Sony DSC-V3 for daytime infrared

[Chuck Olson]

Normally this would go in the rec.photo.digital newsgroup, but I need the
expertise of practicing electronics engineers and technicians who are
familiar with the IC resources of current industry.

There are a number of modifications done commercially to cameras for
operation in the infrared end of silicon photo sensor response. Mostly, the
modifications are semi-permanent - - "semi" because for another $350
expenditure, the modification can be reverted back to normal. However, Sony
has produced cameras that have "Nightshot" capability, which mechanically
rotates away the hot-mirror filter and permits full-spectrum operation of
the sensor, and of course for normal use it rotates back into position. For
reasons of their own, Sony has severely limited the aperture and shutter
operation in Nightshot mode so that it cannot be used as a daytime infrared
camera even if externally filtered for IR only. The hard part that cost so
much money to convert most cameras to IR and back again is already done in
these Nightshot cameras, so we just have to find out how to actuate the
filter mechanism for temporary IR operation while the camera is in an
unrestricted mode.

In
http://geektechnique.org/projectlab/254/how-to-turn-a-digital-camera-into-an-ir-camera, a
comment by macona in response #84 describes how he modified his DSC-V3, "I
found that the solenoid is controlled with a little surface mount (SO-8)
H-bridge chip. I installed a tiny switch that sends a pulse to the chip
which manually moves the filter out of the way. Just cycle through the
nightshot modes to return to normal." This is wonderful - - a modification
that might only cost $100 if done commercially, and one that's very likely
within the capability of most electronic techs - - a new way to make money
providing a service. If only macona had given a little more detail on how
far to disassemble the V3 and what chip and contact to connect to his
switch, a new industry might have blossomed.

Now we need to know what that SO-8 chip might be, and what it takes to drive
the filter servo to the "mirror out" end of its operation range. It has to
be capable of operating with the 3.6V NP-FR1 lithium battery. Could the
servo be entirely self-contained in the SO-8 package? Are there unused
inputs that could be energized with a momentary switch closure or capacitive
pulse? Come on, all you genius electronics guys - - who makes the chip, and
where can we get the data sheets on it? Who wants to dig into the
modification and publish the procedure with pictures (and become famous)?

Chuck

 

[Spehro Pefhany]

Normally this would go in the rec.photo.digital newsgroup, but I need the
expertise of practicing electronics engineers and technicians who are
familiar with the IC resources of current industry.

There are a number of modifications done commercially to cameras for
operation in the infrared end of silicon photo sensor response. Mostly, the
modifications are semi-permanent - - "semi" because for another $350
expenditure, the modification can be reverted back to normal. However, Sony
has produced cameras that have "Nightshot" capability, which mechanically
rotates away the hot-mirror filter and permits full-spectrum operation of
the sensor, and of course for normal use it rotates back into position. For
reasons of their own, Sony has severely limited the aperture and shutter
operation in Nightshot mode so that it cannot be used as a daytime infrared
camera even if externally filtered for IR only. The hard part that cost so
much money to convert most cameras to IR and back again is already done in
these Nightshot cameras, so we just have to find out how to actuate the
filter mechanism for temporary IR operation while the camera is in an
unrestricted mode.

In
http://geektechnique.org/projectlab/254/how-to-turn-a-digital-camera-into-an-ir-camera, a
comment by macona in response #84 describes how he modified his DSC-V3, "I
found that the solenoid is controlled with a little surface mount (SO-8)
H-bridge chip. I installed a tiny switch that sends a pulse to the chip
which manually moves the filter out of the way. Just cycle through the
nightshot modes to return to normal." This is wonderful - - a modification
that might only cost $100 if done commercially, and one that's very likely
within the capability of most electronic techs - - a new way to make money
providing a service. If only macona had given a little more detail on how
far to disassemble the V3 and what chip and contact to connect to his
switch, a new industry might have blossomed.

Now we need to know what that SO-8 chip might be, and what it takes to drive
the filter servo to the "mirror out" end of its operation range. It has to
be capable of operating with the 3.6V NP-FR1 lithium battery. Could the
servo be entirely self-contained in the SO-8 package? Are there unused
inputs that could be energized with a momentary switch closure or capacitive
pulse? Come on, all you genius electronics guys - - who makes the chip, and
where can we get the data sheets on it? Who wants to dig into the
modification and publish the procedure with pictures (and become famous)?

Chuck

All this effort just to cop a peek through women's clothes? Why not
use the old-fashioned methods?

Best regards,
Spehro Pefhany

 

[Chuck Olson]

http://geektechnique.org/projectlab/254/how-to-turn-a-digital-camera-into-a n-ir-camera, a

All this effort just to cop a peek through women's clothes? Why not
use the old-fashioned methods?

Best regards,
Spehro Pefhany

http://www.speff.com

Nice, Spehro - - you and Sony think alike.

 

[[email protected]]

Normally this would go in the rec.photo.digital newsgroup, but I need the
expertise of practicing electronics engineers and technicians who are
familiar with the IC resources of current industry.

