect of the trace. (BTW, this is not apparent on the Agilent).  But then, I have no idea where I am on the voltage (x) axis.

Would be nice if there was a way to drop a marker on the trace so it can be moved left or right and have some knowledge of where you are.

If I set the Agilent to 302mV V sens. I get traces that can be measured through the graticule (but not cursors).

The other thing about the Siglent is that it reduces the working area of the screen by maybe 50%, horizontally.  Keeps the vertical.

Most of this is academic, but I would like to better understand the different behavior of the Siglent against the other two scopes.  I suppose it is the difference in technologies, but what ARE those differences?

Why the scaling issue?
Why the loss of screen territory?  (Does the same on all 3 graticule settings)
Why such a broad trace only on the curve tracer (great traces on all other applications)?

The scaling issue may be that I need to calibrate the Siglent to the 501A individually which is too much effort if I have other means and these other means do not require individual calibration.  And it does address the graticule or trace thickness (which increases measurement error).  Nor does it explain why the other 2 scopes can be used at the same settings.

Since I am setup to learn EE (using Kahn, USN and Diligent YouTube) I purchased an Analog Discovery 2 w/ Curve Tracer module.  I have not yet set it up, but the tutorial indicates scope-like trace on CT mode.

I really like the Siglent, and realize it could well be it is the wrong tool for the application.  I bought the BK 1520 to see if analog behaved better, which it did.  I had the Agilent as a starter scope before I found out about the Siglent here.

I am driven by understanding "difference".  And without difference, there is nothing to measure.  So this strikes at the very core of my being.

THANKS!!!
Regards,

Dewey
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Use of SDS1204XE as Curve Tracer (Tauteg??)
« Reply #1 on: October 09, 2023, 01:29:38 pm »
You can find some clues for proper use of a DSO as a curve tracer here:
https://www.eevblog.com/forum/testgear/fooln-around-with-dso-awg/

Mike goes into greater depth here:
https://www.eevblog.com/forum/projects/dso-awg-based-curve-tracer/
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline watchmakerTopic starter

  • Frequent Contributor
  • **
  • Posts: 612
  • Country: us
  • Self Study in EE
    • Precision Timepiece Restoration and Service
Re: Use of SDS1204XE as Curve Tracer (Tauteg??)
« Reply #2 on: October 09, 2023, 05:14:14 pm »
Tauteg,

Thanks, but those links do not address my question.  Attached are four shots of four of the curve tracers.  They are:
  • The Discovery with its curve tracer attachment
  • The Discovery with the BK 501a as trace generator
  • The Agilent with the BK 501 as trace generator
  • The Siglent witht the BK 501A as trace generator

The Discovery/Trace Module gives the best display with cursors that measure correctly.
The Discovery in xy with the BK generator is ok and better than the Siglent but not measurable with accuracy
The Agilent with BK 501A is a clean trace (but flickers) and the cursor measurements match the graticule (H=1V, V=100mA)
The Siglent gives a stable, not useable trace using the BK 501A.  The lines are too thick, I have to set it to V=150 to measure with the graticule and the cursors provide meaningless data (at least to me).

Also note the lost screen area on the Siglent.

I did not get the BK1520, but I presume you will take my word that it yields the same information as the Agilent, it just flickers too much with the BK501A.

I am just starting my journey, but this discrepancy is going to bother me.  I do not NEED a CT at this in point in my studies, and I have two ways to get good measurements (Discovery and Agilent).

In order to know how to use the Siglent, I would like to understand the way it behaves in comparison to everything else.

THANKS!

1895064-01895064-1




Regards,

Dewey
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Use of SDS1204XE as Curve Tracer (Tauteg??)
« Reply #3 on: October 09, 2023, 08:21:24 pm »
Respectively watchbreaker (get my handle right and I'll get yours correct) this issue is staring you plainly in the eyes if you study your screenshot against those Mike posted in the thread I linked.

Look at this one and examine settings and channel 0V settings, V/div and their offset to have traces on the XY display.


Instead you have 0V positions near or at 0V on the XY display.


Further you are using AC coupling for channels and trigger which you need change all to DC.
Also note Mike is using the 20 MHz BW limit, found in each channels menu and just this alone will reduce trace noise dramatically.
Pro tip:
Rather than use the Save/Recall menu for screenshots I recommend you instead use the blue Print (file) button, the shortcut to grab screenshots to a USB stick which also captures menu settings for the scope.


