# Any recommendations for a good cheap USB oscilloscope?

Gold Member
2020 Award
There has been some discussion on another thread about possible problems with mains spikes and lighting LEDs. I have decided that I could use an entry level scope to capture and measure any that are around on my supply. My regular analogue scope is hopeless for this purpose.
I am sure that there will be some home experimenters on this forum. Is there a particular make that I should love / avoid? The Hantek range is high profile. Are they OK or is it just 'marketing'?

krater

Hi Sophicentaur,
I was considering purchasing the Hantek HT6022BC20MHZ usb scope, however, I read some reviews on Amazon and 25% of reviewers gave it one star. Evidently the software is atrocious. One of many problems is the trigger is off to the left of the screen. Inexpensive but probably unfit for your purpose. Be careful unless you like collecting useless junk.

Peace,
Fred

I have been looking at the Analog Discovery module for a while - promised to to my self a year ago, but still no.

JRMichler
Here's a possibility: https://www.omega.com/pptst/OM-USB-1208HS_SERIES.html. I have used Omega data acquisition products several times over the years with excellent results. I have no experience with the oscilloscope software that is packaged with the data acquisition boards.

Omega is a reseller. Their pricing seems to be the same as you would pay elsewhere, and they normally have everything in stock. I believe that their USB DAQ products are from Measurement Computing, with Omega labels: https://www.mccdaq.com/data-acquisition/low-cost-daq.

Gold Member
2020 Award
Hi Sophicentaur,
I was considering purchasing the Hantek HT6022BC20MHZ usb scope, however, I read some reviews on Amazon and 25% of reviewers gave it one star. Evidently the software is atrocious. One of many problems is the trigger is off to the left of the screen. Inexpensive but probably unfit for your purpose. Be careful unless you like collecting useless junk.

Peace,
Fred
Thanks Fred, I will bear that in mind. Unfortunately, the problem with Amazon reviews is that you need to know a bit about the reviewers themselves in order to get real use out of them. I have read 'bad' reviews of purchases which just reflect how inept some people can be. Personally, I have had many products which have worked find - when I have learned how to use them (RTFM etc.). I looked at all the Amazon Hantek reviews and found that over 50% were 4 or 5 stars. It we were discussing a washing machine or vacuum cleaner, I would probably take more notice of the bad comments. This is one reason why I was after personal assessment of the devices by PF members - who tend to know more about how many beans make five. Pity you haven't actually bought one and learned to tame it.
When you think how much you have to pay for an Arduino processor plus peripherals, the £50 for a working system is very cheap and it's not surprising you don't get £1k's worth of product.
@JRMichler: Also, thanks for that. The omega page shows mostly "big boys' " equipment that would require a fair bit of development work to make it work.
I have been looking at the Analog Discovery module for a while - promised to to my self a year ago, but still no.
That unit looks very useful and not a ridiculous price but it is more than I want to pay - I want to monitor mains spikes in a hunt for a way to protect my LEDs against whatever it is that's giving the such short lives. I could buy a lot of replacements for £200.
I may need to rethink what to do about this. eBay doesn't have much in the way of used PC scopes. It makes me wonder why dissatisfied users have not flooded eBay with them.

anorlunda
Staff Emeritus
There has been some discussion on another thread about possible problems with mains spikes and lighting LEDs. I have decided that I could use an entry level scope to capture and measure any that are around on my supply.

Whoa. Slow down. The kinds of spikes that might fry a LED do not occur every cycle. They might occur 3 or 4 times per year when one of your neighbors switches a big load, or when the line crews are doing maintenance and repair, or when lightning hits a power line miles away from your home.

When I was a fire fighter, I recall incidents where the wind knocked down a high voltage line. As the wire was falling, it momentarily touched a lower voltage line below. Those incidents fried TVs and other electronics in houses for miles around. Most people could not imagine what caused it.

There are nearly 2 billion cycles in a year. You can't watch all of them on your oscilloscope. So I think an oscilloscope is not the right tool for the job.

To capture those infrequent spike events, you need something that measures the highest instantaneous voltage, then retains that measurement until it is copied somewhere, or until you manually reset it.

An Arduino or a Raspberry Pi might make an excellent data logger, but even then you can't just measure voltage on an analog input in a software loop. The spike might last only a microsecond and your software may not be able to loop a million times per second. Some kind of analog latching circuit needs to be part of your solution.

Check the Arduino or Rasberry Pi, forums for "power quality monitoring" projects.

dlgoff, NTL2009 and donpacino
Borek
Mentor
If you want it cheap there is always DSO138 DIY. With all its limitations it already proved itself useful to me on several occasions.

