Inside the Siglent SDG 5000 series SDG5082 80MHz Arbitrary Function Generator. Same as the SDG5122 and SDG5162
Forum: http://www.eevblog.com/forum/blog/eevblog-497-siglent-sdg5000-function-generator-teardown/'>http://www.eevblog.com/forum/blog/eevblog-497-siglent-sdg5000-function-generator-teardown/
Photos: http://www.flickr.com/photos/eevblog/sets/72157634773277340/
Datasheets:
PLD: http://www.latticesemi.com/lit/docs/datasheets/cpld/DS1002.pdf
TCXO: http://xinwei-e.com/cn/product/TCXO/BT0507.html
USB: http://www.stericsson.com/technical_documents/CD00264885.pdf
SRAM: http://www.cypress.com/?docID=41873
DAC: http://www.analog.com/static/imported-files/data_sheets/AD9780_9781_9783.pdf
VCO: http://www.analog.com/static/imported-files/data_sheets/ADF4360-7.pdf
Driver: http://www.ti.com/lit/ds/symlink/ths3095.pdf
Isolator: http://www.analog.com/static/imported-files/data_sheets/ADUM1410_1411_1412.pdf
PSU: http://media.digikey.com/pdf/Data%20Sheets/NXP%20PDFs/TEA1610T_P.pdf
Comparator: http://www.analog.com/static/imported-files/data_sheets/ADCMP561_562.pdf
OPAMP: http://www.alldatasheet.com/datasheet-pdf/pdf/394398/INTERSIL/ISL28210FBZ.html
EEVblog Main Web Site:
http://www.eevblog.com
EEVblog Amazon Store:
http://astore.amazon.com/eevblogstore-20
Donations:
http://www.eevblog.com/donations/
Projects:
http://www.eevblog.com/projects/
Electronics Info Wiki:
http://www.eevblog.com/wiki/

Hi Welcome to tear down Tuesday Today we've got a brand we haven't torn down before. This is a Sigin Sdg 5082 or 5000 series function function SL arbitary waveform generator. It's quite a decent uh bit of Kit goes for over four digits, so let's not muck around. let's take a look inside.

No. I haven't even peeled the Uh tape off the front, haven't even powered it up after taking it out of the Box You know what we say here on the EV blog? Don't turn it on, Take it apart actually I will stand corrected on that uh 4digit uh comment or over $1,000 Yes, it is over $11,000 for the $60 MHz version, but this is the 80 MHz version I Believe the only difference is the um, you know, maximum output uh frequency. The sample rate 500 Meg samples per second is uh, the same between units and uh, so so this one the 5082 um, you can get this for under uh 600 bucks. uh, street price I think it goes for um, just slightly over 800 uh from the official rep here in Australia that's in Australian dollars with the bit of a weak Aussie dollar at the moment.

but uh yeah, quite an affordable Uh function gen actually. And of course it uh competes directly with the Uh Ry Gold DG 4,000 series which we've seen in quite a few videos now, but this one they claim they one up the Ryo with it with this easy pulse technology in this thing apparently. so uh, which allows for lower phase noise lower uh SL Jitter and uh, greater resolution on in terms of being able to set the Uh duty cycle. the pulse width uh, stuff like that so you can get really fast edges apparently at low uh output frequencies.

So anyway, um yeah, it does have USB This isn't going to be a review of course, but but uh, we haven't seen this before so we probably shouldn't uh, probably should just give a minute here to uh, give a brief overview I have not powered it up as I said haven't used it so I don't know how it operates clearly. These Uh buttons all light up all the soft buttons. Very small little keypad here so it doesn't take up a huge amount of room. nice indenting on the big knob there I Rather like that and yes, it's pushable.

It's got a secondary uh select function. really nice channel one, channel two I believe the second output has uh, much greater memory I think six. don't quote me. 16k on Channel 1 and like 512 on channel 2 or something for some reason.

Lots of soft buttons along the bottom, big graphical color LCD display on the thing and a couple of Uh mode buttons and well, that's a bit. Obviously Got a fan on the side so it does heat up a bit and on the back we've got a USB uh device external uh trigger gate FSK burst multifunction Uh, we've got modulation in, we've got syn out, we've got 10 MHz external, 10 MHz reference in excellent, so you can hook it up to your Rubidium standard or something like that. Fantastic. and the uh reference output if you need to synchronize the internal reference with something else.

But apart from that, it's pretty. Spartan Universal AC input voltage There is one thing though that irks me a bit. just these little tacky gold QC Pass QA Pass by you know, tester number 01 and QA operator 01. It's just.
you know, they're the sort of type of exactly the look and feel of the stickers you get on. just the real cheap, crappy stuff that comes out of China it it, you know I Know it's only minor, but it just gives me that impression. But anyway, the unit does feel very good. So uh, we got no problems whatsoever.