I do understand any new instrument requires time to become familiar with all the OSD markers and UI in general but persevere and get to better know your instrument then you can get the result you seek.

Looking forward to seeing your future work.
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 
The following users thanked this post: egonotto

Offline watchmakerTopic starter

  • Frequent Contributor
  • **
  • Posts: 612
  • Country: us
  • Self Study in EE
    • Precision Timepiece Restoration and Service
Re: Use of SDS1204XE as Curve Tracer (Tauteg??)
« Reply #4 on: October 10, 2023, 01:49:43 am »
Tautech,

My apologies, I have no idea how that got into my head. I presume you know it was not intentional.

Goddamn it.  I had taken the time to write a reply and I got timed out.  Also, this message system has no undo button !

Moving on.  Ya larned me some stuff!!  Thank you.

It seems we have a little in common.  We both like to clear away the bullshit.

It looks to me like you get joy from helping others learn.  That has been the focus of my website since 1999 (Front Page!!), although not updated sine we moved to New Hampshire.

That site (www.historictimekeepers.com) is the reason I got invited to a year in Switz at WOSTEP at the age of 56.  I try to give the "kids" some hints about what to avoid and various ways to accomplish the job.  I even have some hidden documents  that I share only with those who are intently interested.  As most might guess, "watchbreakers" are not particularly knowlegeable about their craft, let alone first principles.  I even have one document l that describes watch precision in terms of "Q".  You might be shocked to know many watchbreakers do not know their craft, let alone first principles. |O

I do appreciate you taking the time to help me with this.  I am not one who feels "entitled" to the knowledge of others.  Sharing that knowledge is a gift to the recipient.

OK.  I am 70 and not feeling challenged by my craft. I am starting to get hand tremors and know the time I have left is limited.  So, SMDs are out.

I decided to create a course of self study in EE.  Art of Electronics, USN NEETS (which I was taught in HS before they eliminated circuit analysis), Diligent You Tube and Kahn Academy (mainly for calculus at this point).

Because of this group, I avoided some serious mistakes in building my learning lab.  I have built my lab around an Instek GPS 4303, the Discovery, three Fluke 8040a's and HP 3478a that all agree, the Siglent scope for under $1K including my Amazon wall power strips (learned not to screw around after 40 years of evaluating useless tools in watchmaking).  I also have a Lambda LK 351-FMOV that has served me well over the years but is now best used for chest presses.

In the process, I have been shocked by how the world has changed.  Online compilers for Fortran,Cobol,and Python! When did THAT happen?  I remember punch tape, punch cards, and Word Star!  Budgeting for computer time.  I cannot believe the equipment today compared to 1970 when the Simpson 360 was the"gotta have".

So, I AM serious and appreciate your help.

I think your advice enabled me to get useable information off the Siglent when used as a CT.

But I still have two questions.  One is the thickness of the lines in the curves.  Is this a resolution issue?  I understand this is a beginner's scope.

Why the loss of screen real estate?

I am still confused about Sa/sec/Mpts/KPI.  Is there a place I can read up on them?  Seems like it might be useful to understand.

Below is screen print (thank you) of the trace that seems to work.  I can measure with the cursors.  BTW, I was not worried about the relative position of the 0,0 because all measurement is relative.  And I was lining up things to interpret the graticule.

Is there a way to improve the resolution short of replacing the scope I just bought?

Any thoughts on the limitations of the Discovery Analog 2?  Seems to solve the AWG and digital measurement issues.

Regards,

Dewey1895424-0






Regards,

Dewey
 
The following users thanked this post: tautech

Offline tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Use of SDS1204XE as Curve Tracer (Tauteg??)
« Reply #5 on: October 10, 2023, 02:26:28 am »
Tautech,

I think your advice enabled me to get useable information off the Siglent when used as a CT.
Much better.  :clap:

Quote
But I still have two questions.  One is the thickness of the lines in the curves.  Is this a resolution issue?  I understand this is a beginner's scope.
Hardly. We have sold many to real EE's doing development work on all manner of retail projects for the NZ market.

Engage the 20 MHz BW limiter for each channel, it's in each channel menu.  ;)
These are seriously sensitive DSO's and when trace noise appears sometimes you need suppress it with a BW limiter. Here the scope is no way at fault and only demonstrating its high sensitivity which as a competent operator is no issue to manage.
Quote
Why the loss of screen real estate?
Scaling. Notice how the XY graticules are square instead of rectangular in scope mode.
IMO this is a most correct XY display especially when both inputs are using the same V/div setting.
For something like you are doing it could be seen as a disadvantage but for most XY work it is not.

