I do not have an oscilloscope of any kind at the moment. I love the idea that if I used a USB-based scope, I could put the image up on the big TV and see it clearly while poking the circuit with the probes (the TV is hooked up to the PC already).

So... are these any good?
http://www.ebay.co.uk/itm/Hantek-60..._Measurement_Equipment_ET&hash=item1c3de69f18

...is there anything better around the same price point (ish) ?

I really only tend to do logic-based stuff, with PWM, on PIC processors. I've only ever used the low end PICs (PIC16F690 etc,) though I have just started looking at the bigger chips (PIC18F etc.). I don't anticipate needing high speed stuff over 20MHz. Being able to look at I2C comms etc. at 100kHz or 400kHz is probably the most common use, or looking at audio frequency signals (20Hz-20kHz).

 

Note that with a 20MHz bandwidth, the scope risetime is likely not much better than about 20ns, thus a 20MHz digital signal will tend to look much like a saw-tooth waveform. Generally you want at least 5 times the bandwidth to look at a digital signal so a 20MHz sinewave bandwidth is only good for digital signals up to about 4MHz. You have to decide if that's adequate for your needs.

 

I've got a Hantek 1008. It's an 8 channel USB scope rated @ 250kHz. I use it for low speed monitoring of analog and digital signals in machine control systems. even the paltry 250kHz is more than I need; half the time I have it set @ 5sec/div, using it as a trendplot.

Your needs and youe scope-of-interest are a bit different than mine, but I suspect that at least most of the pros and cons of the 1008 are equally applicable to the 6022, so I'll go ahead and speak as if all Hanteks are created equal, and let you do the leg work to confirm/deny.

Pros:
1. extremely inexpensive, therefore excellent "value" (bang for your buck)
2. lightweight, compact

Cons:
1. The bandwidth is shared among all active channels. if I am using all 8, my bandwidth per channel is 250kHz/8 = 31kHz/channel, and that's probably giving credit where it isn't due. I suspect your dual channel is probably 10MHz/ch if using both.
2. The scope chassis ground is tied to the USB ground, which is tied to your laptop's ground. If you do something stupid, you can A: "float" your laptop several hundred volts above ground potential (if not plugged into the wall) or B: if plugged into the wall, turn your laptop and power adapter intoa New Year's light show.
2. The software is half-ass. I won't waste the rest of my evening listing off all nails in its coffin, but I'll give you #1 - the waveform radically changes when the volts/div is changed. I mean, not like a bandwidth issue that rounds off the corners of a square wave, but like showing the expected half-sine waveform of the output of a 60Hz phase angle fired power controller at one volt/div setting, and turning it into a sawtooth at another volt/div setting, and into a square wave at yet another volt/div setting. It's not an input voltage limit either; for example my 10V half-sine looks good @ 1V/div, and wacky @ 5V/div. When I see something like that, it makes me wonder if the thing is ever really telling me the truth, or if its just showing me what it thinks I want to see; just (badly) taking a handful of samples and cross-referencing them to a handful of pretty pre-recorded waveforms loaded in memory from the factory. The lower volt/div settings usually show me more what I expect to see, so I end up using 10X or 100X probes so that I can see my 24V rotary encoder square waves in 1V/div or less. I bury my head in the sand and tell myself it was still a good buy @ <$100.

 

Ok, good point about the bandwidth. I think even if it were 1MHz it's probably going to be more than I need... although the PIC runs much faster, the comms lines etc. I want to be able to monitor are sub 1MHz, and I'll probably use 1 channel most of the time.

I don't do high voltage... which as far as I'm concerned is anything much over 12VDC... yes you can call me a chicken if you like, but iI regard myself as very much a novice to the electronics - my background is programming, hence the PIC programming. Hopefully that means if I just remember to tie the chassis to ground, I'm good.

The waveform thing is a concern though... do you think that's a limitation of bandwidth, or something else kicking in? I must admit if I saw the scope doing that I'd be wondering if it were faulty, after all, isn't displaying waveforms well within its bandwidth the whole point of a scope?

 

If you're going to be scoping comms lines, it might pay to have something that can decode them, like this: http://www.linkinstruments.com/mso19.html

 

Ok, good point about the bandwidth. I think even if it were 1MHz it's probably going to be more than I need... although the PIC runs much faster, the comms lines etc. I want to be able to monitor are sub 1MHz, and I'll probably use 1 channel most of the time.
... do you think that's a limitation of bandwidth

At 48MS/s, a 1MHz sine wave will have 48 points for each period. You should easily see you signals.
The problem with strantor's instrument might be just with that particular instrument and not with every one.

 

I recently got this scope from http://www.banggood.com/buy/6022be.html for fieldwork. It has it's limitations, but if it breaks not much is lost. If you use it with Win8 you need to download drivers from the Hantek site. E

 

If you're going to be scoping comms lines, it might pay to have something that can decode them, like this: http://www.linkinstruments.com/mso19.html

Take a look at the Beagle from Totalphase as well. Once you get the I2C hardware working, you'll appreciate having a protocol decoder and logging software. Its important for I2C to ensure that the analyzer be fast enough and otherwise non-intrusive. I've used some that catch up by holding the I2C clock - legal in I2C but leading to situations where it works with the analyzer connected but not when it isn't.