Author Topic: Siglent Technical Support join in EEVblog  (Read 288854 times)

danshtr and 1 Guest are viewing this topic.

Online colorado.rob

  • Regular Contributor
  • *
  • Posts: 241
  • Country: us
Re: Siglent Technical Support join in EEVblog
« Reply #750 on: June 26, 2017, 06:19:19 pm »
One issue I had on the analog side (when using only analog input) was that the trigger indicator would go missing and I could not bring it up on the screen.  It would tell me what the trigger level was, and I could see the effect of changing it on the trace, but the dotted bar would not show up.  Is this a known issue?  I wish I knew how to reproduce it but it "just happened" a couple times. Turning off the scope would make it work again.  Is this a known issue?

If remember correctly, trigger cursor disappears when AC-coupled. This was quite annoying. It still indicates on the screen current trigger voltage.

That would make sense as that is something I was doing at the time.
 

Offline sng61

  • Contributor
  • Posts: 5
  • Country: ru
Re: Siglent Technical Support join in EEVblog
« Reply #751 on: June 27, 2017, 03:58:30 pm »
Hello.
I recently purchased model SDS 1202X-E. Very fast and functional. But very uncomfortable interface. Showing only 4 measurements, and very small font size. Before that, I used the oscilloscope Hantek DSO 5102B in my work, and there was a very convenient display of all measurements, 8 of them at once - and all in large font. I propose to make the font of the numbers more 2 times, and the name of the parameter itself - to leave small. So it will be much clearer to read and it is more convenient to work with the device.

For comparison: (Hantek 5102B and Siglent 1202X-E):



And one more drawback - if there is no measurement of some parameter, then do not show three stars ("***"), but just leave an empty space. So there will be less unnecessary meltesheniya on the screen, which without it is well loaded with information.

A big request in future firmware is to fix this small - but very unpleasant lack of such an excellent device.
« Last Edit: July 02, 2017, 05:46:57 pm by sng61 »
 
The following users thanked this post: boggis the cat

Online colorado.rob

  • Regular Contributor
  • *
  • Posts: 241
  • Country: us
Re: Siglent Technical Support join in EEVblog
« Reply #752 on: June 28, 2017, 01:24:20 am »
One issue I had on the analog side (when using only analog input) was that the trigger indicator would go missing and I could not bring it up on the screen.  It would tell me what the trigger level was, and I could see the effect of changing it on the trace, but the dotted bar would not show up.  Is this a known issue?  I wish I knew how to reproduce it but it "just happened" a couple times. Turning off the scope would make it work again.  Is this a known issue?

If remember correctly, trigger cursor disappears when AC-coupled. This was quite annoying. It still indicates on the screen current trigger voltage.

That would make sense as that is something I was doing at the time.

Confirmed.  When AC coupling or LF Reject is selected, the trigger cursor disappears.
 

Offline rfspezi

  • Regular Contributor
  • *
  • Posts: 111
  • Country: 00
Re: Siglent Technical Support join in EEVblog
« Reply #753 on: June 30, 2017, 07:34:00 pm »
On my SDS1202X scope i got offset jumps when switching between different voltage ranges and time bases.
For example on channel 1 using 200 mV/div, when crossing the timebase boarder from 5us->2us using a single channel and from 2us->1us using both channels.
I did the self calibration several times without any change.

Here are some examples of my offset values:
(In every vertical scale i stepped through all time-bases and took the worse looking one)
----------------------
-9 mV @ 5 mV/div
-10 mV @ 10 mV/div
-7 mV @ 20 mV/div
-8 mV @ 50 mV/div
-8 mV @ 100 mV/div
+40 mV @ 200 mV/div
+40 mV @ 500 mV/div
-120 mV @ 1 V/div
-320 mV @ 2 V/div
+400 mV @ 5 V/div
-800 mV @ 10 V/div
-4000 mV @ 20V/div
+4000 mV @ 50V/div
-8000 mV @ 100V/div

Are these offsets normal?
In the datasheet on page 6 they state:

