Best Oscilloscope under $300?

[Aldo22]

I’m not sure I should be suggesting the following since I have never used one and there have been a number of complaints on eevblog about it, but for its price it seems like something that could be considered as a beginner scope:  the Hantek DSO2D10.  Also the review posts aren’t always 100% negative - and it seems that there’s a lot of help available on eevblog too.

You can get one on Amazon for $200 (after an 8%off coupon is applied by checking a box).  It also comes with a signal generator which would be useful in school labs.

If you want to buy a Hantek DSO2X1X, then take the cheapest DSO2C10, they are all the same in terms of hardware. I got mine for $130 and it's worth that.
There are a few flaws, but you still get a hell of a lot for the money.
Lots of features. I'd actually recommend it for learning oscilloscopes.
If you just want to learn electronics and just want the scope to work without having to deal with it too much, it might not be the best.
But the price is an unbeatable argument. You can always buy a more expensive one later, once you've got an idea of what it's all about and what you really need.
You can't even get a decent "boat anchor" for this price here.

[tggzzz]

I wonder how we managed with digital signals before digitising scopes were available? Oh yes, we made the stimulus repeatable (always and still a good idea for many many reasons!) or used a analogue storage scope (horrible things !)

Where that isn't possible, use a scope to look at the analogue waveforms to ensure they will be correctly interpreted as digital signals (i.e. signal integrity). Then, for digital signals, flip to the digital domain with very inexpensive logic analysers or use a protocol analyser such as a bus pirate.

I am not saying that it is a problem to work with the old tech. But unless your interest lies specifically in learning how things was done in the past, I am arguing that the beginner is better of using a modern entry level DSO from a reputable vendor (and specifically the Rigol DHO800 or Siglent SDS800X HD). Those instruments are not perfect, but good enough to avoid most surprises and with a good range of features. It is something you will be able to keep for many years before possibly outgrowing the instrument.

Older A brand instruments may also be good options. The problem is that the beginner has no way to know if he is getting a good deal or not. There is a lot of overpriced and outdated stuff out there.

I am advising against going too cheap. The new stuff that is significantly cheaper than the before mentioned Rigol and Siglent is often "too cheap". If you are on a budget and need to make ends meet, I would say you need to go the used route. Yes that might even include an old CRO, because any instruments is better than no instrument.

We more or less agree, but...

I don't think a beginner should be concentrating on learning any tool. They should be concentrating on the operation of their electronic circuit. Tools change with monotonous regularity; physics lasts a lifetime

If budget is an issue, then old scopes are virtually free nowadays. Provided they are working, they will enable a beginner to learn a hell of a lot, and when their capabilities have been exhausted throwing it away will only "lose" $20 or so! At that point the beginner will know what they need and can ensure their next instrument (scope or otherwise) is suitable.

And for the price of a new Rigol etc, you can get an extremely capable older scope plus a logic analyser plus.... (Major exception: capturing single shot events),

BTW I'm glad nobody is suggesting digitizing scope that were low end professional scopes before, say, 2010. Those digitizing scopes are uniformly awful (capture depth, aliasing, etc).

[mwb1100]

If you want to buy a Hantek DSO2X1X, then take the cheapest DSO2C10, they are all the same in terms of hardware.

Cheapest I can find (admittedly not an extensive search) was $150.  The sig gen isn't "included" with the DSO2C10, but my understanding is that the sig gen hardware is sometimes (often?) installed and can be enabled with a bit of hacking.

You are right to point out that model and the fact that there are vendors cheaper than Amazon.

[Aldo22]

Cheapest I can find (admittedly not an extensive search) was $150. 

I bought mine in a christmas promotion.
I think $150 is still okay, but I wouldn't pay much more. There should be a reasonable difference to the price range of a Rigol or Siglent, which are of course higher quality (I assume).

The sig gen isn't "included" with the DSO2C10, but my understanding is that the sig gen hardware is sometimes (often?) installed and can be enabled with a bit of hacking.

Yes, I think nowadays you have to be very unlucky if you get one without the AWG hardware.
I wouldn't even call it "hacking". It's very simple and safe.

The scope has so many features, it's like a book on oscilloscopes. You open a new chapter every day. 
I find it suitable for beginners, not for professionals.

[Muxr]

Anyone who buys 2ch scopes today dont know what they are doing and should instead become a nurse or something.