The tilting baale actually. uh, it's not too shabby. So anyway, let's see if we can crack this open. It's got.

uh, these rubber surrounds. There we go. They just pop off so they don't Looks like they don't hide any screws or anything. Um, they could actually come off I Probably prefer.

um, like the full rubber boot like you get on. uh, the Agilant gear and stuff like that. But it looks like we have two screws here and it looks like that back will pop off and then the whole shazzy there will just slide off. warranty void if broken.

We can deal with that no problem whatsoever. Yeah, there are methods to actually, uh, get those off. but big whoop. So let's crack this open.

We got a standard Phillips here. now we haven't by all accounts they're not. uh, not too bad at all. So expect a reasonable um, standard of quality in terms of the design and the uh, build quality of the manufacturing quality of the PCB they must have captive screws in there I think no, doesn't it just popped out and all.

Better pick that up before I lose it and it looks like is that going to slide out Or have we got another one holding it in? No. Maybe we have to take the handle off. No. Turns out that we didn't.

Um, there we go. Got some nice uh, they've got. they've remove the paint there just so that they can get some extra shielding on the case. That's rather nice and I can see some nice RFI gaskets here I got it upside down.

Oops, all the electrons will fall out. Got to be careful. Oh that looks really nice inside, but there we go. Got some very nice Um gaskets there.

nice attention to detail for the uh Emi shielding on that thing and this folks looks really, really good. Now at uh, first glance I was going to say I don't mind uh, the build, the design and layout and build quality of this thing and well, yeah, I don't mind the you know modular board, uh construction and and uh, stuff like that. It looks, uh, fairly neat. they've you know, they've got metal shielding in the middle.

here. they've got, uh, clearly the um, the Main's power supply on this side here going across all separated. nice big mechanical clunking switch which I don't know if that's the real yeah, it looks like the real deal cuz we got Main's wiring running over here and going like that. anyway.

I was going to say I don't mind that and I don't but the first thing I noted was the cheap of the metal in this and then I looked closer and even from a distance I'll get my macro lens out but it looks like the Shazzy is Rusted along here and here. a there you go look at that. That is just truly awful. It really is to get a factory fresh unit straight out of the box and then have you know, rust on top of the damn thing, have to get a screwdriver and try and get that a scrape.
But that is just truly horrible. And it's not just in one spot either. Let's go to the other side and there's the other side. I mean look at that.

it is just horrible. It really is. Oh man, that's just. ah, thumbs down.

And then the rear panel. they've just clearly just welded that on instead of you know, screwing that in place. So it just it just looked, looks and feels really cheap quality. I Don't like it at all now.

I Know diddly squat uh about welding. So uh I clearly can't comment on that. So maybe someone with uh World well, any worlding experience might be able to comment on that, but it's just so cheap and nasty. And combine that with the Rusted shazzy.

You got to be kidding me. And they've done the same on the front panel. here. you can see the rust on the uh, left hand side side of the connector there on the side.

and then they've just, uh, spot welded those uh tabs there. I mean it's all actually folded up from the bottom of the shazzy, but they're OB um, it's all one sheet from the bottom and folded. but then at the uh top of the shazzy up there, they're just spot worlded together. Then if you have a look in there, at the cable uh, penetrator that goes through the front shazzy for that ribbon cable there, it's just, ah, just punched out and folded and there's some rust on that too.

Looks horrible. so you can see what they've done. This is all one sheet and they've obviously cut out all of this bottom for the Uh PCB here. uh, not entirely.

Uh, sure why you would need the bottom on their exposed actually. Um, but anyway, they've decided to do that and it's folded over here. They've obviously just put the uh uh Deco sticker on the back there, so they folded that down, folded it over there and then they've uh, spot welded the the sides and also at the Uh front here as well. and I can't really get the camera angle in there.

It's uh, very difficult. but right up under there you can see a huge big uh, scrape or file marks or something plus more rust. Well, let's forget about the shocking rusted Shazzy for a minute and have a look at uh, the electronics as in depth as possible because uh, it actually does. ADD First glance, it looks really good, the soldering and the construction and the design quality.

As I said, they've clearly got like all the major uh, all the major processing and uh, you know, display stuff and uh, logic stuff. most of majority of it, uh, happening on this board over here, which is opto isolated. There we go. we've got some Analog Devices uh isolators.
We'll take a Serial isolators. We'll take a look at there so they're jumping that. That's why they've split the boards in there. You can see that the output uh, B andc of course they're actually uh to the input or the output.

