Is this FNIRSI-5012H 100MHz bandwidth 500MS/s Handheld pocket oscilloscope any good?
Review + Teardown
http://s.click.aliexpress.com/e/Lirr17DW
Mystery of x1 Oscilloscope Probes revealed: https://www.youtube.com/watch?v=OiAmER1OJh4
Reverse Engineering forum thread: https://www.eevblog.com/forum/projects/reverse-engineering-fnirsi-5012h/
Forum: https://www.eevblog.com/forum/blog/eevblog-1260-$70-100mhz-oscilloscope/
EEVblog Main Web Site: http://www.eevblog.com
The 2nd EEVblog Channel: http://www.youtube.com/EEVblog2
Support the EEVblog through Patreon!
http://www.patreon.com/eevblog
AliExpress Affiliate: http://s.click.aliexpress.com/e/c2LRpe8g
Buy anything through that link and Dave gets a commission at no cost to you.
Donate With Bitcoin & Other Crypto Currencies!
https://www.eevblog.com/crypto-currency/
T-Shirts: http://teespring.com/stores/eevblog

Hi I've got an interesting bit of low-cost Teske kit for you today. 100 megahertz little portable Oscilloscope isn't it? Kind of cute and it's only 70 US dollars for a hundred megahertz bandwidth. So if this I have no idea how performs some about to find out. but if this is any good it's It's quite a remarkable bargain.

And although this is not quite pocket-sized look at the thickness of it, but you know it's got a real BNC on it only single channel. but these little pocket oscilloscopes and things. they're real, you know, dinky little toy things and the bandwidth snot there and all the rest of it and not really partial of those. but this one with a proper BNC on it, a hundred megahertz bandwidth you know, building lithium-ion battery or the rest of you know, a reasonable sized screen I guess and reasonable looking controls for what it is then if it works for 70 US dollars, 400 megahertz bandwidth Wow from a company called FN IRS iana per there apparently the company that maker got it from their official store I guess you'd pronounce that Finn C 4 no see something like that? 512 H And they make lots of other gadgets.

They also have a previous sort of like little pocket portable oscilloscope but this one looks much better. The interesting thing is is that now I'll put up some photos here, have a look at what it looks like on the website and it it just looked fake. It looked like photoshopped and they are. So I Was curious to see what this thing was like and if it performs so let's go.

You get a nice little are zippered pouch then it's got no branding on it. It comes with a probe it's a yeah, it's just a typical one. Hundo times 10 switchable probe. it's what you'd expect, you know seems okay and USB cable and an instruction manual and the instruction manuals all in English Here Arm is a versatile, highly practical, cost-effective handheld oscilloscope for the maintance industry.

and R&D sampling rate a 500 Meg's per second which is enough for a hundred megahertz bandwidth you've got five times. No worries, it can withstand building high voltage Protection could withstand 400 volts continuous and I hate an 100 volts peak and that'd be in times 10 mode. Of course it's got a 320 by 240 2.4 inch display up to 64. Meg Storage can capture 2,000 waveform pitches.

The storage process is simple and fast. just one touch, no cumbersome tips or 10 choices quietly. it is very convenient to save the current waveform. So yeah, a bit of Chinglish happening here, but it's ok.

Thumbnail browse in detailed viewing page turning deletion 3000 milliamp hour battery High-quality 10 hours for a full charge I'm not gonna bother to what test that I'll just you know, take it for granted and he scared any shock. Shockproof. comprehensive protection of the fuselage. Love it.

So their suspects: 3 nanosecond rise time 128 K Storage def dorage so the actual sample memory is 128 K afterglow that's persistence mode Suit 1 to 8 Adjustable 12 kinds of parameters. There you go. you can screen and you can pause this and have a look. Common problems and stuff like that, but you know they've done over reasonable enough job there.
I Think Oh Peter we own signal measurement of my boss to buy IGBT Wow They've got like example things: I Don't believe it has any art protocol decoding Eva and it comes in the classic Chinese package in here. Of course they don't waste any packaging material over there. They reuse everything. the angle of display.

It does dim a bit at lower angles like this, but it's still still holding up. Okay, it's best front on the top angle is, you know it. The graticule vanishes sickly, vanishing at really high angles. stuff like that, but you know it's doing a reasonable job.

You'd have to go outdoors and have a look. As far as the user interface here goes: I Don't mind it, it could be better, could be worse. But anyway. I Like that they've got a 50 percent trigger button, a dedicated Ac/dc button.

you can see that change up the top dedicated x 1 x 10 pro button. That's very nice. You can see that change from 5 volts to 50 volts up there. and no Nf1 Nf2.

They're obviously a programmable function to measure don't know if you can change those and a dedicated edge trigger button which looks like it just goes positive, negative like that. There's no alternate trigger function as in, like positive and negative, it looks like it's either one or the other. It's a bit limiting. Now, this mode button here is not what you think.

Basically these buttons here. It would have been nice if they had some silkscreen on here to tell you, but once you get used to what, it's easy. So up and down of course changes are this fairly intuitive. it changes your vertical scale like that.

and of course, obviously this one changes your horizontal scale. Now this is a bit weird to Scott your regular one to five sequence up here, but when you get down low twenty-five nanoseconds, twelve nanoseconds, and six nanoseconds. that would have to do with the sample rate and the memory yard depth and stuff like that. but that's the lowest six nanoseconds per division.

It's a bit oddball, but anyway, and it goes up to 50 seconds per division. But if you hit the mode button here, you can see that change and you might be able to guess that moves your waveform around. You know, would have been nice to be able to just Center it back. but you know, beggars can't be choosers.

So a basic one kilohertz sine wave here. Of course we can change our triggering edge. that seems to work fine and it's a little bit off. It's actually four volts peak-to-peak on my seat.

Gentlemen, get in for point One two. It's bang on on the frequency though. What's interesting is if you put it in stop mode like this, watch this. It seems to have some data outside the display window either side, so that's interesting.
It's capturing more than what's on the screen And it's got an auto set button. Let's try that. I've moved the waveform around, so auto set that and bingo, It's scented it. It didn't.

It's entered the vertical, didn't Center the horizontal back. There's a few little user interface things you could argue I Don't think right now it's 4.08 so it's getting close. So I'm not going to quibble over 4.08 versus four volts on my cycle engine. That's fine.

now. your trigger button actually changes your trigger mode. It's not a trigger menu and also rudimentary, so that'll go into our single-shot trigger mode. We're in stop mode now, but then you can put it in normal mode like this and jeez, it's a fair bit of jitter in normal mode.

I Don't know why there's our trigger level there. Jeez, that's pretty how you doing anyway. that does move our trigger level. Look at that.

Wow Oh, and then it's going G to e like that. So now what? Why is it not? Yeah, it's just it stopped. Of course you'd expect that in normal mode. but yeah, not impressed by normal mode there.

It's a bit how you're doing, but you're put in normal arm auto mode and that's you know. That's fairly rock-solid but of course we can't change our trigger point in auto mode, which sucks why. And as for those functions, it looks like you know if you wanted RMS It just changes. Basically, you've only got the two parameters on the screen there at any one time, but that's alright.

It's just a little pocket scope for getting you know basic waveform measurements. This is absolutely right off the bat. It's no substitute for a bench scope. anyway.

What do we get in menu here? We've got View Wave: There's no inter button, you press mode. Yeah, you're press mode and there you go. What they go: it's showing the waveform that's a saved waveform. kinda shows that there's an icon view.

That's kind of nice. So there we go. There's a 1 kilohertz square wave and we're going there and we view our wave like that. There it is, Let's select that and it's popped it back up.

It's kind of neat. Can you do measurements on that afterwards? Oh look, you can just scroll through them but like there's no measurement cursor functionality in it. Like after you've done that, you can't go in. It's just a screen capture I Can't actually go in and analyze any of the waveform or data so me might have some limited misuse, but it's disappointing.

Time measure for example, we can do the frequency. Okay, so well let's just do time. Oh that's showing it there. I Thought that was cursor mode.

Aha Auto A 50% That's my boy. We may not be able to adjust our trigger point I Suspect Now can we try? Yeah, there we go. In auto mode. There you go.

Auto 50% so we can change that now. And Bingo There we go. Yep, Cool. Ah, no worries.

I take back that that It was a bit limiting. Afterglow afterglow of course is going to be our persistence. So I Dodd that just sets the level. How do we actually set the mode? Oh, we might put my standard way for me and standard 1 megahertz sine wave with 1 kilohertz 100 % amplitude modulation on it.
and well, it's not handling that very well. Is it my scopes? don't? That'll That'll be quite common and that operates like most scopes. It can't handle the trigger on that. Let me change the trigger level up to here somewhere.

No, we're not going to be able to see that modulation. I Don't think. no, that's not going to be terrific. and it's not triggering on that.

A lot of other scopes have trouble triggering on that too, but we're not. It looks like it doesn't have the on-screen memory depth to actually, um, show anything there. Let's put it in stop mode and then see if we can zoom in. Now there's your problem.

