AN: Open Instrumentation Project Progress
PeterH
open-source software and low cost hardware for electronic
instrumentation.
Currently available instruments include an oscilloscope, waveform
generator
and network analyser. Source code is available at sourceforge/oip.
Information on the hardware is available at www.syscompdesign.com.
Recent developments in the project:
- The Mac OS-X port appears to be operational, thanks to a contribution
by
Seneca Cunningham. We hope to have the beta test version up on
Sourceforge
shortly. (If you'd like to push that along, send me email and we'll
move it
up in the queue.) This means that the oscilloscope and waveform
generator
are now available for Windows, Linux and Mac OS-X operating systems.
There are three new papers on the Syscomp web page:
- An illustration of oscilloscope single-shot capture operation,
documenting
contact bounce in an electromechanical relay.
- Notes on BNC adaptors and cables, including a pointer to building
cables
using RG-174U (small diameter coax).
- Instructions on Downloading and Executing code from Sourceforge.
(This is
specific to OIP software, but may be useful for other downloads using
the
Tcl/Tk language.)
The vector network analyser which uses the oscilloscope and waveform
generator plus GPLed software, is described in the September issue of
Circuit Cellar magazine.
The oscilloscope software has the following new features:
- Software Zoom (aka Waveform Browser) captures 32k samples and allows
a zoom
into any portion of that data record
- Vertical Calibration allows the vertical sensitivity to be
calibrated, to
yeild improved accuracy over the raw hardware
- X-Y Plot displays in the foreground while displaying X vs time and Y
vs time
in the background. Allows the display of Lissajous figures.
The waveform generator software now supports 0.1Hz resolution. (The
hardware always supported that resolution, now the software takes
advantage
of that.)
Peter Hiscocks
Syscomp Electronic Design Limited, Toronto
http://www.syscompdesign.com
647-839-0325
Joerg
> The Open Instrumentation Project (OIP) has been formed to support
> open-source software and low cost hardware for electronic
> instrumentation.
> Currently available instruments include an oscilloscope, waveform
> generator
> and network analyser. Source code is available at sourceforge/oip.
> Information on the hardware is available at www.syscompdesign.com.
>
Just an idea: What mankind really needs is a simple spectrum analyzer
for EMC pre-compliance work. Nothing fancy, just something that ranges
from 150kHz to 1GHz and can show the ballpark numbers. IOW where people
can see how good their chances are before they head off to that
expensive lab test. This tool would also be useful when they have come
back from the EMC lab with a black eye and must chase those nasty clocks
that pushed them over the cert limits.
It doesn't have to be much, maybe a glorified scanner without a front
panel but with a USB connection and nifty display software.
--
Regards, Joerg
Steve Sousa
"Joerg" <notthis...@removethispacbell.net> wrote in message
news:lGTUg.19397$Ij....@newssvr14.news.prodigy.com...
> Just an idea: What mankind really needs is a simple spectrum analyzer
> for EMC pre-compliance work. Nothing fancy, just something that ranges
> from 150kHz to 1GHz and can show the ballpark numbers. IOW where
> people can see how good their chances are before they head off to that
> expensive lab test. This tool would also be useful when they have come
> back from the EMC lab with a black eye and must chase those nasty
> clocks that pushed them over the cert limits.
>
> It doesn't have to be much, maybe a glorified scanner without a front
> panel but with a USB connection and nifty display software.
>
Wouldn't a bunch of antenas and an anechoic chamber be necessary too?
Best Regards
Steve sousa
whi...@gmail.com
Joerg wrote:
> Just an idea: What mankind really needs is a simple spectrum analyzer
> for EMC pre-compliance work. Nothing fancy, just something that ranges
> from 150kHz to 1GHz and can show the ballpark numbers.
Err... isn't this a job for an old grid dip meter? Speaking of which,
why
aren't there any NEW grid dip meters? They're as useful a test tool
for RF as strobe lights are for mechanics, but all the ones I see are
leftovers from 30+ years ago.
