Choosing USB Scope or standalone

[LapTop006]

Most of the cases where you want 4 channels with Arduino are "digital" in nature (SPI etc.) and he did say he has one of the Ebay logic analyzers which should cover that off reasonably well.

Here's a case that came up recently where I used all four analog channels of my scope.

Ethernet PHY chip isn't working, I know it might be power related (I'm overdrawing from the nominal spec of a rail, but pretty sure it's ok), so I put all four rails for this chip on the scope while running through tests. If they weren't rock solid I would probably have wanted to add some additional signals to better time-correlate the drops, but in this case it wasn't power.

Now sure a basic Arduino circuit (let alone single chip) probably doesn't have even two rails, but you do find uses for those channels very quickly, and sure you can often get away without them, but darned if they don't make things quicker & easier.

[Fungus]

Now sure a basic Arduino circuit (let alone single chip) probably doesn't have even two rails, but you do find uses for those channels very quickly, and sure you can often get away without them, but darned if they don't make things quicker & easier.

Yep. If you have them, you will use them.

If you've never had a four channel 'scope then two channels might not look like a limitation. If you've had four channels, though, it does. Two channels would be a big step backwards for me.

One other thing to consider if you really have no money is one of those DSO Quad things. The user interface is awful but they're small, they really work, and they have four channels.

I used one for about a year before I got my DS1054Z and I still take it to Arduino club (where small size+battery power is a big deal). The difference between a DSO Quad and no oscilloscope at all is huge.

[DDunfield]

Most of the cases where you want 4 channels with Arduino are "digital" in nature (SPI etc.) and he did say he has one of the Ebay logic analyzers which should cover that off reasonably well.

Here's a case that came up recently where I used all four analog channels of my scope.

Ethernet PHY chip isn't working, I know it might be power related (I'm overdrawing from the nominal spec of a rail, but pretty sure it's ok), so I put all four rails for this chip on the scope while running through tests. If they weren't rock solid I would probably have wanted to add some additional signals to better time-correlate the drops, but in this case it wasn't power.

Now sure a basic Arduino circuit (let alone single chip) probably doesn't have even two rails, but you do find uses for those channels very quickly, and sure you can often get away without them, but darned if they don't make things quicker & easier.

I'm not saying "don't get a 4 channel scope" - I have a 1054Z and use it frequently. There are times when it's really handy. But I still often use my other 2-channel scopes, and my 1-channel scopemeter --- I don't recall ever being out and about with one of my other scopes, and not having the 1054Z threw up my hands saying "Can't do without it" - that's not to say I haven't "wished it were here".

In all fairness, in the testing described above could have been done with a 2-channel scope by repeating it, and measuring two pairs. It could even have been done with a 1 channel scope by repeating it 4 times, once to look at each rail. But I completely agree that being able to see all four rails at once is "just easier", especially if you need to do this kind of test frequently. Although ... free-running and "just looking" may not be the best way to find small rail glitches... I might end up looking at them individually anyway.

The reason for my previous statements is that I was trying to offer information applicable to the stated budget. I'm just not aware of any decent 4-channel scopes which would fit (anyone?).

Now sure a basic Arduino circuit (let alone single chip) probably doesn't have even two rails, but you do find uses for those channels very quickly, and sure you can often get away without them, but darned if they don't make things quicker & easier.

Yep. If you have them, you will use them.

If you've never had a four channel 'scope then two channels might not look like a limitation. If you've had four channels, though, it does. Two channels would be a big step backwards for me.

One other thing to consider if you really have no money is one of those DSO Quad things. The user interface is awful but they're small, they really work, and they have four channels.

I used one for about a year before I got my DS1054Z and I still take it to Arduino club (where small size+battery power is a big deal). The difference between a DSO Quad and no oscilloscope at all is huge.

I've considered the DSO-Quad for the portability, but so far avoided it as much of the feedback I've seen from people used to traditional scopes is pretty bad - which leads to a question since you are familiar with both .. Would you prefer to have the DSO Quad or a decent 2-channel bench DSO if it were the only scope on your bench on a daily basis?

Dave

[DDunfield]

Just want to add:

If it's mostly for "Arduino" then get four channels. Two channels isn't enough.

While I agree that 4 channels is really nice, not sure I agree 100% in this case, as I don't think he can stay within his stated budget and get a decent 4-channel scope.

Most of the cases where you want 4 channels with Arduino are "digital" in nature (SPI etc.) and he did say he has one of the Ebay logic analyzers which should cover that off reasonably well.

I disagree. I don't use 4 channels for SPI decoding, I use 4 channels for debugging, eg. toggle a pin in an interrupt handler so you can see it's happening after a signal is received, etc.

Those logic analyzers are good for decoding data but they don't update in real time for watching signals.

Yep.. 4 channel is nice for this You can see 3 things plus your toggle. With a 2-channel scope, I tie the "toggle" pin to external interrupt, then I can look at 2 other things. If I need to see more, I use the LA and have to start a new capture every time I want to see what is going on.

They wouldn't actually have to add much to the LA to accommodate an almost real-time update (like DSO above a certain speed):

1) Let you set sample rate and buffer depth to accommodate short/quick captures.

2) Provide an option to automatically re-arm the trigger following a capture.

3) Leave the last display on-screen (and not cover it up with a "Waiting" window) while you are waiting for the next capture to complete.

Then... every time you toggle a pin, you would get a picture of up to 7 (or 15, 23.. depending on the LA size) other signals activity around it (as defined by your sample rate, buffer size and pre-trigger size).

