Electronics > Projects, Designs, and Technical Stuff

Photodiodes saturating in ambient light

<< < (7/10) > >>

aussie_laser_dude:

--- Quote ---Roughly speaking, you can calculate the bandwidth of the photodiode by its capacitance, and the load resistance:
F_H = 1 / (2 pi R C) .......

This is a good reason to employ PD bias: the capacitance drops maybe 3x or more under bias....
--- End quote ---

F_H = 1/(2*pi*1kohm*25pF) = 6MHz (25pf capacitance @ 3V) vs 2MHz (70pF capacitance @ 0V photovoltaic mode).

Looks like the ~1 kohm resistor that guys here have suggested is a pretty good value to use with this photodiode.


--- Quote ---1. Think about this as a current transformer and everything becomes much clearer. In fact its really important to think of this whole thing in terms of a signal current.
--- End quote ---

hmm, I think I understand things a little, for equal magnetic flux change between the two transformer coils you'll need an equal current change (for 1:1 case), so on secondary coil we'll have our standard V = IR resistor load electronics happening which we can amplify by whatever method.


--- Quote --- The solution there is a circular polarizer, which should work well given the narrow linewidth of a typical laser.
--- End quote ---

I've actually done something kind of similar in the past, pulsed laser + circular polarizer + wollaston prism with dual balanced photodetectors + lockin amplifier (and other stuff) for detecting incredibly small changes in light polarization. Was pretty good for measurement of sub ps electronics. It is a rare and pleasant surprise to now be measuring a raw signal that can be displayed on an oscilloscope and not be buried in the noise, this is a luxury indeed  :D.

It's going to take a few days for me to comprehend all of this stuff in detail, but decided to give a crack at a simple design with light/current/voltage calculations and stuff. Op amps sound like a good idea to me, the high input impedance makes them a bit easier to understand ;D, an npn BJT amplifier design will take me another day to learn.

If there's anyone out there who likes to criticize designs, here's a good opportunity. There's got to be a really obvious, really dumb and really wrong mistake in the attached photo...

Siwastaja:
IMHO, don't waste time on polarizing filter, you may be able to get some (like maybe 2-3x?) of extra signal-to-ambient ratio out of it, but with a simple bandpass filter you easily get 20-30x!

Yes, laser is heavily polarized and ambient light mostly isn't, but laser is also a small peak in wavelength, which ambient is not. The latter feature is easier to extract a lot of benefit out of it. If you want to use polarization to perform attenuation of ambient, the physical alignment of source and measurement becomes critical.

twospoons:

--- Quote from: aussie_laser_dude on May 13, 2020, 08:34:15 am ---
If there's anyone out there who likes to criticize designs, here's a good opportunity. There's got to be a really obvious, really dumb and really wrong mistake in the attached photo...

--- End quote ---

Ah, yeah. The 30M resistor is ridiculously high and will cause you nothing but trouble. Scale everything back by 100x, or use a T-network of more reasonable values.

Also you can do this with just one opamp and one resistor. And this will get you your best bandwidth.
- neg input is set up as a virtual earth, so any current in the secondary of the transformer will be forced (by the opamp) to flow in R1.  So VOUT = I x R1  : thats your transimpedance gain equation.
- The tfr secondary sees a very low impedance, which is what you want for best bandwidth.

You will note this is the same as your circuit topology, but with R2 set to zero. R1 becomes unnecessary, as it is 'shorted' by the virtual earth.  Remember this is a current signal.

You dont actually have to use a transformer - an inductor will do too.

StillTrying:
How much current does the laser produce from the photodiode, your 5mV with a 1k load suggests only a 5uA change, that seems very low, I'd expect (hope :)) for 10 to 20 times that.

How much current does the ambient light produce from the photodiode.

With a few numbers on the wanted and unwanted current values it's quite possible to simulate photodiode circuits reasonably accurately.
https://www.eevblog.com/forum/projects/charge-sensitive-amplifier-for-photomultiplier/msg2871826/#msg2871826

How is the timing of the 10us light pulse produced, by the laser crossing the PD, or a 10us electronic on-time.
If it's a 10us 'laser turned on time', the 10us signal from the PD should be very square, because there's enough bandwidth.

Why do the pulses have to be 10us wide, I'd probably go for a bit less to make it a bit easier to separate the pulses from fast unwanted ambient light changes such as reflections off a shiny object, but I'm only guessing what your final want is. :)

aussie_laser_dude:

--- Quote ---How much current does the laser produce from the photodiode, your 5mV with a 1k load suggests only a 5uA change, that seems very low, I'd I expect(hope :)) for 10 to 20 times that.
--- End quote ---
  Well spotted! I've fixed the incorrect value, 5mV was due to forgetting the x10 scope probe attenuation |O. It's currently ~100mV for 0.2mW laser light for both experiment and theoretical calculations (only part of the 1mW beam will strike the 2mm wide array given laser beam width).


--- Quote ---How much current does the ambient light produce from the photodiode.
--- End quote ---

  See my Reply #30 attachment. Even with a $100 optical bandpass filter sunlight is still ~3x higher than the laser, this is better than the 50x ratio without the filter. A UV laser would have been a good choice since sunlight photon flux is low at that wavelength but I suppose it's nice having a laser that can be seen. btw I'm impressed at the deductive skills of some individuals here, too many smart people haha :-X


--- Quote ---- The tfr secondary sees a very low impedance, which is what you want for best bandwidth.
--- End quote ---
OMG... Witchcraft! I had no idea op amps could amplify a current directly. I'll need to study transimpedance amplifiers in detail. I would have thought the current in the secondary coil is being forced through a very high resistance load, but apparently it just goes around the op amp. My mind is blown.

Didn't realize the complexity on op amp design, all kinds of noises and special circuitry tricks, could spend days learning the absolute basics (pretty interesting though!).


--- Quote ---You dont actually have to use a transformer - an inductor will do too.
--- End quote ---
1. I need to get over my fear of transformers and grow as an individual
2. I suspect that a slow changing background light source over ms will be filtered better by a transformer. My thoughts are that the inductor will take time to change from one DC current to another when a shadow passes the detector causing some "leaked dc" into the amplifier over milliseconds?

Anyway, I think that's the two subgoals reached! Got a photodiode circuit that will work in the sun and got a pre-ADC amplifier. Cheers guys! Thanks heaps twospoons for the help with different designs! Will have to re-read some posts to understand everything. 

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod