Photodiode amplifier

[Ray Xu]

Hi, I am having trouble making the receiver part for my laser range
finder, but I do
have a head start (FYI, I am not an analog person). Currently, on my
board,
I have a PNZ334 photodiode connected to a current to voltage converter
(using a op-amp; feedback is 100K), and connected to 2 stages of
amplification (I THINK that they are 15x and 100x). The problem is in
the
current to voltage converter; it is not sensitive enough to even pick
up the
slightest refracted laser beam from a convex lens in a dark room. I
also
did a separate test with the photodiode alone, and it was very
sensitive to
the small laser refracted beams. I am using photovoltaic mode and not
photoconductive, because I tried it before, it is too hard for me
(unless
someone can tell me how to build it). Can anyone please tell me how to
fix
this, or better yet, how to design a better current to voltage
converter?
Thanks.

 

[whit3rd]

On Aug 16, 11:12 am, Ray Xu <ra...@tx.rr.com> wrote:

Hi, I am having trouble making the receiver part for my laser range
finder, but I do
have a head start (FYI, I am not an analog person). Currently, on my
board,
I have a PNZ334 photodiode connected to a current to voltage converter
(using a op-amp; feedback is 100K), and connected to 2 stages of
amplification (I THINK that they are 15x and 100x). The problem is in
the
current to voltage converter; it is not sensitive enough

Is the amplifier saturating?

What purpose does the current/voltage converter serve?
Wouldn't V proportional to logarithm-of-current
output be more useful? Alternately, if you're going to
mix down a modulated signal, can you just steer the
photodiode current with a Gilbert cell mixer?

 

[Ray Xu]

On Aug 16, 8:50 pm, whit3rd <whit...@gmail.com> wrote:

On Aug 16, 11:12 am, Ray Xu <ra...@tx.rr.com> wrote:

Hi, I am having trouble making the receiver part for my laser range
finder, but I do
have a head start (FYI, I am not an analog person). Currently, on my
board,
I have a PNZ334 photodiode connected to a current to voltage converter
(using a op-amp; feedback is 100K), and connected to 2 stages of
amplification (I THINK that they are 15x and 100x). The problem is in
the
current to voltage converter; it is not sensitive enough

Is the amplifier saturating?

What purpose does the current/voltage converter serve?
Wouldn't V proportional to logarithm-of-current
output be more useful?  Alternately, if you're going to
mix down a modulated signal, can you just steer the
photodiode current with a Gilbert cell mixer?

Well I was working with the Seattle Robotics Society on this too, and
the person said I am driving it into saturation. He never said on how
to fix it. Because the photodiode has high impedance, it'd be hard to
interface it directly to the stages of amplification. I've never
heard nor I have any ideas on "logarithm-of-current" (Could someone
please tell me what it is?).

And one more thing, I am a middle school student working on my
project.

 

[JosephKK]

On Sat, 16 Aug 2008 11:12:22 -0700 (PDT), Ray Xu <rayxu@tx.rr.com>
wrote:

Hi, I am having trouble making the receiver part for my laser range
finder, but I do
have a head start (FYI, I am not an analog person). Currently, on my
board,
I have a PNZ334 photodiode connected to a current to voltage converter
(using a op-amp; feedback is 100K), and connected to 2 stages of
amplification (I THINK that they are 15x and 100x). The problem is in
the
current to voltage converter; it is not sensitive enough to even pick
up the
slightest refracted laser beam from a convex lens in a dark room. I
also
did a separate test with the photodiode alone, and it was very
sensitive to
the small laser refracted beams. I am using photovoltaic mode and not
photoconductive, because I tried it before, it is too hard for me
(unless
someone can tell me how to build it). Can anyone please tell me how to
fix
this, or better yet, how to design a better current to voltage
converter?
Thanks.

From this post alone i already expect that your problem area is optics
not electronics. Try working on the optics please.

