need help with oscilloscope selection

Hi,
I wanted to have digital oscilloscope. I was using CRT based and it is gone now.
Please suggest some cheap and best oscilloscope to use. Which oscilloscope is good PC based or DSO oscilloscope?

The oscilloscopes cost between $250 – $1mil.
What do you want to spend? N. of Channels? Bandwidth (MHz)? Samples/sec (Gs/s)? Mixed signal (analog+logic analyser)? Memory (few kB – 150MB – unlimited (pc based)?

Good forum to read / ask: http://www.eevblog.com/forum/testgear/

FWIW…

I bought a RIGOL DSO a few years ago which had a 16 channel logic analyser.

But I never use the logic analyser, I bought a PC dongle logic analyser and its much more useful, as the software that can analyse the data, not just view the 1’s and 0’s

I have a Rigol DS1052E which is quite nice.
But meanwhile I would by a the more capable Rigol DS1054Z because it has 4 channels, has a larger Display and can be hacked for higher bandwidth.
This one
http://hackaday.com/2017/08/08/touchscr … illoscope/
sounds funny to play with.

@ChrisMicro

I have the 100Mhz version with logic analyser DS1102D, its a good scope, but the DS1054Z looks really nice, but I’m not sure how many people need 4 channels.
I’d rather have 100Mhz and 2 channels, but the 100Mhz Rigol scopes are considerably more expensive

Thanks,

Lower bandwidth is Ok for me right now like 10Mhz or near to that. But till what level this frequency will be helpful?
i don’t know that much. because for very less time i used oscilloscope.
2-4 channel (not yet decided). Because i used 2 channel only and i had 2 channel only.
not required to have integrated LA in that for me.
i’d love handy useful oscilloscope.

I’d rather have 100Mhz and 2 channels, but the 100Mhz Rigol scopes are considerably more expensive
Ähm .. well.. there is this hack:
http://hackaday.com/2014/11/12/how-to-g … l-ds1054z/

An oscilloscope is an investment for say 5-10y.
100MHz is an absolute minimum today when messing with mcus (for example you will hardly see 36MHz BluePill’s SPI clock), 300MHz (2+Gs/s) is good for next few years.
For example DS2072A is a good candidate – hackable till 300MHz, large screen, a lot of features, dual channel.
Except Rigol there are other “cheaper” brands, but usually they cannot be hacked so quite expensive at 300MHz.

Hi,
I’ve a Rigol MSO2072A. It should be hackerable too… but I haven’t done it for now… (it came with all the sw options included for free… it was a special offer…). I wanted an LA too so I’ve chosen this one.

Here some photo and video using the LAN remote control (very useful…) and some advices how to install the needed sw (I used a VM over a linux host): http://www.stm32duino.com/viewtopic.php?f=45&t=858

as stm32f407ve has a rather fast ADC (3×12-bit, 2.4 MSPS A/D converters) and that it has quite a bit of ram i’m thinking if it might after all be quite feasible to do a cheap storage scope, i.e. 2.4 msps x 3 ~ 7.2 msps, that would enable sampling signal frequencies up to 3.6 mhz being a nyquist rate
the boards could be the stm32f407vet black or stm32f4 discovery
http://wiki.stm32duino.com/index.php?ti … 4xx_boards
for ‘continuous’ sampling, a usb 2.0 ulpi chip/module and interface may be necessary to push the usb limits of 200 mbps, but i’d guess it’d need really good usb cables and construction (and i don’t think there is ‘arduino’ style libraries for it yet, it’d likely mean using st’s cube mx hal codes for that)
https://www.ebay.com/sch/i.html?_nkw=USB3300&_sacat=0
the data could be simply transmitted to the pc over usb-serial and plotted say using processing
https://processing.org/
or even simpler plotted on the arduino ide using the ‘serial plotter’ under tools

my guess is for a logic analyser with simply 0 and 1, a f407 may be able to just pull that off at 42 mhz (using the SPI pin), i’m not sure if it could go any faster with gpio pins

stm32f303, stm32f3 discovery has even more impressive adc capable of 4adc x 5msps, unfortunately the f3 discovery board is a stm32f303vc which sports only 48k of sram
http://www.st.com/en/evaluation-tools/s … overy.html

and of course the $10 stm32duino-o-scope is still pretty much there
http://www.stm32duino.com/viewtopic.php?t=107

on another note, i’m half way wondering how it may be possible to ‘sync’ multiple BP/MM or F407 black so that we’d literally be able to ‘add’ the ADC msps, then if you have 10 BP, u’d get 20 msps scope while 15 f407 would give 108 msps ADC speeds

ChrisMicro

I didn’t realise the newer Rigel scopes were also hackable.

I know the 50Mhz scope of the same series as mine can be hacked to run at 100MHz, but I paid the extra for the 100MHz because I recall noticing in one of the teardown videos that there was a hardware difference in the input electronics, before the signal got to the sampling hardware.

Actually, my scope has a fault on Channel 1, where it starts to get noisy after it’s been turned on for a few mins
So I think I must have fried the input analog circuit at some time.

I was intending to see if it was fixable, but I don’t have time at the moment, and I can still use channel 2 without any noise.

You can have a look at this:
https://www.ikalogic.com/scanaquad-logi … enerators/

Actually, my scope has a fault on Channel 1, where it starts to get noisy after it’s been turned on for a few mins
So I think I must have fried the input analog circuit at some time.
Hi Roger,
this could be probably quite helpful to repair your input amplifier:
https://www.youtube.com/watch?v=lJVrTV_BeGg

@Roger: David from eevblog is your neighbor
Send him the oscope for the repair and in turn there will be a new episode ready

@pito

Yes. Dave is only about 1000km away from me if I went by road.

@ChrisMicro

I have the older scope, and from what Dave says in that video, they have changed the design of the input stages.

and there is this PRUDAQ that’s based on beaglebone black
https://github.com/google/prudaq/wiki
https://groupgets.com/manufacturers/get … cts/prudaq
it is capable of 20mhz x 2
but at $159 it still feel tad premium/pricy
the good thing would be that beaglebone black has 512 megs of onboard ram which means it can capture rather ‘continuous traces’ for quite a while