EEVblog
ฝัง
- เผยแพร่เมื่อ 30 ก.ย. 2024
- What's inside the Micsig 100MHz 1GS/s isolated dual channel handheld oscilloscope?
What mystery ADC does it use?
David2 also makes an appearance.
www.micsig.com/...
Forum: www.eevblog.com...
Datasheets:
www.silabs.com...
www.ti.com/lit/...
www.intersil.co...
www.ti.com/lit/...
www.issi.com/WW...
EEVblog Main Web Site: www.eevblog.com
The 2nd EEVblog Channel: / eevblog2
Support the EEVblog through Patreon!
/ eevblog
EEVblog Amazon Store (Dave gets a cut):
astore.amazon.c...
T-Shirts: teespring.com/s...
Donations:
www.eevblog.com...
Projects:
www.eevblog.com...
Electronics Info Wiki:
www.eevblog.com...
💗 Likecoin - Coins for Likes: likecoin.pro/@...
"The penalty you pay when you go for a BGA FPGA" - untethered poetry.
+FFcossag I never knew I had poetry in me...
+EEVblog Hi Dave can you make a video how to read burned resistors ?
or explain me cuz i am in trouble
Thanks :]
+Frank Alvarez You'd probably have to use the components it's connected to to infer possible resistance values, then try them until one works properly...
+FFcossag I liked the "coz these things aren't magic" comment myself! (at approx 20:53)
For the signal routing, it would be interesting to see if you can detect a difference in the noise floor between the two channels. If you can and its got an FFT function it would interesting to see the spectrum.
Didn't see a current shunt or fuse - but has a current range - what's the spec ?
Maybe those mosfets are an electronic fuse ?
The "unsoldered pin" and "fallen out pin" is a cheap way to do DIY keying now you can't plug in certain revs of the board or accidentally reverse polarity in testing cables. Also the 0505S is indeed a 5v to 5v converter but the 'S' stands for single supply, so not a dual +/-5.
Listening to an expert talk about a subject you know little about sounds like this:
Erflob concracker spoodlewhiff napperzolg. Inflecamotive zarry bounding, fetzimter noogie splatwasher. Immertle goonknacker, comined tetraflunastypill, immernobsplurfer.
Dodgy soldering for sure! If that is - like I think it is - a retail product and not a pre-series demo, then with those unsoldered pins and a few other suspecting bits I'd be pretty upset if I'd bought one. Personally I don't care how good something can be if they disregard things like QA and assembly inspection. I hope this video gave them something to act upon but I wouldn't spend my money on a Micsig product as it is now.
Design seem sound enough but QA is a problem.
That thing seems to be a mixed bag, great design but poor manufacturing I guess! Lots of little things you pointed out, like the 2 little transformers (or common mode chokes) one looking a bit broken, C123 is missing and the solder did not reflow right, that bit of solder shard shorting 2 pins on that 8051 near the end, etc! I wonder if it would have even been fully functional before you took it apart!
Very nice videos.
There is any posibility of making any video regarding the bad usb killer ? And maybe made a circuit where you can plug any usb device and get a reading of the voltages to avoid putting a usb killer in a computer ?
6:00 Altera CLONE! Obviously a fake.
Looks like a good job all-around. Not much to whine about thank God. The whining has almost made me unsubscribe a few times.
Why can't they throw in a nice 2048x1536 display on it. For that astronomically high price, they need to at least use a more modern resolution for their display.
It also should have more memory (e.g., 1GB per channel would do nicely).
+Razor2048 2GB of SRAM? Are you mad? I'd say the display is fine, but for the price they should probably manufacture their product a bit better.
+Razor2048 hard to find industrial reliability grade displays at the higher resolutions. remember, this is not a toy.
Eeeek!! never mix analog inputs with switching psu's, it must be very low impedance.
5:58 seems like the processor is a clone.
Lol!
It all looks very pre-release to me. Expensive stuff left in, screening cans left off, bodge components. Full of "yeah I wouldn't do it like that" design decisions. The enclosure design looks a bit 10-years-ago to be honest, it looked good then (although not sure about the colour) but a bit cheap now. Anyone else noticed how the logo looks an awful lot like the Microsoft logo typeface?
Dave Jones drinking game update: Every time he says "no wuckers" take a small sip. See if you can stand up at the end of the video.
any 8051 fanboys ?? :D
I would have expected David II to be David 1, and you the Dave 0 ;)
Cripes there is some variability in quality control there! The meter inputs are very poorly soldered, and I suspect the person who did that also dropped the blob of solder into the front end electronics. They need to see your soldering videos, Dave! Maybe the soldering technician was off sick, and the window cleaner had to do the job instead.... chinese factories!