Quote
Offset Accuracy:
    ±(1%* Offset+1.5%*8*div+2 mV): ?2 mV/div
    ±(1%* Offset+1.5%*8*div+500 uV): ?1 mv/div

Can anybody give me an example how to calculate the maximum offset error for e.g. 200 mV/div?
Many thanks!
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 15952
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: Siglent Technical Support join in EEVblog
« Reply #754 on: June 30, 2017, 08:01:16 pm »
Please check you have the latest FW installed, it came out in April and addresses Self Cal optimisation.
http://siglentamerica.com/gjjrj-xq.aspx?id=5203&tid=15

Come back and tell us if it's better or not.
Avid Rabid Hobbyist
 

Offline Paul Moir

  • Frequent Contributor
  • **
  • Posts: 865
  • Country: ca
Re: Siglent Technical Support join in EEVblog
« Reply #755 on: July 01, 2017, 06:56:07 am »
rfspezi, I think I had a similar problem with mine.  It went away after I did a self cal after letting it warm up for over 30 minutes.  Previously I had done a self cal after only letting it warm up for a few minutes.
There seems to be something odd going on during the warmup.  If I crank the gain up at startup, I'm seeing the traces jump about 1/3rd a division in distinct steps every 10 or 20 seconds, starting around div and a half out.  After about 30 minutes they settle down to bang on 0v.  This isn't an analog drift but a very distinct jump to a very precise position.  Sometimes closer to zero but sometimes away, but always eventually ending up at 0.  I've tried it with terminators and it does the same thing.   I suspect it is some sort of warmup compensation routine that actually doesn't do any good.
 

Offline PA4TIM

  • Super Contributor
  • ***
  • Posts: 1116
  • Country: nl
  • instruments are like rabbits, they multiply fast
    • PA4TIMs shelter for orphan measurement stuff
Re: Siglent Technical Support join in EEVblog
« Reply #756 on: July 01, 2017, 08:07:02 am »
I made 2 reviews about the SHS-1062 number 1 https://youtu.be/R5lTLuclk1M
While doing that I found a bug. In scope mode when I use the run/stop the trace disapears after a short while (I noticed that a few days after the review while using it)

Yesterday I powered it up while it was in in log mode. I pressed on scope and the 2 modes "merged"  I had to repower it to go away.

The accuracy is not bad but the RMS value in scope mode is a bit off, but only at lower voltages

Continuity sucks but that is not a bug. The DMM IC is made to use a LCD display and I think they use the rs232 to talk to the rest of the scope. Maybe that is to slow. It takes several seconds to beeb

The menu structure can use a big update. Sometimes you need the menu button to select, sometimes the "right arrow" etc

The inverse color can use better combinations to make it better readable (more contrast instead lightgreen, yellow)

But beside this faults I really like this scope, a pity it has to go back to the dealer next week.
« Last Edit: July 01, 2017, 11:59:08 am by PA4TIM »
www.pa4tim.nl my collection measurement gear and experiments Also lots of info about network analyse
www.schneiderelectronicsrepair.nl  repair of test and calibration equipment
https://www.youtube.com/user/pa4tim my youtube channel
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 15952
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: Siglent Technical Support join in EEVblog
« Reply #757 on: July 01, 2017, 08:18:48 am »
I made 2 reviews about the SHS-1062 number 1 (YT link)
While doing that I find a bug. In scope mode when I use the run/stop the trace dissapears after a short while (I noticed that a few days after the review while using it)

Yesterday I powered it up while it was in in log mode. I pressed one scope and the 2 modes "merged"  I had to repower it to go away.

The accuracy is not bad but the RMS value in scope mode is a bit off, but only at lower voltages

Continuity swucks but that is not a bug. The DMM IC is made to use a LCD display and I think they use ther rs232 to talk to the rest of the scope. Mabey that is to slow. It takes several seconds to beeb

The menu structure can use a big update. Sometimes you need the menubutton to select, sometimes the " richt arrow" etc

The inverse color can use better coninations to make it better readable (more contrast lightgreen, yellow)

But beside this faults I really like this scope, a pitty it has to go back to the dealer.
Thanks for the vids, watched a few days ago and forwarded to Siglent.
Yep, Continuity is bad, painfully slow and Siglent are well aware of comments on this going back some time.