You can't zoom in. We got no extra data. So all you've got so much for your hundred and twenty 8k sample memory. That's what it said.

It had right. 128 kilobytes storage depth. Well, where is it? We're stopped our waveform and we can't zoom into that. We can move it around.

That's hopeless. Anything else we've got on here, but looks of it is brightness and that's it. So yeah, this thing is really rudimentary and I'm disappointed that you can't 128 K memory would have been useful, but it looks like it's only got that the display memory. You know the 320 samples across and that's it.

I Can't find a way to zoom in on that. That is very, very disappointing. Yeah, well, there's La. Listen, I'm feeding in 100 megahertz for volt peak-to-peak sine wave.

There's some offset there as well, so of course if we go to zoom in on, that should come good. There we go. But no, there's certainly certainly some offset there that's not coming from my generator. And if we do this this, we have some average sampling.

Like it's almost as if it's in an average sampling mode or something. That that's really quite weird. Anyway, there's six milliseconds per division. It's measuring it hundred megahertz.

but for 60 or for 70 millivolts? Um, it's a four volt peak-to-peak signal. So we're looks like we're not getting our megahertz. our bandwidth. Of course, a bandwidth of a scope is specified at three DB down.

So 0.707 of four volts. Get your calculator out. You can work out that little bit of DC offset there. So let me wind the frequency down.

Yeah, it's going up. Let me whoa. If we got an averaging mode turned on hi. this is the T800 model day from a few days into the future because I actually finished this video and I released it to my patrons as I often do.

Our patrons and forum supporters and a couple of people are pointed out that the reason I was getting a low bandwidth. They reckon I've come a gutter because this times 1 times 10 switch here was actually a physical Hardware attenuator inside and in times 1 mode it's only got like five megahertz bandwidth. and that way I'll send a terrible bandwidth. Well, that's actually not the case, but as we'll see, but this actually led me to do further tests of this thing.
and it does kind of sort of meet its bandwidth spec up the top here. how did Meg But yeah, no, that's not the full story. but of course that would be completely abnormal because no scope time, no real scope has a times one times ten Hardware attenuator inside. it's just a software multiplier on the screen, but it turns out this one actually does work the same ways.

even though it does actually say in manual down here about this five megahertz bandwidth like this. They're actually talking about the the Times Ten probe and I've done a whole video on times One or what limited bandwidth on Times One probes I have to link that in. But yeah, it turns out this is just a software multiplier like all Ova scopes, but this thing as you're gonna see it doesn't really kind of sort of meets its bandwidth. but it doesn't it's it's figment of your imagination.

So if this video is a bit disjointed and kind of doesn't feel a bit right, sorry about that. It's kind of a mix and match of like old and new clips. So I didn't want to reshoot the whole thing because I that's not what I do. Anyway, let's go.

What I've got now is a proper connection with 50 ohm terminated okay and I'm feeding in one volt with the 50 ohm load at a hundred Meg's so he's supposed to get one volt and we confirm that on a keysight 200 Meg scope, hundred megahertz, one volt peak-to-peak with a one to one software probe option because we're not using A / 10 probes so we don't have the software option enabled. Okay, but we plugged the same thing into here and we don't get it. It's just silly buggers. There you go: 312 millivolts Wow Okay, so what I'm going to do now is just calibrate this probe.

So I've got just a one kilohertz square wave here and I'm gonna give it a little tweak a dick to make sure it's right. otherwise we'll come a gutter and we'll get out a lot of wrong amplitude and I've set the outputs no longer the 50. Ohm So there you go. Eric and that's a square Us: Look at that.

Well, what the heck is that? I am down at the minimum like there's something there. Okay, and six nanoseconds that is not working. Okay, it's doing some funny business there. And there's also talk on the eevblog forum that this does not have proper 500 mix samples per second.

And as we'll see in the teardown, yeah, that's probably the case. So let me wind the week down on that frequency. Okay, check this out. I've gone back.

I've actually I changed it to ten volts peak-to-peak ere. we get our noise square wave like this, which we compensated for. Okay, it's okay in that range. Watch this.
Look at that. this range here when you change ranges, your compensation changes. So this this front in is so dodgy brothers that you can't even come and set your probe and have it be the same over all of your voltage input ranges. That's just nuts.

So I've got to now get in there and recalibrate that I can't even I can't even I can't even compensate it I can't even compensate it on that highest voltage range. Are you kidding me? That's just nuts Wow Not nice. This is this is just silly. Okay, so let's check the bandwidth in times 10 mode and to do This properly using the supplied probe, we're feeding in a 1 kilohertz square wave.

