Entry-level oscilloscopes are a great way to get some low-cost instrumentation on a test bench, whether it’s for a garage lab or a schoolroom. But the cheapest ones are often cheap for a reason, and even though they work well for the price they won’t stand up to more advanced equipment. But missing features don’t have to stay missing forever, as it’s possible to augment them to get some of these features. [Tommy’s] project shows you one way to make a silk purse from a sow’s ear, at least as it relates to oscilloscopes.
Most of the problem with these lower-cost tools is their low precision due to fewer bits of analog-digital conversion. They also tend to be quite noisy, further lowering the quality of the oscilloscope. [Tommy] is focusing his efforts on the DSO138-mini, an oscilloscope with a bandwidth of 100 kHz and an effective resolution of 10 bits. The first step is to add an anti-aliasing filter to the input, which is essentially a low-pass filter that removes high frequency components of the signal, which could cause a problem due to the lower resolution of the device. After that, digital post-processing is done on the output, which removes noise caused by the system’s power supply, among other things, and essentially acts as a second low-pass filter.
In part 2 of the project, [Tommy] demonstrates the effectiveness of these two methods with experimental data, showing that a good percentage of the noise on a test signal has been removed from the output. All the more impressive here is that the only additional cost besides the inexpensive oscilloscope itself is for a ceramic capacitor that costs around a dollar. We were also impressed: [Tommy] is a junior in high school!
Presumably, you could apply these techniques to other inexpensive equipment, like this even cheaper oscilloscope based on the ESP32.
You said: “Presumably, you could apply these techniques to other inexpensive equipment, like this even cheaper oscilloscope based on the ESP32.”
But, in that case it is an STM32
They make an oscilloscope without a low pass filter?
In the words of Luke Skywalker, “what a piece of junk!”.
Just adding a single C_filter to make the second amplifier a low pass filter is really great!
Brand new Rigol DS1054Z is $350. Minimum wage in Germany is about $14/h. That’s 25 hours or about 3,5 days of working at McDonalds. Do you really want to waste your life screwing around with subpar “tools” instead of getting the real thing? (even if it’s just Rigol)
Reminds me of people who each month slave away for hours, every day, for two or three weeks grinding virtual gold in an MMO just to earn enough rl-life cash for another month of subscription; when they could earn the same amount by working for just a day.
Right…
You must have rich parents, I recognize that version of logic.
‘every highschool student should work 8 hours every day in McDonalds and buy tools for school only with the earnings.’
Not entirely agreeing with the poster above, or even going to talk about the monetary aspect but something else entirely. I think we really should not be encourage people to settle for inferior tools. Or at least “toy” tier ones.
Human lives are much more valuable, it makes no sense to waste hours and hours being frustrated because your tool wasn’t intended for the task. Its a source of misery, we don’t want that.
Yes, its a great learning exercise to manage to work with what you have, but I say its not comparable to getting a proper tool to finish the job. That’s a real achievement. I will give [Tommy] a pass though, students should waste time, lest they become adults who micromanage time
It’s a mistake to say “My free time is worth $20/h” and use that to justify why you should just buy the more expensive thing, because it’s free time. You’re not paid to sit around.
You can choose to spend your free time to create value for yourself, or to spend value you have to replace with your not-free time. Whether you consider that misery and suffering is up to your attitude – you can’t make that judgement for someone else.
Humans do two things special.
Language and tool using/making.
Buying crap tools is a step in becoming a good tool maker.
Learning to recognize ‘time sinks’.
Part of that is valuing your own time, perhaps not at your billing rate, but something.
Even if getting paid, the wise will not waste time attempting to polish a turd.
Give the MBA what she want’s…
Cover turd in gold spray paint and glitter, find new job.
All costs are opportunity costs.
What else could you have done with time/money?
Play video games and fap?
Proceed with hacking on garbage for learning.
‘Sitting around time’ is priceless, what price sanity?
A 100 kHz scope with a decent analog front-end is still good value for the money if you’re dealing with “DC to 20 kHz” as the author says. The first scope I started with was worse than that.
OF course it’s nice to have good tools, who does not like good tools? And which HaD reader doesn’t feel joy at having a good quality tool. But normal people only have so much money and you have to live with not being able to buy the best of everything.
And frankly how many students would use an oscilloscope enough to warrant to spend that money on it? I expect only a very small number.
Perhaps they should spend that money on some quality headphones for instance, to enjoy good sound which they might spend considerable more time doing. And that’s just one thing of the millions of things where you can opt for quality.
Now from our HaD vantage it would be pretty nice if every high-schooler bought a high quality oscilloscope, and then us having our pick in the flooded second-hand market as they dump them after high-school! :D
Incidentally, I agree the oscilloscope in the article is pretty damn dated, 100KHz max? Come on now.
