GDS-2000A series New economic oscilloscope by GW Instek comes to market

[ddavidebor]

well, it does not double.

logic analizer usually have 100kohm internal impedance, while oscilloscope 1Mohm

[grego]

Well, nobody still has found out what's the GW Instek's logic analyzer like. It might be better than in my DSOX2002A (not enabled by code yet.) At least the GW will have longer memory, I think.

We've played with it a bit here. A boon is that it comes with a few serial decode abilities for free with the analyzer. So that's pretty cool. An unfortunate limitation is that you can't decode serial buses using analog channels. To see analog channels and decode you have to double probe the signal, which of course doubles the loading effects on the DUT. Big bummer there.

We are working on FW 1.14 and waiting for 1.15 as we speak so perhaps that changed?

1.15 is available now but the notes don't refer to that, just a couple of simple fixes.

[mike1305]

Yep, just installed it. Nothing much changed. According to their changelog:

V1.15 Change log

Fixed position adjustment error for average mode

Fixed the data missing issue during the GPIB high-speed data transmitting

Added horizontal expand mode (optional APP installation required)

Trying to figure out what that last one is.

[Bored@Work]

Yep, just installed it. Nothing much changed. According to their changelog:

V1.15 Change log

Fixed position adjustment error for average mode

Fixed the data missing issue during the GPIB high-speed data transmitting

Added horizontal expand mode (optional APP installation required)

Trying to figure out what that last one is.

Download it and try http://www.gwinstek.com/en/product/productdetail.aspx?pid=3&mid=7&id=1349
Select the Download tab, then scroll down to Software, last entry in that section.

The Expand by Center option will scale the
waveform from the center of the display when
the waveform is scaled using the TIME/DIV
knob.

The Expand by Trigger Pos will scale the
waveform from the trigger position when the
waveform is scaled using the TIME/DIV knob.

[Hydrawerk]

The Horizontal Exansion APP doesn't seem to be much useful... It even doesn't look like an APP.
But GW Instek added an info about the DVM APP. It might be more useful, but there's still too few information... I am not sure if you need a DVM when you have eight auto measurements... And what's the DVM's maximal frequency? Is it trigger independent?
At my DSOX2002A the DVM runs only to 100kHz  , it's trigger independent and you even have to buy this option...  But I got it for free together with Wavegen. 

[mike1305]

Thanks for the context there guys. I notice they have an APP section in the scope. Seems like an interesting way to add extra features later. Although I agree, this horizontal app should just be embedded in the UI somewhere instead of a standalone application.... 

[Electro Fan]

Well, nobody still has found out what's the GW Instek's logic analyzer like. It might be better than in my DSOX2002A (not enabled by code yet.) At least the GW will have longer memory, I think.

Hydrawerk, have you had any experience using the LA in your DSOX2002A?  Or nothing yet?  If yes, what do you like about it, what would like to see different - any and all thoughts on the LA functions would be welcome.  If you have already posted somewhere regarding the LA (or know of posts on the LA by other users) please let us know the link(s).  Thanks, EF

[Yaksaredabomb]

I remember speculation about whether installing two function generator modules would allow both outputs to be used at the same time.

At least a couple sellers say yes, it's possible:

http://www.tequipment.net/InstekGDS-2304A.html
"Two function generators can be used at the same time to provide dual output signals."

http://www.bellnw.com/manufacturer/Instek/GDS-2072A.htm
"Two function generators can be used at the same time to provide dual output signals."

HOWEVER,

Page 4 of the "Quick start guide for DS2-FGN" here http://www.gwinstek.com/en/product/productdetail.aspx?pid=3&mid=7&id=1349 says differently:
"Note: Even though it is possible to install two function generator modules, only one function generator option can be used at any one time."

And yet, the "GDS-2000A brochure" (found at the same url) says exactly the same thing found on the sellers' websites.  I'm sure the sellers are simply making direct quotes from page 2 of the brochure.

To add to the confusion, not only does Instek appear to both be saying only 1 can be used at "any one time" (even with 2 installed) AND both can be used at the "same time" if 2 are installed, they ALSO appear to be saying both outputs can be used with a single module installed!  See page 23 of the "GDS-2000A user manual" (same url), where it says:
"The function generator outputs are used with an optional function generator module."

Okay, so it's a small stretch to interpret the last quote as definitively stating one module will allow both outputs to be used at once - but if this was the only quote a person had to go on I do think it's reasonable to interpret it that way.  "Outputs" are used with "an" optional "module" (singular).  If that doesn't mean multiple outputs are used with only one module installed, then I think it really ought to have been written more clearly.

Three different messages in the manufacturer's own documentation!  What a mess!

Of course, if I had to guess which document is correct I'd assume it's the "Quick start guide for DS2-FGN".

[Bored@Work]

Of course, if I had to guess which document is correct

Forget the documents. Reality trumps document, every time.

And Dave showed us the reality. A single output at a fixed pin of the function generator. No chance that module can put out two separate signals.

Further, Dave's hi-res stills http://www.flickr.com/photos/eevblog/8872848654/#sizes/o/in/set-72157633773967624/ show that the GEN1 output is directly connected through to both expansion sockets. (unless they do some  magic under the sockets)

So no chance two of that shown function generator modules can work at the same time. Maybe not even if you hack one and change its output pin to GEN2. The oscilloscope software probably doesn't support it.

On the other hand, the GEN2 output (the one not used) seems to receive some special treatment between the two sockets (R1506, R1507 are visible and R1506 apparently allowing to connect GEN2 to GEN1). I can't see if the resistors are populated or if there is maybe some magic going on at the other side of the PCB. However, giving Instek the benefit of the doubt, it seems they could make another function generator module with two outputs, or another single output module that could use GEN2.

[marmad]

So no chance two of that shown function generator modules can work at the same time. Maybe not even if you hack one and change its output pin to GEN2. The oscilloscope software probably doesn't support it.

In Dave's video, the FG module is first using the GEN2 output. Later, after he's swapped the modules in back, the FG module is using the GEN1 output.

[Bored@Work]

So no chance two of that shown function generator modules can work at the same time. Maybe not even if you hack one and change its output pin to GEN2. The oscilloscope software probably doesn't support it.

In Dave's video, the FG module is first using the GEN2 output. Later, after he's swapped the modules in back, the FG module is using the GEN1 output.

The software said it would. Big difference. And Dave had to reboot because the instrument was acting funny.

With a simple, fixed output pin on the module, and the GEN1 connection going straight through to both sockets I leave it up to you to explain how you think that magic can happen.

[marmad]

With a simple, fixed output pin on the module, and the GEN1 connection going straight through to both sockets I leave it up to you to explain how you think that magic can happen.

I was just responding to your statements about the GEN2 output. I'm not saying one way or the other if the DSO can support 2 FGs.

On the other hand, the GEN2 output (the one not used)...

[Hydrawerk]

Hydrawerk, have you had any experience using the LA in your DSOX2002A?  Or nothing yet?  If yes, what do you like about it, what would like to see different - any and all thoughts on the LA functions would be welcome.  If you have already posted somewhere regarding the LA (or know of posts on the LA by other users) please let us know the link(s).  Thanks, EF

Well, I don't have the digital probes, but I could start the 30-day trial so i tried the LA a bit. You can connect signals directly to the digital connector of your scope. It works. You don't need any special digital probe with built-in electronics. But now my trial has gone and I am not gonna buy the LA option licence DSOX2MSO... http://www.home.agilent.com/en/pd-1953247-pn-DSOX2MSO/infiniivision-2000-x-series-oscilloscope-mso-upgrades?cc=CZ&lc=eng

[Hydrawerk]

Unfortunately, this is the only picture I took. It was in the demo mode, so there was nothing connected to the digital port. Well, the DSOX2002A logic analyzer has quite small memory, only 50kpoints. I don't know if it can be upgraded but probably not. DSOX2MEMUP upgrades only analog channels I think. On the ither hand, it probably maintains the high waveform update rate even if you turn on all channels. And it's 1GSa/s per each digital channel.
Note that GDS-2000A logic analyzer has 2Mpoints per channel. It's much better.

[Yaksaredabomb]

...
Nothe that GDS-2000A logic analyzer has 2Mpoints per channel. It's much better.

I'm really liking the "2M record length (per channel)" in the picture you attached, but given the discrepancies in Instek documentation I mentioned here I'll believe it when someone verifies it.

[marmad]

I'm really liking the "2M record length (per channel)" in the picture you attached, but given the discrepancies in Instek documentation I mentioned here I'll believe it when someone verifies it.

Well, there seems to be 16M x 16 {32M x 8} of DRAM available - so assuming they mean 2M maximum when none of the analog channels are on, it's possible. Greg would be able to verify this reasonably quickly.

[Yaksaredabomb]

I haven't seen anyone ask this yet, but maybe I've missed it.  Could a GDS-2000A owner please confirm the external dimensions of the scope?  Including knobs and all protrusions, ie, what are the inside dimensions of the smallest rectangular box that would just barely fit the scope?
 
The specs I found list 380mm x 220mm x 145mm, which is about 15.0" x 8.7" x 5.7" (inches).  I'd rather have real-world, end-user measurements though, please.
 