Another let down is the awful Chinglish font they use on the display. Not sure where that font came from, but is is horrible and hasn't changed since the 1980's. It would cost next to nothing to fix that, but the archaic character set isn't something the Chinese seem to worry about.
On the plus side, the board components and layout are state of the art and mostly very well thought out, with no cost cutting. Like I said, shame about the hand soldering.
As for that header with the pin sticking out. It might be a way of ensuring the pins are properly lined up when the boards are pressed together. Some old PC IDE headers were a bit like that, with one of the pins blocked off to prevent misalignment when assembling. Just my theory!
thicc
@28:30 I was laughing out loud. So funny. Easy to do ''haste pace'''etc
+mart fart Should have used the Fail button there ^^
I wonder why they did not just run the coax a few cm longer before they go into the board.
Assuming that coax is more noise resistant than board traces.
Properly terminating coax into pcb traces can take up enough space that its easier to run a proper pcb transmission line the extra distance.
+Tilman Baumann Maybe he was tired.
+Tilman Baumann Maybe they already had enough layers in the board to isolate the signal properly, what with the MILLIONS of micro controllers and the big BGA(caution, some exaggeration).
Silabs makes 8051's with interesting peripherals for test gear. There is a thread about a µV/mΩ meter kit which uses a C8051F350. This is a brother of the microcontrollers in the video, it has a has a 24-bit adc and a 1-128x pga on board!
Love how you had trouble assembling it, happens to me from time to time and I just get very annoyed when it does happen. So I know how you must of felt when it kept happening to ya'., lol.
+EEVblog Happy BTTF day. Sorry to hear about the movie theater plans - they have I and II showing here. I and III were probably the better ones.... "What the?!" Anyways, catch you next time! Edit: Of course, it has to do with the whole 'Oct 21, 2015' - facepalm
+EEVblog how can you say that they aren't cutting corners when you see that most of the solder joints are rubbish?
There's a second unsoldered pin shown at 22:00.
Great teardown video by the way, looks like a handy tool to have.
I wonder, in digital scopes when you move the trace up and down is the scope only moving the trace on the display or is it changing some analog offset value before the ADC? Also as a German, what's wrong with calling it a multifunctional device? I mean it is multifunctional, no? Does it just sound weird or...
Hey Dave! Just found your EEVblog, I'm completely hooked! Keep up the great content! -new Fan from Syracuse, New York
Nice teardown as always - thank you Dave!
The Pin 2 of the Silicon Labs F330 (@24:18) seems to miss some Solder. :D
A bit concerned with the missing solder under the F330 on one of the pins @24:00. Guessing it is a ground pin and the solder has gone in under the circuit to a ground tab. It is not pretty.
How does this hand held scope compare to the 20 year old ths720 series hand held scopes. It would be interesting to see a tare down of one of these.
My question is, (and ,maybe I missed it, but is there a current cutoff, if is there is a thermal overload, should the fan fail or something like that? Anybody notice anything
Dave, you should do a teardown of a servo amplifier (motor driver). It would be cool !
Fuck being first.
Despite our "Western" bias, the Chinese are perfectly capable of making good quality products. The problem is exactly what you see here: Quality costs money. If this scope were half the price of anything competing on the market, then it would be a no-brainer, but as it stands, the only thing to separate this scope from any other in the price range is the range of features it offers, and how those features meet the needs of the people buying it. Overcoming brand loyalty is a big challenge; hopefully Micsig can bring a lot to the table. :)
The rather inconsistent quality control evident in this tear down perfectly justifies the bias.
Dave you should you a fundamental Fridays session explaining memory types and how to read the datasheet.
SHARC is still the leader in DSP :D
Did you at least forget to put the screws in, so putting all the other bits you forgot to put back would be easier? :P
I have had those days... You need something to just take apart and scrap... GREAT stress relief! :D
Always awesome videos, would not mind more repairs.
Thanks :)
Am I the only one who thought they did a bad job of hiding their traces under that metal pot?
Why couldn't they had just did a uniform sanding on the top? Or better yet, silkscreen a big white block over it? Looks bodgy as all hell.
And that white goop, eugh, there has to be a better way. Thinking back to iFixIt's mantra of _If you can't fix it, you don't own it_, Micsig automatically lost me as a customer.
