Can you build a 10MHz USB oscilloscope for under $50?

[Damien P]

I'm always concerned before I start a project that if something goes
wrong, I won't have the gear to work out what the problem is. I also
noticed that scopes are expensive and large. I reckon you can build a
USB scope for under $50. (It should probably be called a "signal
analyzer" rather than a scope, since I'd doubt it will be very
accurate.)

Here's my idea:

http://sourcegate.wordpress.com/2010/04/15/a-10mhz-usb-oscilloscope-for-under-50/

Does anyone think this is possible? Would anyone be interested in
trying this out with me? I might struggle a bit with the high
frequencies involved though.

[Ken]

http://www.dpscope.com/

--
You received this message because you are subscribed to the Google Groups "HackerSpace - Adelaide, South Australia" group.
To post to this group, send email to hackerspac...@googlegroups.com.
To unsubscribe from this group, send email to hackerspace-adel...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/hackerspace-adelaide?hl=en.

[Paul Schulz]

Hi Damien,

Count me in..

[Kim Hawtin]

morning all =)

Damien wrote:
> I'm always concerned before I start a project that if something goes
> wrong, I won't have the gear to work out what the problem is.  I also
> noticed that scopes are expensive and large.  I reckon you can build a
> USB scope for under $50.  (It should probably be called a "signal
> analyzer" rather than a scope, since I'd doubt it will be very
> accurate.)
>
> Here's my idea:
>
> http://sourcegate.wordpress.com/2010/04/15/a-10mhz-usb-oscilloscope-for-under-50/
>
> Does anyone think this is possible?

yes, absolutely. assuming you don't include the cost of power supply
or the case it should be doable.
actually you should be able to power it from the USB power, 100mA
should be ok, but it will be noisy.
i had to put filtering on my DC receiver to cut down the RF interfernce.

> Would anyone be interested in trying this out with me?

yes please. i'm building something similar, although I use a sound
card as the input to the peecee.
having an inexpensive way to sample about 30-60 kHz is what I'm interested in.
but you will need a bandwidth much much higer for an oscilloscope.

> I might struggle a bit with the high frequencies involved though.

this is where you cheat a little and have an intermdiate frequency
(IF) like a radio.
in radio speak, you tune your variable frequency oscillator (VFO) to
cover the portion of the band you are interested in.
you can do this with a direct digital synthesis (DDS ) chi[p like the
DS-1077 for under $5;
http://www.littlebirdelectronics.com/products/Programmable-Oscillator-%252d-8kHz-to-133MHz.html

using a DDS you can programmatically adjust the frequency you want to sample at.
you might want to divide the signal down to get better resolution for
a given band.

you can also make a spectrum sampler, you can scan across the band and
sample the frequencies in use.
do some math on that and show whats happening. like a spectrum
analyser, but will probably be a bit less accurate.

alternatively, we could get one like this and build something similar?
http://www.littlebirdelectronics.com/products/Digital-Oscilloscope-DIY-Kit-%252d-Fully-Assembled.html
http://www.littlebirdelectronics.com/products/Digital-Oscilloscope-DIY-Kit-%252d-Through%252dhole.html

Now, if you are actually looking for a CRO, there are a few second
hands ones about...

72,

Kim
--
"Art without engineering is dreaming; engineering without art is
calculating." --SKR

[Ken]

You can see how someone else has done it at http://www.pdamusician.com/dpscope/files/dpscope_v1_1_schematic.pdf

I have one of these things.  I could bring it along to a meeting if anyone is interested.

Ken.

[Robert Hart]

Kim Hawtin wrote:
> you can do this with a direct digital synthesis (DDS ) chi[p like the
> DS-1077 for under $5; http://www.littlebirdelectronics.com/products/Programmable-Oscillator-%252d-8kHz-to-133MHz.html
>

Nice find Kim, just what I need to modifiing my SDR cheers.

Any luck in finding a low cost 128Khz sound card?

alternatively, we could get one like this and build something similar?
http://www.littlebirdelectronics.com/products/Digital-Oscilloscope-DIY-Kit-%252d-Fully-Assembled.html

I like this little kit would useful for building into other projects :-)
as it has its own screen

Robert

[Steven Pickles]

it's not usb powered, but there is this hack to turn a bt878 based tv capture card into a dso.

http://www.domenech.org/bt878a-adc/index-e.htm

there is a driver to treat the input as an audio device which can then be viewed by xoscope in linux.

pix.

[Kim Hawtin]

Robert wrote:

> Kim wrote:
>> you can do this with a direct digital synthesis (DDS ) chi[p like the
>> DS-1077 for under $5;
>>  http://www.littlebirdelectronics.com/products/Programmable-Oscillator-%252d-8kHz-to-133MHz.html
> Nice find Kim, just what I need to modifiing my SDR cheers.

i haven't got one working yet, but i think its worth a try. mind the birdies.