 

[JosephKK]

On Sun, 17 Aug 2008 05:50:14 -0700 (PDT), Ray Xu <rayxu@tx.rr.com>
wrote:

On Aug 16, 8:50 pm, whit3rd <whit...@gmail.com> wrote:
On Aug 16, 11:12 am, Ray Xu <ra...@tx.rr.com> wrote:

Hi, I am having trouble making the receiver part for my laser range
finder, but I do
have a head start (FYI, I am not an analog person). Currently, on my
board,
I have a PNZ334 photodiode connected to a current to voltage converter
(using a op-amp; feedback is 100K), and connected to 2 stages of
amplification (I THINK that they are 15x and 100x). The problem is in
the
current to voltage converter; it is not sensitive enough

Is the amplifier saturating?

What purpose does the current/voltage converter serve?
Wouldn't V proportional to logarithm-of-current
output be more useful?  Alternately, if you're going to
mix down a modulated signal, can you just steer the
photodiode current with a Gilbert cell mixer?

Well I was working with the Seattle Robotics Society on this too, and
the person said I am driving it into saturation. He never said on how
to fix it. Because the photodiode has high impedance, it'd be hard to
interface it directly to the stages of amplification. I've never
heard nor I have any ideas on "logarithm-of-current" (Could someone
please tell me what it is?).

And one more thing, I am a middle school student working on my
project.

The basic diode has a log transfer function, and the transfer function
is voltage as log of current. Make of it what you will.

 

[Jim Thompson]

On Sun, 17 Aug 2008 11:50:21 -0700, JosephKK <quiettechblue@yahoo.com>
wrote:

On Sat, 16 Aug 2008 11:12:22 -0700 (PDT), Ray Xu <rayxu@tx.rr.com
wrote:

Hi, I am having trouble making the receiver part for my laser range
finder, but I do
have a head start (FYI, I am not an analog person). Currently, on my
board,
I have a PNZ334 photodiode connected to a current to voltage converter
(using a op-amp; feedback is 100K), and connected to 2 stages of
amplification (I THINK that they are 15x and 100x). The problem is in
the
current to voltage converter; it is not sensitive enough to even pick
up the
slightest refracted laser beam from a convex lens in a dark room. I
also
did a separate test with the photodiode alone, and it was very
sensitive to
the small laser refracted beams. I am using photovoltaic mode and not
photoconductive, because I tried it before, it is too hard for me
(unless
someone can tell me how to build it). Can anyone please tell me how to
fix
this, or better yet, how to design a better current to voltage
converter?
Thanks.

From this post alone i already expect that your problem area is optics
not electronics. Try working on the optics please.

Sounds like something to pass on to S.E.K ;-)

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

My simulation, struggling to converge, is telling me something
Maybe my idea of a ring of current mirrors is flawed ?

 

[Ray Xu]

On Aug 17, 12:07 pm, Paul Mathews <opto.nospam.at.whidbey.com> wrote:

Ray Xu <ra...@tx.rr.com> wrote in news:ef831cf3-db5f-413c-9473-
32ac762f8...@z66g2000hsc.googlegroups.com:



Hi, I am having trouble making the receiver part for my laser range
finder, but I do
have a head start (FYI, I am not an analog person). Currently, on my
board,
I have a PNZ334 photodiode connected to a current to voltage converter
(using a op-amp; feedback is 100K), and connected to 2 stages of
amplification (I THINK that they are 15x and 100x). The problem is in
the
current to voltage converter; it is not sensitive enough to even pick
up the
slightest refracted laser beam from a convex lens in a dark room. I
also
did a separate test with the photodiode alone, and it was very
sensitive to
the small laser refracted beams. I am using photovoltaic mode and not
photoconductive, because I tried it before, it is too hard for me
(unless
someone can tell me how to build it). Can anyone please tell me how to
fix
this, or better yet, how to design a better current to voltage
converter?
Thanks.

Think of the photodiode as a 'light to current' transducer. You probably
want a 'light to voltage' transducer, so you need to feed the current to
a resistance or, alternatively, to a 'transimpedance' amplifier. Of
course, like most current sources, photodiodes have their limitations.

The first limitation to overcome is the fact that photodiodes have a
semiconductor junction in parallel with the current source. If you allow
much voltage to develop across the junction in its forward direction,
most of the photocurrent will shunted through the junction rather than
into your amplifier. If you maintain a reverse voltage across the
photodiode, you're in 'photoconductive' mode (which also has the benefit
of reducing photodiode capacitance, more later). However, reverse
voltage also gives rise to leakage currents, which can be a problem in
some systems.

So, another way to keep the junction from conducting is to keep the
voltage across the photodiode low, preferably near zero. For obscure
reasons, you are then said to be operating in 'photovoltaic' mode. The
classic negative feedback 'transimpedance amplifier' is often employed
for that purpose, and I'm guessing that's your circuit. The circuit can
be as simple as a photodiode, an opamp, and a feedback resistor (plus
power supply, bypass capacitors). This circuit is 'driven into
saturation' (as some people would say) when the output voltage of the
opamp reaches its maximum or minimum level permitted by its internal
design and the supply voltages. Since the output voltage of a simple
transimpedance amplifier is given by V = Rf * I, where Rf is the
feedback resistance and I is the input current, you can 'bring a
transimpedance amp out of saturation' by reducing the feedback
resistance. In your case, this might involve reducing the 100K resistor
to 10K.

However, I've left out a very important part of amplifier design:
stability. Simple transimpedance amps are generally unstable at high
frequencies, especially when the input source is a large area photodiode
operating in photovoltaic mode. The instability has to do with the way
that the capacitance of the photodiode slows down the negative feedback.
Oscillation of the transimpedance amplifier is often mistaken for some
other kinds of problems, and you should be sure that all of your
amplifier stages are stable before going much further. The simplest way
to tame oscillations in this case is to add a parallel capacitance
across the feedback resistance. Usually, a capacitance that is a small
fraction of the photodiode capacitance will settle things down.

It's best to use an oscilloscope and a proper probe to sniff around your
circuit. In the dark, every point in your circuit should be 'quiet',
with only dc levels on any opamp pin. If you let your photodiode 'see'
some fluorescent light fixtures, you might find some 120 hertz 'ripple'
showing up. Any other kind of oscillations mean you have stability
problems.

After you've quieted things down (if necessary), tell us more about what
you're trying to do and come back with a better description of the
optics. At that point, we'll be in a better position to suggest things
like feedback resistances, number of stages, lensing, frequency
response, etc.

Paul Mathews

Thanks Paul for the reply. I'll go look for some transimpedance
opamps on Monday. Just yesterday, I re-looked on the concept of
photoconductive and re-designed the current to voltage converter to be
in photoconductive mode (I need it to be like this because I am making
a laser range finder; this portion of the circuit needs to be precise
and fast). For the stability part, I haven't seen this problem at all
since I started building this, but I added on some 10pF caps just in
case it ever does happen, for some reason. There's also a 50Hz
lowpass filter I designed before and I'm planning to add it to the
receiver's end. For now, I'm not going to be modulating the laser
above 20Hz. And to avoid saturation in the opamp stages, would it be
best to have more stages of opamps and to cascade them, or to try and
put all the gain in 1 or 2 stages of amplification?

Anyways, I'm building a phase shift-based laser range finder. On the
receiver side, I have a 555 built oscillating at about 10Hz
(eventually going to replace it with a MCU). That 10 Hz is used to
drive a LA modulation-capable laser diode module from Meredith
Instruments. For the phase shifting part, I am using a THS3202
(mounted on a adapter board) as a high-speed comparator. I choose
this because of it's high slew rate, bandwidth, and that the rise and
fall times is 480 pS. The THS3202 will compare the sent modulation
against the received modulation, and the received modulation should
have some lag proportional to the distance from the object. On the
comparator's output will be a pulse train, and (if I remember
correctly) a DPRG (http://www.dprg.org/) said to feed it through a RC
network which will turn it into a analog voltage and where a ordinary
MCU can read it and determine the distance. And for tuning, I may add
a programmable delay chip in between the laser modulation wire and the
output of the 555 (or MCU) to "cancel out" the circuit delay. For the
hardware, I'm going to mount the laser and photodiode parallel to each
other but on the same "level", and the object is going to be placed
perpendicular to the laser and photodiode. The convex lens will be
placed on top of the photodiode. I hope this helps.

 

[Ray Xu]

On Aug 18, 9:40 am, Paul Mathews <opto.nospam.at.whidbey.com> wrote:

Ray Xu <ra...@tx.rr.com> wrote in news:59b3ca8c-a90a-40a6-8c63-
4b7a8f3a0...@m36g2000hse.googlegroups.com:

Thanks Paul for the reply.  I'll go look for some transimpedance
opamps on Monday.

