EEVblog
Вставка
- Опубліковано 5 жов 2024
- Teardown of the new $399 Siglent SDS_1104X-U Four Channel oscilloscope.
A cheaper alternative to the SDS1104X-E
And what's all this magnetic shielding stuff anyhow?
Video on Standard Deviation: • EEVblog #1223 - Oscill...
Forum: www.eevblog.co...
Subscribe on Library: lbry.tv/@eevbl...
EEVblog Web Site: www.eevblog.com
The 2nd EEVblog Channel: / eevblog2
EEVdiscover: / eevdiscover
Support the EEVblog through Patreon! / eevblog
AliExpress Affiliate: s.click.aliexpr...
Buy anything through that link and Dave gets a commission at no cost to you.
Donate With Bitcoin & Other Crypto Currencies!
www.eevblog.co...
T-Shirts: teespring.com/s...
"Low cost" = $400 (Month's wage where I live)
That's exactly why people save up money.
Same with cars, same with houses, same with anything.
@@poptartmcjelly7054 Man, Idk about you, but I wouldn't call $400 cheap. Saving isn't a factor here.
@@anonymousarmadillo6589 Cheap is relative. If going by your logic, you wouldn't call a $400 car cheap?
@@anonymousarmadillo6589 when there's nothing cheaper this becomes cheap in comparison.
I make about 2000EUR a month after taxation and stuff and I do consider 400 alot of money.
It doesn't matter how much you earn but how you respect them, that's way multimillionaires end up bankrupt because their bank account is "endless"... in some cases.
I think after all these years a new budget scope shootout is warranted, see if the Rigol is still the king
Yes definitely!
I'd love to see Dave do a shootout of the $400 range DSOs actually
Dave, Hello,,, its time for a new shootout.. i know how you love them...lol
@eevblog Yes definitely!
Please do another scope shoot out Dave, we need you to compare them! And perhaps Keysight will refresh their scopes soon, i hope? That would also be something nice to look forwards to.
love from Kenya bought sds1202x-e after watching your channel loved it
Please, more comparing two or more circuit boards of the same brand, with your excellent and always clear explanations on the details of the components and their purposes and effects on the circuit boards... You are ONE OF THE UA-cam KING OF ELECTRONICS and always thankful for your great work.
So much crowding at the lower end: A very nice thing to see!
I'm also blown away by what's happening in the low-end of the spectrum and network analyzer market, especially the OWON XSA 1000 series.
I got my butt kicked at work while trying to get a new system through EMC certification. Had multiple issues below 1 GHz that none of our prior systems had exhibited, and two nasty spikes between 5 and 6 GHz. We have never done in-house pre-scans before going to folks like TUV SUD for a dress-rehearsal pre-scan then a qualification scan. This time did THREE pre-scans before we were ready for our cert scan.
For the cost of just ONE external pre-scan, we can equip a surprisingly decent in-house lab for 1 GHz and below, including an EMC-capable spectrum analyzer (the OWON), EMC software (form Tekbox), antennas (from Tekbox and others), LNAs (from MiniCircuits), and a set of near-field probes (Tekbox).
I would love to see a comparison to the classic 1054z
Especially in UI responsiveness.
Please do a comparison video on FFT in SDS1104X-U to SDS1202X-E
this!!
Wouldn't it be more logical to compare SDS1104X-U to SDS1104X-E? 100 MHz 4 ch. vs. 200 MHz 2 ch. would depend a lot on what you use it for.
@@werediver Yeah, what you said is logical. But the FFT configs in 1000X-E series are same also I have 1202x-e.I don't have 1104x-e. Also the bandwidth in 1104x-e is just firmware limited and the ADC distribution per channel in 1000X-E series is same. So while typing I didn't give much importance for the bandwidth and number of channel.
Dave's "get off my lawn" moment: asking whatever happened to video ratings. A change that was made back in 2009.
Your mom used to rate me.
@@EEVblog Oh shit! LOLOLOLOL. Yes! more "middle school disses" Dave, please! You got him, Dave... You got him good. 🤣😂
Recently on the main UA-cam page it asked me to rate a recent video I had watched from 1 to 5 stars, and even click some tags to describe how it was. Never seen that again.
@@gblargg I've had it ask me to rate suggestions like that a few times. Thing is, it keeps asking on videos I haven't watched yet...
@@EEVblog OOOOOOOOOOOOOOHHHHHHHHHHHH!
So, you get what you pay for. Makes sense. Thanks Dave as always!