Thanks for reporting the bugs, I'll forward your post to Siglent right now.
Avid Rabid Hobbyist
 

Offline DazA1

  • Contributor
  • Posts: 46
Re: Siglent Technical Support join in EEVblog
« Reply #758 on: July 02, 2017, 02:19:54 pm »
@tautech

Did you have time to look at the problems i and others have had on your own SHS806?

Do you know if these problems are being worked on?
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 15952
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: Siglent Technical Support join in EEVblog
« Reply #759 on: July 03, 2017, 09:46:41 am »
@tautech

Did you have time to look at the problems i and others have had on your own SHS806?

Do you know if these problems are being worked on?
I know Siglent see all that's in this thread and more that I and others point them to.
TBH I think the SHS FW needs some freshening up as HH's are gaining in popularity, not only for portability but isolation from mains ground. The FW we know today is much the same as when I first sold Siglent and needs smartening up some to be competitive.
What they will do I have no idea but they have in the past reworked FW like in the SDS2000 to end up with a better  GUI.
By all means put some suggestions forward.
Avid Rabid Hobbyist
 

Online colorado.rob

  • Regular Contributor
  • *
  • Posts: 241
  • Country: us
Re: Siglent Technical Support join in EEVblog
« Reply #760 on: July 06, 2017, 03:23:01 pm »
I own a growing collection of Siglent test equipment.  Two of the items that I have are the SDM3065X and the SDG2042X.  They appear to use the same (or very similar) rubber bumpers on the front and back.  These bumpers have a bump on one side (top or bottom) and an indent on the opposite side into which the bump will fit.  This allows gear to be securely stacked on top of one another.  Or, rather, it would if Siglent bothered to always install the bumpers so that the bumps (feet) were down and the indents were up.  However, on the two items mentioned, they are opposite one another.

Siglent, please standardize the installation of the bumpers, preferrably with the bumps facing down on all of your equipment.

Also, please consider securing the nuts on the inside of the case.  I swapped a pair of bumpers on the SDG so the two would fit as described and had a hard time getting a couple of the nuts to stay put while I put it back together, and almost lost one inside the gap between the case and faceplate.

One other item: consider adding graphing options to the frequency counter on the SDG2000X series.
 
The following users thanked this post: tautech

Offline tautech

  • Super Contributor
  • ***
  • Posts: 15952
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: Siglent Technical Support join in EEVblog
« Reply #761 on: July 07, 2017, 08:47:19 am »
Siglent, please standardize the installation of the bumpers, preferrably with the bumps facing down on all of your equipment.
They have and I knew this too.  :palm:

From factory tech support: (abbreviated)
In the last year or so the bumpers have been standardized but you have got models from pre and post standardization.

IIRC a later model I looked at did have captured nuts so to overcome the problem you describe.
Avid Rabid Hobbyist
 

Offline N8AUM

  • Regular Contributor
  • *
  • Posts: 130
  • Country: us
Re: Siglent Technical Support join in EEVblog
« Reply #762 on: July 07, 2017, 06:19:26 pm »
Probable bug

Setup to recreate:
  • No signal applied
  • [Default]
  • Ch 1 [V/div] to 500 uV/div (must choose a low setting, as hard to notice at 2 mV/div and greater)
  • [Acquire] > Average 16
  • [Measure] > Clear, then add Type > Mean
  • [Menu On/Off]
  • [Clear Sweeps]


Note 'spike' at zero on timebase

[Trigger Setup]
Change from Edge, back through to Edge (not important what is selected, or even if a different trigger type is selected)


Spurious 'spike' is now gone.


Question about this behaviour

After pressing [Run/Stop] to halt acquisition, the display seems to change to the last waveform, rather than maintain the averaged waveform.