We're compensated it properly and we're using 50 Ohm output termination here. We've got a 50 Ohm output terminator and we're probing it correctly over here and it. Trust me, it's in the X 10 position over there matching X 10 on there. So we are now in the ideal position to measure there's bandwidth here.

Let's give it a ball. and there you go with the 5 volt peak-to-peak we expect. At times Point 707, we expect three Point Five three. so it's actually slightly bigger than that.

so it's gonna have a bigger bandwidth. So if we keep going, we expect that. your job 5 3. Yeah, let's call it there 107 megahertz.

So it does actually have a hundred megahertz bandwidth. Of course, we can't use the probe. The probe is hopeless. We can't compensate it on its highest voltage range so that we change to the 10 volt range or we just can't do it.

And then of course you got this. Complete silly buggers. If you go up a range I Mean that's just nuts. You could easily think that that is your actual waveform that's going on there because it should go off screen.

That is just nuts to have to have it do that. It's just that that is inexcusable. This is a ten megahertz now. Okay, so if we, well, that's normal mode just doesn't trigger, it's just unbelievable.

Anyway, if we go down a range, look, we get our expected results. So it goes over a range like that. And we're not getting in the silly buggers because it's we've got the necessary sample rate and memory to do it. But yeah, I'm at the high frequencies.

Now you're gonna come a cut sir. Look, this only happens at a hundred megahertz. Watch this. Go to 101.

Doesn't happen. Go to 99 Doesn't happen. This thing has a ridiculous software sampling beat problem. I Was struggling to find a correct terminology for this.

but yeah, this thing is completely coming to guts. Are trying a sample at these high sample rates and it's just it's just going nuts. It can't do it. And as for this times one times ten button, it doesn't physically change the waveform so it's not doing it.

It's not a divider on the input I Looks like it's just a software multiplier just like on a regular scope and this thing definitely has a strange peaking in the response. I'm in times 10 mode. but I'm feeding directly in via coax with 50 ohm it terminated. Okay, so that this thing will be rock-solid we're taking there all the cables out of the equation and watch this.
If I lower this frequency, would expect that amplitude to pray stay pretty constant, but look jumps all over the shop like that at about 45 megahertz. Crazy just to prove to you that's not my function Jim 57 Meg Drop it down 50 45 It's rock-solid There's nothing wrong with that Jen at all. So it's this puppy and it is one sick puppy. So if you ask me where the usable bandwidth of this thing lies I would say look here.

it is right. It's you know. 20 May 20 micro. it's going up, it's going up, its going up.

Everything's fine, right? So I would say usable bandwidth before that front end and sampling system starts doing funny business. 42 megahertz. Let's say like 40 Meg Beyond that, yeah, it just goes crazy now. either I'm on some serious drugs or that waveform is moving and bouncing around and going.

Whoa. 40 Meg It's just gone. Boo! And that's in auto mode and of course single shot capture mode. There we go.

I that's why. Look, you can see that there's there's not a lot of samples in there. that's not a particularly good sine wave is it. And normal mode is absolutely useless as tits on a ball.

And if we take it up a range. oh yeah, we can really see that start to jump around and deepest funny business. So I'm gonna say it's not even 40 mega Hertz that's starting to look more rock-solid Isn't it 20? So yeah, can we say that's it's doing alright Anyway, yeah, you can't use it as a normal scope because it just doesn't defaults per division. It's I'm feeding in 5 volts.

Trust me. Look over here on the key side: I can change my probe. It's not. It's not 50 volts there we go.

it's it's it's 5 volts. But this thing shows like 3 and a half. even at 20 megahertz. it's just no, no, no, no, no, no, no, no no No yeah.

100 megahertz my ass. I've got a bit of hope for this, but no, no, no, no. So this is really starting to disappoint now and well. calibration.

So maybe that was the offset thing. So let me run that pulled our probe in USB Yes, I Have success. There we go. Ok, so I calibrates itself.

Let's try that again. Wow I Tell you what at 10 megahertz, look, can't even can't even trigger. Okay I'm in normal mode. Of course it works fine.

Everything works great in this magical auto mode. Come on, it's smack in the middle. Look, give me a break. I'm gonna see if I wind the frequency down.

Does that eventually come good? Two megahertz, one megahertz? No, it's still. it still doesn't come good. Wow it can't even trigger in normal mode. One megahertz again if we switch that over to auto.

Anyway, I'm at twenty six megahertz gonna wind up the wick and it seems to fix the DC offset there so that calibration seems to have done the business anyway. Okay, I'm gonna go into single-shot mode I've got my trigger level set above there so it's armed. Let me wind the wick up see if it captures. Yep, there we go.
works. So it does single shot capture. Looks like a micro for charging that looks really deep. you might have trouble.

Yeah, as I suspected, they provide like that, looks like an extra-long micro. USB So that's that's kind of annoying. And the power on/off switch. So if you want to see it boot, there you go.

Don't know if you can do firmware, upgrade, or whatever. No idea. Anyway, that's it. Looks like nude.

There's nothing else on the bottom. it's just a plastic jiffy type box. Let's crack it open. And we have self tappers of course.

But you know that's what you'd expect. So that's yep. Battery in the back. Oh, that actually looks quite neat, doesn't it? I Like that there's our battery just whacked in the back.

You could actually retrofit a bigger one. There's a bit of unused space in there, so Jesus Not much doing is there. the analog front-end I Do like how they've like laid it out. look at the nice neat rows there.

I Didn't like somebody. probably took a bit of pride in that. I Don't mind that is there anything on the bottom? and nope Zippity Doo da R on the bottom. There you go.

Just got our membrane key pad. there's our bean see and no LCD And that's all she wrote. So the main processor here is an ARM Cortex rm4 2 Giga devices 32 F 4:07 and this is a bit of a monster. It's got like I believe it has like three Meg of Flash in it.

like 192 K of SRAM looks like an SPI Win Bond flash memory over there. That would be for saving all of the waveforms and this is interesting. Look at this as a mystery device over here with absolutely no markings on it. I've looked under there and it doesn't look like it's been scrubbed off.

It looks like it's actually been supplied without any silkscreen at all and that's clearly your analog to digital converter. You've got your output here and we've got a wiggle-wiggle-wiggle year trace there, which a serpentine trace. Let's just start length matching. That'd be the sample clock.

so anyone want to reverse engineer the pin out there and try and figure that out? All right, Well, let's go check out the input down here. We've got ourselves a solid state relay. This is our AC coupling cap. You can see it's just basically shorting out that AC coupling cap and that's driven via directly from the micro down here.

So exactly as expected, we've got some duo protection here and all this stuff here. This looks like an input divider with compensation caps across there. Why have they got to? Hmm. Anyway, got ourselves a real fair dinkum relay there.

That's pretty good and these have looked at the datasheet they would be photo must realize would they. And then we've got a whole bunch of trainees down here. Looks like they're driving these relays over here. and we've got ourselves a little 5 pin slot 23 amplifier in the way between the output of all of our front-end switching and the ADC.
So that's just because you need a driver for the ADC obviously. So the signals coming in like this, it's been AC coupled like this and it's been divided. This input divider is this two here. and then you've got this long divider chain here and it's all compensated.

It's got double compensation caps on each one, and then you've got the input capacitance of all of your switches here. Which these tabs one of these taps. so we've got one in the center of that. one two, three, four five, six seven one two three four five six seven eight.

Oh, we've got eight. Oh what? Ah, there'll be one at the top. And of course, so yeah, it's it's tapping off all of those. and then you've got the input like you've got the capacitance of these as well, upsetting the applecart.

And when you've got a huge big divider like that, then you're gonna come a gutter. Um, it's just this front end is just not designed for high frequency like we saw like forty six odd megahertz before it really starts going silly buggers and nonlinear. But then you couple in the fact that this thing is clearly not sampling at 500 Meg's samples per second because there's no way that this micro what does it. I think it's runs at 168 Meg or something at megahertz in something isn't Anyway, you can't use this.

generating the sample clock for the ADC over here. Um, somebody on the Eevblog form is out reverse engineered this by the way, and they provide a link down below for the reverse engineering thread. I Didn't know about this before I started shooting this video and they've identified that as an analog devices or a clone analog devices part. I'll put in the part number for this thing.

You can't generate 500 make samples per second with the micro. It's just not possible. Look, there's one oscillator over here driving the micro. The micro drives the clock for this.

They're just not doing this right. This is not a proper oscilloscope architecture and okay, you might call it. You know you might be able to eat more than 20 mega out of it or something. but I would say no.

It's basically gonna be like a, you know, 20, maybe 40 at the toss, Maybe 50. Yeah, absolutely pushing it. But as we're seeing the compensation on this comes the gutter. Nothing's accurate in this thing.

It's just hopeless. Nothing much doing there. That'd be a charger. Not even gonna go and look that one up.

