SIGLENT SDS1102CM and EasyScope

[Kozmyk]

Well, I went for the SIGLENT SDS1102CM in the end.
Suits my pocket and 2Mpts of memory to boot.
I haven't had long to play with it but it seems fine as far as I can tell using an old synth resonant filter as a sine source and the probe square wave o/p.
The replacement for my old blown signal generator still hasn't arrived yet.

The only annoying thing with it so far is the EasyScope 3.0 software.
The bundled CD version is in Chinese, it links OK but I can't read the buttons.
I downloaded English versions from SIGLENT, ATTEN and LeCroy but none of them will link properly with the scope
I get one Green light only when I test the connection and then the Link Fails when I hit the connect button.
Looks like communication is only going one way.

I've emailed SIGLENT but in the meantime I think I'll try installing the English version to a memory stick and then copy files over to the main install on my HD one by one and see if I can get it to work AND speak English.

Any constructive ideas about how to fool EasyScope into speaking English and working at the same time will be most welcome.

[Kozmyk]

You'll be relieved to learn I'm sure that I've sorted it.
The Chinese version of EasyScope 3.0 Ver 3.01.03.01.09 can indeed be fooled into operating in English.
The installer won't proceed when another version is installed so I first installed an older English language version Ver 3.01.03.01.02 onto a USB stick, then removed the stick and uninstalled what was left through Vista Control Panel.
I then installed the Chinese 09 but replaced the set.ini (that carries the language command) from the stick along with some likely looking .dll files which probably hold the English characters.
Second or third time of bodging and she sprang into life in full English AND talking to the scope.
Result!!
Not an elegant solution but hey ... brute force and ignorance has it's uses

[saturation]

Happy for you.  Give us a review of its performance when you can.

[Zero999]

I'm be interested to hear a review too. It does look tempting, perhaps a little too good to be true.

[Kozmyk]

Aarrgh! A review? No pressure then ...

I guess I could take a few pics, I'm not rigged for video but I think I could squeeze 640 x 480 20fps out of my Canon A470 still camera.
Bottom line will be that it works.

I'll give it some thought.

Here's the spec page to be going on with:
http://www.siglent.com/en/product_show.asp?articleid=159
OR
http://aprilatsiglent.en.ecplaza.net/2.asp

I would say that Siglent replied promptly to my email enquiry about the EasyScope software.
That's a good sign in my book.

 

[saturation]

Yes, if you can, if you do I think you'll be the first and only Siglent review on eevblog, if not many forums or review sites too.

Aarrgh! A review? No pressure then ...

[Zero999]

Going from the website, there are three models: 60MHz, 100Mhz and 150MHz. I wonder how hackable they are?

I couldn't find the 150MHz model on ebay. The lowest prices I could find for the 60MHz and 100MHz are $350 and $420 respectively. If I bought one, I'd probably just get the 100MHz rather than hack because it seems more trouble than it's worth.

[Kozmyk]

I think I may have found the catch in the bucks vs features vs quality on this scope.
The waveforms displayed are unstable, jittery.
Even low frequency signals move around a bit.

I've recorded a bit of it compared to my Kikusui 5020 20MHz analogue scope.


Apologies for the video quality. I needed to get in closer and use less zoom but it was too late at night to go shifting furniture.

I don't know what to do about the scope now.
I'll try some other sources but I think it's the scope that's at fault not the signal source.
Am I being too fussy?
Maybe it's a shortcoming common to all these budget DSOs?

[johnnyk126]

I think that this jumping on digital scope comes from the difference in display technology.

On cheap digital scope without extra features enabled pixels are only on or off. Waveform is redrawn few times per second and there is no fadeout (unless persistence is enabled). There is only one waveform drawn at a single time, and that causes this jumping, because consecutive redraws are in different places and previous waveform is instantly removed. Sharp pixel image enhances this jumping badly.

On analog scope because of persistence of phosphor jittery waveforms produce blurry lines or double images. They seem to be stable because they are not turned on and off instantly.

Try to enable persistence and play with intensity/brightness settings, this should it bring more to analog behavior.

[Kozmyk]

Do you think that it's Just aliasing type noise?
At the trigger point, which on the DSO is shown in the middle of the screen, the display is rock steady.
The jitter appears to get worse as you get further away from the centre.

Altering persistence does cover some of the digitisation noise but not the jittering of a whole vertical line on a square wave

I'll certainly play around with it some more before I decide if I like it or not

[Zero999]

I think that this jumping on digital scope comes from the difference in display technology.

On cheap digital scope without extra features enabled pixels are only on or off. Waveform is redrawn few times per second and there is no fadeout (unless persistence is enabled). There is only one waveform drawn at a single time, and that causes this jumping, because consecutive redraws are in different places and previous waveform is instantly removed. Sharp pixel image enhances this jumping badly.

