Owon VDS3104 - a surprisingly decent USB oscilloscope.

hi we've got today is this this is the on VDS 3 104 USB scope it's really unusual in a hatch having four channels they do a range from these they do some turn for channel and 25 50 60 and 100 megahertz versions and they seem to break into two two classes depending on how much memory it's got the language on got five K in the hind which you've got ten ten examples that depends on your channel it's basically ten mega samples for two channels or five for our four channels so it's all the same memory architecture on those so but it falls into these two categories on the lower one they do do a version that's got an isolated USB there's also an option for a LAN connection so for the high-speed ones you should barely get isolation because the 10 base T LAN connector is it has built in isolation it should be possible to get an isolated scope by using that where the uni isolation really depends on what some applications you're dealing with I'll see you in the land one that the channels won't be isolated from each other and unless you're doing fairly specialty stuff that's can actually be more of a pain than the benefits that means you actually actually have to connect all your programs rather than just relying on the the ground for example if you're probably being around something that's connected to say that that same computer's USB port then a lot of time you can just rely on the usb ground to Sony low frequency applications just to avoid out the tipple the ground clips on comes before these these are pretty much the standard probe you get most Chinese scopes jobs are perfectly reasonable and your total connectors on the front you've got the full channels you've got these plastic I don't like plastic BNC connectors a little bit can be a bit fragile not the end of the world but there to get potentially prone to getting bashed and the case has got these sort of soft bumper feet things which to be nice I think a little bit pointless firstly they the gallery stick on very well they're quite prone to just falling off and they don't really add me give you a little bit of a sort of foot foot grip but they don't really protect the BNC so to be honest IV Bessel just three the three these are one puts a little low profile stick on that roller feet on there we're doing just as good a job and reduce the bulk a little bit on the back ends so on the back we've got the probe calibration they've got this connect to this is can be configured to be triggering trigger out or pass file for the mask test now it's a one these little connects I'm not actually sure strictly which which flavor of small characters it might be SMB or something by I found a copy in my bag of you know random coax connector so it's not something that's too obscure but you're not even going to use that so not a big deal there's an unpopulated comport unpopulated LAN port on this one and the other thing you notice is they've used the wrong USB connector now I'm assuming this is because of the height now there's some argument ok well that's probably a bit more rugged than a micro USB but yeah micro USB are so common now I think maybe if they perhaps use the micro USB and proudly tested to give a bit of strain relief that might have been a better choice than using this base it means you've got to use and understand a cable and just having to you're having one of a cable you have to have on hand this is the cable comes with it I think it's the you know the incorrect eight-way thing but also it's got this dual plug because the powder on this is a little bit more than what you can legally take out the single USB port I've measured it about 860 milliamps that seems to be fairly constant regarding what you're doing now and my laptop seems to be quite happy providing that out of a single single port but because you're you know you the whole point about this is it's also a lightweight portable scope use on the go USB scopes aren't really any substance you can probably top scopes you know we really odd knobs but we all take something slightly small light and you're probably using real laptop so it's not ideal if it has to use two ports but I'm sure you could even if you need to putting makeup up plump up a power split arrangement to run it for an external adaptive who really wanted so just a bit annoying but you never take someone that amount of care and then there's probably not a huge amount you can do with it I'd imagine perhaps the the lower arm ones might perhaps some take a load off power to run directly from a single port but exactly on my laptop at least it's does seem to work ok on a single port nothing to complain and there's also a lead there so I just done a quick test what it appears to be this looks like it turns green once the internal process has booted so if we to plug it in wait for a few seconds and I did hear a relay click last time ago I wonder if maybe they using latching relays and that already lacks but say then goes green if I reduce the the power supply voltage instead leave is a switching regulators as I'm reducing the supply voltage the current is increasing at 5 volts it's about 770 millions it's not active smog or anything talking to it so like a little bit when it's actually working or juice is slowly reduce the supply voltage it drops out at 3.3 volts so it's when it's got a switching regulator and as its near dropout it takes out of an amp and I suppose one advantage of end supplying the USB cable is that you know you know you can get decent quality once if this is putting an amp then you know one that a lot of USB cable these days have substandard conductors so this feels or fairly to Freddy's or solid chunk of actually measured it but eventually other place apply this cable so they know that you can get a decent decent low resistance you're not losing too much on the cable the case is very simple just let single aluminium extrusion just screw in each side with the plastic cap so it feel you've got quite a nice feel to it it's nice and compact it's not smaller than a lot of the other PC scope so it's just you know the case on the board and very little wasted space what you'll probably want go do it would have been nice if they'd have supplied it in just something like like a failure sort of thin sort carrying bag that just gives you enough space