The white goop is there because the transformers are heavy. The fact, that they are Chinese doesn't mean they don't know what they are doing! And it's much cheaper to hand grind the markings of the part. If they wanted to spend more money for hiding it, they would have ordered the transformers with their brand and part number, but not to listen to your suggestions.
You seem impressed. But, the loose piece of solder, the unsoldered pin, and the "bodgered" resistors/caps make me surprised that it actually functions properly. Don't these defects scare you off, especially at a >$1000 price point?
So they want to be the TME market's equivalent of Huawei...
Is this a demo unit? The soldering quality on those inputs could improve during production and also they could correct the bodges so it looks all nice.
ok i have a 1958 heathkit o-12 scope i was wondering if there was any way to get a reading from a Jacobs ladder u have been trying but the Jacobs ladder really screws with everything in my lab and the scope is not even giving me a reading i think is true the trace just goes from scanning to like this messed up blob in the middle of the screen any ideas??
Hah the first thing i noticed on the battery is the CE logo. And its the China Export logo wchich is designed to look like CE logo fo equipment that matchec EU standards. Sneaky ;)
For a company who claims to want to play with the big boys, they sure have a QC not worth a shit.
$1400 for this workmanship is unforgivable.
Was expecting it to not turn cause you forgot the battery, lol.
It is a shame that quality of soldering is really bad..and also there is a lot of dirt all around...that isn't very good also..but it looks like good scope..if they would fix thissoldering..then all would be nice..also imput protection is well a bit low..
Come on, how safe does it need to be. If someone is that clueless they need a fully plastic protected oscilloscope plug they shouldn't be playing with one.
Does the (ground?) copper to the left of the ADC at 11:20 look a bit shoddy (bad etching or something)?
23:52 SIL C8051F330 Pin 2 (Ground) not soldered properly?
DAVE, you a big BTTF fan, where was the DeLeorian Time Machine Taredown ?
Enough with all those oscilloscopes, multimeters etc. Totally unintresting. These are not the only el. devices that exist !
I wonder if MicSig watch this and take note ?
How would it compete with Siglent SHS1062 which would you choose ?
Why have they not use a +-5cm longer coax cable? Coax is cheap and isolated.
Those Q24 and Q25 are probably shunts for current measurement.
I try to watch the adverts to support the channel but that Tesco Mobile is too annoying for me
6:00 the corners of that chip are all broken. What can be inferred about the manufacturing process here?
+John Drachenberg I think that might be tooling marks from manufacture.
Is that a cap or a battery on the cover image I’ve had them lie about caps before and just put in a battery and labeled it as a cap.
anyone else notice that "MICSIG" written on the back of the unit is in microsoft font.
could the front end adc be running in paired and use the two spare data lines for overflow?
20:26 is that C123 ripped off? :O
+Jaroslav Malec
That looks weird. It doesn't look ripped of (you'd see some sharp edges in the solder then), but it doesn't look like it was never populated either (double negative on purpose ;) ). What really worries me though is, that the solder did not reflow properly on the whole pad. Many of the other solder joints look dull, too. So maybe they didn't heat the board long enough and if there was a cap on the pads it never reflowed correctly and just fell off when washing the board. Would need a second unit to compare, but doesn't instill a "quality" feeling at all.
and take a look at r141
+superdau I must totally agree, it does look like improper soldering process.
+Jaroslav Malec U41 a resistor on a SO8, this product is full of last minute work.
+Bernard Lim Indeed, it's everywhere.
The TMS320 is still quite a capable DSP. In the early days of production they were used in high-end 3D graphics systems and even found their way into Midway's first 3D video game hardware where they served as both the main processor and the geometry engine at the same time.
Are you the voice of Wheatly from Portal 2? Anyone agree?
Nice oscilloscope... but will it run mario? :D
is that a zebra laser printer i see in the background?
Have I missed something? This is rubbish!
do not like 50 fps looks all jaggy and strange
Will you please stop making videos for about, say, two months so I can catch up on my binge viewing? I'm only up to EEVBLOG #Two Hundred something.... I've learned so much the last month or so that my brain is getting full I think...
I just bought a couple Siglent Chinese O-scopes. It's very good, and as an American I'm pretty worried that China will be not only making the cheap stuff, but the good stuff too. We're doomed.
I recently used AD7799BRUZ ADC's on a recent NASA project. They are 24 bit ADC's (which is overkill for a lot of stuff), but super easy to use and very stable. For that matter I suppose the refresh rate is way too slow for a o-scope. I've been wanting to make a multimeter specifically for aviation applications and might dust them off for that.
That pin in the header socket (J57?) looks like an improvised polarising key. Shouldn't be needed in a PCB to PCB arrangement like this, but they may use ribbon cables somewhere in testing/faultfinding.
