[sci.electronics] Batty monitors and cheap video

adf@watcsc.waterloo.edu (Andrew Dellano Fernandes) (09/28/89)

First off... a possibly silly question.  Anyone out there know of any sort of   overheating problem a monitor can have that causes the screen to loose          interlace?  Instead of scanning the lines _next_ to each other, it scans them
on top of each other.  I'm reasonably sure that it is a heat related problem,
and it occurs on both the NTSC and analog RGB inputs.
Thanks for any help.

Now, regarding a cheap lo-rez video input, a mech eng friend of mine once
told me of this thing called a RAMera.... 

Warning: this is totally untried.


First of all, take a 4kX1 bit DRAM and somehow expose the silicon.  I suppsose
you could pop the top of  an old ceramic or plane the top layer off a plastic.
Anyway, fit it with a lens... and that is almost it.

To use the thing as a camera, just charge up the thing to all 1's, wait for a
bit, and then check which bits are discharged by the ambient light.

The reason you can't go above 4K is 'cause manufactures like to break up that
nice sqaure area of capacitors above that density.

Anyway, it sounded interesting.


-Andrew. ("adf@watcsc.uwaterloo.ca")

peg@psuecl.bitnet (09/30/89)

> Now, regarding a cheap lo-rez video input, a mech eng friend of mine once
> told me of this thing called a RAMera....

Yes, and you can see why he is an ME, can't you?  ;-)

> First of all, take a 4kX1 bit DRAM and somehow expose the silicon.  I suppsose
> you could pop the top of  an old ceramic or plane the top layer off a plastic.

Do you have a clean room handy? :-)

> To use the thing as a camera, just charge up the thing to all 1's, wait for a
> bit, and then check which bits are discharged by the ambient light.

Well, this actually works, in a manner of speaking.  The devices are called
CCD's, charged coupled devices, and are used for picture tubes in most
camcorders and video cameras.  As for making one from an actual RAM, I
seriously, seriously, seriously doubt that you can get ANYTHING useful by
opening up a RAM chip, but you are welcome to try.

If it helps, though, I recently saw single-dimension CCD's in a catalog.
These were strip type image sensing devices, that I think they use in
optical scanners (like the hand held scanners for computers) and in
optical character recognition devices.  The catalog was either Newark
or Allied Electronics.  I just looked for a while and  didn't find
them, but I know I didn't imagine them.  I seem to recall they were
fairly cheap ( < $100 ).  That might be something for you to fiddle
with, if you are interested in image sensing.

Also, I seem to recall that Don Lancaster spent some time discussing
these devices (as used in camcorders), their availability, etc.  Look
through back issues of Radio-Electronics, or get Don's address from
one and write him.  BTW, Don writes the Hardware Hacker column.

Well, good imaging!!

Paul

wiml@blake.acs.washington.edu (William Lewis) (09/30/89)

In article <57732@psuecl.bitnet> peg@psuecl.bitnet writes:
>> Now, regarding a cheap lo-rez video input, a mech eng friend of mine once
>> told me of this thing called a RAMera....
>
>Yes, and you can see why he is an ME, can't you?  ;-)
>
>> First of all, take a 4kX1 bit DRAM and somehow expose the silicon.  I suppsose
>> you could pop the top of  an old ceramic or plane the top layer off a plastic.
>
>Do you have a clean room handy? :-)
>
>> To use the thing as a camera, just charge up the thing to all 1's, wait for a
>> bit, and then check which bits are discharged by the ambient light.
>
>Well, this actually works, in a manner of speaking.  The devices are called
>CCD's, charged coupled devices, and are used for picture tubes in most
>camcorders and video cameras.  As for making one from an actual RAM, I
>seriously, seriously, seriously doubt that you can get ANYTHING useful by
>opening up a RAM chip, but you are welcome to try.

   Nope, RAM chips work, and are cheaper. Less sensitive, no doubt, but
for normal room lighting it works. I described this to the
original asker of the question, but since it has sprung up in public here,
I may as well post...