There are a number of modifications done commercially to cameras for
operation in the infrared end of silicon photo sensor response. Mostly, the
modifications are semi-permanent - - "semi" because for another $350
expenditure, the modification can be reverted back to normal. However, Sony
has produced cameras that have "Nightshot" capability, which mechanically
rotates away the hot-mirror filter and permits full-spectrum operation of
the sensor, and of course for normal use it rotates back into position. For
reasons of their own, Sony has severely limited the aperture and shutter
operation in Nightshot mode so that it cannot be used as a daytime infrared
camera even if externally filtered for IR only. The hard part that cost so
much money to convert most cameras to IR and back again is already done in
these Nightshot cameras, so we just have to find out how to actuate the
filter mechanism for temporary IR operation while the camera is in an
unrestricted mode.

Inhttp://geektechnique.org/projectlab/254/how-to-turn-a-digital-camera-..., a
comment by macona in response #84 describes how he modified his DSC-V3, "I
found that the solenoid is controlled with a little surface mount (SO-8)
H-bridge chip. I installed a tiny switch that sends a pulse to the chip
which manually moves the filter out of the way. Just cycle through the
nightshot modes to return to normal." This is wonderful - - a modification
that might only cost $100 if done commercially, and one that's very likely
within the capability of most electronic techs - - a new way to make money
providing a service. If only macona had given a little more detail on how
far to disassemble the V3 and what chip and contact to connect to his
switch, a new industry might have blossomed.

Now we need to know what that SO-8 chip might be, and what it takes to drive
the filter servo to the "mirror out" end of its operation range. It has to
be capable of operating with the 3.6V NP-FR1 lithium battery. Could the
servo be entirely self-contained in the SO-8 package? Are there unused
inputs that could be energized with a momentary switch closure or capacitive
pulse? Come on, all you genius electronics guys - - who makes the chip, and
where can we get the data sheets on it? Who wants to dig into the
modification and publish the procedure with pictures (and become famous)?

Chuck

Two optical solutions come to mind. One is to add an external IR
filter. You may find once you filter the light, you don't have an
aperture problem. The other is to use a neutral density filter. A
polarizer might have enough light loss. I suspect the IR filter is the
best solution.

While I haven't tried to photograph women through their clothing, i.e.
you intended application, I did buy an IR filter from
http://www.freestylephoto.biz/e_main.php
for use with IR film.

It looks like the smallest R72 they sell is 49mm, so you will probably
need a step up (or is it down) ring. These are quite cheap, but you
need to read the literature carefully to figure out which direction it
works.

 

[Chuck Olson]

Two optical solutions come to mind. One is to add an external IR
filter. You may find once you filter the light, you don't have an
aperture problem. The other is to use a neutral density filter. A
polarizer might have enough light loss. I suspect the IR filter is the
best solution.

While I haven't tried to photograph women through their clothing, i.e.
you intended application, I did buy an IR filter from
http://www.freestylephoto.biz/e_main.php
for use with IR film.

It looks like the smallest R72 they sell is 49mm, so you will probably
need a step up (or is it down) ring. These are quite cheap, but you
need to read the literature carefully to figure out which direction it
works.

Yes, that is the only way that avoids tearing into the electronics of the
camera. Typically they find an R72, plus an ND8 and an ND4 work to bring the
IR exposure down to a useful value despite the forced camera settings in
Nightshot. See http://www.pbase.com/nh/nightshot for various combinations of
filters and corresponding results. But that pretty well limits the camera to
tripod use. The improvement you get with the electronic hack is full control
for handheld use rather than being stuck with 1/30 @ f/2.8. With the
proposed mod, the camera can then be restored to normal color response just
by cycling the Nightshot button - - the best of both worlds, and worth going
to some trouble, I feel.

I understand the DSC-V3 requires an extension tube with a proprietary 47mm
thread at the camera end to hold the filter since the lens pokes out when
zooming. I've seen two versions of the adapter tube, for either 52 mm or 58
mm filters, and expect to use an existing 58mm R72 that I bought long ago
for my Nikon Coolpix 950. The winter white rendering of green sunlit foliage
by an IR camera is an extraordinary effect - - something every digital
camera ought to be equipped to do.

 

[Chuck Olson]

Fortunately, Sony publishes a Service Manual and in the Level 3 section,
most of the questions below are answered there. The H-Bridge driver is an
LB1938T-TLM-E and I found a full data sheet by Sanyo just doing a Google
search for LB1938T. The camera is powered by a 3.6V Li-Ion cell, but there's
a large DC-DC converter system in there that produces a variety of voltages,
one of which is +5V which runs this chip, among others.

One large question that remains is, what will happen if I drive the desired
input of this chip to +5 while it's connected to an output pin of the
microprocessor? Presumably both inputs are held low in "standby" between
programmed actuations of the filter motor, so driving one high is
potentially a contentious operation. Are microprocessor outputs typically
"3-state"? It they aren't, will the output driver in the microprocessor
survive this abuse? The H-Driver chip will see an input high if it can be
pulled up to +2V, so I would like to add a resistor in series with the
switch in the effort to be "kind" to the microprocessor, hoping it will live
a long, eventful life. If the outputs are not "3-state" then what typical
output resistance at "low" can I expect to encounter from a typical
microprocessor in this battery portable application?

