Whats the best osciloscope preferably handheld for the money?

Page 4 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Rubycon

Madame President
Aug 10, 2005
17,768
485
126

bobdole369

Diamond Member
Dec 15, 2004
4,504
2
0
forget the multimeter.
What I need is an DSO preferably handheld. I will be working on RF mainly UHF. What do I need?

While your thread title and the first 4 pages of this thread were conflicting - I present now what I understood the thread was asking for. A handheld oscope under $100. This won't do at all for UHF though, at least not anything besides the audio path and maybe one or two of the mixer stages.

BOBDOLE DELIVERS!

http://www.google.com/products/cata...=X&ei=0bboTv6MLsTVgQeF98mBCQ&ved=0CIoBEPMCMAA

I hope your budget is more than $100 now.

A scope for $100? You'll be lucky to find a busted 30 year old scope for that.

Your best bet will be to get one of the software scopes that you run from your computer. I think those are less than $1000.

DONE, see above for a $60 handheld oscope. It's absolute garbage and you build it yourself.
 

bignateyk

Lifer
Apr 22, 2002
11,288
7
0
While your thread title and the first 4 pages of this thread were conflicting - I present now what I understood the thread was asking for. A handheld oscope under $100. This won't do at all for UHF though, at least not anything besides the audio path and maybe one or two of the mixer stages.

BOBDOLE DELIVERS!

http://www.google.com/products/cata...=X&ei=0bboTv6MLsTVgQeF98mBCQ&ved=0CIoBEPMCMAA





DONE, see above for a $60 handheld oscope. It's absolute garbage and you build it yourself.


1 MHz? *Shutters* D:

This is why I'm glad I work for a company with lots of $$ for equipment. We have about 10 million dollars worth of high end agilent equipment to work with.
 

PottedMeat

Lifer
Apr 17, 2002
12,363
475
126
I am not even aware there are handheld osciloscopes.

heh i bought one off ebay for ~$100 about 10 years ago. made in germany and was about 15 years old at the time. it had a really hard to see lcd screen, ~5Msamp/ch, and a weird 24V adapter. thing was barely usable.

at least i was able to dump it for ~$10 less than i paid for it.
 

bobdole369

Diamond Member
Dec 15, 2004
4,504
2
0
1 MHz? *Shutters* D:

This is why I'm glad I work for a company with lots of $$ for equipment. We have about 10 million dollars worth of high end agilent equipment to work with.

OH yeah, and at that price point I guarantee it isn't even close to 1mhz, and its probably only good to see if a signal exists. I wouldn't trust a single measurement from it.
 

Rubycon

Madame President
Aug 10, 2005
17,768
485
126
Waiting for the OP to change his/her mind again, asking for spectrum analyzer. :biggrin:
 

Modelworks

Lifer
Feb 22, 2007
16,240
7
76
I haven't purchased one, but this caught my eye a few weeks ago. ARM based pocket oscilloscope for <$200. 72MHz bandwidth and a built in signal generator.


Edit: This link has more specs listed: http://www.micro4you.com/store/tools/instruments/ds203-oscilloscope.html


They are crap for anything but very basic task. The specs might seem good at first but when you take them apart you see how bad things are:
Sampling rate 30S/s – 72MS/s

72Million sample rate does not equal 72MHZ scope. The reason is because of how ADC work and how the output is determined. The fact they are trying to promote it as a 72Mhz scope is enough for me to know that they don't have a clue what they are selling and that means the rest of the product is probably junk too.

The other thing is bandwidth and sampling rate are linked but separate features. To sample a 72Mhz signal accurately you also have to have the pathways into the ADC capable of delivering the signal unchanged in order to not change the signal you are monitoring.

Here is a quick overview of ADC and how the specifications work.
http://zone.ni.com/devzone/cda/tut/p/id/2709
 

Modelworks

Lifer
Feb 22, 2007
16,240
7
76
Those looking for used test equipment check out gov liquidation. The stuff is surplus military mostly but if you know what you are buying you can get some good deals. A lot of this stuff was really expensive when purchased, $10K scopes are common items. You can often pick them up as 2 or 3 in a lot for $300 and because they are all the same models salvage the parts to a functional item.
http://www.govliquidation.com/test-equipment.html
 

Rubycon

Madame President
Aug 10, 2005
17,768
485
126
They are crap for anything but very basic task. The specs might seem good at first but when you take them apart you see how bad things are:
Sampling rate 30S/s – 72MS/s

72Million sample rate does not equal 72MHZ scope. The reason is because of how ADC work and how the output is determined. The fact they are trying to promote it as a 72Mhz scope is enough for me to know that they don't have a clue what they are selling and that means the rest of the product is probably junk too.

The other thing is bandwidth and sampling rate are linked but separate features. To sample a 72Mhz signal accurately you also have to have the pathways into the ADC capable of delivering the signal unchanged in order to not change the signal you are monitoring.

Here is a quick overview of ADC and how the specifications work.
http://zone.ni.com/devzone/cda/tut/p/id/2709

Of course DX is like Harbor Freight in HK but that does not mean I would not want one. ;)
 

NutBucket

Lifer
Aug 30, 2000
27,045
559
126
Input impedance isn't the problem, its the DC voltage that kills them.