I don't think this is as big of a handicap. Like what percentage of times do you really use more than 2 channels? Half the time I'm too lazy to setup 3+ probes on the scope, I just move the probe if I need to see another signal. And with digital scopes you can store the traces if you need to compare things. And with all the triggering options available you can get the similar result.

Like sure 4ch is nice, and everyone should probably get a 4ch scope as it doesn't cost too much to get the extra 2ch. But if budget doesn't permit, 2ch will be absolutely fine.

I dunno, maybe because I started on a 2ch scope, I learned to work around the limitation, but I certainly would be fine with a 2ch scope, if that meant I can also buy a decent multimiter or another piece of important equipment needed to start my hobby lab.

[Atlan]

Purely theoretically, a 2-channel oscilloscope on channel three, because it is a trigger input.

[tggzzz]

Anyone who buys 2ch scopes today dont know what they are doing and should instead become a nurse or something.

I don't think this is as big of a handicap. Like what percentage of times do you really use more than 2 channels? Half the time I'm too lazy to setup 3+ probes on the scope, I just move the probe if I need to see another signal. And with digital scopes you can store the traces if you need to compare things. And with all the triggering options available you can get the similar result.

Like sure 4ch is nice, and everyone should probably get a 4ch scope as it doesn't cost too much to get the extra 2ch. But if budget doesn't permit, 2ch will be absolutely fine.

I dunno, maybe because I started on a 2ch scope, I learned to work around the limitation, but I certainly would be fine with a 2ch scope, if that meant I can also buy a decent multimiter or another piece of important equipment needed to start my hobby lab.

Just so.

Learn to use your skill and imagination - something that's important in any engineering job. (An engineer is someone that can do something for $1 that any fool can do for $5)

Besides, decent probes aren't cheap. Probes frequently needed by amateurs cost several hundred $/£/etc. Some professional probes cost several month's salary

[Aldo22]

If budget is an issue, then old scopes are virtually free nowadays.
...
Provided they are working,

Maybe in England, but here you can't get anything in working order for less than ~100.-
If you buy it online, you have no idea what you're getting.
I really don't think a beginner should start with such an old piece of equipment. Does a beginner even recognize the faults?
For me, it's also out of question because they are too big and too heavy. You need space on the desk and a healthy back for these.
No thanks! 

[nctnico]

I don't think a beginner should be concentrating on learning any tool. They should be concentrating on the operation of their electronic circuit. Tools change with monotonous regularity;

And for the price of a new Rigol etc, you can get an extremely capable older scope plus a logic analyser plus.... (Major exception: capturing single shot events),

You are contradicting your own logic here  Tools change and what was useful yesterday, is less useful today. There ain't no such thing as an extremely capable older scope. Logic analyser is the same story; Tektronix (and maybe Keysight as well) has even stopped making these as nobody has a use for a logic analyser nowadays. Modern day oscilloscopes have become super versatile and for somebody at the beginning of their career, it makes sense to start with today's technology as much as financially possible.

The other day I read an article about 2 interns (studying fine woodworking) who where working on restoring old ships. Great fun but none of the crafts/wood joining methods used to build old ships is used today.

[vk6zgo]

I don't think a beginner should be concentrating on learning any tool. They should be concentrating on the operation of their electronic circuit. Tools change with monotonous regularity;

And for the price of a new Rigol etc, you can get an extremely capable older scope plus a logic analyser plus.... (Major exception: capturing single shot events),

You are contradicting your own logic here  Tools change and what was useful yesterday, is less useful today. There ain't no such thing as an extremely capable older scope. Logic analyser is the same story; Tektronix (and maybe Keysight as well) has even stopped making these as nobody has a use for a logic analyser nowadays. Modern day oscilloscopes have become super versatile and for somebody at the beginning of their career, it makes sense to start with today's technology as much as financially possible.

The other day I read an article about 2 interns (studying fine woodworking) who where working on restoring old ships. Great fun but none of the crafts/wood joining methods used to build old ships is used today.

There are still a lot of applications for a good analog 'scope, but they are not the things DSOs are good at.
Analog circuitry is like  "the ghost at the feast"  & still comes back to haunt us.