Companies are no longer allowed to use stickers to void warrantees here in the US. You’re allowed to open the product to look inside. Modifying the product is still a no no for maintaining the warrantee, of course.
I thought that hasn't been allowed since the seventies, but the FTC has just issued a clarifying statement in 2018. And apparently this only applies to products over $15.
Rigol 1054Z shootout please!
They now use the same simple "Calibrated" sticker on the SDS 1104X-E as well (mine is almost 2 years old).
But it is still Calibrated! 👍
The IC in MSOP above the OPA4872 is LMH6551
All those variable resistors look to be in a suspiciously similar position for something "calibrated" :p
lol u right :D
That is interesting, huh ?
Yeah, but what and how? Don't recall seeing a pot on a front end before.
@@EEVblog maybe just a function of the cheaper components needing a little tweaking to get them into spec? I only have a cro myself, and that does have trim pots in the front end/attenuator to adjust the 0 offset of the input, about 5 separate ones in fact XD so all the different pieces would work properly-magnification, inversion etc.
Didn't know the modern ones were missing them.
Man scope model numbering is as bad as Intel CPU naming. Also they are all wanting a piece of the 1000 model number pie. Siglent and Rigol both playing the 1k naming game trying to confuse everyone.
Would love a shootout between this, rigol, and other options. Thinking about getting a scope soon.
Would like to see a video on if your recommendation has changed from the Rigol 1054z.
I just ordered a 1054z yesterday, just got the tracking number watching this video. Hope I made a good choice!
@@alanduncan3710 It will serve you well.
@@EEVblog so excited!!
I own a siglent 200MHz and have for 2 years. It works great.
The Rigol DS1054Z be like: Finally a worthy opponent!
Totally 😀
Its old at this point. No comparison.
@@Okurka. Since a year or two Rigol is selling the DS1054Z with all software features unlocked including the higher bandwidth.
@@Okurka. Actually true, my bad. The bandwidth upgrade is not included. Most other options are however and i think you can still install it.
@@Okurka. Yes Riglol works. Installed it a few weeks ago. Thats why i was wrong. I did not check which features exactly were installed when i received it. Just saw that 'all' the features had been unlocked. But installed Riglol anyways.
The mag shield would be to keep relay transients from displaying a glitch. Would not do anything if the relay were operating.
Got a rigol 1104, that I bough just recently for £400 and I’m pretty happy with it. By now all software bugs are ironed out and all features unlocked
I just bought the 1104x-e a couple months ago. Saw this video pop up and thought "man, please tell me they didn't come out with a cheaper UPGRADE!?"
Whew, thanks for the teardown. Still happy with my purchase.
Just bought the E a few days ago after comparing the specs with U. Didn't see a point to downgrade to save 78€.
What's the difference between siglent sds1204x-e & sds1204x-u
"Let's void the warrantee on that bad boy"
Apparently in Yankee land simply opening the product can't be used as grounds to void the warranty.
@@gblargg Yup!
Bought the Rigol a few months before this video dropped. Sure the Siglent has a bigger screen but honestly, I prefer the 'encrusted with buttons' look of the Rigol.
It's much more impressive looking. Which is really all that matters.
I am kidding, but seriously, I much prefer buttons instead of trying to remember which menu to click on or hunt through touch interfaces.
The 'Measure' button is always in the same place on the 1054Z.
When I outgrow it, it will be demoted to Lissajous duty for my home studio.
Dave, I'm so glad you took one for the team with that sticky magnetic shielding - I was just about to get a roll and stick it literally everywhere on everything I have 😂
Interesting scope. Honestly, a 100MHz is a very good place for a modern scope to be fair.
The Magnetic shielding in the channels seems rather tiny to be fair. Maybe they just had issues with RF bouncing around in the can and stuffed in an absorber to handle that. Could stuff in a sine wave and do an FFT and see if the noise floor changes. Or if any peaks and such disappear.
If only the scope had a trigg in.... I don't like offering a channel for just a simple trigger.
Also, were sitting thinking, "I would really like to watch a scope teardown from Dave..." And were going to head over to the channel to look for one that I haven't watched 3 times already...
Guess I were lucky.
Perhaps the magnetic shield isn't to prevent interference from getting in, but rather getting out? Try switching the range to make the relay activate and see if it couples to other channels with and without the magnetic shield.
Oh no! I've got to say it......... "that's not a knife"
I was just about to buy a scope for the first time in my life, and I was going to go with the 1054z. Seeing this though, and looking on the specs and through the manual, I'm pulling the trigger on buying this one instead. It's 100 MHz over the 1054z's 50. I also like the way it's built in serial analyzer functions work compared to the 1054z. Being a newer unit means it's likely to have the firmware updated for longer than the now long in the tooth 1054Z.
Relay has a big coil, the shielding on top shields some of the EM fields, but not magnetic fields. I suspect they put the mu metal on top to prevent magnetic pickup of the relay coil, which can then potentially couple into the ground/power supply lines. Origin of the magnetic field is probably from the power supply as you mentioned.
On the off chance the weird transition at 16:48 was intentional and not an encoding error or something: I liked it!
Wow, no idea what caused that. It's in my render, so it was Vegas doing that.
It’s funny that you mention rating videos at the end. UA-cam has started doing that again, but it’ll appear randomly in my feed. As I’m scrolling, it’ll prompt me to give a rating out of 5 stars to a video I’ve previously watched
Me with my Tektronix 2246 100MHz 4ch scope, looking at this 🤯!! Maybe time to upgrade 🤔
@Repairman i agree. When I upgrade, it'll be something better than this. My Tektronix has given me decades of trouble free use. If it had memory, I'd probably hold off on the upgrade path right now.
TI's digikey pricing is super high on TI's website you can get that for less than $5 and I am sure if they are buying tens of thousands they are getting even better volume pricing.
I actually bought this as my first scope about 2 weeks ago! It works really well for my beginner-level applications, it's easy to use and quite capable if you don't really need high bandwidths all the time. My only problem with it so far were occasional hangs at boot, where you need to restart it for it to start up fully
Sounds like os issue
I think the performance for the money is incredibly but I just cannot get over the build quality. How can they design such amazing electronics and still build wobbly knobs!
I have Siglent 1104X-E that has the same knobs as this one and nothing is wobbly what are you talking about..
A whole bunch of EE students just got really moist.
PGA in the older front end was an AD8370, not AD8350. Not a huge price difference, but I figured I'd mention it for the record.
All thumbs up Dave 👍. Great overview. Interesting to see the changes but I’m not sure they have saved that much on the BOM there to cover that saving. I’m guessing there is a cut in dealer margin as well as often entry level products expect the dealer to take the hit as well on margin to bring people into the brand. The Rigol must have been getting enough traction over them to bring this to the table.
Would be interesting to see if the new X's come out with that optimised power supply from the U.
New modern shootout. Yes please, and thank you.
LoL you can't even get toy guns in Australia! Does using your finger gun's count?
I chose to get the GW Instek GDS-1054b for about $100 less. I like that it has dedicated controls for each channel, and you can hide the menus to use more of the screen for the waveforms.
It has half the bandwidth (50Mhz).
$100 cheaper with no bode, shared sampling, Lower quality front end, no wifi, no network update, less memory????. No thanks.
Where's the rust?!
A long time viewer. Yolo.
"Thanks Siglent for sending me this."
"We're going to do a teardown to see if we can hack this to get more bandwidth without paying for it."
I'm sure they told themselves that it was for R&D, as they swallowed their antacids...
Was that really magnetic shielding tape, or is it thermal transfer tape to help transfer heat from the relay to the case (relay coil in enclosed box may make it heat and cause more front-end drift)
Thermals was my first thought too.
Very rarely would a thermal pad be magnetic, and being near a relay which generates and absorbs magnetic fields, it's almost certainly not for thermal effects.
I'm not going to remember my magnetics in detail at this hour, but I'm thinking that the magnetic shield may be only relevant when the case is completely closed. That's when the switching power supply is going to be closest to that relay coil. My thought is that it may be inducing a voltage in the coil that somehow backfeeds through the coil's transistors into the main circuit.
Just a random thought... Could it be the distance between relays and power supply that makes a difference and begs for magnetic shielding? What will happen if you assemble the scope fully without magnetic shield and run the same test?
The price difference here in Europe is a bit too small, the SDS1104X-E is 498 Euros while the SDS1104X-U is 417 Euros, thats a delta of 81 Euros.
I dont know if this is worth it..
i think the shielding for the relays is for when the relays are active and injecting signal into other channels
This is my first assumption too. To reduce the coupling through the electrical shield. But the answer from Paul Ste. Marie below makes me doubt... disturbances induced by switching power supply might be not negligible. More tests would be required to conclude ;)
Personally I would pay the extra $100 to get the E model. Obviously they are trying to compete with the Rigol 1054Z on price.
In comparation to X-E ($500) this one should be like $300 max to consider buying this X-U.