What about a data logging multimeter with peak detect?

donpacino
Gold Member
2020 Award
Whoa. Slow down. The kinds of spikes that might fry a LED do not occur every cycle.
I thought I could put the scope on single shot (perhaps 50ms sweep) with a pre trigger of a few ms. Wouldn't that pick up a spike? I was thinking that a 20MHz-ish bandwidth scope (what you get for that sort of money) would show a blip. Looking at the display every few hours / days would show what happened at that one instant. You could do it with an old analogue storage scope as long as you could come back before the screen had bloomed into nothingness.
Do you really think my idea is a non starter? Best to strangle it at birth than find it's a non-runner.
Hantek scares me so I decided to buy Xprotolab Plain http://www.gabotronics.com/oscilloscopes/xprotolab-plain.htm. Cost: $20 . I plan on setting it up next week. Insha Allah I'll let you know how it goes. Fred I am in no rush so I can wait to see how well you get on with it. What about a data logging multimeter with peak detect? I haven't used one of those. Would a peak reading re-set after a while? To allow recording more than one peak. If you want it cheap there is always DSO138 DIY. With all its limitations it already proved itself useful to me on several occasions. That looks pretty 'entry level' and it's in kit form (isn't it?). Also cheaper than I was think of paying. Opinions on the Picoscope?? anorlunda Staff Emeritus I thought I could put the scope on single shot (perhaps 50ms sweep) with a pre trigger of a few ms. Wouldn't that pick up a spike? I was thinking that a 20MHz-ish bandwidth scope (what you get for that sort of money) would show a blip. Looking at the display every few hours / days would show what happened at that one instant. You could do it with an old analogue storage scope as long as you could come back before the screen had bloomed into nothingness. Do you really think my idea is a non starter? Best to strangle it at birth than find it's a non-runner. That depends. What are you triggering on? Is it practical to leave it set up for 3 months waiting for a trigger? Is it really the waveform you want to see, or just the value of the peak voltage? Borek Mentor That looks pretty 'entry level' and it's in kit form (isn't it?). Also cheaper than I was think of paying. Yes and yes, requires an evening with a soldering iron, but in general it is quite simple (if I made it, everyone can ). Science Advisor Gold Member 2020 Award That depends. What are you triggering on? Is it practical to leave it set up for 3 months waiting for a trigger? Is it really the waveform you want to see, or just the value of the peak voltage? I could leave it waiting for it to trigger on a significant spike whenever I wasn't using the PC. I could test all my electrical appliances without needing to be right by the logger. A trace of an actual spike could be informative. I could also log the voltage variations, which would be useful in itself. I remember, years ago, I was losing filament bulbs after only a very short time. The mains volts were peaking well over 250V and it was only when the company installed a logger (pen and paper roll) that they actually believed me. Of course, the LIARS told me there was nothing wrong with my supply but I had read the trace and told them so. Within a day or two, the volts were something much more reasonable. The house was right next to a substation which fed a couple of hundred houses, downstream. They had jacked up the transformer volts so that people at the far end of a resistive cable were getting a reasonable voltage. I never went to those houses to ask if their lights were only red hot after the reduction! Science Advisor Gold Member 2020 Award The picoscope looks a far superior unit (the cheapest is about £70) and the opinion is that the software is pretty good too. I think I will need to find another excuse for buying one before I actually part with the old mazuma, though. It will have to go somewhere on the list of priorities but not at the top. I haven't used one of those. Would a peak reading re-set after a while? To allow recording more than one peak. https://www.amazon.com/dp/B010Y71G1K/?tag=pfamazon01-20 Something like this cheap and cheerful object. It sends data to an app on your phone - the idea being, you check it every day or so for the max reading. I haven’t used one of these, but my Fluke 87 has a 250 us peak reading with max/min recording. I cancel the auto power off, and power it from a wall adapter. That way, it will pick up the max peak voltage, but won’t record how many peaks per session. I think the Owon would. Common surge protection clips any voltage over 400V - is this what you would consider a damaging level for your lights, or perhaps 10% above the usual peak of 330V, say 363V? I’d be very interested to see what you find, but I suspect it may be a while before you catch some juicy transients. How would you solve this - can you get global surge protection for whole lighting circuits? You’d need it to be connected all the time. What about modifying an LED light, that is used most of the time, with a suitable MOV? This should clip transients for the whole circuit. Better still, an earthed light fitting with three MOVs between L, N and E. Science Advisor Gold Member 2020 Award https://www.amazon.com/dp/B010Y71G1K/?tag=pfamazon01-20 Something like this cheap and cheerful object. It sends data to an app on your phone - the idea being, you check it every day or so for the max reading. I haven’t used one of these, but my Fluke 87 has a 250 us peak reading with max/min recording. I cancel the auto power off, and power it from a wall adapter. That way, it will pick up the max peak voltage, but won’t record how many peaks per session. I think the Owon would. Common surge protection clips any voltage over 400V - is this what you would consider a damaging level for your lights, or perhaps 10% above the usual peak of 330V, say 363V? I’d be very interested to see what you find, but I suspect it may be a while before you catch some juicy transients. How would you solve this - can you get global surge protection for whole lighting circuits? You’d need it to be connected all the time. What about modifying an LED light, that is used most of the time, with a suitable MOV? This should clip transients for the whole circuit. Better still, an earthed light fitting with three MOVs between