While Running


After Stop

This seems to not be what you'd want to occur.  Is this a bug?

This is normal, it is triggering on noise
 

Offline ebastler

  • Super Contributor
  • ***
  • Posts: 2976
  • Country: de
Re: Siglent Technical Support join in EEVblog
« Reply #763 on: July 07, 2017, 06:24:45 pm »
Probable bug
This is normal, it is triggering on noise

You are wasting your time there, I'm afraid. If you follow the thread along further, you will notice that "boggis" has been quite impervious to that explanation.
 

Offline N8AUM

  • Regular Contributor
  • *
  • Posts: 130
  • Country: us
Re: Siglent Technical Support join in EEVblog
« Reply #764 on: July 07, 2017, 06:50:06 pm »
Yet another possible "BUG" or is this some special feature lol
I just purchased a SSA 3032X and noticed this. I my brother has has the SSA 3021X and I  asked him if he can duplicate my findings which he did. While measuring any signal around -50dbm with RBW at 300Hz, span around 50K and having MIN HOLD enabled I get the following display. Am I missing something, none of my other SA's do that. I also upgraded firmware from V1.2.8.2. to V1.2.8.3.

Any ideas ?

N8AUM  Vidas

   
« Last Edit: July 07, 2017, 06:52:56 pm by N8AUM »
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3068
  • Country: fi
  • Starting with DLL21
Re: Siglent Technical Support join in EEVblog
« Reply #765 on: July 08, 2017, 07:40:22 pm »
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 
The following users thanked this post: ted572, N8AUM

Offline boggis the cat

  • Regular Contributor
  • *
  • Posts: 150
  • Country: nz
Re: Siglent Technical Support join in EEVblog
« Reply #766 on: July 09, 2017, 06:54:28 am »
This is normal, it is triggering on noise
There is a change in behaviour.  The 'new' behaviour then sticks, and is what I consider correct.

So no, it isn't "normal".  Either the initial behaviour or the subsequent (different) behaviour may be normal, but they can't both be.


Edit:

Retesting this seems to indicate that it is only real noise sources (e.g. shorted or open terminals) that produce the aberrant triggering.  It is also progressively less pronounced as you move to longer timebases.

Using a pseudo-noise source (multiple modulated sine waves, from a Feeltech Arb device) or a repetitive signal yields the expected triggering behaviour, after "Default" settings.  No fiddling with the trigger menus is necessary.

In practical terms, this isn't likely to be particularly significant.


I did also try to reproduce the other triggering issues reported, and could not get the issues described -- although the triggering behaviour does seem a bit marginal in some cases.  There is a significant 'dead band' of around half a division where edge triggering doesn't seem to work (checked using sinusoidal and similar repetitive waveforms, exists at the the base of the signal for positive edge and vice versa), and the line indicating the trigger level does indeed vanish under certain circumstances.  So there are some behaviours that are not ideal that could be looked into, as well as the trigger indication bug to be fixed.
« Last Edit: July 10, 2017, 01:43:55 am by boggis the cat »
 

Offline KeBeNe

  • Contributor
  • Posts: 34
Re: Siglent Technical Support join in EEVblog
« Reply #767 on: July 09, 2017, 08:26:51 am »
Yet another possible "BUG" or is this some special feature lol
I just purchased a SSA 3032X and noticed this. I my brother has has the SSA 3021X and I  asked him if he can duplicate my findings which he did. While measuring any signal around -50dbm with RBW at 300Hz, span around 50K and having MIN HOLD enabled I get the following display. Am I missing something, none of my other SA's do that. I also upgraded firmware from V1.2.8.2. to V1.2.8.3.

Any ideas ?

N8AUM  Vidas

the same here (3021X + 3032X FW 1.2.8.3.)
« Last Edit: July 09, 2017, 08:50:20 am by KeBeNe »
 

Online colorado.rob

  • Regular Contributor
  • *
  • Posts: 241
  • Country: us
Re: Siglent Technical Support join in EEVblog
« Reply #768 on: July 10, 2017, 12:49:09 am »
I am looking at Siglent's generic Programming Guide For Digital Oscilloscopes Series (warning: PDF link) that is available from the SDS2000X documentation section and from the global documentation area on Siglent America's web site. I do not see any instructions for accessing the AWG functionality.  Nor do I see any commands documented for the logic analyzer portion of the device. Please provide an update for this.

