Siglent SDS800X HD 12 bit DSO's

[IC_Toaster]

I have bought a Siglent SDS804x HD oscilloscope some days ago. I feel a little lost in the using of digital oscilloscopes and I have a doubt whose answer is easy for sure.
In the attached image we can see that the signal frequency is 999.77 Hz (1 KHz theoretically) but on the upper side of the screen, with yellow characters, is shown a frequency of 2.08 KHz.
Sometimes the trigger fails to detect the signal and the signal shifts horizontally.
Why this difference in the measure and how is the yellow value obtained?

[electr_peter]

Yellow freq measurement is from HW counter. Signal looks noisy, counter likely picks up more some of that and gives higher reading. Depends on situation if HW or SW freq is more useful, but in general HW counter is nice to have.

From 800X HD user manual:

18.12 Hardware Frequency Counter
The hardware frequency counter is a stand alone, hardware based frequency measurement function. Differing from the “Frequency” parameter (see section “ Horizontal Measurement ”) which is measured by software, the hardware frequency counter directly use s the trigger signal as the input of a hardware counter, which yields 7 digit frequency measurement accuracy. The accuracy of the “Frequency” parameter measured by software is relative to the sample rate. The lower the sample rate, the worse the accuracy. In situations where the sample rate is less than 2 times the input frequency, the results will be incorrect . The hardware frequency counter does not have this limitation. Its result is always

[IC_Toaster]

Yellow freq measurement is from HW counter. Signal looks noisy, counter likely picks up more some of that and gives higher reading. Depends on situation if HW or SW freq is more useful, but in general HW counter is nice to have.

From 800X HD user manual:

18.12 Hardware Frequency Counter
The hardware frequency counter is a stand alone, hardware based frequency measurement function. Differing from the “Frequency” parameter (see section “ Horizontal Measurement ”) which is measured by software, the hardware frequency counter directly use s the trigger signal as the input of a hardware counter, which yields 7 digit frequency measurement accuracy. The accuracy of the “Frequency” parameter measured by software is relative to the sample rate. The lower the sample rate, the worse the accuracy. In situations where the sample rate is less than 2 times the input frequency, the results will be incorrect . The hardware frequency counter does not have this limitation. Its result is always

The signal is noisy and probably this is the reason because the frequency measurement has desviations. Just I try to understand in which way the noise affect both measurement of the same parameter.
As I understand from the user manual, the counter depends on the trigger signal as you quoted, and the frequency parameter depends on the edges detected:

Period: Time between the middle threshold points of two consecutive like-polarity edges
Freq: Reciprocal of the period

If the counter obtains a value which is the double from the true frequency, does it means that the trigger is triggered twice on a period?

[2N3055]

Page 88 of User manual explains how to set trigger Noise reject and HF reject.
Play with these two options to get more stable trigger.
If you get it hardware counter will show more accurate reading.

[markus_jlrb]

BTW,

why we had the sig dissymmetry in the screen image?

First thought would be, because we had a DC offset and the DC coupling enabled,
but perhaps after doing AutoSetup the sig become symmetrical.

Could you verify this on your new scope please.

I made same observation on mine device, see attached screen shots.

Markus

 

[2N3055]

People, this scopes have axis labels.
Go into Display Menu-> Axis label Setting.
I use Moving labels, so numbers follow position on screen
It cuts down on confusion.

@ markus_jlrb, IC_Toaster has 1 V of offset set on channel 1. And on that image trigger is set for 1.66V.

[IC_Toaster]

Page 88 of User manual explains how to set trigger Noise reject and HF reject.
Play with these two options to get more stable trigger.
If you get it hardware counter will show more accurate reading.

Both options gave me a stable adquisition and a reliable frequency measurement.
Thank you for your advise.

The steep from analog scilloscopes to digital one is big.

[2N3055]

Page 88 of User manual explains how to set trigger Noise reject and HF reject.
Play with these two options to get more stable trigger.
If you get it hardware counter will show more accurate reading.

Both options gave me a stable adquisition and a reliable frequency measurement.
Thank you for your advise.

The steep from analog scilloscopes to digital one is big.

Glad to be able to help.
Yes, you are right, it needs some learning and getting used to, but it is so much more capable...
Take care!

[IC_Toaster]

BTW,

why we had the sig dissymmetry in the screen image?

First thought would be, because we had a DC offset and the DC coupling enabled,
but perhaps after doing AutoSetup the sig become symmetrical.

Could you verify this on your new scope please.

I made same observation on mine device, see attached screen shots.

Markus

Have you measured the mean value? For sine signals it is the DC value.
What source are you using?
My only signal generator is the Zoyi ZT-703s and this generates a 2.5Vpp sine with 1.25 Vdc..

[IC_Toaster]

People, this scopes have axis labels.
Go into Display Menu-> Axis label Setting.
I use Moving labels, so numbers follow position on screen
It cuts down on confusion.

@ markus_jlrb, IC_Toaster has 1 V of offset set on channel 1. And on that image trigger is set for 1.66V.

You are right, it is possible to activate the axis labels.

[IC_Toaster]

Page 88 of User manual explains how to set trigger Noise reject and HF reject.
Play with these two options to get more stable trigger.
If you get it hardware counter will show more accurate reading.