And then we've got some awesome 4hc 148. So they'd be driving the keypad. That's about it. and then Sam Um, like a pair on city.

good night voltage regulation there. Yeah, so basically they're just asking this poor little front-end circuit to do too much for a hundred mega that's just coming to gutter. I Did like the fact that it has awesome software offset calibration and things like that. but yeah, this is not by any stretch a professional oscilloscope front-end that you'd find in your I goals and your siblings and your key sites and everything else.
Which you know, we used to pretty simplistic front ends these days, but no, they're just. no, this is too overly simplistic. so I can't possibly give this thing a thumbs up. It just doesn't do the business and it's unfortunate because it had a lot of potential and it's a great like the actual build quality and feel of it and everything's fine.

I Don't mind the user interface and things like that. I I think you know and the button interface and stuff it, it has a reasonable amount going for it. Unfortunately, they've just come a guts are on a couple of hurdles here and maybe I'm this new hacked firmware. This completely reverse engineered firmware from scratch might be able to, you know, make this into a more usable platform.

perhaps. But yeah once again of whether or not is there anything else this capable point even this capable in quote marks for you know, 70 US dollars is a good value. I Don't know unless I did a shootout and I don't have any of those IDs Oh no knows RDSO quads left to actually do a comparison shootout I gave those away to beginners. So as a standalone product, no, it does not get a thumbs up.

In fact, it virtually gets a thumbs down. Um, because they've just goofed this and it does not meet its claim. So it's got to be a thumbs down. But it might be a usable platform for some people.

You know, if you just want to view some waveforms or whatever out in the field and you know, stick under 20 mega or something and you don't care about the actual reading? No, no, this DTI anyway might be useful for someone. I'll provide a link down below anyway. but yet I'll provide a link to the reverse engineer in the forum of this. It'll be interesting to see what any additional firmware can do for this baby.

But yeah, that and the fact that there's just no hardware triggering on this. It's all done via software triggering. So yeah, you can you know improve the software in this regard. But and it needs to be improved because the the stock software on this thing is absolutely horrible.

So that that front end, there's not much saving that you're not going to get a hundred Meg out of that for sure. And the sample clock is just. it's not going to do the business. No way.

I Don't even have to measure that. Don't even have to measure the sample clock to know. Oh God. Alright, I'll measure the sample clock now.

First things first. this is at 6 milliseconds per division, you can see that's not sampling all the time. it's sampling in a little burr. soom into that at the moment.
But that's a 34 Hertz 34 waveforms per second. But hey, you don't expect much more for a like a purely micro based system like this. So you can see that we've got two o'clock lines here. and of course all these data lines coming out here.

This is for 180 C This one's for another. so there's actually do lady scenes side here. So the one I was just probing is this one over here? Let's probe this one here and have a look. You can see this one's actually opposite polarity to this one here.

so that's interesting. Oh well, you know it doesn't really matter. they're just maybe that's just a software quirk. They're just idle in the line I instead of idle in at low.

Anyway, if we zoom in on this. Okay, it was single-shot camp and capture that and we're looking at yep. 121 megahertz there. So 500 mix samples per second? I Asked: 25 nanoseconds per division.

Still hanging in there at 50 and 100 nanoseconds? Yep, it does not change until 500 nanoseconds per division and it changes quite drastically. It's dropped down to 25 megahertz and the opposite ones not coming through at all. So obviously it needs both of the ad C's in there to get the highest sample rate at the highest time based settings. but even then it's definitely not 500 Meg samples per second.

It's only 125 make samples per second. But with you Joe ABCs into, leave those to get 250. But look how it's getting the throughput of the data into the SRAM into here at this sort of rate. And of course it's got nowhere near the 128k of memory even though this thing has 196 k of SRAM inside of it.

But yeah, it's a it's the most rudimentary system you can possibly imagine. and of course the slower time by settings. they only need to use the single ADC. So yet now this thing has just come coupled with the input amplifier here, which is pretty how you're doing with its long compensated divider chain.

You're always going to come a gutter on that and just direct sampling straight into your micro using your GPIO pins and I'm surprised they push it. They could push that to 125 mega. It's actually so yeah. but maybe some firmware can help improve this thing.

but there's no way this is ever gonna be a hundred megahertz bandwidth scope. Not a chance. maybe twenty. So there you have it.

that's the Fennessy 51 to H. High-performance no reduced performance little pocket oscilloscope and well, I was a foolish to expect much for my 70 US dollars. but anyway, it did have a potential, but unfortunately arm it's just let down by it's relatively performant and poor performance and lack of options. you know and here it's only single.

Channel but hey, that would have been good if it did actually have a decent hundred. Meg front end on there and if I had some decent options in the software and the triggering worked better. and the sample memory you can actually zoom in on your waveforms and just like yeah, it's it. let down at the final hurdle I think but yeah I don't know there might be an issues for this you how like the form factor.
kind of like the look and feel I kind of like the layout and the options that had it. just it. it was just let down like just huge showstoppers. almost like you know the the sample memory not being able to zoom in and it not working properly in just normal trigger mode.