My immediate quest is to get a screen dump over VXI-11/SCPI .  To that end, I am trying to understand the "SCREEN_DUMP" and related "HARDCOPY_SETUP" commands.  But SCDP just seems to hang the VXI-11 session when using PyVISA and PyVISA-py.  Screen capture seems to work for EasyScopeX, so there must be some way to do this.  Based on a network capture, this should work:

Code: (python) [Select]
#!/usr/bin/env python

import visa, sys
from pyvisa.constants import StatusCode

rm = visa.ResourceManager('@py')

 # Siglent SDS2204X
scope = rm.open_resource('TCPIP::sds2204x')
sys.stdout.write(scope.query("*IDN-SGLT-PRI?"))
sys.stdout.write(scope.query("*CHS?"))
n, status = scope.write('SCDP')
if status == StatusCode.success:
    print("Success...")
    data = scope.read_raw()
    sys.stdout.write(data)
scope.close()
rm.close()

Unfortunately, it just hangs at "scope.read_raw()".  Running the same code on my Macbook Pro using NI VISA drivers results in this error in the same place:

Code: [Select]
pyvisa.errors.VisaIOError: VI_ERROR_INV_PROT (-1073807239): The protocol specified is invalid.


Siglent,

1. How do I download a screendump image using SCPI from an SDS2204X?  Please docurment the SCDP command adequately.
2. What is the format of WF parameter "DESC"?
3. How do I access the digital channel data via SCPI of this MSO?
4. How do I control the scope's AWG via SCPI?  This is apparently possible since EasyWave can talk to it.
5. How do I read frames from the segmented buffer via SCPI?
6. Shouldn't the "MSIZ?" query return "MSIZ 140M" on the SDS2204X?  It returns "MSIZ 14M".
7. You'll notice that I am using two undocumented SCPI commands in my code.  How about documenting these for those of us playing along at home?

The SCPI command interface for this scope is woefully inadequate.  Look at Rigol's SCPI documentation for comparable scopes: a well thought-out command structure, and detailed documentation.  Siglent's SCPI support and documentation seem rather amateurish in comparison.  Please provide SCPI controls for the AWG and digital inputs.  And please update the documentation.

Rob
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 15952
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: Siglent Technical Support join in EEVblog
« Reply #769 on: July 10, 2017, 01:05:58 am »
There's a later version of the Programming Guide here:
http://siglentamerica.com/support_download_28

I don't know (haven't checked) if there's more info in it to help you solve your issues.

Note, the User manual is very recent too, maybe more recent than what you've been using.
Avid Rabid Hobbyist
 
The following users thanked this post: colorado.rob

Online colorado.rob

  • Regular Contributor
  • *
  • Posts: 241
  • Country: us
Re: Siglent Technical Support join in EEVblog
« Reply #770 on: July 10, 2017, 02:55:34 am »
I'll have to read through this.  At first glance, it doesn't appear that any of my issues have been addressed.  The SCDP command is still completely undocumented.  I see nothing for the AWG or digital channels.

That said, I have SCDP working with PyVISA-py now.  It still fails with the NI-VISA drivers on OSX.

https://gist.github.com/mobilinkd/8a07cc124946c87715c6a1458118411e

This will do a screendump over the network from a Siglent DSO.  It should be easily adapted to use USB.  I would like to get this to the point that it automatically transcodes the image to a reasonable format like PNG.
 