Um, the output side here. so they're actually reference to there. So the ground reference on the uh, these inputs and outputs on the rear terminal of course will be ground referenced to the main function gen outputs there. but uh, the USB of course will be is isolated because that's all part of the main uh processing circuitry in here and then isolated by our Analog Devices isolators there.

of course it's got its own separate power coming across here from the main Uh board and then the main output board here of course has it all. Has looks like multiple supplies of course coming from our power supply over here and our power supply over there which you take a look at looks really good, but it's really difficult to get down in here and take macro shots of the chips and everything uh without taking the boards out. So uh, I'm going to have to do that. I'm going to have to basically strip this thing completely down basically the screws uh down in here so I'll have to do that.

We got some nice nice little right angle uh board Mount uh. coaxes coming off here. really good to the output uh terminals so that looks really quite nice and well laid out. so I'm quite impressed now.

I don't mind what they've done with the Uh Earth in here at all. You know they've taken it through here. they've got it up to here. It's nicely, uh, crimped and uh with a you know a slight crimp connector going on to there.

It's all rather uh, nice and I like the layout of course of the board in there which will go into, but of course this is a PCB Mount I connector. So the first thing I think of as well, there's through hole pins going to the bottom side. how much clearance to the case? Now if we have a look on the bottom side here you can see there is bugger all clearance there and at first I thought oh my goodness that is just you know Awful. It's going to short out and everything but the uh but because this is a separate physical board in here it looks like it does look like they do have larger standoffs in there so that's sort of.

the top of the board lines up with there so it's going to be 1.6 mm below that and then you're going to have the pins extending out. Then you got the thickness of the uh sheet metal which is really thin and crap and uh, rusted of course. So I don't know it may have enough clearance in there. but I hope they've put like an insulating sheet or uh, something over that.

Actually just going back to the Rusted shazzy here for a second you can see like down the side wall. it's almost as if like they've uh you know I don't know all my sheet metal uh, mechanical, uh, processes and stuff like that. but to me it almost appears as though they've tried to like. You know deur this thing or something.
You can see all the scratch marks along the side of it and that is just, uh, you know, caused it to all rust and uh, as I said, I can see a similar thing up under the front panel. but uh, even worse than this. Well, I've got the board out and this really is, uh, quite a nice Peak piece of work. I'm relatively, uh, impressed by this.

As I said the big clunking, uh main switch over here. They've even got nice cutouts in there. high voltage isolation slot routed out of the board in there. Very nice indeed.

so we'll whip that off. and uh, that just only gets in the way. And of course, here's the here's the secondary side and we've got some really nice, uh, sastic. Somebody's gone to town with the uh, hotnot gun there and they've uh, sastic, all that down.

We've got some inductors in there, our output caps. We'll take a look at the brands, there's our output dodes there, and uh, they have. You know, it's designed very well. We got our isolation slot under our Opdo coupler.

there. There's our reference circuitry down in there, and really nothing else on the Uh secondary side there at all. There's our suppression from the primary side to the secondary side and uh, now I Was going to say all the isolation was, uh, quite nice in here here, but you'll notice why they've sort of. you know? look at that.

There's not much clearance in there. Um, it's probably going to be good enough, but look, they've gone to town to, you know, remove all the uh, copper fill out of here and do the nice isolation slots in there. Yet They ruin that by bring in this. Trace I Mean they could have, you know, snuck it sort of down this side and around to the uh uh suppression cap there.

but they haven't. They've just look that's pretty darn close and they've filled that in. They've put a big Square in there. Don't know why they've done that, they've blown it.

but it's you know it's good enough. But yeah, they just missed that little attention to detail. Then we have nice little touch. We've got our isolation slots between the individual pins of the Uh driving transistors on the primary side of the Transformer here, so that's a nice touch.

They knew what they were, uh, doing there, certainly. And also, they've done isolation slots in here for the main diode Bridges as well. and on the primary side here, they haven't skimped at all. of course.

Slow Blow fuse protection. Once again, Celastic down. They got a full cover over that. They just haven't left.

its uh, ass flapping in the breeze there. we've got a themister. we've got Mau protection down there. We've got um filter in common mode choke.

It's all happening. Very nice. no problems whatsoever. The main cap is a Rubicon 105 CR rated 400 volt 220 M MX uh series.
You know it's not a one hung low cheapy. They've chosen a reasonable brand there for the price and for those playing along at home. The main controller is a T an Nxp, uh Ta1 610t and that's actually a Zer volt switching resonant converter controller and that's exactly what they're implemented here and it's hard to see, but these are the output caps are 105 C-rated Uh Rubicons as well, so they haven't skimped and if we get our depth gauge on the post in there, there, we're looking at uh, just over five 5 mm clearance and if we have a look at the depth of that pin on the board almost, it's about 2 mm so that 3 mm clearance is going to be uh adequate there to the Shazzy for uh, normal use I Can't remember the RS uh, the exact values from the regulations off the Uh top my head, but yeah, it's probably going to be good enough I Really would have liked to have seen some an insulation sheet under this Uh board. between all of this uh Main's input circuitry and the Shazzy though, and they've gone to the trouble to actually hold this ribbon cable down onto the Uh flat Flex connector down there with some tape.

Look at that, they've actually taped it all the way down there which goes onto the connector. so to get that out I can't just lift the tabs as usual I've got to actually peel off all that tape. How annoying. And we've got the main processor board out Does look quite good as I said.

and uh, let's take a look at some individual devices and the soldering quality and no surprises on the main processor front here. Analog Devices black Fin DSP which we see time and time again in all of these. uh embedded Scopes the Riol ones and all sorts of Brands we uh, you know, know and love they've got Blackfin DSP devices in them Analog Devices must be laughing all the way to the bank and right down the bottom a populated JTAG header connector. just rare to go hack away and at first uh, check I can't really ID that Texas Instruments Part there.

Oh, clearly it's the main uh clock driver PLL so you know 25 MHz main Crystal generating the higher Uh frequencies required for the system operation and the only other major device on here is this Lattice semiconductor. Mark XO PL It's not even a CPL it's just a a sort of a low-end p. Although these PS are actually quite fast, this one will do 3 88 uh mahz. it's only got 640 lookup uh tables.

It's got 320 macro cells. It's got 6K of uh, high-speed SRAM in there. but you know that's about it. And here's a classic example of where you would use a CP or PL over an Fpga.

For example, you just need some sort of uh, you know glue interface logic between the main uh, black thin processor. Here we've got our uh main memory in here. You can see the tracers going down here, so obviously it's doing some sort of memory uh, controlling there. It's obviously uh, got data linked into the processor down here, but then it goes off to uh this main connector which is all the serial lines which go over that Optical interface.
We saw back at the start that uh controls all of the output so that is all the serial interface for that sort of stuff and it also controls look all these traces all tied into the front panel stuff. so it handles all that. So it really is is a big part of the system. glue logic here and then coupled into all that.

We've got our HX Dam here. fairly high density uh one I think it's 4 Meg by 16 bits and our spansion flash memory, which of course contains all of our processor code and it looks like we just got a couple of uh, low-end switching Regulators down there from the main DC input here. and then we've got an ISP 1763 a USB on the go. That's pretty much it there for the main Uh processor board.

oh I forgot a Um programming header up here for the lattice PL as well. but apart from that, well, that's all she wrote, so that's um, quite well designed and laid out. No issues there whatsoever. Let's go have a look at the main output board and the main board here, as I said, looks to be very welld designed and laid out.

They uh, seem to know what they're doing here. Got our regulation circuitry up the top here. these are all obviously low Dropout Regulators for all the various rails. Um, which is why they can get away with no uh regulation on the output of, uh, the converter up there because, uh, the main, um, uh, the main power supply board because then they're just, uh, locally regulating them down here.

and uh, of course, they're going to use multiple Taps there because they want the lowest possible uh Power dissipation in these devices only using the PCB as the heat sink in there. So that's why they've just got the multiple Uh voltage rails there. Of course, because you've got rails for your analog stuff, it might be say, you know, plusus uh, 12 Vols or uh, something like that. Plus, you need your 3.3 and other 5vt and other rails as well and uh, heat sink on the main device here I can't seem to get that off so sorry I am not going to uh, try I do want this thing to actually uh, work when I put it back together so um, unfortunately, it uh is probably stuck on with some of that heat sink glue so that's not coming off real easy.

Um, we've got a TX DQ here which will take a look at a number. a chip with its number rubbed off. You got to be kidding me. The as we said, the serial optoisolated interface coming in here.

they got ground playing around. Looks like we've got some more switching power supply stuff up here. Relays everywhere. Um Quality NEC uh Brands So they haven't actually skimped on those at all.

Some miscellaneous analog stuff happening around here, and it looks like two identical channels for both Channel 1 and Channel 2 here. and then it looks like we have a cap and a a transient suppressor going down to Main's Earth on both Uh Channel 1 and Channel 2 outputs. Now, of course, I could Hazard A pretty good guess of what's under this heat sink. It's going to be a Zyl Linkx or an Altia or maybe some other brand.
uh, Fpga. Um, you know it's almost for sure that they're going to have an Fpga in there implementing their uh, True Pulse uh technology or Easy Easy Pulse technology. uh, sorry. and that's coupled into our SRAM On either side.

