# Testing limitations of a low bandwidth cheapie oscilliscope

Learning basic electronics. Have a few multi-meters.
I'm researching "cheapie" less than $100 oscilloscopes. I read that the most important spec is bandwidth. I've noted that$40 to $100 scopes have severe bandwidth limitations ranging from 200K maybe up to ~1.0 MHz with reasonable (+/-5% ?) accuracy. I've noted one ~$80 scope that advertises 100 MHz bandwidth, but is reasonably accurate only below less than half that, maybe 30 to 40 MHz. I'm wondering if all the cheapies greatly exaggerate their specs like this. They call it "Chinese specs", I've learned.

What I would like to know is 1) what type of testing can be done with reasonable accuracy with such low bandwidth scopes, and 2) what types of testing should be totally out of the question/avoided.

I understand the guideline that a scope should have a bandwidth 5 times the frequency of the things being measured. My question deals with what types of circuits generate frequencies that are in this lower bandwidth range.

My scope functionality question pertains to testing these kinds of circuits: Beginning hobbyist testing such as in Platt's Make: Electronics or Basic Arduino type systems.

$100 is my absolute limit; preferring$40 to $80 range. This is just for a "toe in the water" "explore the basics of scopes" sort of thing. I know that if I spend "just a mere$200 more" I can get a "real scope." And no, don't want to mess with used or eBay.

Here are 4 scopes in my price range:

• I am sure you mean megaherz but you wrote milliherz everywhere. – Justme May 19 at 17:23
• Just don't buy these. Spending your savings on one would only delay the point at which you could get a real scope which probably starts in the low 200's. In the meantime maybe consider a cheapie logic analyzer and if you can find a demonstrated working analog scope on the used market for twenty bucks or something. – Chris Stratton May 19 at 17:25
• Right...meg·a·hertz (mĕg′ə-hûrts′) n. Abbr. MHz One million cycles per second. Used especially as a radio-frequency unit. Also called megacycle. – Gfmucci May 19 at 17:47
• I'm 74. No new electroncis career planned. And no used scope wanted. And I have very limited work/storage space. I'll be pleased as punch with the very basics. – Gfmucci May 19 at 17:49
• Forget all single channel devices. 2 channels is a must because you must be able to compare input and output to decide what a device under test does to the input signal. Two separate single channel devices isn't a substitute because there's no common timing. If you limit all your jobs to audio band signals you can work with 1MHz oscilloscope, probably even with 200 kHz. Logic signals need much more. – user287001 May 19 at 18:20

After a quick look at those links I would say that I have absolutely no idea what the performance will be like. The specifications given aren't remotely trustworthy. Maybe they will be decent when used well within their specs, probably they won't, and the only way to find out would be to buy one and test it.

If I was looking for a really cheap scope, I'd get a Picoscope 2204A. They are £99 ex tax in the UK, including basic probes. Pico are a respectable company (their top end scopes are £15000), and they have a reputation to protect, so the scope will work properly and will meet the specifications given. I wouldn't normally recommend a specific product on here, but as far as I know Pico are the only respectable company making anything like this.

Picoscopes are USB scopes, which are not exactly like full size scopes. Some people like them and some hate them. The interface is very different, but you can do all the same measurements. One disadvantage is that if you do something very silly, like accidentally connecting the scope ground to a high power source, you might wreck the connected computer as well as the scope.

The cheapest picoscope has a bandwidth of 10MHz (and samples at 100MS/s). With that you'll be able to probe pretty much anything you encounter on an Arduino or similar apart from the clock, which might be 16MHz.

• I don't have a computer to sacrifice. I'm bound to do something very silly sooner or later. I would need a stand-alone cheapie. I determined Pico has no plans for it to work on an Android system, so that leaves my tablets out of the potential options. – Gfmucci May 19 at 18:20

In general, you will do RELATIVE measurements.

At high frequencies, the scope probe capacitance (10pf to 20pf, for 10:1) will cause big errors.

In an amplifier with 10,000 ohm Rcollector with 10pf total capacitance when operating, will have 10^4 ohm * 10^-11 farads == 10^-7 seconds time constant, or 1.6MHz 3dB frequency.

Using a 20pf scope probe gives 1.6/3 = 530,000 Hertz bandwidth.

What types of measurements excite you?