Online colorado.rob

  • Regular Contributor
  • *
  • Posts: 241
  • Country: us
Re: Siglent Technical Support join in EEVblog
« Reply #771 on: July 12, 2017, 12:59:27 am »
I am trying to measure two audio waveforms on CH1 and CH2 on my SDS2204X.  CH1 is a filtered and amplified version of CH2.  I have the scaling of the two channels adjusted so that the waveforms are superimposed on one another.  I have the trigger mode set to Auto.  Acquisition mode is Eres @ 1.5 bits.  When the scope is in Run mode, the two waveforms are closely correlated as expected.  When I press Stop, the waveforms are uncorrelated.  The two channels do not appear to be stopping at the same time.

If I put the trigger mode to Normal and put the trigger level near the limits of the signal so that it is not triggering frequently, the waveform is always correlated when triggered, but I can still get an uncorrelated waveform by pressing Stop if more than one trigger event occurs in rapid succession.

This is driving me nuts. This seems like a bug to me.  Am I doing something wrong?  Have I missed some subtle detail in the User Guide?  Or do I have no right to expect that Stop will stop both channels at the same time?

* Addendum: this only seems to happen in Eres mode.
« Last Edit: July 12, 2017, 01:03:35 am by colorado.rob »
 
The following users thanked this post: tautech

Offline tautech

  • Super Contributor
  • ***
  • Posts: 15952
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: Siglent Technical Support join in EEVblog
« Reply #772 on: July 12, 2017, 01:31:46 am »
C.rob
I have forwarded your last few posts to the factory and some to the USA that best fits your timezone, I will send reminders and this one.............
T

Reply from the US: (travelling ATM)
Please ask him to write us at
info@siglent.com

« Last Edit: July 12, 2017, 02:04:48 am by tautech »
Avid Rabid Hobbyist
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3068
  • Country: fi
  • Starting with DLL21
Re: Siglent Technical Support join in EEVblog
« Reply #773 on: July 12, 2017, 05:01:00 am »

There is a significant 'dead band' of around half a division where edge triggering doesn't seem to work (checked using sinusoidal and similar repetitive waveforms, exists at the the base of the signal for positive edge and vice versa),

Please can you define it more accurate. Do you have any  documented and trusted data for this?

Do you perhaps mean that trigger hysteresis is bad thing? There is trigger level hysteresis - and there need be.  I believe that mostly peoples in most situations do not want zero hysteresis for edge trigger.



This oscilloscope do not have freely adjustable trigger hysteresis setting(1) as example in some Rohde&Schwarz models. Siglent have selected some average good compromise for this.

Normally there is around <0.3 vertical div hysteresis  "window". When trigger is DC coupled, as usually is, and if edge trigger is rising edge, hysteresis treshold is below triger setting level and "trig" treshold is at trigger level. If falling edge is selected trigger hysteresis treshold is overside trig level setting and final "trig" treshold is trigger level. For trigged acquisition signal need cross both these treshold. After then waveform position is fine adjusted to trig time position. Fine positioning have  much more resolution than sampling interval due to "oversampling" - fine interpolation.

(1)There is one selection what change trigger hysteresis. In trigger set menu there is  "Noise reject" on/off. Normally (default) it is off. If user turn this on, trigger hysteresis is more wide. Around nearly 0.8 div. (upper or lower side of trigger level set, depending if rising or falling edge is selected.)

Benefits of R&S RTO digital trigger system (pdf)
Page 10 about trigger hysteresis. Also whope paper is good to read for better understanding digital trigger system. Albeit this Siglent model is lot of more simple machine but still it follow these basic principles.

Knowing how equipment really works is very extremely important for avoid misconceptions and false claims.

Previously in this thread was also other total mess where signal was noise and wondering how it show peak in trigger position. Jus due to lack of enough knowledge and experience about how these things work even in basic entry level things.






Then:
Quote
... and the line indicating the trigger level does indeed vanish under certain circumstances.

Can you define it with facts anbd details (data! instead of opinions). And the reasons why it should be different, taking into account all the relevant factors.



Here some simplified case what perhaps make this trigger hysteresis more clear.
Only show Edge and rising. Falling edge trigger: this hyst is also just opposite direction.

