D-I-Y densitometer?
Malcolm Stewart
anyone made their own transmission densitometer? Or know of any links on the
web? I appreciate that I'll probably need calibration filters but I guess I
could borrow those from my local lab.
Thanks for any pointers
--
M Stewart
Milton Keynes, UK
Tom Phillips
Malcolm Stewart wrote:
> With "white" LEDs and a wide range of photo sensors now commonly available, has
> anyone made their own transmission densitometer? Or know of any links on the
> web? I appreciate that I'll probably need calibration filters but I guess I
> could borrow those from my local lab.
Phil Davis in Beyond the Zone System had recipes for ones made from garden variety
lightmeters and spotmeters.
Might give you some ideas.
john walton
(www.taosinc.com) programmable light to frequency counters. (I have these
for sale if you care at www.tech-diy.com/store.htm) . The reason I chose
this chip is that it is very linear and has about 10^5 range. For a
densitometer you need at least 10^3 range (which is at least a 10 bit analog
to digital converter.) If you go to the above webpage it links directly to
the PDF at TAOS' site. (I have RGB detectors too if anyone cares.)
I use 2 converters, one to measure and reference the output of the LED
emitter and one to measure the output through the media being analyzed.
Even though the power supplies are regulated, and don't consume a lot of
milliwatts, there is a tempco function (as with all PN junctions) which must
be compensated.
The output of the device is read by a PIC -- you can use a 16F84 or a Basic
Stamp II -- I measure the pulse duration and have an algorithm to convert
the output to logarithmic. The output of the TSL230 has to be smoothed a
bit and the device is IR sensitive, in a clear case, so some precautions
must be taken to preserve linearity.
Lots of ways to calibrate, the most elegant, of course, is to use the
inverse square law (if you have a point-source!). I have a big ol MacBeth,
however and the two devices compare remarkably well.
Jack
"Malcolm Stewart" <malcolm...@megalith.freeserve.co.uk> wrote in
message news:aqikna$kl5$1...@newsg4.svr.pol.co.uk...
Sherman
message news:aqikna$kl5$1...@newsg4.svr.pol.co.uk...
Malcom,
Do a websearch for "building a densitometer" and you will turn up a few
interesting things. However if all you are interested in is b&w the process
is very easy.
You can use any CdS Photocell (theses aren't linear in response but that
isn't necessary for this purpose).
Wire an LED (or other small light source) to a battery or transformer to
provide your light source. Remember for b&w it doesn't need to be color
calibrated or anything. Next use a multimeter across the terminals of a
photocell to read the change in output. Calibrate your "densitometer" using
a calibrated step tablet such as those available from Stouffers. Write down
each reading from your multimeter for each step in the tablet. Since your
power source isn't stabilized you may want to calibrate at the start of each
session.
For this to work you should have the photocell mounted in a tube of some
sort (I've purchased rubber gasket-type tubes for about twenty-five cents
US) and the light source should also be in a tube. Place your negative over
the light source tube and put the photocell tube down onto the negative.
Read the multimeter to determine the change.
Note that most photocells provide a relatively linear response up to a
density of about 1.6 or 1.7 or so but for b&w photographic purposes that is
about as high as you need. In the linear portion of the response it would
be possible to make a scale that reads directly in density and tape it to
your meter if it is a needle type. Otherwise just use your conversion
table.
I have made one of these and it has an accuracy of .1 or so up to a density
of 1.6 or so. With some electrical knowledge it would be possible to
improve on this. Also using a photodiode rather than a photocell will give
a very linear response for anything you can measure but requires a more
sophisticated circuit to provide black current and even things out.
However, these days it is possible to get good deals on densitometers on
eBay. I got a Speedmaster transmission and reflection densitometer for $65
US and $20 shipping.
HTH,
Sherman
http://www.dunnamphoto.com
john walton
high, or remain that high, since the luminosity of the LED is affected by
temperature. I just did a circuit (which is going to be published) which
uses two photo-receptors and a servo-amplifier to compensate for the changes
of luminosity with temperature. The linearity is 0.01%
I'll publish a schematic, as it's not that complex.
"Sherman" <sherman-r...@dunnam.net> wrote in message
news:LHaz9.2745$tW4.3...@newsread1.prod.itd.earthlink.net...
> "Malcolm Stewart" <malcolm...@megalith.freeserve.co.uk> wrote in
> message news:aqikna$kl5$1...@newsg4.svr.pol.co.uk...