Joerg
>
>>Just an idea: What mankind really needs is a simple spectrum analyzer
>>for EMC pre-compliance work. Nothing fancy, just something that ranges
>>from 150kHz to 1GHz and can show the ballpark numbers. IOW where
>>people can see how good their chances are before they head off to that
>>expensive lab test. This tool would also be useful when they have come
>>back from the EMC lab with a black eye and must chase those nasty
>>clocks that pushed them over the cert limits.
>>
>>It doesn't have to be much, maybe a glorified scanner without a front
>>panel but with a USB connection and nifty display software.
>>
> Wouldn't a bunch of antenas and an anechoic chamber be necessary too?
>
Not necessarily. When I do field work I mostly won't even have access to
a shielded room. There just isn't any on site. Sometimes I rent an
antenna and other times I just use my EMCO probe kit.
Much of my EMI diagnosis for clients happens after a failed EMC test.
Then I already have a baseline, the plots from the lab. So I know that,
for example, 319MHz was 8dB above the line. Now I'll take the analyzer
and go after that. Trying to find what generates it and where it comes out.
The first thing to do before any changes to the unit is to log my own
baseline. It doesn't matter so much what kind of antenna I use, even if
it was a coat hanger from the dry cleaners. It's important to
meticulously document positions, distances, angles etc. Also other
objects in the room. That's why I take my digital camera if the company
allows it.
After finding the root cause(s) and remedies a subsequent scan should
show a drastic reduction. Not just 10dB, considering the uncontrolled
environment that would too close for comfort, but something well north
of 20dB. What I prefer to hear from clients later after the 2nd EMC run
is something like "they couldn't see 319MHz anymore".
Joerg
Nah, those are too flimsy for that. You need something where you can
connect a probe or antenna to and where you can narrow the detector BW
to 10kHz or so.
Mine is the Heathkit HD-1250. Still works fine and it's probably the
most "modern" dipper that was made (came as a kit though). My oldest is
a Boonton Megacycle meter. Now that is a true grid dip meter because the
detachable head contains a real mini-tube that has a grid. The Heathkit
contains a FET.
Why no new ones? Look what comes out of universities these days. Most of
those folks can't even solder. With hams there also is a disturbing
trend towards not building anything with their own hands.
mrda...@gmail.com
Looks like you've just volunteered to design one!
You've been conspicuously absent from the "favorite drinks" thread, I
must add... :-)
Michael
Joerg
>>
>>>The Open Instrumentation Project (OIP) has been formed to support
>>>open-source software and low cost hardware for electronic
>>>instrumentation.
>>>Currently available instruments include an oscilloscope, waveform
>>>generator
>>>and network analyser. Source code is available at sourceforge/oip.
>>>Information on the hardware is available at www.syscompdesign.com.
>>>
>>Just an idea: What mankind really needs is a simple spectrum analyzer
>>for EMC pre-compliance work. Nothing fancy, just something that ranges
>>from 150kHz to 1GHz and can show the ballpark numbers. IOW where people
>>can see how good their chances are before they head off to that
>>expensive lab test. This tool would also be useful when they have come
>>back from the EMC lab with a black eye and must chase those nasty clocks
>>that pushed them over the cert limits.
>>
>>It doesn't have to be much, maybe a glorified scanner without a front
>>panel but with a USB connection and nifty display software.
>
> Looks like you've just volunteered to design one!
>
Maybe some day. But I am not a USB expert.
> You've been conspicuously absent from the "favorite drinks" thread, I
> must add... :-)
>
Kind of busy right now and I guess most has already been said. My favorites:
1. Beer (the heavier, darker ales and stouts)
2. More beer.
3. Schnaps (the stuff my wife makes from gin, sugar and blackberries)
mrda...@gmail.com
Joerg wrote:
> Hello Michael,
>
> >>
> >>>The Open Instrumentation Project (OIP) has been formed to support
> >>>open-source software and low cost hardware for electronic
> >>>instrumentation.
> >>>Currently available instruments include an oscilloscope, waveform
> >>>generator
> >>>and network analyser. Source code is available at sourceforge/oip.
> >>>Information on the hardware is available at www.syscompdesign.com.