That missing top on the choke is probably there only to give a flat surface for pick and place, nothing functional.
i am seeing a "404 not found" when clicking EEVblog. is the site down?
Same here - I was also checking comments here for the 404 Also I get a 403 In US
+Roger Yep, here too. Seems that Dave didn't pay the bills this month.... ;-)
Fantastic stuff as usual. PS Watch Earth 2 available on VEEHD for free for 70s Apollo style nuclear bomb filled action. Bet this one passed you by. Some of it is excellent, some funny of course. Thanks Dave.
Seems like the 410 is the same sort of thing but not isolated. It's also much cheaper. I'm looking into a first scope for automotive use, so it's unlikely to be seeing above 15v or so. The built in meter function is quite interesting too.
Do I really need an isolated scope? What are the advantages/disadvantages?
21:05 - I laughed so hard when I noticed that.))
I thought I'm the only person guilty of soldering pasive components directly to the ICs, but apparently it's Ok even for a few thousand dollar devices. xD
mosfet discharge, caps range?
Closing the scope back... tripple face-palm worthy ;)... Been there, it pissses me off every time I do it.
+EEVblog I've never seen those tiny flush ICs before (the ones around the TI chip). Is that just the casing or is there something up with those?
I'm sure I saw a spare screw on the desk when you finished... ;-)
I would love to have any of your things like a scope..since i have Never had one...I envy you for having all those toys....have a good day, or night...great video...
so, how many times did you need to rebuild it? I figured that you fixed the unsoldered pin before you put it back together the first time...
But yeah, just hasn't been your day, it seems
I would really like to hear more about Dave2 3D printer board project, perhaps a video could be made about it? Thanks!
nice latest FLUKE clone ;)
+InXLsisDeo I mean appearance. Housing and buttons. same as Fluke ;)
+InXLsisDeo The apprentices at my company were bought tool kits - including a Fluke voltage tester (T90 I think) and it does have some nice build, I trust it is safe internals (But I have not done a teardown)...but it's continuity test buzzer takes over .25 seconds to respond...duff design.
It also has an odd interrupted beep if you hold the continuity closed for a second or two --- how is that helping me diagnose a intermittent fault when I wiggle a cable ??.
You don't always get the best tool from the best companies.
(I expect some programmer thought it would be good to ad the intermittent beep, and the code became unresponsive unless the continuity is 'solid'...so it waits a bit to debounce the probe tip contact. SIlly.
I was waiting for you to test the actual frequency response......
Did you forget?
Is that supercap similar to the one used in the fluke 298/7 causing problems?
Checked in to see if you had a back to the future special :) Maybe a flux capacitor teardown??
Wow, new video already ?? Thank you crazy Australian Guy !!
A cool feature with this scope is that is has a math function that perform series/parallel inductance and capacitance reactance calculation for inputted component values..
@26:06 what is that smd component soldered that way between the two pins of that chip? it looks awful.
"I hate it when you can't get your stand up"... Uhm... Yup... :D
The extra bit on the common mode choke could there purely for Pick&Place purposes.
+Armandas Jarušauskas Agreed.
Just Call him David not 2!!!
+1videoshow Ya and it looks like we can't call him Afro Dave any more. :-(
Are these kind of boards designed by one person alone? They are complex.
+xartpant Sure, as a professional PCB designer I've designed boards much more complex like this on my own. Usually it's a one person job. Does take time of course
+EEVblog I'm sure the software these days can do a lot of the "grunt work" for you (matching trace lengths, etc.) I've seen one some time ago that produced weirdly organic-looking traces.
+GothAlice Modern serial-parallel protocols (ex.: JESD204B) use word-level alignment at the MAC layer to eliminate the length-matching constraints between serial lanes. Expect the art of delay-matching multiple LDVS pairs to start dying off soon as more serial-parallel chip adopt word-wise re-timing between serial lanes.
+xartpant it's hardly a one man show.
Hi, I'm mr. Micsig. How can I help you?
24:02 Dat solder joint...
Lets do some repair videos.
Could you explain what "Jellybean" means please
It basically just means generic, off-the-shelf components. That is to say, components which are not application-specific; indeed, the type of components which are often not specified beyond value/tolerance and rating.
320 solder it, am I right?
I spot lots of MC34063 action, yeah... I have a few of those in my M102v5 production design, and by using slightly oversized inductors and massively oversized caps I managed 10mV or less ripple on a 15V rail.
nice
no translation...why?
i love this funny guy