Funny how software guys want different features in their tools!

Dave

[Fungus]

I've considered the DSO-Quad for the portability, but so far avoided it as much of the feedback I've seen from people used to traditional scopes is pretty bad - which leads to a question since you are familiar with both .. Would you prefer to have the DSO Quad or a decent 2-channel bench DSO if it were the only scope on your bench on a daily basis?

If most of my work was "Arduino" (ie. 5V signals, limited to 16MHz clock) then I'd probably choose the DSO Quad. Being able to see four signals at once would easily compensate the lack of twisty knobs. The user interface would mostly be just changing the time base and looking at frequencies, not too demanding on the Quad's buttons.

(plus I'd have one of those $6 logic analyzers for protocol decoding/looking at longer sequences of pulses - the DSO Quad doesn't do that).

For other types of work, eg. If you're constantly switching between horizontal/vertical controls, the DSO Quad is a pain in the ass. Plus the bandwidth is really limited, maybe 20MHz at best.

[Gyro]

One thing that people tend to have forgotten about these days is the external trigger input. This is usually available on 2-channel scopes, but is absent from lower end 4-channel models, eg. the Rigol ds1054z as far as I can see.

In analogue scope days, the Tek475 (and probably many others) for instance, have a trigger view button to help with setting trigger level. Granted, it's not so convenient in modern 2-channel DSOs, as you need to scope your 'strobe' signal first to confirm its integrity but it does allow you to use the external trigger as a pseudo 3rd input channel in a lot of logic situations.

[Fungus]

One thing that people tend to have forgotten about these days is the external trigger input. This is usually available on 2-channel scopes, but is absent from lower end 4-channel models, eg. the Rigol ds1054z as far as I can see.

The idea is to use one of the four channels for triggering. 

In analogue scope days, the Tek475 (and probably many others) for instance, have a trigger view button to help with setting trigger level. Granted, it's not so convenient in modern 2-channel DSOs, as you need to scope your 'strobe' signal first to confirm its integrity but it does allow you to use the external trigger as a pseudo 3rd input channel in a lot of logic situations.

The low end Keysight can display trigger-in on screen and work with it like an analog trace, it's really a 2.5 channel oscilloscope. I don't know of any other low-end DSO that can display the trigger signal on screen though.

[Gyro]

The idea is to use one of the four channels for triggering. 

I know that. I was merely pointing out a somewhat neglected method of achieving the most from a 2-channel scope. I didn't say that it was a replacement for a 4-channel scope.

The low end Keysight can display trigger-in on screen and work with it like an analog trace, it's really a 2.5 channel oscilloscope. I don't know of any other low-end DSO that can display the trigger signal on screen though.

... and I did point out that it was less convenient because you needed to scope the 'strobe' signal to be used for triggering, to ensure it's integrity first.

[Gandalf_Sr]

Save up and buy a Rigol DS1054Z, I bought an open-box one from Tequipmet.net for <$300 after eevBlog discount.  It's possible to 'upgrade' this to a DS1104Z with all options very easily. When I bought mine, all options were included with the exception of the bandwidth upgrade.

This is an awesome 4-channel scope that will do 99% of what most people need, I use mine every day.

I just checked and you can get the same deal I did right now, possibly better.

[Gadjet]

Save up and buy a Rigol DS1054Z, I bought an open-box one from Tequipmet.net for <$300 after eevBlog discount.  It's possible to 'upgrade' this to a DS1104Z with all options very easily. When I bought mine, all options were included with the exception of the bandwidth upgrade.

This is an awesome 4-channel scope that will do 99% of what most people need, I use mine every day.

I just checked and you can get the same deal I did right now, possibly better.

The thing is, it's not that I don't have the funds it's just that I can't justify the spend on something that won't get a lot of use and here in the UK things tend to cost a lot more than in the states anyway.

[Gandalf_Sr]

I know life is harder in the UK (I lived in the UK until I was 40) than the USA and you don't have as much disposable income but, for around $250 for a nearly new DS1054Z you can't go wrong, even if you have to pay 20% VAT.

USB-based scopes are a waste of money IMHO.

There's apparently an old Czech expression.. "we're too poor to buy cheap stuff".

[Gadjet]

Thanks for all the advice, I went for the Hantek Dso5072p, I bought two, one for a friend at work, and got a discount so paid £188 ea.

[Rohde_TestWalker]

One thing you should consider when deciding whether to purchase a USB scope or a standalone or benchtop scope is ease of use. We think benchtop scopes are easier to use, and that ease of use translates into higher productivity. That may not be a big deal for a hobbyist, but for a professional engineer, having a scope that's easier to use, means that you'll be more productive, and that, in turn, helps shorten development cycles and lowers development costs.

Of course, that doesn't mean that engineers should shun USB scopes. They're certainly the right choice for some applications, especially those that don't require a lot of user interface, such as automatic test equipment applications.

[Andreax1985]

I do not feel comfortable hooking up my laptop to a usb scope. I work with audio amplifiers and do not want to risk frying my notebook in case I do something wrong.

[Gyro]

It depends whether you use a USB isolated one or not but yes, it is bad to accidentally pass ground current through any sort of test equipment.

[bd139]

Yeah. A ground can shift some amps. Nice picture of a cock up I did recently by dropping a probe on a power supply after forgetting to put a locating sleve on it. If this was a poorly isolated USB scope I would have lost the PC it was plugged into.



Made one hell of a bang that did!