Hi All,
I know a lot of these threads get started, at times as the first thing a newbie does when window shopping 'scopes (I've read through at least a dozen by now) but I've not yet seen one which hits the points I want to. I'm going to try and be as detailed as possible to save too much back-and-forth Q&A, but also organize to avoid a huge wall of text.
I have a cheap and nasty tablet oscilloscope that I inherited off someone else who - like me - decided it's not worth keeping, and want to get something a bit more future proof. Now, I have a good idea what I want to do with it, but I'm not convinced I know what I need, so I'm as much after the specs to shoot for as actual models which hit that.
What I want is an oscilloscope to support me in automotive and (more significantly) marine troubleshooting, including generators and power systems. I'll list below what I want out of the scope, in rough order of importance and achievability, what I think that means in terms of performance, and summarise by listing a few options I've been considering.
1. Form factor - Portable preferred. As for USB vs Tablet, I already take dedicated laptop with me for most work for plugging into ECUs (STAR/DICE/etc), so a USB scope isn't a big issue for me, and even a desktop would be workable if justified. Basically, capability first and anything goes, but more portable is more convenient.
2. Basic RF/signal/electrical system troubleshooting - voltage, waveform, and duty cycle of things like cam-angle sensors, O2 sensors, CANBus. Most sensors are low bandwidth, CANBus tops out at 1 MHz I believe, so none of this should be difficult for any half-decent 20 MHz scope. Most of this only needs 2 channels, but rarely I've had situations where it would have been nice to have a third (CAM/Crank/ignition) but it's not required to get the job done. The general sentiment from YouTube guys is 4-channel is very nice to have if the budget will stretch, but I don't (yet) see it as mandatory. The ability to read CANBus words would be neat, but it isn't something I can see myself needing very often (and I would think I could always just take a snapshot and do it by hand, anyway).
3. Long-term logging - There's been occasions where I've had to do long-term (overnight) logging of things like generator voltage stability which I can borrow a dedicated device for (e.g. Fluke power meter) but I'd like to be able to do with my own scope. The total number of samples isn't high, 50k at most, but the samples need to be pre-processed/math outputs (e.g. min/max peak voltage & frequency for whole sample period) not raw. If it has to be raw samples, it'd be more like 1k samples per second for 6 hours or more. This is also another scenario where higher channel count would be convenient, but you can also get by with 2.
4. Time domain reflectometer - being able to roughly estimate where a high resistance/short to ground is along a wire run would be particularly nice to have, ideally to within 0.2m or so. It's not really required for engine looms and the like, since you can just inject a waveform and follow the wire with a probe to find where the issue is, but for wiring that disappears into cavities knowing where to start would save a lot of time. I believe the propagation speed will probably be around 100m/us for everyday wiring, so both ways that would mean 0.2m would work out to 4 nanoseconds time-of-flight. To me this suggests that 1GS/s should give the accuracy I want, and a 100 MHz scope should be enough... but then there's no way untwisted pair can carry a 100 MHz signal so I'm not really sure if that's even relevant. I believe higher end scopes actually come with pulse generators for this sort of thing, but it looks like making a little pulse generator isn't very hard so it doesn't seem important.
5. Starter current compression and timing testing - I'd like to be able to do quick compression and ignition timing tests for engines that are having problems starting. The idea is by looking at the current draw waveform on the starter you can get a estimate for relative compression between cylinders, and combined with ignition probe can check if the timing is at least close. With a sparkplug or injector compression probe you can even check for obvious valve leaks, cam timing issues, and excessive exhaust back-pressure. I don't think any of this would be particularly hard for an oscilloscope (sample rate would be what, 100kHz at most?) but having a good low band pass seems like it would be important to filter out all the noise from commutators and such. This would also be a case where a few extra channels would come in handy so you can check the crank angle sensor at the same time. However, this does also seem like a situation where small portable screens would really get annoying, and having a USB scope that output straight to my nice laptop screen would be good. I also like the idea of being about to export the samples out of the scope program into a Python script or similar and get it to spit out the data I want.
I know the Autoscope IV does this, but I don't want to just trust a blackbox that spits out numbers with unknown assumptions and accuracy. It's also expensive.
6. I want it to be reliable, trustworthy and not frustrating to use. Reasonably snappy controls, no stupid little interface foibles, and as little faking of data as possible. I'm really not a fan of how some scopes (like Hantec) seem to apply post-processing to their measurements to make it look more professional, things like smoothing and even (in reviews I've seen) in-between sample estimation to make a 500 MS/s scope look like a 1GS/s as advertised. I have to be able to trust what it is saying and trust it to work when it's needed.
7. Lastly, and not lastly, budget. As with most people I don't really have a fixed budget, but I'd like to keep it under AUD$1k if possible. Like most shoppers, I'd rather spend less but I also subscribe to the "buy once, cry once" philosophy, so ultimately whatever needs to be spent will be.
Right, so with all that out of the way, here's a few I've been considering.
1. Picoscope ($500-$1700+)- I'm not sure which model would be the right pick here, but I'd probably pick the 8 bit 2 channel 100MHz 2208B for about $1k. Having four channels would be nice, but the 4 channel 2408B is $1700 which is really pushing the budget for something that is mostly convenience (I think). 12 bit is also just an unjustifiable luxury. If I ignored the TDR side of things, I'd probably go for the 50 MHz 4 channel 2406B for $950 (and maybe 50 vs 100MHz doesn't matter here? Maybe 20MHz would be enough?). I don't think I need to spend extra for their dedicated Automotive scopes and software, as they seem like a lot more money for not a lot more compared to a base scope with some cheap-ish leads and sensors. I could just be missing the use cases here, though.
Pros: 4 channel options, seems more reliable/accurate/trusted.
Cons: Expensive.
2. DreamScopeLab ($400+) DS4T1012 150 MHz 2 channel hand-held for about AUD$400 (I think? Website says $250 but I'm assuming that's USD since other sites list for USD$200), or U3P100 100MHz 2 channel USB for about $600. This one is an annoying one, since the newer hand-held is obviously more capable than the USB with the former in some sort of permanent quasi-sale discount. There's definitely times when a hand-held would be nice to have but I'm worried I'll miss the ability to export and post-process the raw samples, and I believe the DS can only export screen captures. I'm also not really a fan of touchscreens in general, even getting my phone to play nice in a hot and sweaty engine room is frustrating enough without having to pinch zoom and swipe the oscilloscope, so I'm not sure if the convenience for quick jobs makes up for the frustration for long ones.
Pros: Cheaper, seem well built compared with most Chinese baby scopes.
Cons: Max 2 channel, not as easy to trust, touch-screen probably annoying for DS
3. Autoscope IV ($1400+)- a lot more than I want to pay but comes with a lot of accessories and convenience factors. I can see the software being useful once trusted, but it feels like most of what you are paying for is dumbing down things for the mechanics who don't want to learn how to read oscilloscopes. I don't think the value is there for me at this stage. The high channel count (8 for the full job, 6 for the lite, plus various dedicated trigger/high voltage/pulse capture probes) is nice but the bandwidth is way lower than similarly priced Picoscopes (the AUD$600 Lite has a max sample rate of 250 KHz which is divided amongst the active channels).
Pro: Convenience, probe & sensor ecosystem, clever software
Cons: Not very capable as an actual oscilloscope. Limited capability in diagnosing CANBus High.
4. Siglent SDS1104X-E or similar desktop scope (~$600+) - most of the time I have a way to get 240V, and space I can stick a full desktop scope, but it would definitely be a lot less convenient than the equivalent USB scope, I think. I also worry about how dirt, grease, maybe even water might contaminate the insides via those big, open vents in the top. From what I can see in the spec-sheet and manual, it looks like it can be used as a pure USB scope, which would mean I could bag or box it in some way to protect it while using it like I would a Pico, but once you factor in the cost of a pelican case or risk for a soft bag - not to mention having to lug something that size around - I'm not sure it makes sense.
Pro: Better value for money, more use outside automotive/marine diagnostics
Con: Vulnerable to damage
5. The rest - OWON, Hantek, Multicomp Pro, FNIRSI, et al. All of these brands offer better claimed performance for less money, and sometimes better thought out things like hand-helds with actual buttons, but always with a caveat or always needing someone to actually check they aren't fibbing or make sure they are built right in a teardown. I really dislike the sort of dishonesty or relaxed approach to the truth that seems built into these cheap electronics manufacturers, and I'd rather just boycott them completely. If you think this is being too harsh or stereotyping too much by all means, I could be convinced otherwise... but.
Pro: Cheaper
Con: Less reliable/accurate (?)
Home
Help
Search
Login
Register