I end the lack of the bandwidth so no I'm not sure I can recommend that unless you got some niche application. Is there better value for 70 bucks? I don't know unless I did a shootout but yeah I was hoping this would be a winner winner chicken dinner but it ain't but you know it might be useful for some people. Littlefield Pocket view away form you don't care about zooming in and the auto modes good enough for you and the bandwidth is good enough for you and and whatnot? maybe. But anyway.

I'll leave a link down below if it does suit your purposes. No, there's nothing else to write over my mom about I Think there's nothing to write home to my mom about apart from the 70 bucks. but even then, no, I can't give that one a thumbs up. Anyway, let me know what you think down below in the comments and over on the Eevblog forum as always which is the place discuss test equipment The number one test equipment forum on the Internet, no doubt about it.

Catch you next time.

Avatar photo

By YTB

28 thoughts on “Eevblog #1260 – $70 100mhz oscilloscope?”
  1. Avataaar/Circle Created with python_avatars Fredy Casilla says:

    "100 Mhz ,my ass " XDDD

  2. Avataaar/Circle Created with python_avatars Greta Laube says:

    Circuits are hard enough to diagnose, let alone if your measuring gear is giving you BS! Nah……I got enough headaches.

  3. Avataaar/Circle Created with python_avatars miguel says:

    It's good enough to test a 32768 crystal. It's alive oh ok! The first sign that it's bullshit is in the specification, 100Mhz @ 500MS/s. it should be 1GS/s minimum for a respectable wave representation. Thanks for the analysis.

  4. Avataaar/Circle Created with python_avatars techVlog says:

    Pilipino loves u bro❤

  5. Avataaar/Circle Created with python_avatars Mark Anderson says:

    Enough of this dodgey talk. It does display to 100 mhz, I never troubleshoot a circuit with frequencies over 10 MHz. I always thought that a low bandwidth meant you coukd not display a proper square wave at frequencies even 10 percent of the bandwidth. A 10 MHz square wave wont look square on a 20 mhz scope. Most of us are looking for signals from sensors and on a bus or enable lines. Of course it wont work like a real bench scope. Most other low end scopes can barely do 400 khz. I did buy one and am happy with it for my uses like radio service or bit chasing in the field. Better than just a dvm on AC for sure.

  6. Avataaar/Circle Created with python_avatars Hirus the Hellhound says:

    Its good for sound viewing though

  7. Avataaar/Circle Created with python_avatars Light Science says:

    I would love to see a vid like this for the 138 pro 30 dollar 200khz scope. 2.5ms/s. It's like 2 inch by 2 inch by a half inch, battery powered.
    Basically I'm wondering when you buy a cheap scope if you stick to a more reasonable bandwidth limit for something so cheap does it perform within spec at that point rather than calling something cheap 100 Meg?

  8. Avataaar/Circle Created with python_avatars Atomicskull says:

    A simple low cost scope like this works fine for R/C where you just need to look at low frequency signals to troubleshoot and a $1000 scope would be overkill.

  9. Avataaar/Circle Created with python_avatars HERMINO JALECO says:

    I buy that only for audio amplifier troubleshooting and its enough for me to have that…

  10. Avataaar/Circle Created with python_avatars JAM łasica says:

    This pocket scope is great.

  11. Avataaar/Circle Created with python_avatars Pranas Momkus says:

    Diedo reikalavimai per dideli

  12. Avataaar/Circle Created with python_avatars al klapaxida says:

    these little do-dads oscilloscope are are for tight spot work like in the car under bonnet work looking at canbus are brilliant
    serious

  13. Avataaar/Circle Created with python_avatars pcrengnr1 says:

    Thx Dave for the rvw. It appears to be a 20 MHz scope that's being pushed to higher freqs or claimed to have higher bandwidth. It's a shame the Chinese feel the need to lie so badly. Maybe you'll do or say anything to get a meal when you're hungry.
    Again, Dave thx for the rvw and teardown.

  14. Avataaar/Circle Created with python_avatars Vandalyst says:

    can you measure mains earth referenced circuits with this ?

  15. Avataaar/Circle Created with python_avatars Alex Rodrigues says:

    You can make a vídeo fnirsi 1014D ok! Thanks 👍!