[tggzzz]

If budget is an issue, then old scopes are virtually free nowadays.
...
Provided they are working,

Maybe in England, but here you can't get anything in working order for less than ~100.-
If you buy it online, you have no idea what you're getting.
I really don't think a beginner should start with such an old piece of equipment. Does a beginner even recognize the faults?
For me, it's also out of question because they are too big and too heavy. You need space on the desk and a healthy back for these.
No thanks! 

Well, Switzerland has always been excessively expensive (except for cuckoo clocks  )

Big and heavy? Bigger and heavier, yes, but absolute values matter as much as relative values. While some justifiably deserve the term "boatanchor", I've used digitising scopes that were marked "two person lift".

You need a trusted source of any old equipment, e.g. a car or house

[Fungus]

Like sure 4ch is nice, and everyone should probably get a 4ch scope as it doesn't cost too much to get the extra 2ch. But if budget doesn't permit, 2ch will be absolutely fine.

I assume you all drive Smart cars then?

2 seats is enough to move any number of people, you just have to make more trips...

[tggzzz]

I don't think a beginner should be concentrating on learning any tool. They should be concentrating on the operation of their electronic circuit. Tools change with monotonous regularity;

And for the price of a new Rigol etc, you can get an extremely capable older scope plus a logic analyser plus.... (Major exception: capturing single shot events),

You are contradicting your own logic here  Tools change and what was useful yesterday, is less useful today. There ain't no such thing as an extremely capable older scope.

You need to come into the modern world, and not be stuck with decades old technology 100MHz was just about adequate 40 years ago with LSTTL logic. 100MHz became outdated in the late 80s. Modern jellybean logic has sub-nanosecond risetimes.

The 24x5 scopes are very good, have 50ohm inputs, and higher bandwidth . The 485 has even higher bandwidth, if you can get hold of one.

Logic analyser is the same story; Tektronix (and maybe Keysight as well) has even stopped making these as nobody has a use for a logic analyser nowadays.

That's a very weak argument. Tek/HPAK have never made equipment for beginners.

Tek/HPAK stopped making LAs for reasons that are irrelevant to beginners: FPGAs and multi Gb/s baud rates are standard and attaching LAs to those is "interesting".

Haven't you noticed the very cheap devices that are readily available? They have limitations, just like all tools, but they are usable for many purposes.

Modern day oscilloscopes have become super versatile and for somebody at the beginning of their career, it makes sense to start with today's technology as much as financially possible.

A beginner should learn the fundamentals of the technology that will last a lifetime.

The other day I read an article about 2 interns (studying fine woodworking) who where working on restoring old ships. Great fun but none of the crafts/wood joining methods used to build old ships is used today.

If those interns are aiming to be technicians assembling houses, you are probably correct.

But if they are aiming to become engineers then a sound knowledge of the strengths and weaknesses (literally!) of different types of timber is likely to be useful. Using the old techniques might be useful.

Directly relevant electronics example... Look at the "old primitive" topologies of radio receivers and transmitters. Then look at the topologies of microwave and terahertz radios. They are very very similar. The reason isn't hard to understand: in both cases the RF components are pushed to the limits of their capabilities.

[tatel]

My 2 cents:

a) OP seems to be looking for a decent oscilloscope to buy it and forget about buying another one for years to come. I think it's a sensible approach, with just one caveat: I would spend a hundred bucks more to get a 4-channel one, quite probably one of the new/coming Rigol/Siglent -unless OP is sure he will never do any digital work. In that case a 2-channel one could be fine. But I find that probability is quite small as of today, so I would really go for a 4.channel device.

b) If budget is really tight, and considering there are other tools that come really handy, I would get the really cheap way and accept the shortcomings unavoidable on cheap oscilloscopes. That way there will be some remaining money to buy some of these other tools.

On the cheap toyscopes available today, I can't see too much value on 2-channel oscilloscopes unable to substract one channel from the other, and real performance is usually worse than the published specs. So:

b1) I would get the really cheap ZeeWeii DSO154Pro, just one channel and useable bandwidth just around 10 MHz, but enough to learn quite a bunch of things and able to remain useful to do field work in the future, say looking at his car waveforms, etc. Probably less than 50 bucks.

b2) a second cheap option would be to get an old 2-channel, 20-50 MHz, about 100 bucks, CRO oscilloscope which will usually be better than the specs and able to substract one channel from the other. OP is in the US and things there are much better than almost anywhere else when looking for a vintage oscilloscope. In the future, it will remain useful to work with PSUs, do analog work, etc.