to put the probe and USB lead and maybe have a little zip compartment for all the little little bits and pieces of the probe various other bits that come with probes is always nice just to be able to keep all that together and also a little compartment for a little bit and otherwise get lost I'm sure you can really find a some sort of case that will fit that it would have been nice to actually supplied that in fact the the case that the probe is coming you could possibly almost get away with using that about the right size so final sale you can actually fit this in and you probably get at least two probes in there in the USB lead before I got to two ridiculous though what comes the words to each exact so this the PCB in the traditional on blue solder resist our son sold of this shield is tax holes on a few places some of that there's also some copper foil between the two shields I think that's probably mostly to make contact with the outer metal case when it slid in yeah fairly straightforward light we also got the input for each channel we've got FPGA there the samsung 400 megahertz online processor there I've got some ddr2 RAM for the acquisition around four main processor flash ROM a two-day ehm CD 1511 from p-type she's pretty standard and know that they're multiplexing all four channels into 180-day so your sample rate does go down with each additional Charlie I'm sure that that have done it so it's like one one you go to a single channel 500 or 250 depending on the number of channels so desktop for four channels copes tend to use two so like the 2g controversial with money today and I for trauma version we - they've multiplexed into single Hz there so I'm just better mind if usual channels that your sample rate is going to be on your quarter and part from that it's all sort of pretty straightforward is a couple of hd500 5s down there just to do some digital i/o for the front end you see there's some capacitors on the apples there again should presume adjuster filtering noise out those which get Gulf down here there's a nine you use their relishes for the LAN option so eath let's magnetics and Ethernet socket there and then unpopulated rs-232 transfer rs-232 transceiver with some connector on there for it and then unpopulated chip there I've no idea what that is one move maybe there's perhaps an options like an ALP generator or something coming out of the socket we populate for that and really not great your little engines all the power supply stuff here some switching regulators eductor so it's you know pretty much what you expect to find in a scope but - the display in the user interface stuff it looks fairly clean your front body can't see any odd wise but there are in rora if you wanted to little mods on here the fusel components christmas treat on top of the other lizards of capacitor down there and the art there is the old one or to some components piggybacked on top of others here are their but nothing nothing too drastic not much on the other sides nothing it's all under the shield just a couple of chips down there the couple of capacitors interesting although they haven't easily that component they fitted the capacitors across the isolation barrier there but we just they didn't have to worry about having different different configurations for the bottom side population they've done it all the same with the front end BNC socket relay there's a solid-state relay and a capacitors that's probably going to be for the ac/dc switching is quite common way of doing it a few quite a lot discrete and there's a MH 6.5 on differential am producing the differentially output to drive the ADC that doesn't seem to be any attempt to sort of equalize all the trace lengths or anything now on scoping this bandwidth I don't think that's really necessary chop synthesizer there to generate clock for the HD and yes pretty much it's pretty much what you expect no real surprises right now beings or Chinese supplier I wasn't really expecting a great deal from the software because yeah that's where a lot of Chinese products disappoint but we're actually been sort of reasonably pleasantly surprised in certain in terms of yeah pretty much everything there that it is in there does appear to work reasonably I did have it crash once I'm not quite sure exactly all that was down to it is involving so fairly large download new phase or large chunk of memory but most of the time not going to go into like a detailed review of everything here in terms of the features it's got pretty much all the features you'd expect you know this or scope and you know the features that are there do seem to work reasonably well there are a few issues with the user interface that I think could be improved but generally is fairly usable in terms of screen size this this laptop got slightly unusual resolution it seems to adapt reasonably well or not if you want really minimize it you start losing some of the channel indicators but you can actually live without those Technic people you can actually for example if your TV you've got something here debugging you need the scope screen as well nothing to be reasonably flexible develop Maximizer one difference between this and most other USB scopes the traditional USB scope basically it's just the front end the ACD converter that chugs everything into the PC and then the PCs you tells everything else this and in some cases it also does triggering based on the sample memory which is pretty awful at the time as you think the internals yeah this is basically a complete scope in a box and they're just really using the laptop as a remote display in user interface which means all the triggering and capturing stuff works pretty much like a decent scope but one thing in particular it means if you're capturing with a large memory depth you know you're not constrained by how fast it can get the stuff into the PC because that's all handled in the box you know there's only there's only so much resolution you're ever going to see so what it's basically doing is transferring data at like the screen resolution in real time so for example if I set changes to full memory to that five mega sample memory depth yeah it's a little bit slower but not you know it's only not transferring that whole that whole memory into the PC it still so reasonably acceptable and if you go to some intermediate