  16. Avataaar/Circle Created with python_avatars Antonio Lanera says:

    É UM BABAXCA, NÃO SABE DE NADA VOLTA PRA UNIVERSIDADE

  17. Avataaar/Circle Created with python_avatars Juan M Ramírez R says:

    Compañero, buenos días…Por un casual no habrá tenido algún problema con él…..El caso, es que tengo el mismo modelo que lo habré utilizado un para de veces para ver una señal de baja frecuencia….El caso, es que al ponerlo en funcionamiento, la pantalla me arroja unos valores erróneos sin tener nada conectado….Al intentar calibrarlo, me arroja FALLO en la Calibración…He intentado en multitud de ocasiones que me presten ayuda el Fabricante, pero ha sido inútil….¿sabría cuál es el problema que presenta?…Un saludo cordial desde Las Islas Canarias….

  18. Avataaar/Circle Created with python_avatars حالاتى status says:

    made in china😉

  19. Avataaar/Circle Created with python_avatars billybbob18 says:

    That programming error you found is very common. It's referred to as the "off by one" error.

  20. Avataaar/Circle Created with python_avatars Charles Kinzer says:

    For the cheapie scopes that all seem exaggerate bandwidth (or conflate real or imagined sample rate with it), however, you get a lot more performance out of this one with its 2 to 5 times exaggeration versus the DSO211, DSO212, and DSO213 claiming 200 KHz, 1 MHz, and 15 MHz respectively with a similar exaggeration. And those cost from about $65 to $210. So, the 5012 is easily the best value (if you just must have one of the "mini" scopes) if bandwidth is important.

  21. Avataaar/Circle Created with python_avatars Charles Kinzer says:

    I got one of these to play with as I am reviewing a number of low cost scopes to advise others for model railroad purposes. (Testing most of the DSOxxx tiny ones and larger Hantek ones like their 2C42 which I think is money MUCH better spent than for any of the "mini" scopes.) The one I got was marketed by Yeapook as the ADS5012H and seems identical. It is a somewhat different animal than the DSOxxx scopes which have ludicrously exaggerated bandwidth claims. The 5012 peak to peak voltage diminishes quite a bit once you get above 20 MHz, but it can still show a semi respectable sine wave to well above 50 MHz (although slightly flat topping above that). Of course, a square wave at those frequencies looks extremely trapezoidal. I found one of the most interesting features being that a button is provided for pretty much each of the day to day operations you do. Even a 1X/10X button is there which is something buried in a menu on every other scope I have seen. By far the easiest scope for a beginner to use. It comes with a 1X/10X probe but like most cheap probes with cheap scopes, it introduces voltage error and has no means for compensation. You need to use a 10X probe for max bandwidth so if accuracy is important, you may have to buy a scope probe that costs as much as the scope. I must say that after flailing at a number of the other mini-scopes and figuring out the trick(s) to access all the menu choices, this 5012 scope was like a breath of fresh air. It even came with a very nice semi hard shell case with room for accessories. Finally, the case has the slogan "A hardy tool makes a hardman". On the Amazon site, the slogan is "A Handy Tool Makes a Handyman".

  22. Avataaar/Circle Created with python_avatars x86guru says:

    I measured only 23MHz of bandwidth on mine. The 100MHz bandwidth claim is simply FNIRSI arrogance and dishonesty. A 25MHz 5.0 volt p-p sine only measures 3.32 volt p-p on the 5012H.

  23. Avataaar/Circle Created with python_avatars Rich says:

    DIY scope project – thanks for the review…I am interested in building a diy scope as a learning experience and this discussed a lot things to consider. I have been searching your other videos for a discussion of the properties of a real scope vs a crap scope (I think you discussed the 2 categories in one of your videos) or even better a design discussion of the blocks that are involved (analog input, A/D, storage, speeds, etc.) have you done one on the topic that I am just missing. I have watched dozens of your videos and have learned much from them thanks for sharing your knowledge.

  24. Avataaar/Circle Created with python_avatars Chase Yeager says:

    It's good for car audio set up especially when tunning you're amplifier for checking clipping.. I only used for car audio amplifier setup for Wave form ahhhhhh

  25. Avataaar/Circle Created with python_avatars gen kidama says:

    lool, thanks for the laugh, and the money saving !!

  26. Avataaar/Circle Created with python_avatars APM Electronics new tech. says:

    Very good from India 👌

  27. Avataaar/Circle Created with python_avatars D D says:

    LOL, the Chinglish is hard to understand even for a Chinese.

  28. Avataaar/Circle Created with python_avatars q zorn says:

    may be it would be a little more fair for a show n-tell from someone who has used the o-scope on some troubleshooting projects? a least the little scopes are getting better and cheaper.. thanks a lot…:)

Leave a Reply

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