Of couse any of the b*) options will have shortcomings: it's just a "toy" or is a "boat anchor" that a "newbie can't be sure about it really working good". And, going with a "b" option, it probably means OP will be getting an entry-level decent benchtop oscilloscope somewhere in the future.

But, if budget can't get stretched to about 400 bucks just for the oscilloscope, OP will have to accept some shortcoming -again unless he's sure he will never want to do any SPI work with ease.

TL;DR: get a 400 bucks, 4 channel new Rigol/siglent 12-bit device

[tggzzz]

Modern day oscilloscopes have become super versatile and for somebody at the beginning of their career, it makes sense to start with today's technology as much as financially possible.

But cheap entry-level scopes often have crippling limitations that a beginner might not appreciate before purchase.

Most have very deep capture buffers, so they can capture far more than can be shown on a <800x600 display.
Some have digital signal protocol decodes, so you can see the characters/digits being transmitted serially.
But the decodes only work with what is displayed on the screen, i.e. a few characters at best.
That means searching for the unexpected/incorrect message is very very very tedious and error prone.

Better to use a separate tool dedicated to that, e.g. a cheap protocol analyser.

Same problem with their FFTs ignoring most of the captured data. Plus their inputs are 8-bits at best, and probably significantly less at "high" frequencies (make sure you understand the scope's ENOB spec and how it affects the spectum). Makes their FFTs little more than a toy.

[Aldo22]

But cheap entry-level scopes often have crippling limitations that a beginner might not appreciate before purchase.

Well, it's not surprising that cheap devices have their limitations.
But if I can get a device for $130 that can do some protocol decoding, (limited but usable) FFT, has a (limited) signal generator and that opens a door for me to get to know the possibilities of modern measuring devices, why should I tie a boat anchor to my leg?

Somehow I don't understand what you are trying to tell us. 
Aren't boat anchors even more limited?

[tggzzz]

But cheap entry-level scopes often have crippling limitations that a beginner might not appreciate before purchase.

Well, it's not surprising that cheap devices have their limitations.
But if I can get a device for $130 that can do some protocol decoding, (limited but usable) FFT, has a (limited) signal generator and that opens a door for me to get to know the possibilities of modern measuring devices, why should I tie a boat anchor to my leg?

Somehow I don't understand what you are trying to tell us. 
Aren't boat anchors even more limited?

Strawman argument: not all old scopes are boat anchors [1] Many are delightfully portable - even though I have a snapped biceps muscle.

Personally I dislike "multipurpose" tools that attempt to do everything and end up doing nothing well. Whenever I've gone down that route I've rapidly found the limitations of one bit and had to spend more money on something without the limitations. Then I find the limitations of another bit and - you can guess the rest[2]. Exception: where the UUT has been carefully designed to avoid stressing the swiss army knife; my projects are never like that

Instead...
I prefer to get a good example of each type of tool.
I dislike using a hammer to insert nails, and prefer to use a screwdriver.

That's why I encourage people to use a scope 250MHz (or better) to debug signal integrity problems, then switch to using digital tools for digital signals (e.g. logic/protocol/LAN/CAN/etc analyser, printf()).

[1] The nearest things I have to boat anchors are desktop/portable 21GHz spectrum analysers. Their new cost was similar to that of a small house; mine (in perfect working order) cost about the same as a lawyer charges for 2 hours work


[2] N.B. the money is irrelevant if something only costs 10 £/$/Eur; just move on and regard it as a learning experience

[Aldo22]

@tggzzz: I understand what you mean, but we're in the "under $300" thread here and we're not talking about the same thing.
I'm talking about a first somewhat "complete" scope that opens a door to a new world.
In the old days, you might have bought a book on oscilloscopes. Today you buy a Hantek (or similar) for the same money and try out what you can do with it.
There is even something to read through the help system (Attachment). 

The price is so low, it's about the same as the surcharge of a Rigol DHO914S on a Rigol DHO914.
If you get tired of it, you can continue to use it as a signal generator. 
I can't see anything wrong with it.

[Fungus]

If you get tired of it

The thing is you KNOW you'll get tired of it and a Hantek has low resale value

OTOH a Rigol DHO804 might be all the oscilloscope you'll ever need (it's all the 'scope I need, personally) and even if it isn't you'll easily be able to sell it for a good price when you trade up to a pro-level 'scope.