> >>>
> >>Just an idea: What mankind really needs is a simple spectrum analyzer
> >>for EMC pre-compliance work. Nothing fancy, just something that ranges
> >>from 150kHz to 1GHz and can show the ballpark numbers. IOW where people
> >>can see how good their chances are before they head off to that
> >>expensive lab test. This tool would also be useful when they have come
> >>back from the EMC lab with a black eye and must chase those nasty clocks
> >>that pushed them over the cert limits.
> >>
> >>It doesn't have to be much, maybe a glorified scanner without a front
> >>panel but with a USB connection and nifty display software.
> >
> > Looks like you've just volunteered to design one!
> >
>
> Maybe some day. But I am not a USB expert.
Atmel seems to make a series of USB controller chips. I stumbled on
them while I was doing research into what microcontroller to use for a
first project. I guess they're similar to using a MAX232 for the
serial port. (They're supposed to shield you from the messy details of
USB.)
http://www.atmel.com/products/USB/
You'll master it within minutes, I'm sure.
Take care,
Michael
martin griffith
Yep, seems like a good idea, Joerg
maybe this
http://users4.ev1.net/%7Ewsprowls/ as a starting point.
Warning engineer designed website, may offend many non engineers
martin
Joerg
>>
>>>The Open Instrumentation Project (OIP) has been formed to support
>>>open-source software and low cost hardware for electronic
>>>instrumentation.
>>>Currently available instruments include an oscilloscope, waveform
>>>generator
>>>and network analyser. Source code is available at sourceforge/oip.
>>>Information on the hardware is available at www.syscompdesign.com.
>>>
>>Just an idea: What mankind really needs is a simple spectrum analyzer
>>for EMC pre-compliance work. Nothing fancy, just something that ranges
>
>>from 150kHz to 1GHz and can show the ballpark numbers. IOW where people
>
>>can see how good their chances are before they head off to that
>>expensive lab test. This tool would also be useful when they have come
>>back from the EMC lab with a black eye and must chase those nasty clocks
>>that pushed them over the cert limits.
>>
>>It doesn't have to be much, maybe a glorified scanner without a front
>>panel but with a USB connection and nifty display software.
>
>
> Yep, seems like a good idea, Joerg
> maybe this
> http://users4.ev1.net/%7Ewsprowls/ as a starting point.
>
That is quite a good approach. Nowadays a USB interface would almost be
a must as most laptops do not offer a parallel port anymore. Also, I'd
probably spend a couple bucks more on the log detector :-)
> Warning engineer designed website, may offend many non engineers
>
It's quite good but bogs down older PCs because it is all placed on one
long page. Better to split it up and use hyperlinks.
martin griffith
Just wondering...
Having a laptop next to a spectrum analyser will corrupt all/many of
it readings, wont it?
martin
Joerg
To some extent, yes. But you'd be surprised how much noise modern and
expensive spectrum analyzers let off. Once I was at a client and nobody
had a clue where a new forest of noise peaks could possbibly be coming
from. There was nothing on the plots from the EMC lab, total silence in
that range. So I got some aluminum foil from the cantina and waved it
here and there. Surprise: When I held it in front of the $20k+
analyzer's LCD screen the noise disappeared.
One can live with problems like that. Furnace filter mesh, fencing
materials, aluminum screen door stock, all that helps as long as you can
see the display through it. One of the first questions when arriving for
an EMI job at a new client is where the next hardware store is located.
USB and power cables would need some #43 ferrite but almost any EMI job
requires to keep a couple of pounds of those at hand anyway.
Robert Latest
Joerg <notthis...@removethispacbell.net> wrote
in Msg. <TAWUg.7458$TV3....@newssvr21.news.prodigy.com>
> That is quite a good approach. Nowadays a USB interface would almost be
> a must as most laptops do not offer a parallel port anymore.
Wouldn't an FT245R from FTDI be an almost drop-in replacement?
robert
pbde...@spamnuke.ludd.luthdelete.se.invalid
>>
>>>can see how good their chances are before they head off to that
>>>expensive lab test. This tool would also be useful when they have come
>>>back from the EMC lab with a black eye and must chase those nasty clocks
>>>that pushed them over the cert limits.
>>>
>>>It doesn't have to be much, maybe a glorified scanner without a front
>>>panel but with a USB connection and nifty display software.
>> Yep, seems like a good idea, Joerg maybe this
>> http://users4.ev1.net/%7Ewsprowls/ as a starting point.