Thanks,

Chuck

Normally this would go in the rec.photo.digital newsgroup, but I need the
expertise of practicing electronics engineers and technicians who are
familiar with the IC resources of current industry.

There are a number of modifications done commercially to cameras for
operation in the infrared end of silicon photo sensor response. Mostly, the
modifications are semi-permanent - - "semi" because for another $350
expenditure, the modification can be reverted back to normal. However, Sony
has produced cameras that have "Nightshot" capability, which mechanically
rotates away the hot-mirror filter and permits full-spectrum operation of
the sensor, and of course for normal use it rotates back into position. For
reasons of their own, Sony has severely limited the aperture and shutter
operation in Nightshot mode so that it cannot be used as a daytime infrared
camera even if externally filtered for IR only. The hard part that cost so
much money to convert most cameras to IR and back again is already done in
these Nightshot cameras, so we just have to find out how to actuate the
filter mechanism for temporary IR operation while the camera is in an
unrestricted mode.

In

http://geektechnique.org/projectlab/254/how-to-turn-a-digital-camera-into-an-ir-camera, a

 

[[email protected]]

Ok, I got my hands on a DSC-V3 that had a bad lens assembly. The
previous owner really tore into the lens assembly, which is now
useless. But it gave me the chance to look at the mechanics of the
"Nightshot" function. Reading Macona's claim and after reading the
article on modifying a DSC-717 at http://www.pbase.com/nh/image/74486953,
I wonder now how he is managing to focus the camera correctly since
the hot mirror stage assembly on the V3 is very similar to the 717's
http://www.pbase.com/nh/image/74527713 and it does state that without
it, the camera optical circuit will not compensate for the refractive
loss of the hot mirror while in normal modes, making focusing to
infinity impossible. It's placement is between the lens and the CCD,
just like the 717's.

At first, I thought it would be easiest to open the circuit with a
small switch ( based on the fact that I do not have the schematic and,
like you, I had no way to ID the SO-8 H-bridge chip or the
pinouts...lazy, I guess LOL). Activate the Nighshot mode, open the
switch on the solenoid's circuit, then cycle to the normal mode, which
would leave the hot mirror in the Nightshot position. This should
allow IR shooting in the full color modes and with full aperture/
shutter control. So I would have thought...until I read
http://www.pbase.com/nh/image/74486953 :
=====================================================================================
The original hot mirror (left) is a small piece of optical glass, size
11 x 13 x 1.2 mm.
I replaced it with a special optical window that transmits light from
+/- 300 to 2000 nm,
instead of just 400-700 nm for the standard Sony hot mirror. After
modification the F717
will record images with light from +/- 350 - 1100 nm (the sensor/glass
combination will
not work outside this range). This range includes most of the near-
infrared spectrum.

It is VERY important to replace the hot mirror with another optical
window with the same
dimensions (especially thickness, the tolerance is about 0.02 mm) and
refraction index,
otherwide the camera will no longer be able to focus (especially at
infinity /wideangle).
If you do not replace the original hot mirror with anything, the
camera will only focus
in Nightshot mode (this is because the optical window is part of the
optical system).

You can try to make your own window from a microscope slide or other
piece of optical
glass grinded to specifications, but it will require a lot of trial
and error because the
window must be perfectly flat. If it is too thick, it will probably
not fit in the stage
(maximum thickness is about 1.3 mm). Note: microscope slides have no
AR coating,
using uncoated glass will cause extra reflection / hotspot problems.
=====================================================================================

So technically, applying the correct voltage to actuate that solenoid
or my supposed open circuit solution will move the hot mirror out of
the way, but it will not compensate for focusing. RATS!
Macona makes only one mention on focusing on another post:

=====================================================================================
Two days ago I also modded my Sony DSC-V3. This camera has the built
in nightshot but it is limited. It only functions in standard and "P"
modes as well as movie mode. Also you can not use the flash in this
mode. I took apart the camera (much further than I had to.) and found
how the night shot system worked. It is simply that the IR filter is
on a solenoid. The camera moves the filter out of the way in night and
back in normal. Seeing as there is proabably no way to get into the
firmware there had to be a hardware hack. I found that the solenoid is
controlled with a little surface mount (SO-8) H-bridge chip. I
installed a tiny swithc that sends a pulse to the chip which manually
moves the filter out of the way. Just cycle throught the nightshot
modes to return to normal. Now I can use the camera as a normal color
digicam and full IR mode by placing IR pass filter over the lens and
flash. Or even better use the hotshoe or ACC plug to control an
external IR flash. Only problem I have had is that the camera has a
hard time focusing at full wide angle. Fix is just zoom in a bit.
=================================================================================================
I am more prone to trust in the Niek Haak article on the 717 than
Macona's zoom focusing fix. And most all IR mods I have seen replace
the hot mirror in one way or another.

If you are able to find a viable solution, I will commend you. But for
now, I can deal with the slow shutter speeds and the ND8, ND4, CPL
combo with a 950 nm filter.

In the mean time, I got a V3 I can experiment with ( possibly adapting
a manual focus SLR lens ) to see what I come up with!

Laters!