Cursors are used here only for show trigger hysteresis treshold levels. (hysteresis window borders)

In first image trigger looks bistable because there narrow hystereis but signal have this sawtooth "noise" and trig conditions are met more than one places.

Second image have wide trigger hysteresis and now signal loogs rock solid on the screen.

Images have adjusted for clarify this thing, not for show "nice looking image". This is also why signal is extremely simply.
For better show more rare trigger position (exist only in first image) there is used 5 second persistence in both images.


« Last Edit: July 12, 2017, 10:02:20 am by rf-loop »
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 

Offline boggis the cat

  • Regular Contributor
  • *
  • Posts: 150
  • Country: nz
Re: Siglent Technical Support join in EEVblog
« Reply #774 on: July 13, 2017, 01:11:00 pm »
There is a significant 'dead band' of around half a division where edge triggering doesn't seem to work (checked using sinusoidal and similar repetitive waveforms, exists at the the base of the signal for positive edge and vice versa),

Please can you define it more accurate. Do you have any  documented and trusted data for this?
No, not using the cheap Feeltech Arb I have at home.  I only checked up to 5 Vpp, too.  The edge trigger drop-out may improve at higher V/div settings.

This is just general observations of behaviour.  Finding the exact points would require better signals, and be a bit pointless as there is no specification.

Quote
Do you perhaps mean that trigger hysteresis is bad thing? There is trigger level hysteresis - and there need be.  I believe that mostly peoples in most situations do not want zero hysteresis for edge trigger.
The algorithm used does have to ensure that a stable edge has been found.  I would think that three data points would be sufficient to establish that, at a minimum.

Without knowing what the 'scope is expected to do it is difficult to differentiate a bug from intentional behaviour.  However, the broader you make that 'dead band', the less useful the triggering is.

Quote
Also whope paper is good to read for better understanding digital trigger system. Albeit this Siglent model is lot of more simple machine but still it follow these basic principles.
It is the specific implementation that is the issue, not the principle.  We don't know much about this.  The specifications are quite sparse.  (This isn't documented in the manner of a high-end Keysight or Tektronix 'scope.)

Referring to a white paper from another manufacturer doesn't tell you anything about Siglent's implementation.

Quote
Knowing how equipment really works is very extremely important for avoid misconceptions and false claims.
You have this backwards.  Testing the equipment yields how it "really works".  If we don't know how it is supposed to work (insufficient documentation), then we cannot tell if it meets its specifications.

Quote
Previously in this thread was also other total mess where signal was noise and wondering how it show peak in trigger position. Jus due to lack of enough knowledge and experience about how these things work even in basic entry level things.
The behaviour I noted isn't correct.  Just because some Rigol 'scopes also behave oddly in a similar manner doesn't make the result correct (and if I recall correctly, Siglent is the older company, so I would guess that the algorithms got 'borrowed' by Rigol) -- you have to consider what the ideal is, and if the result departs from that in a systematic and aberrant way.

I noted that the pseudo-'noise' from the Feeltech Arb I have is triggered on as expected -- all of the sinusoidal-based waveforms are centred as expected, with either rising or falling edge set.  Real noise, such as amplified open input (setting the V/div to small values), produces some odd 'stacking' effects.  In some circumstances, initial results will change by simply opening the trigger menu.

Given that triggering consistently and predictably on noise is not a practical concern (we use 'scopes to inspect signals, where defined trigger points are important), this may not matter for usage.  It may indicate problems or limitations in the algorithms used, however, so noting such behaviour might be of some use to those writing the algorithms.

Quote
Then:
Quote
... and the line indicating the trigger level does indeed vanish under certain circumstances.

Can you define it with facts anbd details (data! instead of opinions). And the reasons why it should be different, taking into account all the relevant factors.
When you adjust the trigger level there is a horizontal line to indicate where that level is, along with the numeric indication.  But that line will disappear under some circumstances.  Also, it seems odd that if you change from DC to AC coupling the trigger level is still settable, but only the numeric indication is present (and persistent, it doesn't snap back to zero).

There are a few odd bugs of this type, and other people have already documented them fairly well here.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf