Re: IR detector system, biasing of photo diode

On Mon, 28 Oct 2024 11:08:50 +0100, Klaus Vestergaard Kragelund
<klauskvik@hotmail.com> wrote:

On 27-10-2024 21:39, Joe Gwinn wrote:

On Sun, 27 Oct 2024 10:49:27 -0700, john larkin <JL@gct.com> wrote:
 

On Sun, 27 Oct 2024 13:42:27 +0100, Klaus Vestergaard Kragelund
<klauskvik@hotmail.com> wrote:
>

On 27-10-2024 13:40, Klaus Vestergaard Kragelund wrote:

On 27-10-2024 13:26, Klaus Vestergaard Kragelund wrote:

On 27-10-2024 03:26, john larkin wrote:

On Sun, 27 Oct 2024 02:19:14 +0200, Klaus Vestergaard Kragelund
<klauskvik@hotmail.com> wrote:
>

Hi
>
I am working on an IR detector that will guide a robot into a docking
station.
>
A IR transmitter on the docking station transmits a beam, and 2 IR
detectors on the robot detects the beam and lets the robot navigate
towards the target. The working distance is a couple of meters.
>
I need it to be insensitive to ambient light/sunlight.
>
The IR detectors are placed in a tube, to narrow in the beam angle and
to avoid sunlight (since it is seldom the sun is actually that low in
the horizon)
>
The IR transmitter will be modulated with 10kHz (TBD) frequency, low
duty cycle. Low duty cycle to be able to drive the LED with high
current, frequency modulated so that the receiver can ignore the effect
of daylight (DC)
>
If the LED on the docking station has higher radiant intensity at the
point of the robot (2 meters away) than possible IR from sunlight, then
that would be perfect.
>
Example of transmitter:
>
https://www.vishay.com/docs/83398/vsmy2850.pdf
>
Has up to 1000mW/sr. Seems my basic calculation for a 15 degree beam,
shows less than 10nW/m2, while sunlight has 1W/m2. So driving a beam
that has higher output than sunlight seems unlikely.
>
I would use a IR phototransistor at 850nm, something like this:
>
https://www.ttelectronics.com/TTElectronics/media/ProductFiles/
Datasheet/OP505-506-535-705.pdf
>
Or a photo diode:
>
https://docs.rs-online.com/9f58/0900766b816d8a09.pdf
>
Fed from reverse 3.3V and into a transimpedance amplifier to boost the
signal with bandpass filter.
>
One can get digital IR detector used in a remote control systems:
>
https://www.vishay.com/docs/82491/tsop382.pdf
>
It has AGC, but digital output. I need analog output to be able to zero
in on the transmitter beam.
>
I have been looking for IR detectors that has the analog output, not
just the digital, but have not found any.
>
If the photodiode detector is subjected to sunlight, I am guessing I
would need very high gain on the 10kHz modulation frequency to pick up
the burried signal in the DC from sunlight.
>
How do I best bias the photo diode for optimum detection of the 10kHz
signal while being immune to the ambient sunlight?
>
I have chosen 850nm which seems to be a good wavelength. The
spectrum at
sea level has some dips due to water absorption.
>
https://sciencetech-inc.com/web/image/49169/
Spectrum%20with_out%20absorption.png
>
Seems like 750nm would be better, since then the IR from the sun is
lower, but does reduced the effective range of the system during
fog/rain. Probably that's why these system do not use 750nm
>
Other considerations?

>
You could drive the LED with a square wave, 10 KHz or whatever. The
photodiode could have +DC on one end and the other end can hit a
parallel LC to ground, resonant at 10K.
>
That takes out the sunlight DC component and adds bandpass filtering.
>

>
That's a very nice idea. The Q should not matter much, just as long as
DC is removed.
>
The photodiode will still be subjected to the high ambient light, but
the gain would be close to zero for the stage after. I would then
still need to be sure the photodiode is never saturated by ambient light.

>
A photodiode won't saturate as long as it has a few volts of DC across
it. It might melt if there's no current limiting.
>

>

Actually, wont a simple high pass filter work equally well?
>
Photo diode with bias -> capacitor to gain block....

>
Like this:
>
https://electronics.stackexchange.com/questions/416184/how-does-this-op-amp-photodiode-circuit-behave

>
The LC tank combines background light rejection and bandpass filtering
and has high signal gain, with two parts.
>
I think there are photodiodes with colored plastic, essentially a
cheap optical bandpass filter. Used in TV remote receivers.
>
The windows in TVs may be optical bandpass filters too. They work with
very little signal from the remote, in high room light.

 
To this I would add a trick.  We know something very useful about the
10 KHz modulation, its exact frequency, given that it is (or can be)
generated electronically, and thus its frequency is ultimately
controlled by a logic-clock crystal oscillator.
 
So feed the amplified signal from the 10 KHz LC tank to a I+Q homodyne
circuit, filter to pass signals from DC to a 10 Hz and compute the
magnitude of the received signal - this is used for figuring out the
direction to the docking station.

>
Very nice, sort of like for lock-in detection, I have done that before,
works great. Would take the tolerances out of the passive components.

Yes, it is a form of lock-in detection, adapted to the lack of phase
data, other than to know that it changes slowly.

But there is another issue, interference from artificial light, such
as streetlamps, at dawn and dusk.  Many streetlamps are strongly
modulated at a harmonic of the local prime power frequency, 50 Hz or
60 Hz.  Also, the receiver itself will have a noise floor that is
essentially constant random noise.

A classic remedy is correlated double sampling (CDS), where the sample
period is exactly one tenth of a second in duration, into which an
integral number of power cycles will fit exactly, and so will
integrate to zero.

Originally, CDS modulated the transmitter to alternate between on and
off, so external light would fall in both odd and even numbered
samples, while real data would only fall in one set of samples.  Here
the TX runs continuously, and it would be a nuisance to synchronize
it, so we need a way to do the same in the RX units.

The simplest approach would be to receive at two modulation
frequencies, 10 KHz and say 11 KHz, alternating between them.  Only
the 10 KHz will have data, while both frequencies will have system
noise and power-frequency modulated stray light.

There are two identical integrators and a switch.  The odd-numbered
cycles go into the first integrator and the even-numbered cycles go to
the second integrator.  These integrator are leaky.  One reports their
difference as the 10 KHz modulation strength.

There are many schemes like this that could work.

Joe Gwinn