Both options gave me a stable adquisition and a reliable frequency measurement.
Thank you for your advise.

The steep from analog scilloscopes to digital one is big.

Glad to be able to help.
Yes, you are right, it needs some learning and getting used to, but it is so much more capable...
Take care!

Yes, digital oscilloscopes are much more capable.
Take care you too.

[eTobey]

I found this in the manual:

... the hardware frequency counter directly uses the trigger signal as the input of a hardware
counter ...

I set up a delay trigger, which i checked on the trigger output, which is as expected, at about 4.5Hz. But when looking at the frequency value of the hardware counter it shows 100Hz.
Interestingly, i have a 100Hz signal between the trigger. So either the manual is wrong, the scope has another issue or its an user error.

[2N3055]

I found this in the manual:

... the hardware frequency counter directly uses the trigger signal as the input of a hardware
counter ...

I set up a delay trigger, which i checked on the trigger output, which is as expected, at about 4.5Hz. But when looking at the frequency value of the hardware counter it shows 100Hz.
Interestingly, i have a 100Hz signal between the trigger. So either the manual is wrong, the scope has another issue or its an user error.
(Attachment Link)

Error in presumptions.
Fact that counter uses data from parts of trigger path is not same as trigger using counter circuits.

On beginning of trigger path there is a comparator block that will detect every transition through threshold. Those transitions are fed to hardware counter block and rest of trigger block. Rest of trigger functions then start acquisition if other prerequisites are fulfilled.. That happens asynchronously with comparator block that keeps detecting edges and feeding it to counter block, and rest of trigger functions ignore it until they are ready again.
Trigger out outputs pulses that are synchronous with actual acquisitions.

Note in documentation that counter is connected through parts of trigger circuit is there to point to the fact that counter won't work properly unless you have stable trigger.

[temperance]

I found this in the manual:

... the hardware frequency counter directly uses the trigger signal as the input of a hardware
counter ...

I set up a delay trigger, which i checked on the trigger output, which is as expected, at about 4.5Hz. But when looking at the frequency value of the hardware counter it shows 100Hz.
Interestingly, i have a 100Hz signal between the trigger. So either the manual is wrong, the scope has another issue or its an user error.
(Attachment Link)

One can clearly see a 100 Hz signal on CH2. Unless the scope is making up some signal it must be you feeding that signal into CH2. You are probably seeing a glitch generated a counter from a micro controller or something. Some controllers like an AVR will output a glitch even with the compare value set to 0.

[SHF]

One question, I got my new SDS804-HD today, connected a GPSDO with 10 MHz as a test signal and the counter at the top right shows 10.00015 MHz! Is that within the normal range?
Firmware V1.1.3.6

[Martin72]

Hi,

The accuracy of the time base is +/-25 ppm, so no worries.
How accurate should your GPS be, to put it another way...

[SHF]

Does anyone know what the time base frequency is?
The SDS1104X-E runs at 25MHz
The SDS800X-HD runs at .....?

[ebastler]

Does anyone know what the time base frequency is?
The SDS1104X-E runs at 25MHz
The SDS800X-HD runs at .....?

Here's a post from rf-loop, a Siglent "insider", stating that it's a 25 MHz reference clock as well:
https://www.eevblog.com/forum/testgear/siglent-sds800x-hd-12-bit-dsos-coming/msg5453810/#msg5453810

I seem to recall teardown photos showing the 25 MHz quartz (larger and presumably more stable than a couple of others on the PCB), and also a discussion whether it might be possible to "force" the clock to sync with an external reference. But can't find that thread, and may be confusing it with some other scope model.

[SHF]

Thank you very much, I have to do some research

[Martin72]

What are you up to, if I may ask?

[SHF]

to synchronize the time base with an external GPSDO

[maxmatteo]

finally pulled the trigger for an sds804x.

unfortunately the fan noise is quite loud :/

has anyone already tested similiar fan noise mods like with the SDS1204X ?

replacing with noctua fan? lowering voltage with diodes or 7809 regulator?

thanks for any insights

[tautech]

finally pulled the trigger for an sds804x.

unfortunately the fan noise is quite loud :/

has anyone already tested similiar fan noise mods like with the SDS1204X ?

replacing with noctua fan? lowering voltage with diodes or 7809 regulator?

thanks for any insights

Fan noise is similar to SDS11/1204X-E as expected with a small fan spinning fast.
The OEM fans are roughly the same noise as Noctua therefore reducing noise is only possible by slowing them down < not advised.
Several threads contain info on fan mods to the 4ch X-E models if you take the time to hunt them out.

SDS1000X HD in the much larger enclosure uses a ~80mm fan so is somewhat quieter.

[Martin72]

What you could try is to find a fan that is quieter at the same air flow rate.
To do this, you would need to know the data of the “Siglent” fan and check other fans accordingly.
Whether this effort is worthwhile remains to be seen, because as Tautech has already noted, the fans will not differ much from each other – size pays tribute to physics.
Another thing you can do, but it will certainly void the warranty, is to remove the fins in the inner metal casing to reduce drag for the flow and thus minimize air noise.
Overall, it's an effort I wouldn't go to.

[2N3055]

Removing fins on fan opening might increase EMC emissions... I said might.. I didn't try.