L, N and E. This is the trouble with PF. People are so sensible and well informed. (Moi aussi) Yes, that solution would probably give me an answer and I would have a real problem arguing against it. However, I would really fancy a better scope than the (second hand analogue) one I have already. There's no excuse for this because I do very little construction or fault finding these days. What I want is for some equally irresponsible person to tell me that the, for example Picoscope is excellent value and works very well. Then I could go out an buy one - and regret it when I find that it doesn't solve my problem. I know that a good digital scope will cost a lot. I could spend my money on improved Astro equipment and get more use from it. But thanks for the advice in all these posts. Mainly, you have saved me from the potential Hantek Black Hole. Meanwhile, I think I will provide my cheaper DMM with a mains adaptor and just keep my eye on the running peak value it measures. That could be sufficient evidence for me. anorlunda Staff Emeritus atyy tech99 Gold Member THANK YOU @tech99 . It is not every day that I get to add an electrical word to my vocabulary. I never heard the word coherer before. That also sounds like a fun project. I bet you could adjust the gap to trigger at different voltages. The threshold voltage seems to depend on the materials and is just a few volts. There is also a "linear" mode of operation allowing detection of signals of around 50mV without amplification. The following notes might be of interest:- The device was used as an early detector of radio waves. It is usually described as a glass tube in which two metal electrodes are placed, the gap between them containing some loose metal powder or sometimes a drop of mercury. There are other types known, some of which resemble semiconductor diodes, but which might or might not employ semiconductor action, and others which are just light contacts. A bias battery of typically 1 volt is connected, no current flowing under rest conditions, but when the voltage is increased by the addition of RF voltage to the battery potential, a direct current starts to flow. This can create a click in a pair of earphones, and if the signal is amplitude modulated, the modulation can be heard. If the RF voltage is fairly large, say 1 volt or more, and the resistance in the circuit is small, the contact can micro-weld itself closed, so that a large current can flow from the battery. This produces a latching action, able to operate a relay, and a mechanical reset was often used in the form of an electromagnet to shake the coherer and break the contact. The operation of the coherer is seen to have two modes: a “linear” detection mode and a latching mode. It does not appear to involve the cohering of the metal particles together, and such a mechanical action seems unlikely in view of the use of the device at 60 GHz by J C Bose in 1895. The linear mode seems to occur due to an oxide film on a contact surface, and the response seems to be an S-shaped curve which is symmetrical with respect to battery polarity. By biasing the device a little way up the curve, an asymmetrical action occurs, and when AC is added to the bias potential, rectification occurs. However, this does not appear to be semiconductor rectifying action, as it occurs with either battery polarity, even though the materials used might resemble those of a semiconductor diode or crystal detector. The action has been said to resemble that of a Metal-Insulator-Metal (MIM) diode, where tunnelling occurs through an insulating oxide layer. My own observations support this, as I found the action to be nearly always symmetrical, and it did not occur with carbon, which does not form an oxide film. A particularly sensitive design uses copper electrodes with a drop of mercury between them, and it seems likely that the oxide barrier consists of copper oxide. It also works with iron, zinc brass etc but not carbon. There is no need for the signal to be AC; the coherer is just responding to an increase in voltage, either as a non linear conductor or a threshold device, and frequency has no relevance. Maybe the first demonstration of radio communication took place in London in February 1880, when Professor Hughes obtained a range of 500m using a mobile receiver having a steel/carbon contact. The mercury coherer was used by Marconi in 1901 for the transatlantic test, in conjunction with a very sensitive earpiece. In my own tests I was able to hear HF broadcasting with a coherer and obtained sensitivity approximately 10 dB inferior to that of a Germanium diode. sophiecentaur and anorlunda The picoscope looks a far superior unit (the cheapest is about £70) and the opinion is that the software is pretty good too. I think I will need to find another excuse for buying one before I actually part with the old mazuma, though. It will have to go somewhere on the list of priorities but not at the top. I purchased this picoscope a few months ago. Loaded the software on my linux system, and played with it a bit, and it seemed to work quite well. I haven't had any specific use for it since then, so can't comment much beyond that, but for the price it seemed good to have available in case the need came up. I gave away my dead, bulky tube Tektronix a few decades ago. I paid ~$140 US for the picoscope. It also has a built in function generator, so that's a nice plus.

https://www.amazon.com/gp/product/B00GZMRZ3M/?tag=pfamazon01-20

While they say 10MHz BW, remember you will want several samples to really see a wave shape. So a 1MHz square wave would only show the 3x 5x 7x 9x harmonics. But that's good enough for many things, just don't expect much more than go-no-go signals at much above 1 Mhz.

sophiecentaur
Gold Member
2020 Award
While they say 10MHz BW, remember you will want several samples to really see a wave shape.
I was thinking that mains circuits would probably limit spike bandwidth to not many MHz - or at least contain significant energy within that limit. This link talks in terms of several μs for transient period.
I know we would all want 100MHz scopes but, at the price. . . .

I purchased this picoscope a few months ago.
--- Can you set it up to monitor your mains, and report how it works? -- Need isolation (which becomes a filter), and set the trigger outside of the regular waveform, turning off a large appliance may be a good enough event to force a trigger..