Here, we got a TX D which we'll take a look at. Mystery device here that's been sort of uh, L you know, rubbed off I don't think it's been laser Ed it's not. Uh, it's not. uh, quite.

Um, you know, straight and perfect enough for that. So they've probably got that on the grinder there and just grind it away. Why? Why jeez, it's not going to stop anyone. if you really wanted to find out.

that chip is, is, you know? Oh jeez, you know, a day work tops? Unbelievable. Give me a break now. This thing actually boasts some pretty good uh phase noise uh Jitter specs. And here's why: it: BT 507a It looks like a uh Chinese manufacturer I've never heard of, but it seems to be a uh Tcxo temperature controlled Crystal oscillator You can see the multiple Footprints they've actually put in there so that they can, uh, choose a suitable device foru production.

uh, later, maybe a high spec uh version or something like that. and right next to our external uh input connectors. Here we've got an Analog Devices ad CMP 562 that's a dual uh PCO ecl. uh comparator.

Extremely quick. We're talking uh 500 P output Ryzen 4 times. Very nice. and there's that Dremel off chip.

You got to be kidding me. It's tied into the uh, presumably Fpga down here and I couldn't readily identify those devices there at first glance. so they're you know, presumably just uh Op amps. They've got two there, um, presumably uh, driving some outputs or some inputs there.

and then they've got another one over here as well. And as we've mentioned a couple of times, here's these: Analog Devices are ADM series 1410 digital isolators and they've decided to use these instead of uh, your regular opto uh couplers that you traditionally find in these applications. These are just nicer, higher spec, high performance. You don't have to worry about uh, various issues and stuff like that, they just work digital indigital out magic kind of devices.

They do cost more though, unfortunately, and it looks like we have some sort of uh, current shunt thing happening here with some sort of uh, maybe instrumentation amp tacked on there. And there's the SRAM we've mentioned uh before. it's a Sirus logic part and it's a 512k by8 bit. uh Ram Of course you need uh, you know, more than your regular 8 Bits because this is a 14bit resolution output uh converter on each channel.

So they got two of these Uh devices. once again, 500. They're both 512k uh devices identical. so I'm not sure why the spec for one of the channels is less than 512k.
and of course our main Dack which is one of the hearts of any function generator of course. uh, they of course haven't skimmed. it's an Analog Devices that TX stack. um, 80 9781.

And as per the Uh data sheet specs for this thing, this is a dual channel 14bit. uh Dack does 500 Meg samples per second per second Exactly as it claims on the Uh front panel. and as they claim in the data sheet for the resolution, not a problem. and you can see it's an Lvds interface.

You can see the controlled impedance pairs going through and their length matching. That's why of course I've done. uh, quite talked about this quite often. that's why they got those squiggly lines on there.

just length matching all of those just to make sure that the signals arrive at the exact time for the deck. and we've got some more fast comparative action around here. Same device we saw on the back panel and then we've got a few. Interel 28 210 low noise jant opamp surrounding a good old 74hc Uh 451 classic.

Got to have some Uh 74 Sl4000 series logic in here somewhere. And of course, driving all this magic is an ADF 4360 and that is a Vco. um, many gigahertz operating range. So that's what's uh, generating the clock and of course matched to that.

You got to have a low noise low Dropout regulator just dedicated to powering that sucker. So they've got a Uh Mcel what is it if I can see it 5209 and that's going to give us a main Uh clock output frequency of somewhere between 2.4 and 2.7 GHz But of course it's got uh, multiple. It's got divided down, uh, selectable divided down outputs as well. but you know clearly there we go that's going up.

there. Little bit of length matching in there to drive Our Deck So a lot of the analog performance of this thing is going to be determined by uh, you know this uh Vco here which generates the high frequency clock for the main deck of course. and how you you know power that you can't just power that willy-nilly from the supply Rail and expect to get good specs. So really, they've um, you know, they've done reasonably well there.

They haven't skimmed at all, of course, but that's what you expect. I mean this thing is, you know it. Specs are very, very good for a um, well, for any sort of oscillator, let alone one in this, uh, sort of price range. And we haven't showed you the relays yet.

There they are NEC UD 2-4 something like that. And anyway, they are n branded. really crappy looking. um inkject printing uh, labeling identifier on the top of those things.