But it’s still interesting to see the hack to improve it though.
Or just a kid who’s driving daddy’s car and living at home eating and living for free, thinking that the money you’re paid is all yours to spend on goodies.
At minimum wage, assuming you’re in no debt, you can easily spend 70-80% of your income in just the necessary expenses like food, rent, utility and phone bills, internet, tax, commuting to work… so it would actually take nearly the entire month to save up enough cash to buy the oscilloscope – assuming you didn’t want to buy anything else. No Netflix subscription, no beer and snacks, no nothing – just saving money to buy a ruddy oscilloscope.
People often have more time than money, so spending $50 on a crap oscilloscope and the time to make it better affords you to buy six other nice things with the money you save.
‘every highschool student should work 8 hours every day in McDonalds and buy tools for school only with the earnings.’
Where exactly did he say the guy should work 8 hours every day in McDonalds? Nowhere.
Have you heard of weekend jobs that school kids have?
And your second assumption that he is obligated to buy the scope for school is also just an assumption. It is hardly likely that a junior high school student in Germany should have to buy one himself. More likely the school’s physics lab will have them.
Parents can help. In Europe parents tend to support hobbies of their children and are willing to spend money on them.
Besides, used 2-channel analog oscilloscopes are relatively cheap, and one can buy a decent model for 30-50Eur. Or one can buy something like Zoyt ZT-702S, or one of those cheap Hantek or Owon scopemeters.
The only reason for a DIY oscilloscope is education. Especially analog front-end design, and DSP programming. 100kHz of bandwidth makes such a DIY scope limited to work on audio circuits only. And one rarely needs 12 or even 10 bits. Besides, you can’t increase bandwidth but you can increase resolution with oversampling, which probably all modern oscilloscopes offer. Mine with 8-bit ADC offers up to 14-bits in this mode.
Personally I think Siglent SDS1104X-U is the best budget oscilloscope money can buy, as it’s fully-featured without hacking. Alternatively one can buy Hantek 2D10 that has two channels, but includes signal generator, which is a good thing to have. Then there are all those cheap USB oscilloscopes. PC software is not the greatest, but these devices are relatively cheap. Many of them are sold as used because they are picked as “my first oscilloscope” and after few months get replaced with something more advanced.
When I started writing articles on electronics for Elektroda.pl, I kept that meager income on separate account and bought my DSO for that money. Then when I started working for two electronics magazines, I spent that income on a good bench multimeter. Tomorrow I’ll be ordering new, better soldering station, as current model is slowly breaking apart after ten years of hard use. But I’m designing my own ultra-precise power supply as there is nothing on the market with specifications or performance I want.
If your budget is limited, you can’t afford to use the cheapest tools.
Subtract taxes and living and recalculate. You’re not left with all the money at the end of the day.
At minimum wage, you have very little left of your pay at the end of the day. That means you actually have to work for months to save up enough cash to spend it on a Rigol, that is if you don’t have other sudden expenses like a broken car or having to go to the dentists…
And then there’s the matter of debt, which is probably the case since people on minimum wages generally can’t afford to just buy things like cars and homes.
Technically you have no savings even if you have cash in hand. If you’re not spending the extra money you have to pay the debt as soon as possible, you’re going to lose more money later. $350 now at 3% interest rate over 10 years is $470. Inflation is going to eat some of that in real terms, but that depends on whether your wages keep up with inflation, and whether the interest rate goes up with some index which is usually the case.
This is why spending $50 on a crap oscilloscope and $300 to pay back your loans is the better deal. You’ll have a scope and save more than its worth in interest payments later. Beggars can’t be choosers.
Well, kind of depends also on wether you make any money with the scope. If not, then it’s a hobby and you spend on it what you can. If you do make money, then it might be better to save up for a better than the minimum, because if you buy something too crappy, you might not make that money after all. It’s a game of balance, obviously, if you need to get that scope now to make some money, then maybe you can’t wait.
I just bought a scope (Rigol). I won’t make any money with it, so i couldn’t justify a 450€ scope, which seemed nice. So i bought a 270€ scope (discounted). Works alright for my needs. It was kind of a let down, with artificial limitations and support seems to have been dropped years ago. Did not know that. Could be the same thing with the more expenssive scope too, don’t know
Well it’s kinda assumed that you don’t buy a $50 scope for any real work, although I do have one for quick diagnostics stuff and it’s really handy. It means you don’t have to lug a Rigol to the worksite just to check some basic stuff, and you don’t mind if it gets broken or lost.
If you’re your own business, then an oscilloscope as a tool is a tax write-off anyways.
And the things you learn from the hours spent on improving tools could help you eventually earn more than minimum wage.