I can't see anything wrong with it.

Hanteks are known for being limited in features, buggy, and crashing a lot.

[Aldo22]

The thing is you KNOW you'll get tired of it and a Hantek has low resale value

So what?
First and foremost, it has a low acquisition cost.
I'm not thinking of selling it. I'd rather give it away when I no longer need it.

Hanteks are known for being limited in features, buggy, and crashing a lot.

I'm not interested in what it's "known for".
I have one and I'm having fun with it and learning a lot.
I'm also aware of its limitations.

Maybe one day I'll buy a Rigol or Siglent, but I certainly won't regret the "investment" in the Hantek.
It's simply not expensive enough to regret it 

[tggzzz]

@tggzzz: I understand what you mean, but we're in the "under $300" thread here and we're not talking about the same thing.
I'm talking about a first somewhat "complete" scope that opens a door to a new world.

I'll sell you a very "complete" working recapped Tek 2465 4-channel 300MHz scope for £300. I have two, and ought to reclaim the space.

After making sure it works properly (possibly including recapping), I'll sell my spare 350MHz Tek 485 with real 50ohm input (not 50ohm//15pF) for the same.

You're not having my working dual-channel 1.7GHz scope that I bought for £20 But that definitely isn't a general purpose scope for a beginner

In the old days, you might have bought a book on oscilloscopes. Today you buy a Hantek (or similar) for the same money and try out what you can do with it.
There is even something to read through the help system (Attachment). 

The price is so low, it's about the same as the surcharge of a Rigol DHO914S on a Rigol DHO914.
If you get tired of it, you can continue to use it as a signal generator. 
I can't see anything wrong with it.

You do realise that 100MHz was just adequate for digital logic in the early 80s, don't you? Modern jellybean logic is much faster than that, and important features on the waveforms will simply be invisible.

You won't get tired of either the 485 or 2465; they are both delightfully easy to use. Unlike some new scopes, those Teks "do what they say on the tin".

The full manual for the Tek scopes are all free.

The 1ns risetime "signal generator" in the Tek 485 is useful for verifying performance, much more so than the traditional one in the 2465s.

[Aldo22]

I'll sell you a very "complete" working recapped Tek 2465 4-channel 300MHz scope for £300. I have two, and ought to reclaim the space.

After making sure it works properly (possibly including recapping), I'll sell my spare 350MHz Tek 485 with real 50ohm input (not 50ohm//15pF) for the same.

Thanks, but with shipping and taxes it will cost me an estimated CHF 500.-
I don't want to spend that at the moment.
I'm just playing and learning. 

[nctnico]

You do realise that 100MHz was just adequate for digital logic in the early 80s, don't you? Modern jellybean logic is much faster than that, and important features on the waveforms will simply be invisible.

You keep repeating this but in the real world you rarely run into situations where you need more than 100MHz. And certainly relatively simple circuits beginners are working on don't require use of >100MHz scopes. Features like single shot capture, deep memory and protocol decoding are far more useful when doing things with microcontrollers. And if the need arises, a typical university will have a high speed oscilloscope available. Or, when on a tight budget, one can buy one of the many older >500MHz DSOs from Agilent or Tektronix.

[tggzzz]

I'm just playing and learning. 

Including, I hope, the first two lines of my .sig 

[tggzzz]

You do realise that 100MHz was just adequate for digital logic in the early 80s, don't you? Modern jellybean logic is much faster than that, and important features on the waveforms will simply be invisible.

You keep repeating this but in the real world you rarely run into situations where you need more than 100MHz.

You keep failing to understand the only reason for needing a scope for digital logic is to ensure signal integrity. Once that is ensured, flip to digital domain tools.

Analogue tools such as scopes make very suboptimal digital domain tools.

And certainly relatively simple circuits beginners are working on don't require use of >100MHz scopes. Features like single shot capture, deep memory and protocol decoding are far more useful when doing things with microcontrollers. And if the need arises, a typical university will have a high speed oscilloscope available.

If they are working on modern digital logic, 100MHz is insufficient. <yawn>No, the signal period doesn't matter; only the transition time is important</yawn>

For the reasons I noted earlier, the deep memory and protocol decodes on low-end scopes can be very unsatisfactory to the point they become unhelpful. Much better to get the appropriate dedicated digital-domain tool.