Now if we have a look at the uh bottom of the board here, here's the four relays for one particular: Channel and you can see they've got all the matching pads and they've physically removed the uh, ground fill underneath directly underneath those pads. Normally you might do something like that to, just, uh, decrease the capacitance of the individual pin to the ground plane. You know you might be able to save, you know, half a puff or something that reduce your capacitance to ground by half a puff or something. But because this is a four layer board, there's going to be an internal ground plane anyway, unless they've actually left off, uh, the ground fill inside the inner ground plane as well.
But you can see they've also done similar stuff around here as well. so maybe they do actually know what they're doing in that respect and over on the more closer to the output side here. They've also done it here as well. So I think they probably know what they're doing there, just trying to reduce that uh capacitance by half a bees dick and the output circuitry.

here. We've got some exposed copper on the bottom here, lots of Vas that tells us that the Uh chips on top have a thermal pad and uh, clearly they're trying to heat sink those. so let's find out what those chips are. Aha, no real surprises.

Texas Instruments THS 3095 uh high speed, low noise, high voltage High current Uh Current mode opamps So uh, these things are maximum. Uh bandwidth 210 MHz sort of. uh, pretty much, uh, perfect for the Uh application required here. now.

I'm not entirely sure why they've sort of surrounded all this with sort of like a big dip, uh, footprint. It's almost as if like another, like a design to have possibly another module in there. or, uh, something like that. Perhaps.

Whether or not, um, they were going to abandon this circuitry and put it on a secondary module. Whether or not there's a second module that plugs in there. Not entirely sure now because this is only the 80 MHz model. Uh, does that mean that the 160 MHz model has something additional in here? I Don't know if anyone with a 160 MHz model of this thing can open it up and confirm.

that would be really nice. So yes, they're the pads of every alternate resistor there that they've removed that ground fill on plus the traces going to them. So clearly they're just, uh, reducing the the capacitance to ground on those individual high-speed nodes. H Very tricky and clearly they meant for a shielding can to go in here cuz take a look at the uh classic uh mounting points there for that shielding C which would go over that whole thing and also, um, there'd be an internal shield in there.

just uh, separating the relay modules from the output drivers as well and the main output. uh, coax connector here. they've got that running internally underneath, so you can't actually see the Uh tracers on there. It's running between the internal ground planes and you can see the isolation split in the ground planes here of course.

uh, separating all of our output. uh, ground reference to our Mains ground reference. Uh, stuff that we had and of course they're our, that's our data Optical isolation barrier in there. all the data coming over.
We can see that it also goes up to here to one of the output connectors as well. So one of the output connectors is Main's ground reference while all the others aren't And if you have a look at the back panel here Tada There we go. We've actually got uh two of them that Main's Earth reference Because there we go. They don't have the white insulating thing and three other isolated ones, so you can actually tell where they come from on that connector because these ones are the isolated ones here on this connector.

plus uh, one. So there'd be two connections on that one plus one on the other one. And obviously the other two are connected to this side of the ground instead of the output side. And as for the Uh, soldering quality on both boards, Well, pretty much.

uh, first class. No complaints there whatsoever. and even the high thermal capacity coaxial connectors? Not a problem. As you've seen, these two boards are designed and manufactured.

Uh, pretty much. uh, first class and certainly more than adequate for the Uh price range, that's for sure. We're only talking a $600 uh unit here for the low-end one. I've gotten just over 1,000 for the 160 High unit.

but um, whoever. Uh, Des This thing certainly knew what they were doing from a function uh, generator point of view, they've used. uh, you know, certainly. uh.

top class uh components. I Don't think they're overclocking, uh, anything at all. Everything uh, seems to be within spec. Don't know what's going on with the rubbed off number chip here H bastards.

But anyway, um, it seems like they really know what they're doing in this thing. Uh, probably I would uh, be fairly certain that, uh, it would certainly meet its specs for its, uh, you know, uh, fairly, uh, demanding specs it's got for its performance. so not a problem whatsoever. Thumbs up on the electronic side of things, but the Shazzy? Ah, come on, you got to be kidding me.

Rust straight out of the box. That could be the new slogan. Unbelievable. It's just flimsy crap, folded metal and spot welded and rust as far as the IE can see.

Unbelievable thumbs down. What an absolute shame because the Electronics is. You know, pretty darn impressive. Not a problem whatsoever.

Just let down by stupid shazzy. Got to be kidding me. oh face palm. and once again, just like we saw with one Uh unit on the top of the board, we've got some current shunt uh Shenanigans going on here and here as well.

so we got a total of three of those. By the way. uh, these output connector boards. Nothing much, uh doing on there at all.

Um, but they do have a couple of surface mount Transformers on there. Quite neat and just for completeness, here is the uh front panel. There were a couple of uh screws on here. nice looking output connectors.
They look really huge and uh, massively solid integrated in there. Really like those. They should last quite some time. We've got some uh Emi tabs up here to connect through to the main.

uh Shazzy There there's our once again they got tape. Nice attention to detail. there tape holding down the flat Flex uh cable connectors there and there so that's very nice. Nothing much, uh doing.