depth iso100 cave still perfectly responsive I see the response time is it's pretty good and again pretty comparable to a low to mid range standalone scope in this sort of class when they do notice a if you do still on the say stop and then start zooming in that's where you do get a a quick delay while it downloads the data so rigid so I do a capture the full memory depth if I now want to say to him into that it does take a little bit of time to actually extract that and I've actually got all four channels on so this is basically downloading data for all the chat all four channels as well so that is a little bit sluggish but once it's in there you can then zoom in and around it settlement units a one-time thing and if you want to just actually work on a live zoomed in waveform it's still reasonably responsive I turn those other channels off they'll pretty speed it up as well but if I start to turn that back to live y'all can be zoomed in on that for capture depth and it still gives a you know a pretty acceptable response time now magic as user interface there's a few yeah some of the obvious things do what you'd expect to them but there's like quite a few things I think that for example why shift you just grab this and move up and down it is a little bit sluggish I suspect what's actually happening here is it's actually sending that value back in a way to the scope and then it's putting it back so there is a bit of a turn around aligned there are some operations that are actually quite sluggish which I think it may be due to the fact that you've basically got a complete scope in a box and it's just sending that bad in so there's probably scope for improving that little bit there are keyboard shortcuts for the really important stuff like say you've got the left hand side of the keyboard these are your 4y sensitivity controls so as you move those up and down that works each channel each of the four channels so you've certainly got you know a separate key for each channel which is what you want and similarly left and right arrows operate the time best now the Y sensitivity works the correct way ie going up makes it more sensitive which is normal convention the left right arrows unfortunately go the wrong way the right how you would expect to actually zoom in and make it faster but it actually goes the opposite way around go go slower touched a minor annoyance and again horizontal positioning you've got you've got here and there's also I think f5 is run stop but there's there's not many other function keys I think that would be scope for adding quite a few more keyboard shortcuts for the sort of things you use more often but it's it is it's a decent start it would be nice to see some improvements um in future and this is your trigger level over here again you just set that in your time and that works quite nicely as with a lot of our own scopes it does start suffering from aliasing so if I just crank that slow the time base down this is a hundred kill outside we were feeding in their way so that's at the full memory depth so still seeing a fairly fair amount of yeah waveform with no no aliasing if I turn the sample rate down to 1k so the sample depth down to 1k you'll start out using fairly quickly and show garbage but it does have even you know if your the limits of memory and you're still getting aliasing it does have peat detects ample mode so if I put it into peat detect mode it will now show that there's at least something in there so though it doesn't show the detail each it does tell you yes you know it is aliasing exhibiting the peak peak measurement so you know you never get that situation where you just get a completely false waveform but you do need to remember to turn the the peak on it is Center in the center that when you when you close the program when it comes back again all the settings are as it was before which again is yeah what you'd expect from a normal scope so that that works fairly sensibly one my little thing I found on the triggering I think there might be some fix hysteresis going on eggs if we look at this this waveform you'll quite happily trigger when we're in about the top half when we set the trigger level to the top half but as we go lower it won't actually trigger it's an auto mode the only thing spawn to trigger on the top half if we increase the sensitivity of it you know the amount in the bottom that's not pregnant seems to be fairly fixates again it's that little sort of about a third of division so I suspect there may be some fixed hysteresis so it's I want you to know about it's not a big deal it's just said that just a minor annoyance it pulls me for awhile as to why I couldn't get things are tricky with excited I had the combination of a fairly low input sensitivity and I was expecting triggering in the middle it's got reasonable range of trigger mode you've got edge video slope and pulse but that all seems to work now one of the things that I find particularly annoying about this user interface a lot the more complex things you thought go into this little popup menu in the got various all those functions in here again it's pretty much all stuff you'd find normal scope but some of the selections are a little bit annoying for example you have got two magical single mode alternate mode and here you know we actually have to pull down this menu to get these options now when you've got four or five that makes sense but there are some menus where there's like only two options in here which have been done better with just your having two separate buttons but probably the most annoying thing of all is certainly some of these entry fields particularly thing up the pulse-width or hold off time you can't just go in there and try to figure in you have to use the up-and-down things it's got also repeat the only thing to change the step size is to click there to go in so that's basically going up the left-hand digit at a time and then you'll go down there to do this off the middle digit and then this one's the last eg which is just horrible I hate that I happen to know that this starts with a pulse width which is a little bit over 90 microseconds so I'd like to get a time line 90 in there was at the moment to get 90 microseconds I've got to go down until we're into the Microsoft sort of microseconds range earlier go but it's yeah it's unnecessary