I haven't payed attention but I think the newer ones are more tolerant of DC than previous generations.
 
May 11, 2008
20,138
1,149
126
This. Simple as all hell. Lets you put all your focus on your work, not your tool.

I agree.
The only thing i find troubling with fluke meters is that these meters have a tendency to auto power off at the time i start measuring again.

I am still waiting for Fluke to add logic that checks the signal level or checks the auto range trigger to reset the auto power off timer
.
For example, when measuring, there will be adc counts (dual slope adc conversion). If added logic would reset the timeout timer of the auto power off function when a certain adc counts are reached or that the logic would use the auto ranging feature to detect that measurement is taking place and thus would reset the auto power off timer... The multimeter would stay on as long as measuring is taking place. The time out time can be even shorter now since the timeout timer is reset at every measurement.

I know the meter can be set to never auto power off when needed. But that is also not always the answer.
 
May 11, 2008
20,138
1,149
126
Avoid digital. Digital multimeters are total garbage. They simply don't work.

You can try this at home. Using a digital multimeter, try to measure the voltage coming out of an AC inverter. It will say 120, then 60, then 150, then 78, then 95, and it will keep jumping around.
Try the same thing with an analogue meter. It will go straight to 120v and stay there. No BS.

nevermind. See next post.
(This post can be deleted)
 
Last edited:
May 11, 2008
20,138
1,149
126
People have come to expect solid facts out of me. When other people say things like "torque wins races," it's up to guys like to me to keep it real and explain why that's the most retarded statement in the history of the universe.

error in digital multimeters:
http://cp.literature.agilent.com/litweb/pdf/5988-5513EN.pdf



Uh oh, I guess Agilent is in on the conspiracy as well. They too think this is a common problem!
/tinfoilhat

Not sinusoidal... That explains a lot. Circuitry inside the multimeter will contain 1.11x amplification adjustment to make up for the error when a dc system must measure an ac sine voltage.
If the signal is not sinusoidal, error in measurement will appear.

The 1.11 is a factor derived by dividing the effective value by the average value of a sine wave signal. For a sine wave, this is 0.707 or 1/1.41 for the effective value and 2/pi for the average value of a sine wave. This sine wave must not contain any DC offset ! I would think that the voltage lines where you live are very electrically dirty.

A visual explanation :
http://www.tpub.com/neets/book2/1f.htm


32NE0127.GIF
 
Last edited:
May 11, 2008
20,138
1,149
126
They are crap for anything but very basic task. The specs might seem good at first but when you take them apart you see how bad things are:
Sampling rate 30S/s &#8211; 72MS/s

72Million sample rate does not equal 72MHZ scope. The reason is because of how ADC work and how the output is determined. The fact they are trying to promote it as a 72Mhz scope is enough for me to know that they don't have a clue what they are selling and that means the rest of the product is probably junk too.

The other thing is bandwidth and sampling rate are linked but separate features. To sample a 72Mhz signal accurately you also have to have the pathways into the ADC capable of delivering the signal unchanged in order to not change the signal you are monitoring.

Here is a quick overview of ADC and how the specifications work.
http://zone.ni.com/devzone/cda/tut/p/id/2709

I agree fully with you. There is Nyquist but in reality, the higher the sample rate the better. Especially without using mathematical tricks which will always introduce artifacts in results. Especially if the signal is unknown and thus the artifacts cannot be removed with after processing to end up with the desired signal.
 
May 11, 2008
20,138
1,149
126
I learned average as .707 and RMS as .636.

Woops for you i guess : :eek:
Effective level is 1 divided by square root of 2 * peak value for sine wave.
Or peak wave divided by square root of 2.
The pdf of ShawnD1 will explain...


You must be making a yoke. :)

But here is a picture for reference :

tm-t_wt3_20.gif
 
Last edited:
May 11, 2008
20,138
1,149
126
oscilloscope

Indeed.

Since tablets are sort of cheap and most people have a windows based laptop :
Perhaps then picoscopes are an idea.
http://www.picotech.com/oscilloscope-specifications.html
Even examples are provided to make your own software or use labview.
http://www.picotech.com/download.html.

I hope to build or at least start the development of a small oscilloscope device that will send data wireless to a pc or android phone (makes for great galvanic separation). I am munching through VHDL (work related and hobby) to be able to build something such as this but with out the display and the ui and with better analog hardware :

There is the DSO Quad from seeed studio.
http://www.seeedstudio.com/wiki/DSO_Quad
This is the hardware :
http://www.seeedstudio.com/forum/download/file.php?id=571

I think it is better to remove the LCD, the user interface and primary focus on functionality. Any current model android phone or Iphone or pc has multiple times the calculation and processing power of the STM32F103 used in this device. Also there will be more memory for storage that way.
Imagine that such a device over a few years will have hundreds of megabytes in memristor memory or Fram technology. Automatic data logging without any loss of data. The list of enhancements that can be made with Fram or upcoming memristor memory is very large.