> Any luck in finding a low cost 128Khz sound card?

no, this is why i have a VFO to wind up and down to scan the band.
the sound card in my main machine covers 48KHz, although i have a SB
live that should do 192KHz.
time to find out hey =)

>> alternatively, we could get one like this and build something similar?
>>  http://www.littlebirdelectronics.com/products/Digital-Oscilloscope-DIY-Kit-%252d-Fully-Assembled.html
> I like this little kit would useful for building into other projects :-) as
> it has its own screen

yeah, as a stand alone device it could be useful, but a real cro might
be a better use of the same cash.
a 50Mz HP CRO went last night at the radio club for $50... 100MHz
digital ones go for 300-400$ on ebay.
i have a 20MHz cro and often find myself needing something that can
really cover up to 50MHz...

72,

kim

--
"Art without engineering is dreaming; engineering without art is
calculating." --SKR

[Kim Hawtin]

Pickles wrote:
> it's not usb powered, but there is this hack to turn a bt878 based tv
> capture card into a dso.
> http://www.domenech.org/bt878a-adc/index-e.htm
> there is a driver to treat the input as an audio device which can then be
> viewed by xoscope in linux.

that sounds like an awesome idea, now, how common are these bt878's ?
i am sure i have a few boards with BT848's on them in the shed.

72,

kim

--
"Art without engineering is dreaming; engineering without art is
calculating." --SKR

[Damien P]

That DPScope looks pretty close, but it seems to work a bit
differently in that it stores the data then sends it to the computer
at its convenience. My idea was a bit different in it streams all of
the input it gets, then it's up to software to make some sense of the
data. Using USB 2 means we can comfortably send 20MB/s. The inputs
stages of that scope might be worth looking at though.

I forgot about that Little Bird scope - at 5 MHz it wouldn't be too
bad. I like their handheld portable one as well, but 1 MHz seems a
little bit low to me.

That $50 CRO sounds like a good deal to me!

[Kim Hawtin]

Damien wrote:
> That DPScope looks pretty close, but it seems to work a bit
> differently in that it stores the data then sends it to the computer
> at its convenience.  My idea was a bit different in it streams all of
> the input it gets, then it's up to software to make some sense of the
> data.  Using USB 2 means we can comfortably send 20MB/s.  The inputs
> stages of that scope might be worth looking at though.
>
> I forgot about that Little Bird scope - at 5 MHz it wouldn't be too
> bad.  I like their handheld portable one as well, but 1 MHz seems a
> little bit low to me.
>
> That $50 CRO sounds like a good deal to me!

i know its not quite what you have in mind, but have a look
at the open high performance software defined radio site;
http://openhpsdr.org/
they start send from 8000 samples per second, IIRR.
each sample can be really large, uses over half the available
bandwidth of USB2.

72,

kim
--
"Art without engineering is dreaming;
engineering without art is calculating." --SKR

[Damien P]

It looks like the Hawkboard has a two parallel ports that can work up
to 50MHz (if I'm reading the datasheet for the processor properly...
see p225 of http://focus.ti.com/docs/prod/folders/print/omap-l138.html
); and the board has a USB 2.0 port. This could replace the FT2232H
chip.

Digikey sells the dev board for the FT2232H for US$27, which is a bit
better than Farnell who wants $65.

[Damien P]

Locally, x-on.com.au sells them for $41, or the chip on its own for
about $10.

[Kim Hawtin]

another scope project;

eOscope ver. 1.2 - LCD Scope 40MSPS
http://www.eosystems.ro/eoscope/eoscope_en.htm

cheers,

kim
--
"Art without engineering is dreaming; engineering without art is
calculating." --SKR

[Damien P]

I've found a suitable ADC that doesn't cost a million dollars: the
tlc5540, which Farnell sells for about $3.50. This solves several
problems with the comparator method:

- Fast enough comparators (100MHz) are expensive. (Oddly enough, op-
amps with a similar response don't seem to be as expensive... what's
the difference?)
- Generating a 8x clock to run the counters from. Crystals above
40MHz don't seem to exist. The FT2232H needs a 12MHz crystal which I
thought could be multiplied using a tank circuit, but this is all
unfamiliar to me.
- There were also issues with the FT2232H, which wants 50ns before it
samples its input, but the counter will have gained 5 values after
this time. Using an ADC seems much safer.

This ADC goes up to 70MHz, so if I want to push it I could sample two
channels at 20MHz each, which is approaching pushing the limits of
USB. But if I settle for 12MHz I can re-use the FT2232H's crystal.

I could probably nick the input stage for the ADC off one of the
scopes linked to in this thread.