Thanks!
 
Edit:  The scope only, probes, power cord, and logic analyser or other protruding options removed.

[Yaksaredabomb]

...  Could a GDS-2000A owner please confirm the external dimensions of the scope?  Including knobs and all protrusions, ie, what are the inside dimensions of the smallest rectangular box that would just barely fit the scope?  ...
Edit:  The scope only, probes, power cord, and logic analyser or other protruding options removed.

Have any owners had the chance to measure its dimensions yet?
 

...
Note that GDS-2000A logic analyzer has 2Mpoints per channel. It's much better.

Maybe most know this by now, but I just found what might be bad news in the manual.  The last sentence on page 114 reads: "Please note that digital channels are not supported for measurement using segments."

And here, I thought the segmented memory would be a big help with capturing a string of messages on a bus.  Of course, the decode only works on the digital channels.  So it appears you can either:

1) Use the analog channels to capture up to 2048 messages and scroll through them one by one (possibly over the course of several seconds or even minutes+).  You then have to decode them yourself.

OR

2) Use the LA channels to capture a single, normal 2M aquisition and the scope will decode whatever messages are found.  This will most likely only be a very short time frame since you'll have a constant sampling rate throughout the aquisition.  If you need the 500MS/s rate you'll only be able to look at however many messages occur in a fraction of a second.  So if messages are separated by more time than that, and the DUT can't be made to only produce parts of sequences, there will be a lot of work with offsets and delayed triggers etc to try to capture messages deeper in the communication.

Also, of possible interest is the manual's section on record length that's already been talked about, with "short" confirmed to be 1Ks.  Page 109 has the typo about short memory being 1M, and page 110 has a table with two columns (normal and auto) duplicated.  Why not just have two columns, "Single" and "Normal/Auto" or "Normal and Auto"?
 
Edit June 6th: My display name has recently changed from "jneumann" to "Yaksaredabomb"

[Hydrawerk]

Have any owners had the chance to measure its dimensions yet?
 

Oh, I believe what they say in the user manual (EDIT) about dimensions.

[Yaksaredabomb]

Oh, I believe what they say in the user manual.

   You're kidding, right?   
 
Joking aside I'd still prefer to hear someone's actual measurements.  Not much different than asking for a screenshot showing the memory depth at a certain sampling rate, video showing the intensity grading with a certain signal, or a hands-on account of any other feature.
 
Edit June 6th: My display name has recently changed from "jneumann" to "Yaksaredabomb"

[grego]

I'm heading out of town this weekend for a week but I'll see if I can get that for you before I go.

[Hydrawerk]

Note that GW Instek provides no photo of scope's intensity gradation like this. http://www.rigolna.com/images/products/DS2000.jpg
Their photo has no gradation shown.
http://www.gwinstek.com/product/images/picture/FC_2012109144736.jpg

[Yaksaredabomb]

Note that GW Instek provides no photo of scope's intensity gradation like this. http://www.rigolna.com/images/products/DS2000.jpg
Their photo has no gradation shown.
http://www.gwinstek.com/product/images/picture/FC_2012109144736.jpg

Then again, Instek's photo shows several digital inputs while Rigol's photo doesn't show any....   

[Hydrawerk]

Dave, this GDS-2000A deserves some more investigation. I believe that it has some strong points... When you don't need long memory (7MB or so), fast waveform update rate or intensity gradation.  The LA might be fine?? 
And hey, is the scope design much simpler, when they did not implement fine vertical or horizontal scale? Owon did the same. 

[Yaksaredabomb]

...  Could a GDS-2000A owner please confirm the external dimensions of the scope?  Including knobs and all protrusions, ie, what are the inside dimensions of the smallest rectangular box that would just barely fit the scope?  ...
Edit:  The scope only, probes, power cord, and logic analyser or other protruding options removed.

Have any owners had the chance to measure its dimensions yet?
 

...
Note that GDS-2000A logic analyzer has 2Mpoints per channel. It's much better.

Maybe most know this by now, but I just found what might be bad news in the manual.  The last sentence on page 114 reads: "Please note that digital channels are not supported for measurement using segments."

And here, I thought the segmented memory would be a big help with capturing a string of messages on a bus.  Of course, the decode only works on the digital channels.  So it appears you can either....

Interesting - this post over in the "EEVblog #474 - GW Instek GDS-2000A Series Oscilloscope Unboxing  & First Impression" topic has a video from TEquipment showing segmented memory working with the logic analyzer.  So yet another direct contradiction to what's in the manual, though this time it is a positive contradiction.