> > With "white" LEDs and a wide range of photo sensors now commonly
>
Dan Quinn
"With "white" LEDs..." I think any color will do as long as a
photo diode match can be made. That would be for B&W.
Very good deals can be gotten from eBay. Search this group for
-tobias-. Search by date. Dan
D Poinsett
current mode rather than constant voltage. In constant current mode the
forward voltage drop that changes 2mV/deg C will be compensated by the drive
circuit and illumination intensity will be constant (at least over the time
period we are talking about). Constant voltage drive, on the other hand,
WILL result in LED intensity changes with temperature.
Since B&W densitometry deals exclusively with ratios of light, "calibrating"
the light sensor before each reading with unobstructed illumination from the
LED will remove any short term temperature sensitivity from many parts of
the circuit and provide the baseline reading for determining subsequent
ratios when some portion of the negative obstructs the light path.. This is
a common practice with all commercial densitometers.
The suggestions to use photodiodes for high linearity over an extremely wide
range of illumination is excellent.
Dave
"john walton" <jdwa...@comcast.net> wrote in message
news:58Wdnflh0qE...@comcast.com...
> Even if you stabilize the power supply I doubt that the acuracy will be
that
> high, or remain that high, since the luminosity of the LED is affected by
> temperature. I just did a circuit (which is going to be published) which
> uses two photo-receptors and a servo-amplifier to compensate for the
changes
> of luminosity with temperature. The linearity is 0.01%
>
> I'll publish a schematic, as it's not that complex.
>
> snip...
Sherman
that
> high, or remain that high, since the luminosity of the LED is affected by
> temperature. I just did a circuit (which is going to be published) which
> uses two photo-receptors and a servo-amplifier to compensate for the
changes
> of luminosity with temperature. The linearity is 0.01%
>
> I'll publish a schematic, as it's not that complex.
>
I'd like to see that circuit! I own a Speedmaster densitometer so I don't
really need to build one but I enjoy seeing how things work. As for the LED
light output changes, I think doing a calibration at the start of each
session after turning on the LED for a minute or two would make it accurate
enough for most b&w film testing.
By the way if you know of any "simple" circuits for a temperature controller
I would be interested in that as well! I made a very simple tempering bath
but since it uses the thermostat from an aquarium thermometer the max temp
is only 93 degrees, insufficient for E6.
Sherman
http://www.dunnamphoto.com
john walton
my Durst Printo -- the probes are waterproof.
Search the websites of National Semi (www.nsc.com) Analog Devices
(www.analog.com)
Sherman
I have tried some digital thermometers for monitoring temps. My first one
worked very well but the second one was off by three degrees. What I'm
really after is a temperature contoller that will automatically turn the
heater on and off and maintain temperature to about 1 degree F or less.
I did find a relatively simple circuit online that uses a Radio Shack
thermister and a four channel IC comparator. I think I'll purchase the
parts today and give it a go! If it works I'll put it on my website.
Sherman
http://www.dunnamphoto.com
Malcolm Stewart
> "john walton" <jdwa...@comcast.net> wrote in message
> news:b8ScnfnGgIM...@comcast.com...
> > Temperature Control -- I use a couple of Radio Shack outdoor thermometers
>
> I have tried some digital thermometers for monitoring temps. My first one
> worked very well but the second one was off by three degrees. What I'm
> really after is a temperature contoller that will automatically turn the
> heater on and off and maintain temperature to about 1 degree F or less.
>
> I did find a relatively simple circuit online that uses a Radio Shack
> thermister and a four channel IC comparator. I think I'll purchase the
> parts today and give it a go! If it works I'll put it on my website.
>
> Sherman
> http://www.dunnamphoto.com
My Jobo DuoLab maintains accurate temperature by heating the water bath
generally, and also by (I think) controlling the temperature of a much smaller
quantity of water heated to a much higher temperature, and in thermal contact
with the larger water bath. I guess this means that the differential of the
higher temp thermostat is effectively reduced thus improving the accuracy. In
practice, when it's at chosen temperatrure the actual differential in the water
bath seems better than 0.1C.
Steven Chambers
switched output for controlling a separate heater. The temperature
controller 7027/2 can be found on www.tunze.com.
Steve Chambers
"Sherman" <sherman-r...@dunnam.net> wrote in message
news:f2Qz9.5914$tW4.7...@newsread1.prod.itd.earthlink.net...
