from the image
The HyperVCam Mobile has its Near Infra-red filter built into the lens
assembly which means that it cannot be removed and used in the modified
version for a microscope. I believe the QuickCam has (or at least used
to have) a separate IR
filter that could be removed and reused. If you are only capturing
grey-scale images then the lack of IR filter may
even be beneficial because the camera will be more sensitive in low
For colour images, however, the near-IR light is picked up by all the
colour sensitive elements and ends up swamping all colour information
(as the elements are very sensitive to IR). There are two solutions
- Fit a near-IR filter - either a reflector
or absorber. These
filters have sharp cut-off filter responses and that seems to make them
rather expensive, i.e. in the region of £100+ for a suitable
filter. For my purposes it looked like canabilising a colour Quickcam
was going to be the cheapest solution.
- Use a non-IR emitting light source - I
prototyped a white
LED (on the end of a straw) in place of the bulb and this gave
promising results. Cost of the solution? Roughly £5
As I was doing this on a tight budget I opted for the white LED
solution. Using a near-IR filter is the preferable solution as the bulb
can be a lot brighter.
White LED illumination
The bulb filament requires placement
with millimeter accuracy so I decided to stick with as much of the
existing holder system to hold the LED. I also wanted to re-use the AC
transformer built into the microscope to avoid excessive wires and
power supplies. The solution I went for was to I build a White LED
The original assembly
My microscope has a Kohler Illuminator system with a bulb that has, I
assume, a proprietary fitting (see image below). The bulb fits into the
spring-loaded holder and the clip holds it in place. A ring around the
bulb fitting fits flush in the holder.
I needed a suitable holder for the LED like the original. I
couldn't find any bulbs with the right connector so made one from a
standard bayonet nightlight bulb and a 15mm tap washer. I removed the
top off the bulb (by filing around it) and pulled out the filament.
Here are the bits and the assembled holder. It does not fit quite as
snuggly as the original bulb but does hold pretty firmly in place.
The circuit I''ve used is the simplest
AC-DC converter possible.
The signal is smoothed as any stobing of the LED may cause banding on
the captured images. The circuit uses a 470microF capacitor, a 330Ohm
resistor, a 5mm white LED and a 50V bridge rectifier. Total cost less
than a fiver. The circuit diagram is shown below.
next job's to get the circuit built and fitted into a bulb... this bit
got a bit fiddly. Fiddly bits? Getting the circuit small enough and
bent enough to fit in around the glass stem in the middle. Getting the
circuit soldered to the filament wires (they're coated with
anti-soldering stuff) and removing the epoxy I put on to hold the
washer in place (it wasn't needed and interfered with its housing in
the holder). Getting the LED seated over the center and getting epoxy
to hold it in place.
I fitted the bulb and pushed the LED around gently before the epoxy set
to get the LED perfectly aligned. Completed, one Infra-red-free light
source for the microscope. The light is reasonably bright but cannot be
dimmed by the variable Voltage supply on the transformer. The
transformer van be varied to output between ~6.5V and 8V so, if I was
to make another bulb, I'd place a tiny variable resistor in too to
allow the Voltage over the LED to be taken down to a point where
varying the transformer Voltage would have a significant effect on its