Quote
I am still confused about Sa/sec/Mpts/KPI.  Is there a place I can read up on them?  Seems like it might be useful to understand.
In the Acquire menu you can change settings for faster performance by reducing mem depth when/if needed however recommend you leave it set to max and let the DSO manage it for you as mem depth changes with s/div settings to keep the scope responsive and not need to buffer max data points.

Quote
I can measure with the cursors.  BTW, I was not worried about the relative position of the 0,0 because all measurement is relative.  And I was lining up things to interpret the graticule.
Yep that's how us oldies that learnt on CRO's do things and only engage measurements when we really need them.  :phew:
Quote
Is there a way to improve the resolution short of replacing the scope I just bought?
Answered above.

Quote
Any thoughts on the limitations of the Discovery Analog 2?  Seems to solve the AWG and digital measurement issues.
Yes well it has a dedicated CT mode for those that struggle doing it with a DSO and AWG.
Really study the great 2 threads Mike gave us as they shine a light on the NOT complexities of how a CT works behind the scenes. Granted higher levels than an AWG can provide are entirely another business however his 2nd thread addresses this with a custom fixture.

Have fun.
« Last Edit: October 10, 2023, 02:28:21 am by tautech »
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline Performa01

  • Super Contributor
  • ***
  • Posts: 1729
  • Country: at
Re: Use of SDS1204XE as Curve Tracer (Tauteg??)
« Reply #6 on: October 10, 2023, 05:06:31 pm »
Might it be that you're running a prehistoric firmware? The thick and ugly x/y traces remind me on the very first FW versions back in 2017, before Siglent did the big re-design of x/y and introduced HW support for this feature?

What strikes me most is the deep memory, which slows down y/y operation considerably. This might come from a wrong setup.
The first thing you need to do is setting up the two input channels correctly in y/t mode. Not only the vertical gain, but also the horizontal time base. The latter should be set in a way that one (and only one!) full signal (scan in this context) period fits on the screen.

This way, the record length should get much shorter than 7 Mpts. Tis is done by shorteing the time base. Should the record length get tooo long even with the correct timebase setting, you can always limit the max. record length in the "Acquire" menu. Don't use more than 70 (better 7) kpts in x/y mode unless you absolutely must because of the complex signal. But the fact that you can display it on the Analog discovery proves that it cannot be that complex.
 

Offline watchmakerTopic starter

  • Frequent Contributor
  • **
  • Posts: 612
  • Country: us
  • Self Study in EE
    • Precision Timepiece Restoration and Service
Re: Use of SDS1204XE as Curve Tracer (Tauteg??)
« Reply #7 on: October 10, 2023, 09:26:42 pm »
Like this?1896306-0

Just to be sure, I updated my FW (6.1.37.R10) and OS=8.3.  Unit was just acquired from Saelig last month.

Also, is there a way to have in the 0,0 set into the lower left so when I push the position buttons that v=0 h=0ma is on the lower left corner?

THANKS.  Tautech, Do you agree?
Regards,

Dewey
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Use of SDS1204XE as Curve Tracer (Tauteg??)
« Reply #8 on: October 10, 2023, 11:44:44 pm »
Just to be sure, I updated my FW (6.1.37.R10) and OS=8.3.  Unit was just acquired from Saelig last month.
Latest FW shouldn't change anything for XY mode, only on very early FW releases.

Quote
Also, is there a way to have in the 0,0 set into the lower left so when I push the position buttons that v=0 h=0ma is on the lower left corner?
Mid XY display is the default 0V channel position and V/div settings set the channel scaling for the waveform to fit the XY display.
Use the channel position control to have the XY chart looking like you want.

Obviously less memory depth used with a faster timebase cleans up the traces same and the channel BW limiter will help further.
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline watchmakerTopic starter

  • Frequent Contributor
  • **
  • Posts: 612
  • Country: us
  • Self Study in EE
    • Precision Timepiece Restoration and Service
Re: Use of SDS1204XE as Curve Tracer (Tauteg??)
« Reply #9 on: October 11, 2023, 12:31:09 am »
Thank  you both.  It is clear I was overdriving the display.  Sometime more is not better!
Regards,

Dewey
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf