TSP #44 - Review & Teardown of the Analog Arts SL987 USB Oscilloscope, AWG & Logic Analyzer 

I like how Shahriar tries to find some positive points, even though this board looks like my first hobby project:D

oh man why did they send a product in this state, they lucky Dave didn't get hold of it,lol !

I can't believe a company would send such a bodged piece of hardware to Shahriar. They could have least sent him a decent one that hadn't been on fire first!

Thank you for the in-depth and balanced review Shahriar, I found it very educational. As you demonstrate within the video, it's easy to be misled by raw figures and therefore not understand the limitations of devices, especially in test equipment that has broad uses. It would be great if they'd loan you a later revision of the board to see how many of the issues, such as the DC offset and crosstalk, were caused by the manual fixes.

Shahriar, thanks for taking such a professional stance towards this review. I am quite amazed at some of the comments from the other viewers saying it was unfair - I think you were as fair as you could be. It appears that some of the viewers don't understand the theory behind your comments and I think this is the reason for the conflict. The product has some definite potential but suffers from a few serious problems. I like that you gave some helpful pointers towards fixing the various problems and I hope that Analog Arts pays attention. I will also give the company the benefit of the doubt and assume they sent you that unit accidentally as it was certainly not a production ready unit. I always look forward to your videos as I learn something new from each one. Thanks for taking the time to make them!

It's remarkable how he manages to find positive aspects about this product. I think he was very kind. But those just registered RU-vid commenters blaming his review are not cool.

I loved this review! I hope you will review other products that are not ready for "prime-time". Sometimes there's more to learn from a product that has some issues as opposed to one that's operating exactly as specified. Fantastic work as usual!

You are being too kind shahriar. This product is nothing more than a scam.

A bad review is still a good review! I still learn quite a few things from a bad product. Thanks Shahriar. There is always something to learn from your video.

A very good review, and also a very good product to demo bad designs. Learn a lot from wrongs. Thank you!

The "floating BNCs" apparently are a function of them being isolated style connectors but even those should have collar nuts to secure them to the panel. Hard to believe they would send you such a bodged unit without at least having the courtesy to let you know you weren't getting a production box - they didn't think you were going to take it apart? ;-)

As always you did a very good job and show us how respectful you are. Thumbs up for this video. But I think you should reserve your time for a more refined product. Another point is that I think someone is kidding you. The factory would never send something like this for a review... All the best!

Hey Shariar, don't be discouraged by the trolls. This was a fair review, and even the trolls know it... If the company is behind these trolls, shame on them! They should have just made the corrections to the spec, roll out another prototype with no ugly botches and burn marks, and sent it to you for another review. I am sure prospective buyers would appreciate that and be more likely to buy that product.

Checked the first five or six comments defending this product, replies attempting to find fault with this review, and all accounts checked were created within 24 hours of this video being uploaded. That was enough to convince me that Analog Arts should be boycotted by RU-vid viewers and video creators. Let's hope other electronics-based RU-vid content providers are alerted to the way this company behaves. So they are spared having to waste their valuable time responding while their video's comment threads are hijacked by fake accounts from dishonest marketing drones. That kind of behaviour is a disgrace. Utterly unacceptable. And so poorly executed.

We all learned a LOT, how many many many things should not be done!

Excellent review. But what were they thinking sending you this bodged up instrument? I think you were very restrained in your criticism. It may need more that a re-spin of the board to bring it within sight of their hopelessly optimistic specifications. Significant changes in the laws of physics might have to be implemented ...

This company made a HUGE mistake sending this unit to somebody who actually knows what they are talking about... I'm sure they regretted it as soon as they saw this video go up. Obviously they never watched another one of his videos, they just assumed he would plug it in, and do some simple measurements... but Shahriar cannot be bought and will always tell us the truth about what he reviews. One of the reason I have been watching for so long!

Since I am about to buy a similar device from them, could you clarify as what specifications (if any) you have problems with? I watched your video twice with no suggestions. I looked at the performance of the device against its specifications (in particular the amplitude performance of the generator.) They do have a 50 Ohms in series with the output of the output amplifier. That I think should explain the amplitude reduction at higher frequencies. Also, they specify the amplitude throughout the operating frequencies. Can you tell me if I am wrong? It seemed to me you had mixed feelings about the device. I just want to know if your feelings were based on posted specifications or just an expectation.

It's a kind of toy, not an instrument. Like, they send a product like this for you tear down and put on RU-vid. And of course their company have Times New Roman on their website, who uses Times New Roman as font? lol

Hi Shahriar, thanks for this review. The PCB nearly made me puke, ESPECIALLY the logic analyzer traces, what a sham!!!!!!!!!!!!!! I design CAN communication lines, SPI and ethernet 100Mbit and I ALWAYS use differential pair, what the h@ll is that PCB design??? Anyway, great review as always!

Great review of what would appear to be a somewhat early version of the product. Caveat emptor....

Around 36:30 ... what's up with the software? It starts to fall over when you reduce the vertical scale of channel one. Keeps stuttering and stopping. That's the sort of thing that would drive me nuts.

It kind of hurts to see that "instrument" connected to such a nice Agilent scope ....

Nice demonstration of the effects of undersampling, that alone made it worth watching you tear into this piece of crap. The only use I have found for these cheap USB scopes/generators is as a low frequency network analyzer, which this device doesn't even seem to do. I have the Analog Discovery from Digilent, 100Msamples 14Bit A/D and D/A converters. The network analyzer has a surprisingly good dynamic range. It covers below 10MHz where the regular spectrum analyzers do not go, so I use it to characterize control loop response. They are honest about specifications and provide full documentation and explanation of the circuits. I would recommend the Analog Discovery to anyone who understands the limitations and doesn't expect to use it as a replacement for a real scope.

- Wrong USB connector - Poor shielding and repair practices. - Virtually no isolation between channels for signal/data buses. - No trace length matching to account for time delays. - Specifications are not properly stated to be within actual operating range - Datasheet is too vague for initial block diagram design - Improper input impedances which cause amplitude skewing at higher frequencies - DC offsets as large as 2V exist which can damage components - BNC connectors relying on PCB solder mounting only without important stabilization which causes signal amplitude skewing. But at least the software is intuitive?

In all fairness a device like that is cool. It provides you with the opportunity to work with multiple instruments at a milli-fraction of the cost. Sure you wouldn't use this instrument to perform open heart surgery, but as learning tool it is ideal. I was surprised they sent you a modded out version for review, BAD choice on their part. But anyway, always a good video, it's too bad you had wasted it on something no value. Cheers

I think you are iranian cause your accent screems out that you are. :) loved your work!! Nice video.

I´d love to see what Dave will do with this piece of scrap. Please send it to him! :)

It's interesting that the Cypress USB interface chip is the same one used as the Saleae Logic 8 logic analyzer.

last time they send you a product. lol

what is the current best lower price spectrum analyzer for fft and very low thd in your opinion?

27:57 that's an interesting looking pale ale:)

send this to Dave on the mailbag so that we can have some laugh :D

would it not have been better to buy a couple of days of your time for a private review / issues list BEFORE they released this product!

Did they ever send a proper board back for an updated review? i was thinking of buying this.

Hello again would you ever consider doing a video on Evaluation and Demonstration Boards and Kits. It seems as though a product like the one you reviewed would be possible on a modded out dev board or multiple dev boards. I'm just curious about dev board and their limitations, they are cheap (maybe not so much). My question is that can a dev board provide the functionality of a dedicated instrument? Cheers

1260$ for THAT?! You gotta be kidding me.. the very advanced extremely good looking casing probably cost around a whopping 1$ to make.. I saw a comment somewhere (the manufacturer?) saying that just get it out of its casing made any calibrations go bad. Yea I´m sure that was the reason..

I wonder if they made a large batch of boards and then discovered many errors - perhaps from complaints by paying customers. They then decided that extensive rework was better than junking all those expensive components or trying to desolder them all.

Notice that they SLASHED the prices on their website after this review... LOL!

I'm confused... at 35:28 you say the device can't show a 270 MHz signal. Looking at the oscilloscope output, doesn't that look like exactly what you'd expect? It's hard to read your computer's screen, but it looks like the sine wave's period is roughly 4 ns. Also, in the future could you provide a screen capture rather than a camera view?

Almost all comments bashing this video were made using brand new youtube-accounts. I wonder why that is.

What a heap of junk! I'm amazed that you actually spent the time to fully analyze the board and review the whole thing! I would have thrown it in the trash the moment I would have seen what was (and wasn't) inside the case.

Perhaps too close a look Shahriar. I'd echo the comment regarding the board hacks. This is not a production quality part, or a review quality part; even it Analog Arts think it is. For the price it leaves a lot to be desired. Whilst they used quality components, they seem to have neglected reasonable mechanical/electrical engineering practices. As I think you said, compare this with the design/build of a Rigol 1054Z. Test equipment should not be a lottery.