Of course this is going off to the front panel. uh, full color uh TFT display. There's our USB connector over there and its own little separate board. But apart from that, eh, Once again, you can see the RF Emi shielding along the base of this thing.

They've got that along, actually all four sides in there. Excellent. They've done pretty well in the Emi side of things here, that's for sure. Once again, you can see the horrible rust on the front panel there where they've clearly like you know, scraped away.

debur this thing. It probably doesn't show up on camera as well as I can uh, see it with my eye, but you can just see the see the scrape marks there. Ah, absolutely horrid. Maybe if I there we go look, you can see them clearly and it's just rusted all the way along the front.

there under that, uh, under that front panel bezel. Unbelievable. And by the way, I don't particularly care for the uh, self tappers they're using to secure this front panel on. but I guess you know it's not something that You' take off all the time.

So eh. Well here we go. It's almost back together except for the top case and uh, hopefully I've plugged everything in and uh yeah. well let's switch this sucker on for the first time and uh, see if it works? Hello Siglent! Woohoo! There we go.

We're up and running, not a problem and bingo we're in. That's pretty fast boot I Like it works. A treat. There you have it.

That's a tear down of the Siglent Sdg 5000 or in particular the 5082 uh, 80 MHz 500 Meg samples per second function, arbitary waveform generator and well, I was extremely disappointed at the quality of the Shazzy in this absolute thumbs down for the mechanical shazzy. Uh, quality and construction. Just awful. It's what I expect to see in one of those $50 cheap ass noname you know power supplies you get on eBay or something, you know, 30 amps at five? you know, 30 volts at 5 amps for 50 bucks.

You know you expect to get a rusted old metal shazzy with the bloody thing, but not in a you know siglin who are you know a quality brand these days? or at least they're trying to be. And their their. The performance of some of their gear is uh, you know, really quite good. and this thing, um, you know is has really good specs for the price.

Great bang per Buck but jeez, the quality of that shazzy is very disappointing. Very flimsy metal and um, and then it's You know it's just rusted everywhere as far as the I can see. Unbelievable. But of course the electronics in this.
absolute thumbs up. Uh, first class. um, not an issue. Couldn't fold it in any way, shape or form really.

So overall, what's it going to get? Well, it's certainly not going to get a thumbs up because it was let down by the stupid rusted crap metal work in this thing. But ah, you know it's offset by the decent, high quality, well-designed Electronics in this thing so well it's up to you how you want to call the overall uh, you know, raiding this thing. Whether or not, uh, a rusted little flimsy metal sh shazz matters to you and whether or not that's actually, uh, typical of uh, these units or not, or whether or not I've got some DT or something. But yeah, it's definitely not going to get a thumbs up from.

Me Overall, um, apart from even with the high quality Electronics in there I just less a bad taste in my mouth of rust. Give me a break. There we go. No, yeah, it even taste like rust.

Unbelievable. Give me a break. So yeah, that's it's oh man, it's almost a thumb. thumbs down.

Overall, you know, barely a thumb sideways? Unbelievable. But if it didn't have the rust, even with the flimsy of the metal work and uh, stuff like that in the uh, you know, the cheap spot welding I you know I Still would have been uh, very impressed by this thing. I think so. Yeah, just let B down by something as silly as just a mechanical thing like that.

Unbelievable. If you want to discuss it, you know where to do it. The E Blog forum is the best place for all. test equipment and siglent are actually um on the EV blog Forum as well.

so it'll be very interesting to see what they say about the uh, crap rusted metal shazzy in this thing. Whether or not it's uh uh, typical of this unit or typical of the uh construction of these things. or whether or not I've just, you know, somehow got one that's just left on the dock somewhere and it was just W rusted away to the hilt I don't know. but anyway, um EV Blog forum is the best place on the internet to discuss test gear.

definitely jump on over there. And I don't mention it much. but uh, if you want to follow what's going on with uh, me and the blog and uh, various other rants on a daily basis, Twitter is the best place to do that. I tweet a lot e blog is my Twitter name and uh, so follow that and yeah, Facebook yeah, whatever.

I've got some do anyway I hope you enjoyed that. uh little test equipment tear down I Know a lot of people get a bit sick of seeing test equipment tear downs, but I personally like test gear and I like tearing them Down So that's what you're going to get a lot of here on the Evev blog. Catch you next time.

Avatar photo

By YTB

21 thoughts on “Eevblog #497 – siglent sdg5000 function generator teardown”
  1. Avataaar/Circle Created with python_avatars pagarb says:

    It looks like the chassis and circuit board were built by one bunch of the guys and the cabinet by another group who weren't as quality conscious..

