Help to identify unmarked phototransistor

Discussion in 'The Projects Forum' started by Tornado Bones, Nov 9, 2014.

  1. Tornado Bones

    Thread Starter New Member

    Nov 8, 2014
    14
    0
    I need help understanding how to identify unmarked phototransistors, so I can order extras for a current project. First, I have no formal training in electronics. However, I'm a hopeless science geek who has been dabbling in electronics for over 40years.
    My current project is an invention that performs a specific task when triggered by a visible light signal. It's a little along the lines of an opto-coupler but I'm using it for a very unique function in a sports application.
    The invention incorporates a simple phototransistor light activated relay circuit that I buy as a pre-assembled board. I don't know the specs on the phototransistor that come mounted on the boards but there are a couple of reasons I need to buy more. I'd appreciate any advice on how to test them so I know what I need.
    The other reason I want to know the specs is that I'm looking for phototransistors that can perform similarly with visible light but with less sensitivity to IR. The intended light signal that operates my devices can originate from a variety of possible sources, BUT the signals will always be based on visible green light. Right now, I'm having problems with interference due to ambient IR from sunlight. The equipment must operate outdoors so I need a fix.
    I'd like to try a different phototransistor to help minimize IR interference. Ideally, I would like to find a phototransistor that is more responsive to visible light between 525 and 580nm-ish. Are color filters ever used for things like this?
    My current phototransistors come already mounted on the boards, so I have to desolder them in order to install them into the specialized sensor assemblies that I build. Sometimes I need spares or extra phototransistors but I don't have any specs to help me buy the right ones.
    I've taken a few shots at buying them online but the ones I received were too different to work as a drop in. Once I found some at Radio Shack that worked great, but with no specs! I bought their last few but they have since discontinued that part.
    How do you test an unidentified phototransistor to order extra ones from Mouser or Allied, etc.?
     
  2. KJ6EAD

    Senior Member

    Apr 30, 2011
    1,425
    363
    Usually a request for help identifying a thing is accompanied by more information about the thing such as a photograph, brand name or part number. Since you plan on using a sensor with better sensitivity to visible green light anyway, I'd not waste effort on a matching replacement but look for a sensor that fits your needs.

    Typical sensing technologies tend to be very receptive to IR so the sunlight interference is no surprise and an optical filter will probably be necessary.

    http://www.edmundoptics.com/optics/...s-filters/colored-glass-bandpass-filters/1924
     
  3. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,779
    1,103
    A green LED might be the basis for a sensor, but would probably need some buffering/amplification. Not exactly a drop-in replacement though :(.

    Edit: A quick skim through some phototransistor datasheets found PT204-6C has a ~60% response to 500nm compared to its peak response around 900nm. Didn't find any which peaked other than in the infra-red.
     
    Last edited: Nov 9, 2014
  4. Tornado Bones

    Thread Starter New Member

    Nov 8, 2014
    14
    0
    Thanks for your responses. The phototransistors I've been using are all plastic and bullet shaped, like a 5mm white LED with two leads. If I did this right, there is a picture attached below.

    Right now, I need to continue to move forward on my project so I'm content to find replacement PTs, even if they are still sensitive to the IR. There are more ways to skin a cat and I'm sure I'll eventually come up with a way to manage the IR interference.

    I think I can guess at the breakdown voltage I need, since I know it is used in a 12VDC circuit. Anybody know how to test the phototransistor to determine the rest of the specs?

    phototransistor Pic.JPG
     
  5. Tornado Bones

    Thread Starter New Member

    Nov 8, 2014
    14
    0
    OK, I still haven't found how to test the specifications of an unmarked phototransistor so I've ordered several different types with different spec values. I'll just have to continue trial and error for now to see if I can find one that works - then I can use that spec going forward. Still interested in any input about how to test these in a way that would take the guessing out of this or help my understand why a given PT is working so I can optimize.