At least it helps you to identify what it means to have a good oscilloscope, so you don’t fall on marketing and pay for nothing later.
Well, you can use this to compare countries based on the worth of work and relative cost of specific items.
For example I would have to work over 67 hours on minimum wage in my country to buy this oscilloscope. Almost three times longer. I live in Poland. And this indicates relation between worth of work and cost of living. The reason many people from Poland work and/or live in Germany is simple – it’s much easier to save money on nice things in country where your work is worth 3 times more in relation to the cost of those things…
I don’t know what the rate is in Germany, but in the USA, about 1% of workers earn minimum wage.
If you aren’t the mentioned spoiled high school kid and you are still making minimum wage you are doing something very very wrong.
Consistently show up on time and sober and you will earn more.
In CA they raised the minimum wage for fast food jobs to $20, vs $15 for other crap jobs.
The fast food industry turned over it’s entire workforce.
Now retail is the job for the mouth breathers, you have to hustle to keep a job a McDs.
Don’t fully agree but I will say if you are serious you will tend to put yourself in situations where serious happens
I got a very nice techtronicks dmm and a 20mhz Kenwood dso (still crt based) totally free by not working a shit no money job as a fry cook … but working a shit no money job as an entry level tech
If you don’t want to spend that much, check for hamfests in your area. I’ve gotten lots of analog scopes from them. A couple of them were even free. If you stick around until the end, the sellers will often give you a really good deal on stuff they don’t want to haul back home. Sometimes they leave free stuff behind when they pack up.
Even the old synchronous sweep tube scopes are more useful than one those cheap toy scopes. At least they display a clean trace up to a couple of megahertz. You just can’t make any time or frequency measurements with them.
Getting a second job requires applying and interviewing and then filling out financial forms, and that will add a number of hours on top of the 25 hours you have to work.
If you quit after 25 hours, it’s not nice to your employer who put in a number of hours to hire you.
Improving a piece of electronics is a lot more fun than working at McDonald’s.
A fair amount of the time in a project like this is experimenting. Once you figure it out, you can write it up and other people can do it faster than you did, so there may be some social utility.
Those cheap and cast off and homemade tools got me started
I started with vacuum tube everything and recapped and calibrated them
Same for my 140’s to 80’s craftsman hand power and shop tools from Craigslist yard sales flea markets and scrap piles
Same for my tillers riding and mowers pull behind trailers generators and pressure washers
Same for my 1960 stake body farm truck
And my mid engined econoline van both free and both rebuilt from scrap Ike’s and elbow grease
Same for my 8n tractor and implements
We start with cast offs and cheap and work our way up reinvesting as we go and fixing what we have
We adapt we overcome and we save our money and spend it on what we cannot otherwise get and only what we need
So I tip my hat to the young man and his scope
I think some of you are missing the point and have not looked at the articles in EDN. A high school student researched and modified an electronic instrument.
Mind that anti-aliasing or lack of in a cheap oscilloscope is not necessarily a bad thing. An operator who knows the limits of the scope can judge whether the signal they’re seeing represents something real. You never blindly trust the oscilloscope, because there’s a lot of subtle things that can go wrong.
For example, if you put a high-pass filter in front of the probe and you’re still seeing low frequency signals coming out, you know that it’s aliasing and you know that there’s something going on beyond the Nyquist limit of your ADC, which means you need a better scope to look at it. It also means you just used you cheap oscilloscope to see beyond what it’s supposed to see.
If you had a slow scope that was well filtered to exclude the high frequency stuff, you would never know that it’s there. It would fool you into thinking that there’s nothing going on; sometimes worse is better.
“The first step is to add an anti-aliasing filter to the input, which is essentially a low-pass filter that removes high frequency components of the signal, which could cause a problem due to the lower resolution of the device.”
Doesn’t the low-pass remove problems due to the limited sampling rate? If you have 1GHz, 8-Bits you can still measure a 100MHz signal. While 20 Bits at 100MHz won’t be enough. A powerful 100MHz signal will destroy everything you can measure on a 100MHz scope. A Low-Pass removes that signal and you can measure something.
Unless you meant “frequency resolution”, but in the sentence before you write “resolution of 10-Bits”… Remember, most cheap scopes were 8-Bits only (like the Rigol 1054z) since Rigol introduced their new 12-Bit series.
yeah this was my real thought too…i’d personally rather have noise and imprecision represented as noise and imprecision. the idea of smoothing it out in software bums me out. of course, if i am not using it as test equipment but rather like to record data from a sensor, then i might want to do all sorts of post-processing to it!
but it makes me wonder how heavily-processed the signal on my siglent is.
There is oversampling, which nearly all scopes do, which utilizes the remaining noise to average out multiple samples to achieve greater precision than what the ADC can otherwise do. This is how your low-end scope achieves its advertised performance values using a fairly cheap and low resolution but fast ADC running at something like 1 GHz sampling rate but actually sporting a 62 MHz bandwidth because it’s averaging 8 samples into one.
Everyone here who had to walk uphill both to and from school, raise your hand!
We had to ski, both in winter and in summer, and it was always uphill and against the wind both ways.
Fun thought experiment, figure out how to walk uphill both ways to and from school. I figured it out one time, but, alas, I don’t remember anymore. I think it had to do with a valley.
I walked up hill both ways, to and from school.
To get to to school, I walked up and over the hill, then down to the school. To get home, I walked over the same hill.
Is it really an advancement to spend your time for “idiot work” in order to buy tools that some large factory churned out, and you don’t really understand the inner workings of, instead of spending your time verifying your assumptions, and really learn something?
I chose exactly the way [Tommy] seems to have chosen, and I later became an acknowledged expert in the field. Today I’m buying the most fancy tools because I spend my time differently, but that doesn’t mean I could not make them if I had to.
2nded.
Also: It’s a high school student. Those normally don’t know about Nyquist… (I didn’t, I was messing around with MCU’s in BASIC, as is was a pre-arduino world back then…)
The Esp32 scope link goes to an STM32 based one.
About this line:
“The first step is to add an anti-aliasing filter to the input, which is essentially a low-pass filter that removes high frequency components of the signal, which could cause a problem due to the lower resolution of the device.”
It’s got jack to do with the resolution. The reason why it causes a problem is because of what analog to digital conversion does to the frequency spectrum of the signal. When you discretize your signal, its frequency spectrum turns periodic. The original spectrum appears to be “copied and pasted” in multiples of the sampling frequency (f_s). If your original spectrum had any components above f_s/2, then they will overlap with the subsequent “cycle” of the frequency spectrum, and you get very weird stuff like this.
The way to fix this is by using an antialiasing filter, removing everything above fs/2.
There is also phase errors or beat errors that happens below the Nyquist frequency, which shows up as increasing low frequency artifacts the closer you get to the limit. These can be calculated out if you know exactly what the input signal is supposed to be, but since you don’t, you have to put your low-pass filter lower than f_s/2. Preferably much lower, like 1/5th of your sampling rate.
What is going in here? From electronic discussion to Economic discussion
These cheap devices will never replace a serious scope. Period, even if just on triggering options, but one has to be particularly short sighted to say that cheap alternatives are worthless, especially in face of the overwhelming evidence of their popularity which, to all but the most particularly dense, shows that there are people who do find them useful.
There is a depressing amount of stupidity and elitism in these comments, many likely from very entitled fools who have never had to really work for an actual living. Apart from real economics, which many do not seem to comprehend, they miss several other important points: first, many people use these cheap devices in diy audio synthesis where the advantages of using several devices in the audio chain outweigh the limitations. Since the important components of these signals are below 25khz, these devices, esp with simple fixes like this, and even cheap low impedance probes are more than adequate and, lets face it, provide the most critical bling and squiggly lines. Second point is just the challenge of improving the value if the device which is one of the valued forces driving education in engineering.
At the
end of the day, what really matters is the opinion of those who find value, not the words of blowhards who are too entitled, inexperienced or shortsighted to see value in anything with limitations.
I have the same scope. I found it almost useless with the included power supply. Replacing that with a 9V battery improved things a lot. Now I see I have more steps to take :)
So much pretense and presumptuous attitude here. Learning is learning. Everyone isn’t infected with affluenza, as some of these statements indicate. Never seen do much “assumption” from what would should be assistance and education. Color me gone.
This place is becoming more like reddit every day….
The need for the political grandstanding by some is unfortunate.
haha unlike the great contentious thread above, i simply love that there are so many takes on ‘low-cost’. when i saw the headline, i imagined my $350 scope.
last time i looked at these very cheap scopes, they were closer to $100 than $25, and many of them had USB instead of a display. i ruled them out because <1MHz seemed irrelevant to my needs — the sort of troubles i have just wouldn’t even look like a blip on them. and most of them had very inflexible voltage range limits. truly, that voltage range thing is a problem i find a lot on low-end test equipment…who wants to buy one low-bandwidth logic analyzer and signal generator that only supports 5V I/O, and a separate one for 3.3V? but once i rule that out, products with level-shifting (such as bus pirate) cost a little more. of course they do.
at $25, it becomes the sort of product that is available to someone like where i was 30 years ago in highschool. i still don’t want it but i can see why people are futzing with them. even now, there’s always the attraction to try out one of these preposterously cheap products.