Sherman
> Sherman <sherman-r...@dunnam.net> wrote in message
> news:f2Qz9.5914$tW4.7...@newsread1.prod.itd.earthlink.net...
> > "john walton" <jdwa...@comcast.net> wrote in message
> > news:b8ScnfnGgIM...@comcast.com...
> > > Temperature Control -- I use a couple of Radio Shack outdoor
thermometers
> snip
>
> generally, and also by (I think) controlling the temperature of a much
smaller
> quantity of water heated to a much higher temperature, and in thermal
contact
> with the larger water bath. I guess this means that the differential of
the
> higher temp thermostat is effectively reduced thus improving the accuracy.
In
> practice, when it's at chosen temperatrure the actual differential in the
water
> bath seems better than 0.1C.
> --
> M Stewart
> Milton Keynes, UK
>
Thanks for your input. The idea of heating a smaller quantity of water to a
higher temperature and using that to heat the main tank is interesting. It
seems that it might supply more even heat to the main tank depending upon
the area of contact. I'll have to think about that setup.
One of these days I would like to get a Jobo. They are expensive but based
on my searches for affordable temperature controllers and the accuracy that
the Jobos achieve I think a lot of their cost is in the heating circuit. My
current setup for b&w uses a plastic tray with a snap on lid in which I cut
holes for my chemical bottles. I use a submersible aquarium heater to heat
the water to 75 degrees F. Accuracy wasn't very good until I added a small
submersible pump to circulate the water. Now it maintains 75 within one
degree. Not too bad for $30 worth of parts.
Thanks again,
Sherman
http://www.dunnamphoto.com
David Swinnard
> snip...
> I just did a circuit (which is going to be published) which
> uses two photo-receptors and a servo-amplifier to compensate for the
> changes
> of luminosity with temperature. The linearity is 0.01%
>
> I'll publish a schematic, as it's not that complex.
...snip
I've been toying with the DIY densitometer, even got the Taos bits
(samples they were so kind to privide) but haven't found the time to put
anything together yet. I would very much like to see you schematic and
hear about your experiences with the unit when you publish it.
Dave
Ken Myrtle
>
> I have one built with Texas Advanced Optical (TAOS) TSL230
> (www.taosinc.com) programmable light to frequency counters. (I have these
> for sale if you care at www.tech-diy.com/store.htm) . The reason I chose
> this chip is that it is very linear and has about 10^5 range. For a
> densitometer you need at least 10^3 range (which is at least a 10 bit analog
> to digital converter.) If you go to the above webpage it links directly to
> the PDF at TAOS' site. (I have RGB detectors too if anyone cares.)
>
> I use 2 converters, one to measure and reference the output of the LED
> emitter and one to measure the output through the media being analyzed.
> Even though the power supplies are regulated, and don't consume a lot of
> milliwatts, there is a tempco function (as with all PN junctions) which must
> be compensated.
>
> The output of the device is read by a PIC -- you can use a 16F84 or a Basic
> Stamp II -- I measure the pulse duration and have an algorithm to convert
> the output to logarithmic. The output of the TSL230 has to be smoothed a
> bit and the device is IR sensitive, in a clear case, so some precautions
> must be taken to preserve linearity.
>
> Lots of ways to calibrate, the most elegant, of course, is to use the
> inverse square law (if you have a point-source!). I have a big ol MacBeth,
> however and the two devices compare remarkably well.
>
> Jack
Looks like a nice sensor. I think you could build a nice
densitometer with just the light to frequency chip,
a frequency reading multimeter like a radioshack
Cat.#: 22-811 which can read frequencies up to 4 Mhz and
an excel spreadsheet to convert intensity to density.
You could just read the source intensity before and after
the density reading to take care of drifts in the source output.
Ken
Markus Keinath
>With "white" LEDs and a wide range of photo sensors now commonly available, has
>anyone made their own transmission densitometer? Or know of any links on the
>web? I appreciate that I'll probably need calibration filters but I guess I
>could borrow those from my local lab.
I suppose you talk about a b/w densitometer?
Then the LED can be ok. As far, as your negtives are not coloured (XP
-2 for instance) and you won´t compare your results with other
densitometers.
I thought about buidling a colour densitometer. I could need it for my
diploma-work. My problem was the measurement of high densities above
4. I get much problems and searched for an other way (and now I am
lucky without a DIY Densitometer).
Markus
--
Keinaths Photohomepage:
Modification, Repair and DIY of Photographic Equipment:
www.keinaths-fotohomepage.gmxhome.de