On analog scope because of persistence of phosphor jittery waveforms produce blurry lines or double images. They seem to be stable because they are not turned on and off instantly.

Try to enable persistence and play with intensity/brightness settings, this should it bring more to analog behavior.

That makes sense to me too and is exactly what I'd expect to see from a digital 'scope hooked up to an unstable waveform.

Do you think that it's Just aliasing type noise?
At the trigger point, which on the DSO is shown in the middle of the screen, the display is rock steady.
The jitter appears to get worse as you get further away from the centre.

Altering persistence does cover some of the digitisation noise but not the jittering of a whole vertical line on a square wave

I'll certainly play around with it some more before I decide if I like it or not

By the looks of it you used an uncompensated probe. Try using a compensated probe and it should be better.

What signal generator are you using? By the sounds of it, it only goes up to 5MHz so it can't be that good. Try using a crystal oscillator and you should get a perfectly stable waveform, if the 'scope is any good.
 

[saturation]

Alas, this doesn't look good at all.  If its tends to do this because it was designed this way, or its designed better and has a  fault, either way I would return it.

Those waveforms should be as stable as the analog scope.  It could waver and jerk for many reasons: subsampling, the clock is unstable, DAC are faulty,  and many other things ... none of which should exist in a well made low end DSO at the low frequency used for the test, IIRC the video shows ~ 1 MHz or so.

Here is a copy of a video we did on the Hantek 3x25 arb generator as mechatrommer hacked the software; it shows clearly the effect of 7 MHz vs 100 MHz;  you can see how stable the waveforms are.

At 100 MHz the Rigol starts to 'wobble' a bit, this is the effect of subsampling as the ratio is 1:5, below the recommended Nyquist 1:10 for most stable images ... [at 2 channels the sampling rate it 500Ms/s.]

Do you think that it's Just aliasing type noise?
At the trigger point, which on the DSO is shown in the middle of the screen, the display is rock steady.
The jitter appears to get worse as you get further away from the centre.

Altering persistence does cover some of the digitisation noise but not the jittering of a whole vertical line on a square wave

I'll certainly play around with it some more before I decide if I like it or not

[johnnyk126]

Do you think that it's Just aliasing type noise?
At the trigger point, which on the DSO is shown in the middle of the screen, the display is rock steady.
The jitter appears to get worse as you get further away from the centre.

LCD display doesn't have any intermediate areas between pixels. In addition resolution is relatively small and pixels large. This probably causes you to see this as very large problem and difference seems dramatic.

Also on analog scope every sweep goes on the screen, so you have smoother transitions. On digital it is not possible due to display speed, so this makes this jumping even worse, because consecutive sweeps are further apart.

Try checking the scope on stable frequency source, such as crystal mentioned above.

What I'm trying to say is that analog scopes are masking a lot of the jumping you can see. The same way needle in analog meter seems smooth and stable when digital meter LCD display is full of jumping numbers.

[Kozmyk]

Thanks for the input guys.

Connection used in the video was direct 50 Ohm BNC cable, no probe.
Terminated or unterminated made no discernible difference.
Changing to compensated probes on x10 still displays jitter above 100kHz on the DSO not on the CRO.

For sure it's a cheap sig gen but not so bad that the CRO can't get a stable display from it.
I understand the principles of sampling and aliasing errors/artefacts but I'm not convinced that is the issue here. I could be wrong of course.

I'll have to break into something around the house and look for a crystal output from something.
One of my RC Txs should have something suitable.

I'm beginning to suspect that it's not fit for purpose.
I'm not looking forward to all the hassle involved in sending it back.

The HF crystal osc test shall be the decider.

I'mma go eat and worry about it later
Do some more video editing ...

[Kozmyk]

Well, this thread has achieved one thing at least.
I've broken my YouTube duck.

I found a 16MHz Xtal lurking in one of my RC Txs and here's the vid.



I had a mess about with persistence and intensity but it's only overlaying the variations and ending up with a Thick trace.



Maybe I'm expecting too much from this type of scope, I dunno.
It's OK for freezing a burst of PWM and examining in detail but realtime waveforms get the thumbs down from me.

[Zero999]

I assume this is using a compensated probe? I know adding too much capacitance can pull crystals off. Still either way that doesn't look good.

The frequency appears to be fine, it's more noise added to waveform. What do you see if you short the probes together and choose the lowest voltage/division setting?

[grenert]

The frequency appears to be fine

The frequency does not appear to me to be fine at all.  There is shifting left and right that seems to be a timebase or triggering problem.  Kozmyk, sorry you took one for the team, but I agree that there are probably better options out there.