   Micron Technology used to, and hopefully still does, market the IS32
Optic Ram. This is a normal 32k RAM chip packaged with a quartz window like
an EPROM. This has two rows of 256x512 cells apiece, with a dead zone
down the middle, but the rectangles are perfectly fine for image sensing.
The address & data lines need to be `descrambled' because the cells aren't
arranged "normally" inside the chip, but this isn't hard. This chip
was the basis of the Oct/Nov '83 Circuit Cellar project (BYTE magazine)
and I would guess the heart of the commercial product called the "Microneye".

   I *have* heard several places that merely popping the top off a RAM
chip (dynamic of course -- the refresh line is your `shutter') will
work, although I would put something across the opening before
imaging seascapes.... 

   The advantages of this (as I originally said in email) are that it's
small, cheap, low power and already a completely digital signal. The
chief disadvantage seems to be that it's slow; each exposure is
black&white (not black-gray-white), so for grayscales you need to make
several exposures and hope the scene doesn't change. And Micron Tech. may
not still be around. Does anyone know? 

   --- phelliax

-- 
wiml@blake.acs.washington.edu        (206)526-5885      Seattle, Washington

henry@utzoo.uucp (Henry Spencer) (09/30/89)

In article <57732@psuecl.bitnet> peg@psuecl.bitnet writes:
>> To use the thing as a camera, just charge up the thing to all 1's, wait for a
>> bit, and then check which bits are discharged by the ambient light.
>
>Well, this actually works, in a manner of speaking.  The devices are called
>CCD's, charged coupled devices, and are used for picture tubes in most
>camcorders and video cameras.  As for making one from an actual RAM, I
>seriously, seriously, seriously doubt that you can get ANYTHING useful by
>opening up a RAM chip, but you are welcome to try.

Sorry, you lose:  it's been done and it works.  Several hobbyist sources,
notably Martin Weinstein's book "Android Design", have details.  How well
it would work with today's RAMs is not clear.  Certainly one would get a
segmented image, since they don't have all the memory bits in one array.
Probably a bigger problem is finding a DRAM in a package you can pop the
lid off of; taking the plastic off a plastic-packaged one without ruining
the chip is almost certainly impossible.

Micron Technology at one point even sold DRAMs with transparent lids as
bargain-basement imagers.
-- 
"Where is D.D. Harriman now,   |     Henry Spencer at U of Toronto Zoology
when we really *need* him?"    | uunet!attcan!utzoo!henry henry@zoo.toronto.edu

fritz@mit-caf.MIT.EDU (Frederick Herrmann) (10/01/89)

In article <57732@psuecl.bitnet> peg@psuecl.bitnet writes:
>> First of all, take a 4kX1 bit DRAM and somehow expose the silicon. \
>
>Do you have a clean room handy? :-)

As a matter of fact, there are a couple of dandy ones in my building.
But the easy way is to buy parts in windowed packages.

>> To use the thing as a camera, just charge up the thing to all 1's, \
>> wait for a 
>> bit, and then check which bits are discharged by the ambient light.
>
>Well, this actually works, in a manner of speaking.  The devices are called
>CCD's, charged coupled devices, and are used for picture tubes in most

Let's be careful with terminology here.  You don't need CCDs to build
an IC imager, as the `RAMera' demonstrates.  A CCD is a kind of
`analog shift register' which moves packets of charge between the
the channels of adjacent MOSFETs.  In the IC `picture tubes' you mention,
CCD are used to shift out the contents of the image sensor array in
a serial format for video output.

I think Ciarcia had a RAMera-like project in Byte a few years back.


				- Fritz
				  fritz@caf.mit.edu

peg@psuecl.bitnet (10/02/89)

In article <3863@blake.acs.washington.edu>, wiml@blake.acs.washington.edu (William Lewis) writes:
> In article <57732@psuecl.bitnet> peg@psuecl.bitnet writes:
>>> Now, regarding a cheap lo-rez video input, a mech eng friend of mine once
>>> told me of this thing called a RAMera....
>>
>    Nope, RAM chips work, and are cheaper. Less sensitive, no doubt, but
> for normal room lighting it works. I described this to the
> original asker of the question, but since it has sprung up in public here,
> I may as well post...
>
>    Micron Technology used to, and hopefully still does, market the IS32
> Optic Ram. This is a normal 32k RAM chip packaged with a quartz window like
> an EPROM. This has two rows of 256x512 cells apiece, with a dead zone

Okay, I stand corrected!  Sorry for being bull-headed!  I had never seen
a RAM with quartz window, and I still can't see ripping the top off of
an ordinary RAM--any way you do it is going to contaminate or destroy
some of the chip.