fiddly they could have improved that but it does have a reset button there the other thing I find a little bit annoying is the trigger level it maintains different separate trigger level values for the different trigger mode so for example if you can now go into edge trigger mode it's actually got a different voltage low because each trigger mode got its own threshold which I can't think of any situation where you'd actually want that generally want to change trigger modes on the same trigger level the other thing is the actual although we can do it on-screen if we do it here it's actually quite horrible and sort of slow and you see it's quite sluggish you do get a line on the screen but it's it's a bit laggy but generally there's no basically unless you know you want a specific value then just do it here so it's generally quicker and easier just to do it by just dragging it and one nice thing I'll alternate trigger you've actually got all the modes for each version so you can say alternatively trigger say by an edge and on channel 2 trigger by let's say a pole a pulse-width I think also because we are the yeah because we've change between seeing more turn again it's also got its own set of trigger settings for this so that we change trigger modes again our trigger levels and so on are our aren't set out properly we need to basically duplicate that so that's a little bit of an annoyance say pretty much everything you can get through this menu and there's shortcuts to certain other things that they use more often we've got they the four three two the four channels we've got some basic things gosh you've got the indicator as to what it wants what the settings are the offset is it's also got an input frequency indicator just shows you what's coming which is sort of quite useful to chose that in all the channels and when you've got like the channel on-off control there and down in this corner we've got our current trigger mode we pop up the trigger menu we enforce a trigger so for in a mode where waiting for a trigger we can just the the time base the delay the memory depth and the sample rate the sample rate just comes from the all the other parameters up here again we've got some basic controls we've got run stop single mode single trigger mode bubbles like autocephalous hit the auto set button that takes a few seconds to figure out what's going on and then also say it's powerful key that he doesn't have an undo auto set the other thing that I haven't quite got the hang of I don't think it's implemented particularly well it's I'm really missing something is the way it deals with the deep memory and that you've got this little bar on the top that shows you to all the whole memory in the section you're in but you the actual sort of navigating around that doesn't seem particularly good so we've got again we've got slightly inconsistent and the parents we've got main then assist set which i think is sets the time window so this is like the sort magnified version and then the actual zoom that zooms in but there's no way of getting a simultaneous display which is very disappointing you're normally what you want to see is you want to see the overview of the whole waveform you know a couple of cursors to show you which bit you're looking at and be actually then want to see the zoomed in section simultaneously but no you have to sort of change that but I think in practice you know you can actually just use them all time based delays to to achieve a lot that in practice just seems not particularly sensible and one thing I think which is ridiculous is that when you set the time base over 100 milliseconds it forces you into this roll mode but also it forces you into Auto trigger mode it actually tells you when when time provided on under milliseconds trigger would have set to auto mode but you know if you're wanting to catch or capture infrequent events you know that would actually be pretty disastrous so that's something it definitely needs needs looking at and again I mean the fact you've got you've just for setting your three basic trigger modes which are like also normal and single you've got this stupid pulldown rather than just all three buttons on odd Laden though that's something which is sufficiently so fundamental I'd actually like to see dedicated buttons to detect that I'm just agree running through what else we've got in here through all the channel selections some of these are available as the sort the quick buns on the bottom we've got some additional functions we've got some inverts 28 bandwidth limits ac/dc coupling and ground and the probe scaling which is again pretty much we expect measurements you've got pretty much you know all the measurements that you want now you can actually make it display everything which is we can make it display specific measurements say 10 n over the screen a you tell you which measurements you want on which channels doesn't seem to have measurement statistics as I can tell but it's got all the little stuff lead generally one and you've miss them you can tell it which channel so we'll show the very soft figures what you want so at least it's not too tough that if you don't quite understand what this seems to be like a non dynamic display that just pops up and shows everything so don't understand why they bother with that and he was quite insulting Modi no we've got whole menu just for setting three sampling mode options which is pretty normal three modes you've got some standard peak detect and averaging markers again work pretty much these are just just work on you just drag them around and your shadow figures a little bit dark yeah so that's again pretty much box standard manual markers display dots versus vectors again that's all pretty standard you've got variable persistence which seems to work reasonably well grid plan grid control there's also XY mode it's got this mobile it splits the screen into three three sections which is just seems a little bit bizarre you've got the main waveforms here this is XY and up here is your the FFT mode so for some reason they've done this you know so you can only display these modes in these dedicated sections which just seems a bit peculiar and in fact there's actually a bun down here which actually splits the screen but unless you've got those loads I'd really don't understand what that