I was quite excited at the beginning of the video, as I am looking for a reasonable priced all in one unit right now, however I am only 10 minutes into your video so far and I am really disappointed with the build quality so far, along with the shocking number of bodges and hacks. I will be interested to see how the performance plays out, but there is no way I can imagine this is going to be good for 1Ghz with that complete lack of symmetry and isolation.

I thought you were very fair and balanced. It would have been very easy to rip that product to shreds but you didn't. Yes, this may not be representative of their final board run. They have a lot of homework to do and I hope they are able to iron out all the glitches.

The fact that this is only USB 2.0 alone is a good indication that there's no way this puppy will sample a 1 GHz signal and stream the samples to the PC at a fast enough rate. Given that they use a 10bit ADC that would result in a maximum (theoretical) sampling rate of 48 MSPS (and that's only a single channel). This will be even lower since 480 Mb/s is the absolute theoretical maximum of USB 2.0 which is almost never reached and because they'll probably have to transmit 16bit samples since there's no 10bit data type. The Digilent Analog Discovery is a pretty good example of what is realistically achievable over USB 2.0.

I'm sorry, but this device looks like someone did a digikey catalogue search for nice parts, then decided to put it on a PCB and get a product :)

Shahriar, as I was saying below evaluation boards at a lower cost and higher specs. ANALOG DEVICES AD-FMCOMMS4-EBZ EVALUATION BOARD, AD9364 RF AGILE TRANSCEIVER™ 70 MHz to 6 GHz Operating Range, 1x1 Transceiver, Single FMC Connector, Supports MIMO Radio. Looks like this can be a scope and spectrum analyzer all in one package. There is software support by a multitude of vendors and the board maker. Am I looking at High frequency SDR, or am I getting it wrong looking for a low cost spectrum analyzer?

I just wanted to share my experience with an SL917 I bought from them. It may be against the flow of other comments but I am very happy with my instrument. Foe $290, it does a lot more for me than those other expensive devices. But whatever.

I can't believe they sell this junk!

Any word from the manufacturer?

Fair review. Product not ready for prime time.

This review confirm that all USB scopes are crap.

28 sloppy Analog Arts engineers disliked this

Nice insight to Chinese junk to measure and see it fall down on spec big time This is a 100Mhz piece of kit very badly made for an asking price of $1,260 Divide by 10 it is still to much Very nice presentation by the way regards John Davis UK

This is not a teardown but a destruction.

👍👍

Ouch! Was this video for a device or a manufacturer teardown?!

You knew he was going to take it apart right? Yikes! Maybe respin the board as rev 2 in the future. I think this is a prototype and they were eager to get some marketing. I haven’t watched the entire video yet, but all my money is going on - no way this meets its spec - we shall see. If it does I’ll buy two immediately.

Are you certain that this awful thing is coming from the original manufacturer? Because this has to be a joke, right? EDIT: now that I finished watching the video (I wrote the first comment when I saw the parts pileup at the start), and I still can't believe they are serious. The thing I learnt: I should have advertised my 20MHz analog oscilloscope as a 150MHz scope on ebay. Because I was totally able to measure the frequency of the 150MHz clock of some device (can't remember what it was) with it. The waveform was sinusoidal and not square as it should have been, and the amplitude was maybe a few hundred mV around zero, and not just positive 0 or 5 Volts. But who cares...

I feel your being way to nice with this one and it's causing the video to sound more like a sales pitch than a review, which I don't believe was your intention. Dave would have absolutely ripped them a new asshole if that was sent to him for review lol

WHOA! what a MESS!

Good review overall but I would have tested it first before tearing it down, that way the results gotten would have been explained better...

I almost couldn't bear to watch.

I can't believe they sent you such a poor example of the device. Bodged everywhere. If I had $1260 I would not buy this and if I did, it would have gone back, I would opt for a 'Proper' DSO I can't afford the price asked and have opted for a Velleman 60 Mhz Scope + Spectrum Analyser (PCSU1000) It's ok for £230 Thanks for the heads up here

Its very expensive to buy cheap instruments. Can't figure out why this thing exists.

Whoohah! Analog Arts, Amateurs without the Art you mean! If a company sends you a product like this then that company is run by a bunch of amateurs PERIOD!