[Kozmyk]

x10 probes shorted gives a 4mV p-p 106.3MHz ripple.
50Ohm terminators  straight over the inputs give same freq @0.25mV p-p

Time base seems a likely culprit to me as the trigger appears to hold steady at centre screen but he trace expands and contracts either side of it.
Could be video timebase has the problem while the front end is steady.
Just guessing.

Arguing this as a fault worthy of RMA is going to be fun.

[Kozmyk]

The makers

I've found SIGLENT, their International Sales Dept in particular, to be very helpful, although a fix is not yet forthcoming.
I had hoped that they might have a firmware update to cover this already.
It's an older model by their standards.
They may yet come up with something.

I must say that apart from the Jitter problem, I really like this scope and it's features.
The split/alternate triggering in particular.
The ability to view two unsynchronised waveforms at the same time is a delight.
I really would like to be able to keep this scope.


The sellers

This is an interesting and encouraging response from the seller.
Encouraging because they appear open to the idea of no-quibble returns.
Interesting because the makers and distributors have encountered jitter problems before with this model.
The jitter they discuss is however at a different frequency, assuming the units to have translated correctly. microseconds often becom milliseconds in translation, I've found.

Here's a section of the seller's comms:

We contacted factories and other distributers and it’s confirmed there would be jittery on 1k square wave. People in China managed to minimize these interference by applying the ground attch to a real one meter physical iron rod buried underground. Or compensation to the lead sometimes helps.
Usually people use STOP button when they use 500ms, with Siglent’s good storage performance it displays clear capture.
It was reported by engineers in China that wave interference does existe on 2mv, 2.5ms.
All these are commented as normal here.
As for RMA, it’s up to you, sir. We would love to offer a replacement or full refund with the item back at your cost since.
Siglent is quite a popular brand in China, making OEM manufacture for Angilent, Tek, Lecroy and such. Highly recommend you take your time before you make any decisions.

That a pretty positive seller attitude I reckon.

Fingers still crossed for a firmware upgrade. 

[saturation]

This is good to know they are aware, are trying to fix it, and responded well to you.

But they already have your money,  you still have a defective product so unless you can wait and have no immediate need to use it that's fine but you can also return it until the repair is made then buy it then with the working firmware/fix; with the way products are, chances are it will be cheaper or there is something new that will come out better for the same price.  Why keep a defective item?

Many scopes have split unsync triggering of 2 channels, at the least the Rigol 1052e has it, it works well and you know it been that way since 2009.

The makers

I've found SIGLENT, their International Sales Dept in particular, to be very helpful, although a fix is not yet forthcoming.
I had hoped that they might have a firmware update to cover this already.
It's an older model by their standards.
They may yet come up with something.

I must say that apart from the Jitter problem, I really like this scope and it's features.
The split/alternate triggering in particular.
The ability to view two unsynchronised waveforms at the same time is a delight.
I really would like to be able to keep this scope.


The sellers

This is an interesting and encouraging response from the seller.
Encouraging because they appear open to the idea of no-quibble returns.
Interesting because the makers and distributors have encountered jitter problems before with this model.
The jitter they discuss is however at a different frequency, assuming the units to have translated correctly. microseconds often becom milliseconds in translation, I've found.

Here's a section of the seller's comms:

We contacted factories and other distributers and it’s confirmed there would be jittery on 1k square wave. People in China managed to minimize these interference by applying the ground attch to a real one meter physical iron rod buried underground. Or compensation to the lead sometimes helps.
Usually people use STOP button when they use 500ms, with Siglent’s good storage performance it displays clear capture.
It was reported by engineers in China that wave interference does existe on 2mv, 2.5ms.
All these are commented as normal here.
As for RMA, it’s up to you, sir. We would love to offer a replacement or full refund with the item back at your cost since.
Siglent is quite a popular brand in China, making OEM manufacture for Angilent, Tek, Lecroy and such. Highly recommend you take your time before you make any decisions.

That a pretty positive seller attitude I reckon.

Fingers still crossed for a firmware upgrade. 

[Kozmyk]

You're right of course.
It's just that a company or seller treating you right and doing what they're supposed to do is cause for celebration, these days. 

Guess who's kicking himself right now that he didn't follow the crowd and get a DS1052E ? 

Comms with Siglent have been good.
We were chatting via MSN this morning, very helpful.
 First time I'd received an attachment via MSN as well. (new English version of EasyScope Ver3.01.03.01.11)
It's worth giving it a day or two to see if Siglent can come up with a fix; give their Sales Dept time to get the query through to the relevant engineers.

Failing that it'll be RMA time.

[dfnr2]

Hi Kozmyk,


This phenomenon is totally normal.  You are looking at the difference between a DSO and an analog scope.  When you acquire any piece of test equipment, including a DSO or analog scope, you do have to learn it's limitations.  That's why HP and Tek used to publish detailed theory of operation and schematics for all their equipment.  Looking at your videos, the "unstable" DSO waveforms exactly match up with entirely comparable artifacts on the analog scope.  Depending on which scope you are used to, your brain will tend to integrate out the artifacts in one, and fixate on the artifacts in the other.  Also, it's why it's sometimes nice to have both scopes in the lab.  But if I had to choose, I'd pick the DSO.

You are looking at a combination of several factors:
  1) System noise, including at the generator output, cabling, input stage, and ADC noise sources
  2) Noise from random sample jitter. 
  3) Signal distortion from asymmetric sample interleaving. 

You can see and measure the effects of all these sources in all DSO's.  High-end scopes (meaning $$$$+) use all sorts of black magic to minimize these various sources.  Careful attention to input stage design, low-noise circuits, careful shielding, etc. help minimize noise.  Careful clock management, high-speed ADC's, careful timing, routing, and high-speed design help with the second two factors in high-end scopes. 

I've attached some images of a 16 MHz sine wave (to match your example), taken with a LeCroy Wavepro scope.  The first shows sampling at 500 Ms/s, probably what the siglent is using. Here, because there are fewer samples per cycle, noise on the waveform causes more gross distortions in the perceived sinusoidal shape.

The second is sampled at 1 Ms/s.  You can see distortions of similar magnitude, but half the width.  The energy of each deviation is smaller.  The eye perceives the underlying waveforms better.

The third is sampled at 20 Gs/s.  The noise is the same amplitude, but the underlying waveform is even more apparent, partly due to the display routine, and partly due to the integrative abilities of our retinas and optic cortex.

The fourth and fifth pictures show the result of turning on persistence for the 20Gs/s waveform, and (last) the 500 Ms/s waveform.

You can see that the appearance of the high end lab-grade scope is similar to your siglent when using similar sample rates.  I have tried another similar rebadged Atten scope (the 150 MHz version), and a Hantek 200 MHz scope.  I found the atten/siglent to have overall much less distortion due to jitter and sample-interleave asymmetry than the Hantek.  This has been discussed in several other forum threads, including the long Hantek thread, and a rigol sinc(x) thread.

So take heart.  You have a great deal on an inexpensive decent quality scope, about equivalent to the low-end Tek, LeCroy, and HP scopes.  Actually, identical to the LeCroy WaveAce scopes, since these are simply rebranded Atten scopes based on the same design.

Best regards,

Dave

(EDIT:) BTW, the measurements at the bottom of the screen are not meaningful, because I had them set to channel 1, and the input for this experiment was channel 2.  I didn't think to switch over the measurements, and just noticed it now.  Measurements would be irrelevant to the point of this exercise, however.

[Kozmyk]

Thank you for your efforts to further explain the effects of waveform digitisation on their resultant display.

The graphs illustrate your point but not the temporal stability or lack thereof that a video might show, but I can't go demanding videos, can I.

I've now seen better stability and display clarity from comparable units so this one is going back, I'm afraid.

I went back and played some more with Persistence but the results did not satisfy.
I appreciate that Expectation vs Perception may be playing a part here.
Altering my perception filter however only succeeded in leaving me with the conviction that I was fooling myself.

On all other aspects I found the Siglent a pleasure to use.
Siglent themselves have been most helpful in trying to find a solution.
Comms with them have been a pleasure.

[dfnr2]

Thank you for your efforts to further explain the effects of waveform digitisation on their resultant display.

The graphs illustrate your point but not the temporal stability or lack thereof that a video might show, but I can't go demanding videos, can I.

I've now seen better stability and display clarity from comparable units so this one is going back, I'm afraid.

What's in your video could be perfectly normal.  However, if the "jumping" is worse than other comparable scopes (on your same setup), then it is possible that the distortions you are seeing may be due to asymmetric DAC timings.  This is what I was seeing on the my Hantek, which had a similar effect much worse than the otherwise comparable Atten.  In the case of the Hantek, this was most likely due to imperfectly spaced sampling due to difficulty timing 5 DACs rather than 4 on the atten. 

I imagine it would be possible for two siglent scopes to perform differently.  You may want to borrow a comparable scope to compare the magnitude of the effect.  It's good to know the dealer is being responsive.  I certainly hope they do the right thing and let you return or exchange the scope if it's not behaving as expected.  Let us know how it turns out.

Dave

[saturation]

I agree fully, plus there are other causes either by design or by production defect.  But a good DSO shouldn't jitter in the horizontal timebase; also the amount of sampling artifact on low frequencies needn't be tolerated as there are other scopes that do better.  As example, the Rigol 1052e videos are far stabler even with sampling artifact on default settings for sampling speeds, without persistence, averaging, or sinx on.

What's in your video could be perfectly normal.  However, if the "jumping" is worse than other comparable scopes (on your same setup), then it is possible that the distortions you are seeing may be due to asymmetric DAC timings.