As for the circuit cellar project:  Not suprising I missed that.  I
am an Amiga user, and Amiga users typically avoid that Byte magazine
like the plague (as they avoid the Amiga like the plague!).

Thanks for the daily humbling!

Paul

henry@utzoo.uucp (Henry Spencer) (10/02/89)

In article <57950@psuecl.bitnet> peg@psuecl.bitnet writes:
>... I still can't see ripping the top off of
>an ordinary RAM--any way you do it is going to contaminate or destroy
>some of the chip.

According to the Weinstein book, it *can* be done.  The Robotics group
at Case Western Reserve tried it in 1978, and found that it was not hard
to avoid damaging the chip (given suitable packages -- plastic-packaged
ones are out, you need something that you can just remove the lid from)
and that exposure to the air was not, in the short run at least, a problem.
For the RAMs of the time, anyway.
-- 
"Where is D.D. Harriman now,   |     Henry Spencer at U of Toronto Zoology
when we really *need* him?"    | uunet!attcan!utzoo!henry henry@zoo.toronto.edu

pcf@galadriel.bt.co.uk (Pete French) (10/02/89)

From article <57732@psuecl.bitnet>, by peg@psuecl.bitnet:
>> Now, regarding a cheap lo-rez video input, a mech eng friend of mine once
>> told me of this thing called a RAMera....
> Well, this actually works, in a manner of speaking.  The devices are called
> CCD's, charged coupled devices, and are used for picture tubes in most
> camcorders and video cameras.  As for making one from an actual RAM, I
> seriously, seriously, seriously doubt that you can get ANYTHING useful by
> opening up a RAM chip, but you are welcome to try.

No - they aint CCDS. CCDs are designer for use as a camera - and you cant
access each bit individualy like you can with RAM, you have to shift them
out serially. In low lighting conditions you can get bits lost in the shifting
which explains the smears accross the picture ou get from CCDs in bad light.

You dont have to open the DRAM yourself anyway - there is a company that makes
normal DRAMS with transparent lids. They are just standard production DRAMS
but can be used as cameras. They are a pain to use however since there are
odities like 50% of the bits are stored inverted and to get a picture you
have to jump about in memory since the bits are not stored linearly accross
the silicon.

This does give a useable picture - dont ask me how, but a friend of mine
interfaced one to a PC as a project and it worked for him.

-Pete.

-- 
       -Pete French.               | "I'm just a vision on your TV screen,
  British Telecom Research Labs.   |  Just something conjoured from a dream"
 Martlesham Heath, East Anglia.    |
All my own thoughts (of course)    |                       -SIOUXSIE.

fwb@demon.siemens.com (Frederic W. Brehm) (10/04/89)

henry@utzoo.uucp (Henry Spencer) writes:
>and that exposure to the air was not, in the short run at least, a problem.
>For the RAMs of the time, anyway.

Most of the area of an IC is covered by a thick (in IC terms) glass
passivation layer.  This layer is a silicate glass and today also contains
a few atoms of other elements to increase its effectiveness.  The only part
that is not covered by the passivation layer is the bond pads.  These bond
pads are usually made of thick aluminum to which gold (usually) wire is
welded.  The other end of the wire is welded to a lead frame made of
electroplated copper, I think (my knowledge gives out around this part of
the device).

All this means that the most likely problem area is bond wires coming loose
because of atmospheric corrosion of the bond pad or maybe the lead frame.
This won't happen overnight.  It will probably take longer than the life of
your experimental (it is experimental, isn't it?)  system.

Fred
--
Frederic W. Brehm	Siemens Corporate Research	Princeton, NJ
fwb@demon.siemens.com	-or-	princeton!siemens!demon!fwb

jones@bach (Clark Jones) (10/04/89)

In article <57950@psuecl.bitnet> peg@psuecl.bitnet writes:
>In article <3863@blake.acs.washington.edu>, wiml@blake.acs.washington.edu (William Lewis) writes:
>> In article <57732@psuecl.bitnet> peg@psuecl.bitnet writes:
>>>> Now, regarding a cheap lo-rez video input, a mech eng friend of mine once
>>>> told me of this thing called a RAMera....
>>>
>>    Nope, RAM chips work, and are cheaper. Less sensitive, no doubt, but
>> for normal room lighting it works. I described this to the
>> original asker of the question, but since it has sprung up in public here,
>> I may as well post...
>>
>>    Micron Technology used to, and hopefully still does, market the IS32
>> Optic Ram. This is a normal 32k RAM chip packaged with a quartz window like
>> an EPROM. This has two rows of 256x512 cells apiece, with a dead zone
>
>Okay, I stand corrected!  Sorry for being bull-headed!  I had never seen
>a RAM with quartz window, and I still can't see ripping the top off of
>an ordinary RAM--any way you do it is going to contaminate or destroy
>some of the chip.

Taking the lid off of a ceramic package with the solder-on lid is merely a
matter of practice using a vise and X-acto knife (_DON'T_ try it without the
vise as you _WILL_ remove some part of your anatomy!).  Most folks get the
knack by about the third package.  (Opening packages is important to the IC
manufacturers, as they want to do autopsies on returned chips to improve the
quality of future chips.)  As for the contamination objection, virtually all
chips have a "passivation glass" top layer that will protect the device from
ordinary indoor contaiminates (copier dust not included :-).  BTW, opening
a plastic package without damaging the chip requires some "interesting"
chemistry to get through the epoxy.  I'd opt for the slightly higher priced
ceramic package if I were wanting to open a "live" chip.

>Thanks for the daily humbling!

Always glad to oblige an Amiga user!  :-)

FYI, "in a former life" I worked for a chip manufacturer... now I work for
a company who's customers are chip manufacturers.  :-)

Disclaimer: The ideas expressed herein are mine and not those of Schlumberger
because they are not covered by the patent agreement!

peg@psuecl.bitnet (10/04/89)

In article <1989Oct2.011535.23529@utzoo.uucp>, henry@utzoo.uucp (Henry Spencer) writes:
> In article <57950@psuecl.bitnet> peg@psuecl.bitnet writes:
>>... I still can't see ripping the top off of
>>an ordinary RAM--any way you do it is going to contaminate or destroy
>>some of the chip.
>
> According to the Weinstein book, it *can* be done.  The Robotics group
> at Case Western Reserve tried it in 1978, and found that it was not hard
> to avoid damaging the chip (given suitable packages -- plastic-packaged

Now I *really* feel silly!!  I went to CWRU for a year.  Gosh, I always
know there was *something* special about that place (besides the
Michaelson-Morley interferometer--which flopped).

Paul

jimc@iscuva.ISCS.COM (Jim Cathey) (10/05/89)

In article <371@galadriel.bt.co.uk> pcf@galadriel.bt.co.uk (Pete French) writes:
>No - they aint CCDS. CCDs are designer for use as a camera - and you cant
>access each bit individualy like you can with RAM, you have to shift them out
>serially. In low lighting conditions you can get bits lost in the shifting
>which explains the smears accross the picture ou get from CCDs in bad light.

Incidentally, CCD's are analog devices, each charge bucket is dumped into its
neighbor every time you hit the shift line.  There must be a slight 
degradation every time you kick the buckets, so this is probably where the
smearing comes from.  (I've never seen one, but I do remember an article
on a CCD reverb unit from old an Popular Electronics that explained the
CCD device used there).

Anybody know how they shut off the light sensitivity of the charge
buckets while the image is being shifted out?  Don't these work by the
impinging photons causing a leakaway of stored charge?  I guess they could
depend on the fact that each scanline of the device is 'exposed' for a long
time, but only read/exposed for a short time so the error would be small.

If this was the case then having the sun impinging on the drain side of
the CCD could wipe out the whole line, even if the rest of the scene
was properly exposable -- the sun being bright enough to wipe out the
charge immediately (or at least degrade it significantly).  This true?

+----------------+
! II      CCCCCC !  Jim Cathey
! II  SSSSCC     !  ISC-Bunker Ramo
! II      CC     !  TAF-C8;  Spokane, WA  99220
! IISSSS  CC     !  UUCP: uunet!iscuva!jimc (jimc@iscuva.iscs.com)
! II      CCCCCC !  (509) 927-5757
+----------------+
			"With excitement like this, who is needing enemas?"

deanr@sco.COM (Dean Reece) (10/06/89)

In article <3254@mit-caf.MIT.EDU> fritz@mit-caf.UUCP (Frederick Herrmann) writes:
>In article <57732@psuecl.bitnet> peg@psuecl.bitnet writes:
>>>First of all, take a 4kX1 bit DRAM and somehow expose the silicon. \
>>>To use the thing as a camera, just charge up the thing to all 1's, \
>>>wait for a bit, and then check which bits are discharged ...
>>
>>Well, this actually works, in a manner of speaking.  The devices are called
>>CCD's, charged coupled devices, and are used for picture tubes in most
>
>Let's be careful with terminology here.  You don't need CCDs to build
>an IC imager, as the `RAMera' demonstrates.  A CCD is a kind of ...

Well, theres been plenty of talk about the RAMera idea, but no references, so
here goes:
----------------------------------------------------
Microprocessor Based Robotics -by- Mark J. Robillard
ISBN=0-672-22050-4   LCN=83-60160   (Copyright 1983)
----------------------------------------------------
The book _Microprocessor_Based_Robotics_ has a chapter called "Vision Systems"
which gives a good discussion about building a RAMera (though not by that name)
It is NOT a step by step building guide, but an idea and tutorial book, so
there is still plenty-o-engineering to do.

The discussion explains how to remove the metal lid and replace it with a
clear cover (Microscope slide covers work well, and they are of optical quality
too!).  The electronics and timing are almost completely absent from the
discussion, but it is a start.

The book uses 16k dynamic RAM in the CerDIP (Ceramic DIP package) and pops the
top off by heating it with a soldering iron.  I am working on building such a
beastie, but I don't want to post just yet (until it works).  Oh, you could use
64k or 256 k DRAMS, except that they are hard to find in the CerDIP package.

 ______________________________________________________________________
| Dean Reece     Member Technical Staff |"The flames are all long gone |
| The Santa Cruz Operation 408/458-1422 | but the pain lingers on"     |
|___________deanr@sco.com_______________|___________________Pink_Floyd_|

bergman@m2c.m2c.org (Michael Bergman) (10/11/89)

It occurs to me that it might be of interest to those as are following
this thread that I believe that the Fisher-Price "toy" video
camcorder uses this type of technology.  Regardless, I think the
original question was "what's a cheap way to get video signals for a
project?" and the Fisher Price camcorder (alone) is around $80. Its
awful grainy though...a real video camera (just a camera) is only
~$130, so this may not be too useful.

--
--mike bergman
	      Massachusetts Microelectronics Center
	      75 North Drive, Westborough, MA  01581, USA +1 (508) 870-0312
	UUCP: harvard!m2c!bergman    INTERNET:   bergman@m2c.org         

sparks@corpane.UUCP (John Sparks) (10/11/89)

<3863@blake.acs.washington.edu> <57950@psuecl.bitnet>
<BERGMAN.89Oct10164617@odin.m2c.m2c.org>
Sender: 
Reply-To: sparks@corpane.UUCP (John Sparks)
Followup-To: 
Distribution: sci.electronics
Organization: Corpane Industries, Inc.
Keywords: 

In article <BERGMAN.89Oct10164617@odin.m2c.m2c.org> bergman@m2c.m2c.org
(Michael Bergman) writes:
 >It occurs to me that it might be of interest to those as are following
 >this thread that I believe that the Fisher-Price "toy" video
 >camcorder uses this type of technology. Its
 >awful grainy though...a real video camera (just a camera) is only
 >~$130, so this may not be too useful.

The biggest problem with the FP camera is that it has a cheapo plastic lens and
uses Audio cassettes to record on.




-- 
John Sparks   |  {rutgers|uunet}!ukma!corpane!sparks | D.I.S.K. 24hrs 1200bps
|||||||||||||||          sparks@corpane.UUCP         | 502/968-5401 thru -5406 
Although the moon is smaller than the earth, it is farther away.