Reasonable oscilloscope screen shots with digital cameras?

Spehro, that is an excellent quality photo. But I'd bet it was on 35mm film.
This is what I used so far but there are downsides.

We live outside urbia so it takes a few days to receive your photos. Too long
for many projects. For an extra few Dollars they'll create a CD which allows
insertion of pics into documents but the numbering scheme is so clumsy you'll
have to hunt on the CD for the right picture for at least 5 minutes.

The other thing is that I only achieved this kind of quality with very good
lenses. Which I have, but once going digital I couldn't use them anymore and
would be thrown back to that little lens that is part of the digital camera.
Unless, of course, I take apart my old SRT100, take the guts out of a digital
camera and kludge the two remnants together.

Regards, Joerg.
 
I'd like to as well. But many scopes don't offer RS232, only HPIB.
quipping each with an NI converter would get very expensive.

Regards, Joerg.
 
Hi Ian,

Actually, glass does measure up pretty well if it was a quality lens in the first
place.

For my master's degree final I built a CCD camera from scratch and since the
affordable sensors in those days were under 1/2" diagonal I know the lens dilemma.
But I was able to find a suitable lens for about $300 that measured up well enough as
to not degrade the point spread function and other stuff noticably. The sub-$100
lenses, of course, didn't measure up.

You'll always end up with "unwanted zoom" here but for scope pictures from a tripod
that is ok.

Regards, Joerg
http://www.analogconsultants.com
 
Thanks, Mike. But the web links just produce a blank web site with a
logo.

Regards, Joerg
http://www.analogconsultants.com
 
"Joerg" <notthisjoergsch@removethispacbell.net> schreef in bericht
news:408D43D4.5C5212E3@removethispacbell.net...
Spehro, that is an excellent quality photo. But I'd bet it was on 35mm
film.
This is what I used so far but there are downsides.

We live outside urbia so it takes a few days to receive your photos. Too
long
for many projects. For an extra few Dollars they'll create a CD which
allows
insertion of pics into documents but the numbering scheme is so clumsy
you'll
have to hunt on the CD for the right picture for at least 5 minutes.

The other thing is that I only achieved this kind of quality with very
good
lenses. Which I have, but once going digital I couldn't use them anymore
and
would be thrown back to that little lens that is part of the digital
camera.
Unless, of course, I take apart my old SRT100, take the guts out of a
digital
camera and kludge the two remnants together.
Click to expand...
I don't understand your concerns about those little lenses.
This picture was taken 5 minutes ago, with Olympus C4000,
which costs roughly $250 or 250 Euro:

www.planet.nl/~f.bemelman/images/P4260039.JPG

No tripod, half a meter distance, no touch up, and using the
automatic setting mode for dummies.

--
Thanks, Frank.
(remove 'x' and 'invalid' when replying by email)
 
Joerg <notthisjoergsch@removethispacbell.net> wrote:
Hi Ian,

Actually, glass does measure up pretty well if it was a quality lens in the first
place.

For my master's degree final I built a CCD camera from scratch and since the
affordable sensors in those days were under 1/2" diagonal I know the lens dilemma.
But I was able to find a suitable lens for about $300 that measured up well enough as
to not degrade the point spread function and other stuff noticably. The sub-$100
Click to expand...
What resolution was the sensor you were using?
 
Hi Frank,

That is an excellent shot. With the brightness turned down a tad more this
should be perfect even for publication in a book. You are right, if they come
out this good there should be no need to use extra lenses.

Regards, Joerg
http://www.analogconsultants.com
 
Hi Ian,

Had to look in my ol' thesis book. It was the Philips NXA series, pretty much the first
samples that came out and it had 604 by 576 pixels. I used the monochrome version because
it was meant for industrial pattern recognition.

It did take a few weeks to find a great lens for it though.

Regards, Joerg
http://www.analogconsultants.com
 
Joerg wrote:
Thanks, Mike. But the web links just produce a blank web site with a
logo.

Regards, Joerg
http://www.analogconsultants.com
Click to expand...
Interesting, they come up fine for me. What's the logo?
Unfortunately, the tripod
site has a lot of popups. Popup filters may take out the page too??
Or your site spam filter may filter tripod stuff.
Wonder if anyone else has the same problem.
I can email 'em if you want.
mike

--
Return address is VALID.
Bunch of stuff For Sale and Wanted at the link below.
Toshiba & Compaq LiIon Batteries, Test Equipment
Honda CB-125S $800 in PDX
Yaesu FTV901R Transverter, 30pS pulser
Tektronix Concept Books, spot welding head...
http://www.geocities.com/SiliconValley/Monitor/4710/
 
Hi Mike,

Yes, this happens a few times with tripod and some others. Maybe my
firewall does that. What I see is only the tripod server logo. When a site
begins spawning lots of pop-ups the transmission gets stopped.

If you'd like to email them that would be nice:

joergsch
at
pacbell
dot
net

Sorry for this but spam got so bad that I have to.

Regards, Joerg
http://www.analogconsultants.com
 
Joerg <notthisjoergsch@removethispacbell.net> wrote:
Hi Ian,

Had to look in my ol' thesis book. It was the Philips NXA series, pretty much the first
samples that came out and it had 604 by 576 pixels. I used the monochrome version because
it was meant for industrial pattern recognition.

It did take a few weeks to find a great lens for it though.
Click to expand...
604 pixels across, so the pixels will be spaced around 60/mm.
That's a .33Mp imager.

The best of the non-full frame as I understand it have about double
the area, and twenty times the number of pixels, so you'r looking
at around 180/mm.
Significantly more challenging.
 
Hi Ian,

Well, not quite. We did some pretty nifty sub-pixel stuff and edge detection. That was what
made the lens question so tough.

I also designed a VME bus interface for this camera while at it and had to pay very close
attention not to degrade any signals here. The sampling of the CCD consisted of fast diode
quartetts and toroids so we could squeeze out much more dynamic range than the data sheet said
the arrays had. Many of the modern CCD sensors are read out with not nearly that much attention
to clock spikes through and charge transfer feed through. Often it just gets integrated into
one lumped value and that's then supposed to be the pixel amplitude.

Regards, Joerg
http://www.analogconsultants.com
 
On Mon, 26 Apr 2004 08:38:04 -0700, the renowned John Larkin
<jjlarkin@highlandSNIPtechTHISnologyPLEASE.com> wrote:

On Mon, 26 Apr 2004 08:05:46 GMT, Spehro Pefhany
speffSNIP@interlogDOTyou.knowwhat> wrote:

Communicating modifications or suggestions to a remote site (a few
minutes drive away or across the globe), recording the way things were
before you take apart that complex mechanical assembly with linkages
etc. You can get color prints made at retail places for really cheap
these days (about US0.20) if necessary. Here's a quick snap (with
tripod, flash turned off, didn't bother dimming the room lighting or
dusting the scope off, no touch-up) from a cheap scope (hmm- trace
rotation could use a bit of trim too) that doesn't have graticle
lighting or display of settings. Using the camera timer is a useful
trick to prevent camera shake.

http://www.speff.com/scope.jpg

Best regards,
Spehro Pefhany

Nice one. How far away?
John
Click to expand...
Maybe 2'-3'.

Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
 
Joerg <notthisjoergsch@removethispacbell.net> wrote:
Hi Ian,

Well, not quite. We did some pretty nifty sub-pixel stuff and edge detection. That was what
made the lens question so tough.
Click to expand...
Ah.
I also designed a VME bus interface for this camera while at it and had to pay very close
attention not to degrade any signals here. The sampling of the CCD consisted of fast diode
quartetts and toroids so we could squeeze out much more dynamic range than the data sheet said
the arrays had. Many of the modern CCD sensors are read out with not nearly that much attention
to clock spikes through and charge transfer feed through. Often it just gets integrated into
one lumped value and that's then supposed to be the pixel amplitude.
Click to expand...
Sounds interesting.

I'm in the midst of a variable rate readout design, that can do lower noise
on 'regions of interest', while maintaining pretty much video rates over
the rest.
If the imager in general is read out at 10Mhz, a 20*20 pixel area might
be read out at 10Khz. (with massive oversampling, and attention paid to
the proper readout function)
Not using conventional sampling, just nice fast A/Ds, and doing the
sampling digitally (driving the CCD at just the right time)

Output probably composite video/USB2.

For the goal of eventually doing video-rate star-tracking.
 
Hi Ian,

That is a very interesting project. When I did my CCD project in the mid 80's I couldn't do any of
that because CCD sensors where very fickle. They had to be clocked at just the right speed and
transition times. Too fast and the SNR would tank, too slow and the charges would ooze out all over
the array. You had a window of 50%, if that. I did slow it down once, just for the heck of it. The
image looked like watching TV after 10 to 15 pints of Guinness.

Regards, Joerg
http://www.analogconsultants.com
 
Joerg <notthisjoergsch@removethispacbell.net> wrote:
Hi Ian,

That is a very interesting project. When I did my CCD project in the mid 80's I couldn't do any of
that because CCD sensors where very fickle. They had to be clocked at just the right speed and
transition times. Too fast and the SNR would tank, too slow and the charges would ooze out all over
the array. You had a window of 50%, if that. I did slow it down once, just for the heck of it. The
image looked like watching TV after 10 to 15 pints of Guinness.
Click to expand...
I'm driving the CCD myself, so can control the clocking and slew-rates.
Then again, in the 80s, A/Ds were much, much more expensive.

I'm hoping for perhaps 4 chips in total, including framebuffer RAM, ...,
with options to solder on either USB, or composite output chips.

The second version goes on a 30mm dia board, with micromachined
gyros, and talks over a couple of wires through the gimbal mounts.

The whole thing including gimbals and drivers fits in a 70mm sphere.
 
Hi Ian,

Actually there were some nice ADCs in the 80's. When I presented this project at an ICPS conference most
of the questions from the audience related to how the AD conversion was done. These converters weren't
marketed aggressively enough and many engineers didn't know they existed. I used an 8 bit converter from
ITT. Then Siemens had some extremely fast 6 bit versions. Both were leaders of the pack at that time but
surprisingly both companies seem to no longer be in that biz.

8 bits wasn't a lot but all I could find within budget. But when we cranked the video amp gain in front
of that we could poke the camera out of the window and discern the license plate number of a car that
zoomed by in the dark. We could not read the plates with our eyes.

The hottest rod today might be the ADS5500 from TI. Check it out, I believe its 14 bits at 125MHz for
under $100 or something like that.

I found that the purity of the clocks to the CCD was the most important parameter. Any phase noise on
those would irrecoverably spoil the broth. For the readout I stuck to microwave techniques and
symmetrically driven samplers.

Regards, Joerg
http://www.analogconsultants.com
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top