Although you can buy 'transimpedance amps', the usual approach is to
simply use a low-noise opamp.

  Just yesterday, I re-looked on the concept of

photoconductive and re-designed the current to voltage converter to be
in photoconductive mode (I need it to be like this because I am making
a laser range finder; this portion of the circuit needs to be precise
and fast)

Your modulation frequency, if I understand, is just 10 Hz, so it's not
clear to me why anything needs to be fast. On the other hand, if you're
planning to have a ranging resolution on the order oa centimeter (for
example), your time resolution needs to be on the order of 60
picoseconds. In other words, for such a low modulation frequency, you
need incredibly high phase resolution.

.  For the stability part, I haven't seen this problem at all

since I started building this, but I added on some 10pF caps just in
case it ever does happen, for some reason.  There's also a 50Hz
lowpass filter I designed before and I'm planning to add it to the
receiver's end.  For now, I'm not going to be modulating the laser
above 20Hz.  And to avoid saturation in the opamp stages, would it be
best to have more stages of opamps and to cascade them, or to try and
put all the gain in 1 or 2 stages of amplification?

You haven't said anything yet about the size of your photodiode and/or
its capacitance. If you were getting away with no feedback capacitance,
I'm guessing your photodiode is small. Large lens plus small photodiode
means very precise aiming is required. Maybe that's why you got no
signal.



Anyways, I'm building a phase shift-based laser range finder.  On the
receiver side, I have a 555 built oscillating at about 10Hz
(eventually going to replace it with a MCU).  That 10 Hz is used to
drive a LA modulation-capable laser diode module from Meredith
Instruments.  For the phase shifting part, I am using a THS3202
(mounted on a adapter board) as a high-speed comparator.  I choose
this because of it's high slew rate, bandwidth, and that the rise and
fall times is 480 pS.  The THS3202 will compare the sent modulation
against the received modulation, and the received modulation should
have some lag proportional to the distance from the object.

Plus a comparatively huge lag produced by the slow receiver.....

  On the> comparator's output will be a pulse train, and (if I remember
correctly) a DPRG (http://www.dprg.org/) said to feed it through a RC
network which will turn it into a analog voltage and where a ordinary
MCU can read it and determine the distance.  And for tuning, I may add
a programmable delay chip in between the laser modulation wire and the
output of the 555 (or MCU) to "cancel out" the circuit delay.  For the
hardware, I'm going to mount the laser and photodiode parallel to each
other but on the same "level", and the object is going to be placed
perpendicular to the laser and photodiode.  The convex lens will be
placed on top of the photodiode.  I hope this helps.

From what you've told us so far, I'm guessing that you'll be needing a
modulation frequency in the 10s megahertz, with correspondingly fast
receiver, but there are many unanswered questions.
Paul Mathews

It needs to be fast so there is the smallest possible lag in the
circuit. And yes, the PNZ334 has a pretty small junction.

 

[Jim Thompson]

On Mon, 18 Aug 2008 22:48:31 +0000 (UTC), dbr@kbrx.com wrote:

Two questions: what comprised the "seperate test" and what do you mean by
fast?

Hul

In sci.electronics.design Ray Xu <rayxu@tx.rr.com> wrote:
Hi, I am having trouble making the receiver part for my laser range
finder, but I do
have a head start (FYI, I am not an analog person). Currently, on my
board,
I have a PNZ334 photodiode connected to a current to voltage converter
(using a op-amp; feedback is 100K), and connected to 2 stages of
amplification (I THINK that they are 15x and 100x). The problem is in
the
current to voltage converter; it is not sensitive enough to even pick
up the
slightest refracted laser beam from a convex lens in a dark room. I
also
did a separate test with the photodiode alone, and it was very
sensitive to
the small laser refracted beams. I am using photovoltaic mode and not
photoconductive, because I tried it before, it is too hard for me
(unless
someone can tell me how to build it). Can anyone please tell me how to
fix
this, or better yet, how to design a better current to voltage
converter?
Thanks.

Lay you dollars to donuts that the amplifier string is railed, thus
"it is not sensitive enough".

This thread belongs on the kindergarten group ;-)

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

Obama, Questioned about the Presidency, it's "above my pay grade"