I got the 1202X-S. It's okay. If you don't use logical analyzer and you want a function gen, then it's great for that.
Standard deviation of the standard deviation is absolutely a thing, but it is more about measuring the consistancy of that standard deviation...
I've never heard the word standard deviation used so many times lol thanks for the video great job as always happy holidays
I have the 1104X-E after your review Dave, fantastic scope!
Also bought 1104X-E with discount for $400 + tax. Winner winner chicken dinner!
@@NilsJakobson Me too...
That's not why the magnetic shield is there.
I expect that the relay is a crappier one, that'd tend to magnetise the shield cover over time when pulled in for long enough, then would probably fail to release reproducibly.
100 Mhz scopes are good to examine signals up to 20 mhz with good fidelity. Daiyve should do a video on scopes capable of dealing with today's gigahertz electronics.
Reducing the BOM cost? Come on, digital oscillospopes are all profit margin :)
The extra $50 could be worth it to get 100MHz out of the box vs. having to get the code to enable it on the 1054. It will be interesting to see how the two scopes compare in features. A lot of features in the 1054 require codes to enable them. If many (or all) of the code enabled features of the 1054 are included in the Siglent without needing to pay extra for them it could be serious competition for the Rigol. Looking forward to the video where Dave turns on this puppy. :)
External magnetic fields can result in depression or elevation of the pull-in
and drop out voltage of the affected relay, possibly causing them to
fall outside their specification.
400 for this scope is actually pretty good.
Given scope prices today are 3-4 times higher at the least.
It's all the scope you need for most electronics repairs..!!!
Good video Dave.
Sadly no Bodeplot function supported. I love that function in combination with my siglent AWG. This is something you seldom see on budget DSO's.
They have to have some product differentiators.
About rating videos, I occasionally get asked (by the UA-cam) to rate a video, watched a Big Clive vid the other day and about a day later got the rate this vid question from UA-cam. Not sure how it works but still a thing.
Something is off, RMS should never be less than the DC offset?
Could it be that when RMS was 250, DC off. was less than that? And when Dave measured DC off. as 300-400, that was after some time? He never showed DC off. and RMS in digital form at the same screen.
The DC did shift around a bit as I was using it.
imho - SDS1104X-E features looks way better then SDS1104X-U... just for extra 100 bux
Such a product engineering is actually quite complex. For once, what would be the specs of the new machine so that we can build it cheaper but GUARANTEE the match for the spec. Thats the hard part - to make the product working as advertised. So simplifying the feature set while still remaining useful. Pretty clever thinking there.
Is it worth purchasing just to save the $100, or is the more expensive model worth the difference. After the teardown, I tend to think the latter. Thank you.
Siglent should move this scope to a tablet touchscreen format, that would reduce the cost to perhaps $US199 with a USB interface and external wall adapter. All test equipment are gradually moving away from mechanical switches to compete touchscreen controls
Ehh, I really prefer real knobs but those days are sadly gone. I don't think removing the current encoder knobs would reduce the cost much, quadrature encoder switches are cheap.
Oh, and there's one huge advantage of real controls over a touchscreen: the ability to hold your finger on a button. This seems invaluable to start/stop buttons for example, so you can quickly time to a certain event.
I like mechanical controls, but they take up valuable space, get dirty with use and add to the BOM and assembly time. I understand next generation of Tektronix scopes are going largely touchscreen with large color OLED screens
Good thing I'm getting to old to care what the next generation equipment will lack. I appreciate tactile feedback meaning I prefer real knobs and buttons. Nothing I hate more than rubbing my fingers all over a screen that I'm trying to read and keep clean. I will never buy a tablet scope.
Why are oscilloscopes so expensive? Should they be? Does it really COST so much to make them?
Yes, you see what hoops they have to jump through to save $100. ADC's, FPGA's and fast memory do cost some money. Let alone all the R&D that goes into something like that. I mean, the amount of scope you can get these days for a couple of 100 bucks is already insane! Large memory depth, fast high resolution FFT's, bode plotting etc. These capabilities are insane! My 1104X-E has been one of the best purchases I've done.
as someone who has been a EE since the 80's, I've seen good scopes cost $40,000! Even decent ones were around $10,000. A $400 scope is barely more expensive than a Fluke meter, which makes it a great bargain. I just bought a Rigol 1104Z last year and love it.
Of course not, my best guess for this scope bom cost is $150, the rest is RnD, advertisement, etc.
This oscilloscope is expensive?? Are you mad? Its dirt cheap for such a precision instrument! Actually I am surprised they can achieve prices as low as this! Look how much is a precision multimeter from Fluke! And thats just a multimeter! Look how much is a good quality desktop multimeter and youre complaining about scope that costs about the same money?
I guess if everyone needed an oscilloscope to survive the prices would be closer to cost.
I am disappointed!
No siglent trademark rust?
Don't bother buying it! 🙄
I am thinking the magnetic shields on top of the relays is to address radiated emission levels well above the bandwidth of the scope. I have used stick-on ferrite shields on ICs to bring products into CISPR radiated emissions compliance with this method. BTW, first time watch on your channel. Excellent job mate!
I would be sad if I buy this one and then I see this comparison.
But Dave you should have put the psu in the assembled position to see any difference in the noise figure. Otherwise the psu magnetics are too far away. Anyways this scope is relatively unimpressive compared to the -E version.
I would never buy another siglent scope again after they sold scopes for almost a year after knowing they had defective compensation circuits
Damn, Dave, you are turning people into oscillloscope junkies :D
Ohh, I was just to order a new scope for Christmas... hmmmm!
Can you summarize the video for me lol? What's wrong with the new one?
Maybe the tape on the relays contacted the top cover for heat dissipation or prevent vibration? Magnetic attraction?
Wonder if the magnetic shield is there to prevent the relay's own coil 'glitching' its own channel or the one next to it?
Those jumper links on the PSU are a footprint for a switch.
Nice enough 'scope* but I think I'll probably save up a little longer for the X version.
It's amazing what you get for the money now, my first 'scope was a single channel Cossor of a few MHz bandwidth and felt like was built on the chassis of a WW2 Russian armoured fighting vehicle. I had bugger all idea how to use it.
Kids these days...
Probably you see in the FFT some spurs without the magnetic shielding.
Interesting to see the RA4 and RA8 on the lower cost version to be at an angle and not 100% horizontal/vertical. Isn't that more difficult to manufacture? I mean the part between 14:15 and 15:00
I a,ways thought of silent as a high end brand that was beyond my budget, hell I gave almost $400 for the 1054z.
So still the Rigol DS1054 then.
Cool stuff..I am learning alot hehe..thanks a bunch
Loud fan, seems minor but my Hantek I had was silent.
14:50 So it looks like the cut-down/redesigned one-way front-end in the SDS1104X-U has better quality signal quality than the other two? The component savings went into a super-expensive mux chip?
19:20 Power supply, bottom left: might be a 110 - 230 selection.
Thanks
Maybe the magnetic shielding is more for stabilizing the switching behavior of those cheapo relais when the board is in the (default) upright position? Is the magnetic field of the tape counteracting gravity somehow?
Just a thought. Cheers!
This comment begs the question, why are you looking at a scope. You def not gonna be using one.....
"Floating head Dave" has improved.
8:44 What's up with U44? Upper right corner. Is it chipped?
Hey 👋 i just bought the Siglent SDS1104X-E and i just now learned that you can hack it to be 200mhz!! How can I do that???
Search for the hacking thread on the eevblog forum.
With Afterpay etc options why would you try to save 100 smackers.. 💯
didnt the rigol ds1054z had more math functions? i missed some on the 1104x-e
The prices are really good these days. In 2014, I got crappy Hantek MSO5062D (2 channel 60MHz with small 800x480 LCD and crappy slow and horrible interface, + 16 digital inputs; 1Gsps; 1Mpts, no networking) for 520$. Sure, it sounds on paper similar, but Siglent new UI is so much nicer and faster in general, has more functions, better maths and statistics, more wafeforms per second, better triggers, better noise, etc. I should probably went with Rigol DS1054Z , instead of Hantek MSO5062D, but well, I tought it will be decent.
Rigol vs. Siglent when?
I was hoping it was AU$399 dang
Anyone noticed that all 4 of the trim pots are all perfectly vertical...like straight from the factory? The sticker should say "Calibrated".
Good rubish for aficionados 👍
Isn't the point of the 10X samples to bandwidth the fact that the filter in an oscilloscope ain't particularly sharp- its only 3db down at the quoted bandwidth.
I.e. It will let through a lot of signals that it cannot reconstruct accurately without aliasing. 10x just means enough db down on signals that will cause aliasing so their effect doesn't muck up the bigger picture too much?
I would love to see a shootout in terms of automation via SCPI ... I do this a lot and its an area the DS1054 lets me down :( ... If the Siglent's have better SCPI comms it might be worth a change