Siglent SDS6000A DSO's 500MHz-2GHz

[tv84]

Where did you find it ? Can´t find it on their chinese site..

If I say it was an epiphany would you believe? 

[Martin72]

Ahh, light is on...

[tautech]

Officially we only know SDS7000A is coming:
https://www.eevblog.com/forum/testgear/siglent-sds7000a-dsos-coming/

We need wait for more news.....

[Ulrich.G]

Where did you find it ? Can´t find it on their chinese site..

Its online now

https://www.siglent.com/products-overview/sds6000-pro/

[egonotto]

Hello

When translated into German, it says "Low noise floor: up to 2 μVrms at full bandwidth of 153 GHz".

Best regards
egonotto

[2N3055]

Hello

When translated into German, it says "Low noise floor: up to 2 μVrms at full bandwidth of 153 GHz".

Best regards
egonotto

That is translator error. English translation : Low noise floor: as low as 153 μVrms at full 2 ​​GHz bandwidth

[Ulrich.G]

On the chinese webpage the SDS 6000A completely disappeared - does that mean, that the SDS6000A and the SDS6000 Pro might be identical hardwarewise? Just reduced in capabilities for the market outside China?

[tautech]

On the chinese webpage the SDS 6000A completely disappeared - does that mean, that the SDS6000A and the SDS6000 Pro might be identical hardwarewise? Just reduced in capabilities for the market outside China?

Not sure it was ever there, only the Pro version IIRC.

Some models were developed in conjunction with LeCroy of which Siglent has had a longtime development partnership and western marketing rights were LeCroy's as part of the deal.
SDS3000 and SDS3000X are examples that we have never seen in the west.

SDS6000A has one ADC/ch which retains full sampling rates when all channels are active.

[Martin72]

On the chinese webpage the SDS 6000A completely disappeared - does that mean, that the SDS6000A and the SDS6000 Pro might be identical hardwarewise?

SDS 6000A is only for "our" market.
For an 8-bit scope it has remarkably good vertical specs and it´s ERES goes up to 16bit..

[Performa01]

On the chinese webpage the SDS 6000A completely disappeared - does that mean, that the SDS6000A and the SDS6000 Pro might be identical hardwarewise? Just reduced in capabilities for the market outside China?

Since I rarely ever check the Chinese Website, I cannot know for sure but would be very surprised if the SDS6000A has ever been there.

The international SDS6000A is near identical to the Chinese SDS6204 H10/12 Pro; that means it has one ADC per channel, even for the lowest 500 MHz bandwidth model. So in this regard we are better off than our Chinese peers.

The big trump of the SDS6000A is, that even though only 8 bits of the ADC can be used, we still get 12-bit performance in terms of INL. These are the best conditions for a great working software resolution enhancement (ERES), because it's easy to increase the resolution (at the cost of bandwidth), but you cannot improve the linearity, which in turn is required for monotonicity of the transfer curve.

[tautech]

New firmware for SDS6000A models

Version： V1.4.8.3
130 MB
https://int.siglent.com/upload_file/zip/firmware/Oscilloscope/SDS6000_V1.4.8.3_EN.zip

Release notes
FFT: supported horizontal log axis
Channel: optimized strategy of adding a trace
Fixed several bugs
          a. Bode plot: load and sweep settings not remembered; no virtual keypad for setting of some parameters
          b. ARINC429 trigger not work on SDS6104 Pro

[Minki]

HI All.........Is that mean the happy time is over? :'( :'( :'( :'( :'( :'( :'( :'(
2: Read Operation
Return Count: 10 bytes
ZDL-HD-1G\n
4: Read Operation
Return Count: 58 bytes
Siglent\sTechnologies,SDS6054A,XXXXXXXXXXXXXXX,1.17.1.4.3.3\n
The key gen not working.........

# Enter here the basic model
# You get this by running "MD5_PR?" at the SCIP prompt
# A sample: You have the SDS1104X-E than enter "SDS1000X-E"
# Some more basic models: "SDS1000X-E", "SDS2000X-E", "SDS2000X+", "SDS5000X", "ZODIAC-"
Model = 'ZDL-HD-1G'

[tautech]

The key gen not working.........

It never did for SDS6000A models.
You must follow different solutions.

[Minki]

The key gen not working.........

It never did for SDS6000A models.
You must follow different solutions.

Using NMAP but did't find the telnet port...... :'(

[JeremyC]

The key gen not working.........

It never did for SDS6000A models.
You must follow different solutions.

Using NMAP but did't find the telnet port...... :'(

Try connect using telnet to port 5024/tcp. If it wouldn't work try scan all tcp ports with nmap.
    nmap -sT  -p1-65535  [ the IP address ]

[Minki]

Try connect using telnet to port 5024/tcp. If it wouldn't work try scan all tcp ports with nmap.
    nmap -sT  -p1-65535  [ the IP address ]

NSE: Script Pre-scanning.
Initiating NSE at 09:57
Completed NSE at 09:57, 0.00s elapsed
Initiating NSE at 09:57
Completed NSE at 09:57, 0.00s elapsed
Initiating NSE at 09:57
Completed NSE at 09:57, 0.00s elapsed
Initiating ARP Ping Scan at 09:57
Scanning 192.168.1.39 [1 port]
Completed ARP Ping Scan at 09:57, 1.99s elapsed (1 total hosts)
Initiating Parallel DNS resolution of 1 host. at 09:57
Completed Parallel DNS resolution of 1 host. at 09:57, 0.00s elapsed
Initiating SYN Stealth Scan at 09:57
Scanning 192.168.1.39 [65535 ports]
Discovered open port 111/tcp on 192.168.1.39
Discovered open port 80/tcp on 192.168.1.39
Discovered open port 5900/tcp on 192.168.1.39
Discovered open port 775/tcp on 192.168.1.39
Discovered open port 55629/tcp on 192.168.1.39
Discovered open port 5025/tcp on 192.168.1.39
Discovered open port 5355/tcp on 192.168.1.39
Discovered open port 5024/tcp on 192.168.1.39
Completed SYN Stealth Scan at 09:57, 2.40s elapsed (65535 total ports)
Initiating Service scan at 09:57

5024 just for SCPI....
Welcome to the SCPI instrument 'Siglent SDS6054A'
>>)
5025 in telnet can connect but nothing....am i missing something?(SSH not work on 5025)

[JeremyC]

Try connect using telnet to port 5024/tcp. If it wouldn't work try scan all tcp ports with nmap.
    nmap -sT  -p1-65535  [ the IP address ]

NSE: Script Pre-scanning.
Initiating NSE at 09:57
Completed NSE at 09:57, 0.00s elapsed
Initiating NSE at 09:57
Completed NSE at 09:57, 0.00s elapsed
Initiating NSE at 09:57
Completed NSE at 09:57, 0.00s elapsed
Initiating ARP Ping Scan at 09:57
Scanning 192.168.1.39 [1 port]
Completed ARP Ping Scan at 09:57, 1.99s elapsed (1 total hosts)
Initiating Parallel DNS resolution of 1 host. at 09:57
Completed Parallel DNS resolution of 1 host. at 09:57, 0.00s elapsed
Initiating SYN Stealth Scan at 09:57
Scanning 192.168.1.39 [65535 ports]
Discovered open port 111/tcp on 192.168.1.39
Discovered open port 80/tcp on 192.168.1.39
Discovered open port 5900/tcp on 192.168.1.39
Discovered open port 775/tcp on 192.168.1.39
Discovered open port 55629/tcp on 192.168.1.39
Discovered open port 5025/tcp on 192.168.1.39
Discovered open port 5355/tcp on 192.168.1.39
Discovered open port 5024/tcp on 192.168.1.39
Completed SYN Stealth Scan at 09:57, 2.40s elapsed (65535 total ports)
Initiating Service scan at 09:57

5024 just for SCPI....
Welcome to the SCPI instrument 'Siglent SDS6054A'
>>)
5025 in telnet can connect but nothing....am i missing something?(SSH not work on 5025)

Minki, I suggested all ports scan, because you mentioned “Using NMAP but did't find the telnet port...... :'(“. 
I was guessing that you tried the default nmap scan which include only the most common ports and telnet server maybe listening on another port.
Telnet server will respond asking you for username and as next for a password. If you got connected it doesn't mean that you’re connected
to a “telnet” server. As an example you may try telnet to “google.com 443” but you’re not connected to telnet server…
I’m guessing the string in red color frame it’s what you tried to paste in the terminal (PuTTY) without seeing prompt asking for username,
the string looks to me a line from /etc/shadow file.

I’m not judging, but if you’re not 100% confident in your IP protocol knowledge and modern UNIX/Linux security model then you should
wait to time when somebody will provide proofed solution, otherwise instead of improving your relatively expensive scope you may turn it to a brick.

[Minki]

Just open up and found some interesting thing...
Some PCBA had Lecory logo on it
1.Power cable PCB

2.AC trigger PCB


And any one can help for tell the location and PIN out of the main MCU JTAG for memdump :'(

[Minki]

Any one can help for locate the JTAG pin out of MCU....

And AWG...?It seem only on sds2000x-hd...?

[Minki]

I finally got it

[Martin72]

Excellent 

[Performa01]

There have been complaints about spurious signals in the Chinese SDS6204 H10 Pro in a completely unrelated thread:

https://www.eevblog.com/forum/testgear/siglent-new-sds800x-hd-first-bactch-unboxing-and-noise-compare-with-dho800/msg5238798/#msg5238798

Yes, this looks nasty. Yet it’s just the price of high bandwidth.

If we try to analyze the 2nd screenshot, we see spurious signals up to about -60 dBV at 200 mV/div vertical gain. Full scale is 1.6V, so the spurious signals are ~64 dB below full scale (-64 dBFS). This corresponds to a ratio of 1600 which in turn is equivalent to 10.6 bits. Quite adequate for a 10 bit machine…

Of course there is more to the ENOB calculation than just these not input related spurs. Yet it shows that they are not the limiting factor.

Btw, there is no point in activating ESR for the FFT.

For comparison purposes, here is the corresponding spectrum of a 12 bit SDS6204 H12 Pro:


SDS6204 Pro H12_Spurs_200mV

The strongest spurious signal is about -76 dBV at 200 mV/div. This means -80 dBFS, hence a ratio of 10000 which would be equivalent to 13.3 bits. Once again this is not the limiting factor for the ENOB.

[gf]

Yes, this looks nasty. Yet it’s just the price of high bandwidth.
...
For comparison purposes, here is the corresponding spectrum of a 12 bit SDS6204 H12 Pro:

Wow, so many interleaved ADC cores?
[Having frequencies of fs/2^N, I'm fairly confident they're interleaving spurs. What do you think?]

Btw, there is no point in activating ESR for the FFT.

Enabling it even reduces the FFT resolution for a given number of points, due to the higher sample rate.
ESR is also not supposed to improve sinc-interpolated waveform display.

I think the primary aim of ESR is rather to get a higher time resolution to improve the accuracy for some types of measurements.
[I guess the same should be possible with the interpolate math function on those models that don't support ESR as a dedicated feature.]

[Performa01]

Wow, so many interleaved ADC cores?
[Having frequencies of fs/2^N, I'm fairly confident they're interleaving spurs. What do you think?]

Well, the frequency spacing is exactly 39.0625 MHz; I cannot see the relation to the sample frequency. On the other hand, the SDS2000X HD is almost completely spur-free, there only is a single spur at 500 MHz = fs/4. Yes there is one at 1 GHz as well, but this is at the very end of the spectrum and outside the specified bandwidth.

Btw, there is no point in activating ESR for the FFT.

I think the primary aim of ESR is rather to get a higher time resolution to improve the accuracy for some types of measurements.
[I guess the same should be possible with the interpolate math function on those models that don't support ESR as a dedicated feature.]

Yes, exactly. Yet there is a difference when compared to the Interpolate math function, because ESR is handled in realtime during acquisition. It takes up twice as much sample memory and the DSO software is unable to distinguish the ESR data from the output of a real ADC with twice the sample rate.

It is a feature introduced by LeCroy, which has its uses in some special cases, but I prefer to have it disabled most of the time.

[gf]

Well, the frequency spacing is exactly 39.0625 MHz; I cannot see the relation to the sample frequency.

5000 MHz / 128 ?