was going on there so you've got the usual math functions and the FFT I'll stick a sine wave in a minute and do the FFT stuff that actually works surprisingly well pass/fail mode we can set up some affecting masks so if you enable that we can set up some pass/fail masks and generate sort of statistics and make it stop on file we've got here okay we've got record playback and again that just records a section yeah effectively it records pretty much what you see on screen you can't like playback and then stop and zoom in on individual waveforms it's pretty much a screen recorder so you need to set the screen up right to do what you want but that's probably quite used when you can play it back at different time scales so you can actually record a burst of data and then step you through slowly and the FFT mode actually surprisingly quick source it's doing the processing in that in the Box are not on the PC sorry first give it a we've given one make out sine waves and see that the frequency peak there we can set various different send frequencies we can crack the center around but say we get this waveform and FFT display in a completely unused section over here yeah say a lot of things when you do it says yes to leave f FC before you even go into a certain menus yes so in your XY mode we've got this sort of dedicated display area which is seen as a bit older don't I just don't understand what that's about we've got this yeah we can flip between that in the form oh but we can't get either the X Y or the FFT to display full screen without um without the my waveform right so in conclusion I think I've done a pretty decent job of this I think yeah the price is not unreasonable is a similar sort of price to conventional scope with the same capabilities which okay there's a lot less in it but it's a much lower volume product the USB scopes will never ever be a substitute for a proper standalone scope you know where you need something which is smaller than light the concept of like complete scope in a box and a remote display rather than just a 2d in a box and let that let the PC handle everything is a far far better approach it means it behaves much more like a proper scope say it doesn't replace a desktop it also doesn't necessarily replace of the handheld scope because it depends really I'll see every job is different for a situation where you need to travel light but have a scope with you I think this is a really good solution if you're debugging something which is maybe also connected to the laptop you might perhaps want to have a second monitor or whatever one thing that I would really love to see is if they actually printed did some software that would look around the tablet because you may well have a laptop and a small tablet or even a phone with you the ability to run it front on a tablet I think will be really good a quick look through the software I don't know huge about these things but I think it looks like it's Java so maybe that's something that they were thinking of but so having the scope just you know just running on let's have nothing would be really nice yeah where it isn't really very good for is if you're having till I go around for example a big installation probing things and fault-finding that's where this you know you just can't be a proper handled scope for that sort of thing but if you're just traveling and you know the issue is just bulk and wait actually cutting the stuff around and then working on desk you know with this you're lugging around a spare display a spare battery and maybe a charger as well so this is actually quite bulky fatiguing off you want to get everything to hand luggage as well this is clearly a lot smaller a lot lighter and obviously this is for channel touch but it's more capable as well and you've got a bigger screen to using a laptop yeah I don't think there's really anything particularly wrong with it the software is quite a lot better than I expected it does seem to do everything that it purports to do there are some things that I would really like to see on the software I'd like to see more keyboard shortcuts I'd like to maybe see some of the more important functions like perhaps trigger mode thing might appear as buttons on it so they could at least click those and see what the current settings is maybe even sample depth perhaps expand this menu a little bit and so I'll certainly like to see a few more keyboard shortcuts for things use off and also the expansion of zoom displays I think could be improved I think all in all I think it's you know a pretty nice effort yeah I've not really looked at many other USB scopes because celty of all the ones that are just nicely in the box horrid eyes they're just going to be pretty hopeless but I think yet as a scope a lightweight alternative escape I actually think they've done a pretty decent job of this and I'd give it probably probably eight out of ten and I think that could easily be nine out of ten if they just spent a little bit more time really polishing the software and maybe did at a tablet version I think possibly the one respecting which it isn't comparable to modern low-end scopes is it doesn't have any form of intensity grading in some modes the update rate does get out to about a couple hundred waveforms a second but it's certainly not doing proper intensity graded display which is becoming standard in low and display low in scope so see that's something which would need probably need improved some improved hardware and so much off you would like to see but get given that limitation my summing in this price range with intensity grading as well will be yeah absolutely super so maybe will perhaps a new generation of that that would be good to see I'm not quite quite sure where or in terms of their competitors on that side of things but I'm sure as I'm when they that's on their mainstream scopes they may well look at migrating that to the USB side of things because this is basically you know she once you develop the full scope entry USP boxes patient in case of taking the display and the user interface out so hopefully we might see that but not really aware of any really any USB scopes that do that at the moment

Channel: mikeselectricstuff

Views: 90,264

Rating: 4.8560886 out of 5

Keywords:

Id: vO712DvGAKA

Length: 30min 48sec (1848 seconds)

Published: Mon Dec 21 2015