  2. Avataaar/Circle Created with python_avatars Monchi Abbad says:

    I like the fact that the did a nice rust-proofing on the spot-welds but on on the cut-edges

  3. Avataaar/Circle Created with python_avatars Ian Board says:

    I think you go on way too much about the rust spots. The metal chassis is arguably the least important thing in the product. The board layout and construction looks pretty good, as do the cables and connectors. I have some of their other products and have been extremely happy with them. I would rather they spend the money on things that matter when making it.

  4. Avataaar/Circle Created with python_avatars modulate says:

    That shazzy.

  5. Avataaar/Circle Created with python_avatars Abhi All says:

    Dave i am to know that need to buy all devices for teardown or it sponsored

  6. Avataaar/Circle Created with python_avatars Zeins Elzer says:

    after this video I will never invest on this crapy cheap made equipment. sorry! this suppose to be a high end equipment, but the try to save as much as they can, saving some Pennys tossing some clients. shame.

  7. Avataaar/Circle Created with python_avatars Yoha Roose says:

    and what happened to quality operator 1

  8. Avataaar/Circle Created with python_avatars lnpilot says:

    "Rust straight out of the box". Great new descriptor for shitty products.

  9. Avataaar/Circle Created with python_avatars Blade Runner says:

    Yes, rust again! I hate it! Guys, invest in a dehumidifier!!! And to remove rust from steel best way is scotch brite (google for it) and some rust protection for further in a form of spray.

  10. Avataaar/Circle Created with python_avatars Kabodanki says:

    rust suck, I lived near the ocean, rust can be a real problem, I remember when I teardown my PS3 after 4 years, there was rust everywhere

  11. Avataaar/Circle Created with python_avatars defaultuser000 says:

    Wow. Those welds are atrocious. I've never seen the chassis of a piece of electronics equipment welded together like that before. These kind of folded metal chassis are usually spot welded together. This process involves two large copper electrodes that pinch the two layers of metal together and then pass a relatively large current between the electrodes for a short period of time. Due to the higher electrical resistance of the steel, compared to the copper, the steel heats up at the pinch point and the pressure of the electrodes causes the metal to fuse together. The pressure and current used depends on the thickness of the metal to be welded. These welds on the other hand appear to be MIG/Wire-Feed welds. And while it is completely possible to get a good quality "spot" weld with a MIG welder on thin sheet metal(Auto-body techs do this all the time), these welds are complete shit.

    I also tend to agree with you that the source of the rust appears to be from somebody deburring the edges of the metal, after it was painted. Again, all the chassis I've seen have appeared to been punched out of a larger piece of metal with some sort of press type machine. This leaves a nice clean edge. If the edges of this chassis originally had burrs then maybe these were cut out with some sort of CNC plasma cutter or something along those lines.

  12. Avataaar/Circle Created with python_avatars OSHbots says:

    I hope he installed the ribbon cable that ties the two boards together before he put her clothes back on. 😀

  13. Avataaar/Circle Created with python_avatars Dan Jones says:

    End of the video…. missing ribbon cable for the data lines…. bugged the crap out of me

  14. Avataaar/Circle Created with python_avatars Tim Rotunda says:

    I started tearing down my fathers electronics when I was 6 or 7 then when I started working in his TV shop at 13 I got to tear down pretty much everything. Keep up the tear downs Dave!

  15. Avataaar/Circle Created with python_avatars Matthew Bucknall says:

    I wouldn't call the MachXO devices low end PLDs. They are most certainly CPLDs and architecturally about as high-end as a CPLD gets before you have to call it an FPGA.

  16. Avataaar/Circle Created with python_avatars Shaun Johnson says:

    I personally enjoy the test equipment teardowns, keep up the good work! I agree, the so-called Q.C. and QA should have caught the bear metal on the chassis and required it be repainted. Rust is bad mojo for any electronic gear.

  17. Avataaar/Circle Created with python_avatars Eric Merrill says:

    If you look at the line thicknesses, I think you are actually just seeing more of the line being exposed by two small dimples in the left edge plastic. I don't think you are seeing the orientation mark.

  18. Avataaar/Circle Created with python_avatars blackstar2008 says:

    Rust can grow, and it leaves a residue on the electronics, not good.

  19. Avataaar/Circle Created with python_avatars edmoore says:

    it is probably for the signals going in the other direction.

  20. Avataaar/Circle Created with python_avatars Stevan Simic says:

    i love rust!

  21. Avataaar/Circle Created with python_avatars Inductive Thinking says:

    I think it would be great to see a performance comparison between this Siglent and the Rigol DG4000, you could do it as a 2in1 review.

Leave a Reply

Your email address will not be published. Required fields are marked *