>That is quite a good approach. Nowadays a USB interface would almost be
>a must as most laptops do not offer a parallel port anymore. Also, I'd
>probably spend a couple bucks more on the log detector :-)
What about an ethernet interface..?
>> Warning engineer designed website, may offend many non engineers
Meaning there is actually something interesting to read and no BS ;)
martin griffith
pbde...@spamnuke.ludd.luthdelete.se.invalid wrote:
>>> Warning engineer designed website, may offend many non engineers
>
>Meaning there is actually something interesting to read and no BS ;)
Sort of, I just think that one long scrolling page is a PITA, this one
is even worse
http://www.redrok.com/main.htm
martin
pbde...@spamnuke.ludd.luthdelete.se.invalid
Reminds me of a person at uni that didn't know how to use an editor so he
wrote all documents in the same file.. until the editor couldn't handle it ;)
Maybe the html author have same problem? =)
Way better..
http://wiki.aydogan.net/index.php/Main_Page
Donald
>
> Wouldn't an FT245R from FTDI be an almost drop-in replacement?
>
> robert
No
Programming a USB parallel port is not like programming for a hardware
parallel port.
Software that uses a hardware parallel port can not just use a USB port.
A total re-write of the I/O in the PC app would be needed.
donald
PS: If anyone has code available to interface to a USB parallel port,
please share any links or code.
Joerg
>>>>from 150kHz to 1GHz and can show the ballpark numbers. IOW where people
>>>
>>>
>>>>can see how good their chances are before they head off to that
>>>>expensive lab test. This tool would also be useful when they have come
>>>>back from the EMC lab with a black eye and must chase those nasty clocks
>>>>that pushed them over the cert limits.
>>>>
>>>>It doesn't have to be much, maybe a glorified scanner without a front
>>>>panel but with a USB connection and nifty display software.
>
>
>>>Yep, seems like a good idea, Joerg maybe this
>>>http://users4.ev1.net/%7Ewsprowls/ as a starting point.
>
>
>>That is quite a good approach. Nowadays a USB interface would almost be
>>a must as most laptops do not offer a parallel port anymore. Also, I'd
>>probably spend a couple bucks more on the log detector :-)
>
>
> What about an ethernet interface..?
>
That would be ok but needs setup. Which means extra work. A well
engineered USB device plugs in and the matching software miraculously
shows up ;-)
>
>>>Warning engineer designed website, may offend many non engineers
>
> Meaning there is actually something interesting to read and no BS ;)
>
True. But a wee bit of skill application in organizing a document would
have been nice.
Michael A. Terrell
My site is the other extreme:
http://home.earthlink.net/~mike.terrell/
--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.
Michael A. Terrell
Central Florida
Joerg
>>
>>Sort of, I just think that one long scrolling page is a PITA, this one
>>is even worse
>>http://www.redrok.com/main.htm
>
> My site is the other extreme:
> http://home.earthlink.net/~mike.terrell/
>
Just FYI: When using the back button in Mozilla all the menu items
except the one I had used to enter a sub-page become black lines, no
more text in there.
Mike Monett
> Hello Michael,
>> My site is the other extreme:
>> http://home.earthlink.net/~mike.terrell/
> Just FYI: When using the back button in Mozilla all the menu items
> except the one I had used to enter a sub-page become black lines, no
> more text in there.
The site map link in the bottom right corner seems to work much better.
It is much less confusing, and it gives better feedback to show where you
have been and where you may like to go next.
You have all the available links on one page, which helps ensure you
haven't missed one that might be interesting.
Unlike the javascript menus on the front page, the site map links change
color to indicate which pages you have already visited. So you don't waste
time returning to a page you have already seen.
The site map doesn't require javascript. This is a continual source of
malware attacks, and although Michael's site doesn't have any problems, you
always forget to turn it off when you leave and go somewhere else. You
don't need javascript to obscure email addresses - there are many sites
such as http://www.addressmunger.com/ that do an excellent job of keeping
your email address from spambots. So there is no need to subject visitors
to the security risks presented by javascript.
Simpler really is better.