Kinda acting as pedantic devil's advocate here, because this device seems like more trouble than it's worth, but: a) You say at 34:34, wiggling the mouse over the oscilloscope screen, "this will of course also give you the wrong value", but the reported F0 (just to the right of the "AC" signal coupling button) is 109.96 MHz, which is perfectly correct modulo a dodgy timebase. Furthermore, while you're increasing the frequency, the frequency shown in the oscilloscope display continues to increase and increase. Clearly Equivalent Time Sampling is being employed here with some degree of success, enabling them to reasonably legitimately claim a higher oscilloscope analog bandwidth than their sampling rate would appear to allow. The spectrum analyser fails, but its spec doesn't claim that it won't, so what's the problem? b) Contrary to what you seem to be saying at 14:08, Equivalent Time Sampling does not require a separate clock input, it only requires a reliable, repeatable trigger point on a periodic signal*. The output from your trigger subsystem is your clock, which is then variously phase shifted over multiple acquisitions to build up a higher effective sample rate signal. You could come up with some convoluted frequency-domain explanation on how each acquisition's nyquist images have different phases depending on the true original frequency, allowing the fully broadband signal to be reconstructed, but it's easier to just think of it as the time domain process that it truly is. Again, relying so heavily on ETS is a terrible idea, with all the associated noise and confusion issues that you quite correctly mentioned, but the oscilloscope definitely performs beyond 100 MHz on periodic signals. * Reference from Tektronix manual: www.dudleylab.com/TDS754C-Manual.pdf , page 3-27.

Sorry they claim to be around since 1992???? This is a JOKE! Their website smells very very dodgy!

This thing really has some "made in china" vibes about it.

Blimey, no ADCantialias!!!!!!!!!!!!!!!!!!!!!!! This is really really pissing me off that they have products around and offer them to the public. What a SHAM!!!!!!!!!! Unbelievable!

As this product stands, it is a complete piece of rubbish, in my opinion almost a scam. The specs are beyond misleading in my opinion, they scream about the 1GHz bandwidth and have a 100MSPS ADC on the frontend (possibly overclocked to 125MSPS) - are you kidding! I think Dave would have some fun with this :)

3 words: What the F*** ?!

What a pile of shit. The hardware designer obviously has no idea, just picking expensive components and having them assembled. The software looks really well laid out though. I presume a software guy did the whole lot. A similar product I have with a different problem is the red pitaya. which is a 50MHz 2ch scope and 2ch function gen as well as giving access to digital IO of the main chip. It's got good hardware that's more than capable of its quoted bandwidth although again there is some crosstalk, but the main problem is the software. It runs a very basic embedded linux (theres litterally about 5 terminal commands on it, my router has more.) running a web server and everything's done through a webpage it generates, completely useless. They haven't even gone to any effort to put the scope controls in a layout where you can see the output as you adjust it, they're all arranged down a sidebar that goes off the bottom of the page and it doesn't trigger in "normal" mode. Other programs have almost no controls except things like "reset zoom" and it is buggy. People are getting good results with it, but only because they're academics interfacing it to matlab. I'm just hoping someone writes a PC program for it like this one.

it's a lot of money for such a crappy HW.

what a pile of junk...I hope no one spends their hard earned money on this company.

review and teardown.... amateur video

This looks like absolutely horrible gear for the claimed specs. You are being far too kind in your review.

Fast Talker. He's in the wrong business.

There seems to be something fishy here. You spend most of the video criticizing fixes that were made on what's clearly a prototype PCB. Any engineer knows that this is a normal part of the design process. In the one decent experiment that you actually do after we watch your hands for half an hour, you completely misunderstand the specifications of the spectrum analyzer. The oscilloscope correctly shows the signal, and I bet if I could read the frequency indicators on the panels, that would be correct as well. Any electrical engineer with brains would know that this is some kind of effective time sampling, which is clearly listed on their datasheet (you yourself even comment on this!). Just because you don't know how they do it doesn't mean that they can't do it. Maybe instead of being baffled by what type of USB connector they use (seriously? they even included the cable) you should spend some time trying to actually understand the product you are reviewing! My best guess? You got an early prototype board that you damaged by disassembling it, causing all the problems like the DC offsets. Any reviewer with any semblance of self-respect would at least correct their mistake by contacting the company for a replacement - I bet you didn't even try to do that. Although with the ignorance you display in this review, it's possible you didn't even realize what you did. Either this review is ignorant (aka you don't know what you're talking about), or this review is not genuine and sincere (aka you intentionally misrepresented them). Unfortunately most casual viewers on here are going to give you the benefit of the doubt and not think for themselves.