If it is normalize, then the datasheets make sense. However, the manufacturer should state this. Having designed chips and written datasheets, you never put a parameter in a datasheet that doesn't have clearly stated test parameters.
My application if more hacking and educational that a product. I've messed with pyroelectric film and would like to understand photodiodes a bit more. However, I'm the type that tries to match real life to theory, so I need to have the theory well understood.
Glad to see that you want to get some understanding of the subject.
Quite obviously Henry does not undertstand much about IR and is completely befuddled - just ignore his remarks else you will start making incorrect conclusions from the start.
Have a good read of the manufacturers datasheets and app notes (Vishay quite good for this) -they are quite a fund of knowledge.
Anyway just to help:
sensitivity is not normalised to a unit area of receptor. The output is quantified for a given quantified incident radiation (typically watts/sq metre).
spectral sensitivity has nothing to do with sensor area and is generally normalised to its design (peak response) wavelength.
candela and lumen have nothing to do with IR. Watts, watts per sq metre, and watts per steradian have very much to do with IR.
over this small IR range (850~950nm) no one wavelength intinsically has a faster 'system speed' than any other. Device manufacture does - eg double heterojunction IRleds are faster than large single junctions devices. Not all devices are made at all wavelenghts.
calculate Z out is nonsense. Just understand what you are doing properly. You may decide to operate a photodiode at zero volts.
"RHRRC" schrieb im Newsbeitrag news: snipped-for-privacy@f1g2000cwa.googlegroups.com...
Rumbbling here.
I made a working system in those days:
230kbit/sec manchester encoded on carrier 230kHz. A couple of TSHF5400 ir transmitter diodes driven my a small PowerMOSFET. BWP34 as receiver. transimpedance amp with LF357. A little bandpass-filtering to fight against flourescent lamps. Such a system works reasonable in a white painted 30m2 room with backscattering of the walls. It was not optimum but also not optimized.
So the reader can decide if I know enough.
Yes, this is first to read!
That is the same! If reading the various infos on the Net you will mention that there is a difference between physics people thinking and electronics engineers. They use different metrics.
There is a difference between absolute (yours) and relative (most mentioned in datasheets) spectral sensitivity. As I sayed several times: chip area is normalized!
If you need a point-multipoint ir connection best is to diffuse (or at least radiate on a "plane") the light! That implies no angle definitions. Unfortunately many manufacturers of ir transmitting diodes use the human optical system definitions. That is not my fault! You need at least these things if you want to calculate the effect of having a light bulb in the system as noise source.
I suggest looking in the datasheets: TSHF5400,
formatting link
datasheets (low-speed typs), and others... See what process technology is used at which wavelength and decide yourself. The difference between the two wavelength "systems" is often ten times.
It is impossible to get current without voltage. Photo diode in current mode: If you use a transimpedance amplifier you have almost zero voltage at the photo diode but it is NOT true zero! The problem is that the PIN photo diode have it's most capacitance near zero voltage! So decouple the photo diode with an capacitor and make a current source to drive the photo diode. Such can be an carrier frequency related band-pass LC or a gyrator construction. The other side of the medal is that shot noise goes up with this circuit. That is why low-speed systems (below 1kHz) use the photo diode more in voltage mode. This opens the use of photovoltaic cells as receiver...
One (admittedly of several) of the reasons that most IR remote controls (eg for the TV) are design to work at sub 60kHz frequencies is to avoid hitting a harmonic of the 'electronic ballast' compact fluorescents. There are other significant considerations in designing a IR data (remote) control bfor your TV, DVD, etc but of the billions manufactured all employed very significant bandwidth restriction and none used 230kHz etc etc.
So the reader can decide.....
No it is not.
If you are befuddled about the subject you more than likely think this is true but it is not.
The units used for radiation measurement are universally adopted and certainly used by electronics engineers working in the field which is why the manufacturers of the devices use these units
You obviously do not understand any of this. One wonders what "chip area is normalised" would (could) be normalised to and why it would be mentioned in a datasheet.
That is just rubbish. No manufacturer uses the human optical system definitions for IR radiation - that is just rediculous to even guess it is so even in your befuddled state.
The Lumen is an a univerally agreed quantity of elefctromagnetically radiated energy - the quantity of energy varying with wavelength. A Candela is one lumen distributed uniformly throughout one steradian. The CIE provided tabulated quantities and wavelenghts for the lumen. One lumen at IR wavelengths is more energy than exists in the universe (i.e. is infinite) which is why such units are not used for IR by anyone.
Yes. If you look at low speed devices they are slower than higher speed devices.
I'll just repeat
calculate Z out is nonsense. Just understand what you are doing properly. You may decide to operate a photodiode at zero volts.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.