Regards,
Mike Monett
Antiviral, Antibacterial Silver Solution:
http://silversol.freewebpage.org/index.htm
SPICE Analysis of Crystal Oscillators:
http://silversol.freewebpage.org/spice/xtal/clapp.htm
Noise-Rejecting Wideband Sampler:
http://www3.sympatico.ca/add.automation/sampler/intro.htm
Michael A. Terrell
Try the "Home" button on each page instead of the back button.
Nico Coesel
An I2C controlled TV tuner, some analog stuff to do filtering / level
detection and a microcontroller with integrated ADC is all that is
needed. The firmware doesn't need to be fancy: set the frequency, read
the detected level. Done!
--
Reply to nico@nctdevpuntnl (punt=.)
Bedrijven en winkels vindt U op www.adresboekje.nl
JeffM
>>>>Warning engineer designed website, may offend many non engineers
martin griffith wrote:
>I just think that one long scrolling page is a PITA,
>this one is even worse
>http://www.redrok.com/main.htm
No, thanks.
http://validator.w3.org/check?uri=http://www.redrok.com/main.htm
Failed validation, 1573 errors (Like that was a surprize).
Joerg
>>>>>
>>>>>Just an idea: What mankind really needs is a simple spectrum analyzer
>>>>>for EMC pre-compliance work. Nothing fancy, just something that ranges
>>>>
>>>>>from 150kHz to 1GHz and can show the ballpark numbers. IOW where people
>>>>
>>>>>can see how good their chances are before they head off to that
>>>>>expensive lab test. This tool would also be useful when they have come
>>>>>back from the EMC lab with a black eye and must chase those nasty clocks
>>>>>that pushed them over the cert limits.
>>>>>
>>>>>It doesn't have to be much, maybe a glorified scanner without a front
>>>>>panel but with a USB connection and nifty display software.
>>>>
>>>>Looks like you've just volunteered to design one!
>>>>
>>>
>>>Maybe some day. But I am not a USB expert.
>>
>>
>>Atmel seems to make a series of USB controller chips. I stumbled on
>>them while I was doing research into what microcontroller to use for a
>>first project. I guess they're similar to using a MAX232 for the
>>serial port. (They're supposed to shield you from the messy details of
>>USB.)
>>
>>http://www.atmel.com/products/USB/
>>
>>You'll master it within minutes, I'm sure.
>
>
> An I2C controlled TV tuner, some analog stuff to do filtering / level
> detection and a microcontroller with integrated ADC is all that is
> needed. The firmware doesn't need to be fancy: set the frequency, read
> the detected level. Done!
>
TV tuners don't typically reach below 35MHz and not above 850MHz. Some
(but not all) also show a big void between VHF and UHF. And their
dynamic range, oh, don't get me started on that one...
jasen
> donald
>
> PS: If anyone has code available to interface to a USB parallel port,
> please share any links or code.
how about this:
* This driver is for USB parallel port adapters that use the Lucent
* Technologies USS-720 chip. These cables are plugged into your USB
* port and provide USB compatibility to peripherals designed with
* parallel port interfaces.
you'll find it in the linux kernel (kernel.org among other places)
drivers/usb/misc/uss720.c
Bye.
Jasen
pbde...@spamnuke.ludd.luthdelete.se.invalid
>> detection and a microcontroller with integrated ADC is all that is
>> needed. The firmware doesn't need to be fancy: set the frequency, read
>> the detected level. Done!
>TV tuners don't typically reach below 35MHz and not above 850MHz. Some
>(but not all) also show a big void between VHF and UHF. And their
>dynamic range, oh, don't get me started on that one...
Any other "sources" for this than tv's ..?
Joerg
Not really. Except for scanners but the cheap ones aren't very good.
Nico Coesel
You asked for something affordable... An additional mixer can be used
to convert the higher frequencies.
Joerg
>>>
>>>An I2C controlled TV tuner, some analog stuff to do filtering / level
>>>detection and a microcontroller with integrated ADC is all that is
>>>needed. The firmware doesn't need to be fancy: set the frequency, read
>>>the detected level. Done!
>>>
>>TV tuners don't typically reach below 35MHz and not above 850MHz. Some
>>(but not all) also show a big void between VHF and UHF. And their
>>dynamic range, oh, don't get me started on that one...
>
> You asked for something affordable... An additional mixer can be used
> to convert the higher frequencies.
>
Sure. I was just mentioning this because the "Done!" statement ain't
quite true :-)
Mixing up the <35MHz range isn't a big deal, same for extending to the
WLAN range at 2.45GHz which would greatly increase the market potential
of such a solution. The problem I see with tuners is that many are very
poor performers in terms of dynamic range. Pretty much any TV out here
falls off the cliff when a pilot announces that he or she is turning
left base or final. None of my other RF gear does.
vasile
I like you, spoiled childs !
:)
The initial topic was about a 2MHz analog bandwidth oscilloscope and
you've jumped directly to 1Ghz spectrum analyzer !
My opinions: TV tuner is quite nasty for a digitally controled VCO on
46-850MHz (this is the real range for most of the tuners). It require
also a VGA because the oscilator amplitude could vary about 10-20dB in
the whole frequency range.
One tested example is here:
http://hem.passagen.se/communication/uv916rec.html
On ISM band (2400-2500Mhz) there are a few free projects about spectrum
analyzers which could be modified (not easily, nor simple) for
300MHz-1Ghz range:
http://www.dunehaven.com/lcsa.html
This kind of spectrum analyzer will be "lazy". As lazy as TEK492 (and
who have such old beast knows what I'm talking about).
Even the Spectran analyzers are suffering from the same ill:
http://www.aaronia.de/ (click on english on the left side of the page)
This laziness is mostly because of the USB interface which has used the
fast standard (12Mbit per second) instead the high speed of
480Mbit/sec.
Spectran not able to capture all WLAN "glitches" for example.
So... what to do ?
greetings,
Vasile
Joerg
>>>>An I2C controlled TV tuner, some analog stuff to do filtering / level
>>>>detection and a microcontroller with integrated ADC is all that is
>>>>needed. The firmware doesn't need to be fancy: set the frequency, read
>>>>the detected level. Done!
>>
>>>TV tuners don't typically reach below 35MHz and not above 850MHz. Some
>>>(but not all) also show a big void between VHF and UHF. And their
>>>dynamic range, oh, don't get me started on that one...
>>
>>Any other "sources" for this than tv's ..?
>
>
> I like you, spoiled childs !
> :)
That's engineer thinking. Always looking for better, more, lower cost :-)
> The initial topic was about a 2MHz analog bandwidth oscilloscope and
> you've jumped directly to 1Ghz spectrum analyzer !
>
Well, what could you really do with a 2MHz scope? <yawn>
Even my very first scope had 8MHz BW.
> My opinions: TV tuner is quite nasty for a digitally controled VCO on
> 46-850MHz (this is the real range for most of the tuners). It require
> also a VGA because the oscilator amplitude could vary about 10-20dB in
> the whole frequency range.
That could be calibrated out though, same with the input filter sloping.
> One tested example is here:
> http://hem.passagen.se/communication/uv916rec.html
> On ISM band (2400-2500Mhz) there are a few free projects about spectrum
> analyzers which could be modified (not easily, nor simple) for
> 300MHz-1Ghz range:
> http://www.dunehaven.com/lcsa.html
> This kind of spectrum analyzer will be "lazy". As lazy as TEK492 (and
> who have such old beast knows what I'm talking about).
> Even the Spectran analyzers are suffering from the same ill:
> http://www.aaronia.de/ (click on english on the left side of the page)
> This laziness is mostly because of the USB interface which has used the
> fast standard (12Mbit per second) instead the high speed of
> 480Mbit/sec.
> Spectran not able to capture all WLAN "glitches" for example.
> So... what to do ?
>
Aaronia probably got the closest to a low cost solution but they do not
offer a no-display and no-battery version. They should IMHO but they don't.
itsneu...@yahoo.de
the problem is not the USB (its 2.0 and fast enough) its the circuit
itself. Thats why Aaronia is showing up a NEW unit in some weeks: The
SPECTRAN#3. There are lots of rumors on that unit but whats for sure
now is thats it will be very! very fast and cant be compared with the
"old" Spectran V2. Propably thats what you (and others) are looking for
;-)
vasile schrieb: