EEVBlog #497 - Siglent SDG5000 Function Generator Teardown

[nitro2k01]

Inside the Siglent SDG 5000 series SDG5082 80MHz Arbitrary Function Generator.

[Short Circuit]

oops

[ddavidebor]

Dave, have you noted that this thing has not the EU CE mark, but the china export ones?

[Jebnor]

oops

Umm.. Ok.  What is the oops?

[nitro2k01]

6:47 What's the matter with that screwdriver? Is that dent intentional/useful for something?
8:40 (The cutout) Maybe they're super cheap and place two pieces in a zigzag pattern to save on the metal. Even so, the metal work is only going to cost a fraction of the electronics, so it should be a problem. Maybe they outsourced it completely, or don't communicate too much between the different departments?

oops

Umm.. Ok.  What is the oops?

Here's a hint.

[synapsis]

Looks like he forgot the ribbon cable between the CPU board and the output board?

Those marks are typical of deburring stamped sheet metal with a 90 degree die grinder (with an 80 grit nylock). The welds look like they were touched with a MIG.

[grego]

After watching Dave's video I opened up my new 5082 (I've had it about a week or so) and didn't have the rust issues he did.  Chassis is still a bit flimsy as he showed but I think the rust may have been a bum unit.

[EEVblog]

After watching Dave's video I opened up my new 5082 (I've had it about a week or so) and didn't have the rust issues he did.  Chassis is still a bit flimsy as he showed but I think the rust may have been a bum unit.

The EEVblog curse strikes again.
Was there any exposed metal from the punchout?

[Short Circuit]

oops

Umm.. Ok.  What is the oops?

The missing cable, but what surprised me is no error message at all. No wonder it boots so quickly...

[Rufus]

The EEVblog curse strikes again.

Just goes to show how much those golden QC and QA stickers are worth. Hard to believe all the rust occurred in the box. The bare metal and probably rust must have been visible during assembly and test.

[free_electron]

not rust. leftover from the sheet cutter ( plasma  ).

clockgen : 74HA04 hex inverters  Dave! 

grinded device is too large for fpga config memory.

[EEVblog]

not rust. leftover from the sheet cutter ( plasma  ).

Sure looks and feels like some sort of rust to me. How would a plasma cutter explain all the scrape marks, like the edges have been filed or ground down?

[synapsis]

When they ground the edge to deburr it, they probably took the galvanization off the sheet metal, which is why it rusted. If you look closely, you can see the edge of the coating right where the grind marks begin.

[free_electron]

you can see the backspray on the side in one shot. rust would not grow in perfect parrallel strokes.

12:36 when you talk about deburring.  they cut from the far side towards you. so the plasma exits towards you and causes the slanted strokes. that is spattering : streaks of hot metal. they are all in the same direction ( away from the direction the plasma cutter travels. )

it's too regular to look like grinding.

[TooOldForThis]

I just looked in my SDG5082.  No sign of any rust or grinding on the internal edges.  And yes, the inner case is not battleship armor but that frame just has to hold the guts together until you slide the thing into the heavier outer case.  A thicker inner case would just add weight.  It doesn't distract from having a AWG that can play out cardiac signal with a 20MHz pulse rate.  It's like what an ECG from a frightened hummingbird on meth would look like. 

[AmericanLocomotive]

Super, super minor surface corrosion. Whatever method they used to cut the metal, be it grinding, cutting, stamping, laser/plasma cutting chipped/burnt off the galvanization on that edge, leaving bare steel. Bare un-coated steel will get surface corrosion in a matter of hours just from humidity in the air. The spot welds look fine, and are a cheap and strong way to secure two pieces of metal. I can guarantee you that most of the metal in your car is held together by spot welds.

While the chassis does seem to be made of somewhat thin metal, if it feels solid with the cover on, then who cares?

[99tito99]

Rusty metal out of the box can be a big problem depending on your location.  Tucson 8% humidity not so big, New Orleans 95% could get a cancer and one day have problems with rust flecks all over the boards. Cheers, Mark

[EEVblog]

While the chassis does seem to be made of somewhat thin metal, if it feels solid with the cover on, then who cares?

I care.

[synapsis]

Rusty metal out of the box can be a big problem depending on your location.  Tucson 8% humidity not so big, New Orleans 95% could get a cancer and one day have problems with rust flecks all over the boards. Cheers, Mark

I built a 1931 Chevy car chassis out of bare steel. Never painted it, and it never rusted in three years of sitting in my garage in Tucson. The only place that got surface rust was where my arm touched it while leaning over it.

[orin]

Super, super minor surface corrosion. Whatever method they used to cut the metal, be it grinding, cutting, stamping, laser/plasma cutting chipped/burnt off the galvanization on that edge, leaving bare steel. Bare un-coated steel will get surface corrosion in a matter of hours just from humidity in the air. The spot welds look fine, and are a cheap and strong way to secure two pieces of metal. I can guarantee you that most of the metal in your car is held together by spot welds.

While the chassis does seem to be made of somewhat thin metal, if it feels solid with the cover on, then who cares?

The welds don't look like spot welds to me.  They look more like TIG welds which are an entirely different and much more expensive proposition.  I'd hazard a guess that those welds have lower resistance than a spot weld or screw connection.  I've done TIG welding on far thinner metal than that chassis FWIW.

The rust at this point in time seems to be cosmetic.  My guess is they ground it down to bare metal to make a good connection with their RF gasket material... bad decision as it seems to have rusted almost instantly.

Orin.

[Paul Moir]

Well, there's spot welds and then there's spot welds.  Clearly these are not resistance spot welds.

I'm inclined to agree on the TIG.  There's a lot of undercut there for MIG, and not enough filler.  Never mind there's no evidence of spatter.  Betcha they're plenty strong enough to hold a signal generator together...

At first I thought the rust was the result of a wire wheel deburring, but plasma cutting explains it better.  It's very regular.


 

[staxquad]

invisible inner chassis (for the unDaves)

(MIG or TIG,  , could have been spot  )

can't wait for the review, where true value should be revealed (or not)

[Towger]

Dave, have you noted that this thing has not the EU CE mark, but the china export ones?

That was the first thing I saw on the back. China Export, Siglent will not be able to legally sell the model in Europe.

[tinhead]

i don't see any rusty metal here

https://www.eevblog.com/forum/testgear/tests-siglent-sdg5082-waveform-generator/

but the way how to chassis/enclosure has been made seems to be the same.

[rf-loop]

Dave, have you noted that this thing has not the EU CE mark, but the china export ones?

That was the first thing I saw on the back. China Export, Siglent will not be able to legally sell the model in Europe.

They can sell in europe normally.
Btw, Dave live in Au and there  ChinaExport is just ok or what ever there need. EU directives are not valid in Australia. 

I have also full report documentation about these, other is 52 sides and other 75 sides. Very detailed documents including of course all data how tested and results etc with laboratory  document pictures but these stay behind my body. These can try ask from Siglent, they may give these or not.
------------

About rust or what ever.  I have here opened and documented one SDG5082. No signs of this level of rust what Dave show. But, I can see also very very thin signs of "rust" on some metal cutting. This metal is somehow coated but cutted borders have open iron and it may slightly rust in some environment. Also I can see extremely light signs about this.
I suspect that it have happend in Shenzhen what is sometimes extremely humid/wet place. Perhaps in some time theu have been outside and wet before they arrive to Siglent electronic factory.

Outer case is aluminium.
Rf shielding seals are?
Internal body metal coating is undefined?

It may be better if do galvanic (electrolysis) coating after all metal work is ready and before use these in product line. But more this is like cosmetic problem and only inside unit  -- UNTIL it is used in environment where where condensation may occur.

Internal can do also as old HP, aluminum casting and then the metal machining. But this body is then more expensive than whole this equipment now.

Btw, I did not see front panel teardown..
(oh yes but it can see in my images)
-------------------


But here certificates