    I've been doing a bit of research and found that they do make phototransistors with peak sensitivity around 580nm. They aren't as common but they do exist. Some of them have a greenish tinge to the color of the lens/package, which kind of makes sense because they are probably filtering out the near IR with dyes or pigments. The key limitation I'm finding is that the filtered phototransistors sometimes cut off the sensitivity at the higher wavelengths with optical band pass filters, which only narrows their sensitivity range. If the PT didn't respond to green light before filtering, it wont respond to it after filtering. In other words, filtering may prevent ambient IR interference but depending on the phototransistor, there may or may not be ANY response to green light.

    I have found other products out there that have shifted (not narrowed) the overall range of sensitivity so it peaks at 570 or 580nm. I believe they have used different semiconductive materials to achieve this. Seems like a much better option for me BUT... so far all that I've found have a breakdown voltages of only 4-6V. I put a meter on my circuit and it shows 10V at the terminals where the phototransistor will go. I believe this relegates me to the next higher category of phototransistors, which have a breakdown of 20V or higher. None of these have peak response below 800nm.

    I've considered using a photodiode instead. They go way down in wavelength, even into the UV, but these are a lot more expensive (maybe too expensive) and their dimensions won't easily fit into the other essential components in my sensor assemblies. From what I'm reading, the PT's respond like more of a switch, whereas the PD's are more linear so their output will be more variable. For my application, it's critical that the response be reproducible and sharp. I felt like a PT would hit a sharper threshold to trigger the relay in an on/off fashion than the PD, which would kind of edge up to it. In case I'm wrong, I'll be testing this option too.

    I've also thought about photodarlingtons but I'm not sure how to adapt them. The ones I'm aware of have 3 or more leads and I would be using them to replace a phototransistor with two. I'm thinking two of the leads will serve as the collector and emitter but do I just skip the third lead? Is the third lead a base, as if it were a single transistor? I'm already using a Darlington amp in my sensor circuit to trigger a relay. Seems like I should tweak that before I pay for a photodarlington but I can't afford to change the PC board right now, only the components.

    Thanks in advance for your patience with this newbie and for your help with my questions. I am coming up on a hard deadline to finish my testing or I will lose the rights to my invention.
     
  6. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,779
    1,103
    Which particular specs are you interested in? If all you need to know is 'how does the LED current for a given illumination compare for various LEDs biased in the same way ' then it shouldn't be difficult to set up a suitable test jig.
     
  7. djsfantasi

    AAC Fanatic!

    Apr 11, 2010
    2,802
    832
    What is the light source that you are detecting with the phototransistor? Is it or can it be modulated?

    To avoid false detection, remote control with photo transistors are often modulated (e.g., to 38kHz) to eliminate background noise.
     
    Johann likes this.
  8. KJ6EAD

    Senior Member

    Apr 30, 2011
    1,425
    363
    If I remember correctly, Perkin-Elmer and Hamamatsu are world leaders in such detectors so they're worth a look.
     
  9. Tornado Bones

    Thread Starter New Member

    Nov 8, 2014
    14
    0
    I am an organic chemist and mathematician, with a biology habit and some background in light/optics. Definitely not an electronics engineer. Thanks again for your help.

    The source can not be modulated. That would make it really simple. (I'm already using modulated IR for another project.) My sensors are intended for "field use" to detect a signal from a variety of signal sources over which I have no control. The sources all emit in the visible range with similar intensity and peak output (greenish) around 580nm. That's all the detail I am able to share about the application for now. I can share more background in a couple of months. It will make sense then.

    I basically have two needs. One is just to understand the specs on the phototransistors I'm already getting on the pre assembled boards I'm using because I need extras for the end use. This one is urgent and it is a show stopper for me. Even if I can't find an alternative that is wavelength specific, I still need to just find phototransistors that perform similarly to the ones that are already mounted on the boards I'm buying. I'm down to just buying a bunch of phototransistors (with data sheets) and trying them one by one, since I have no spec to work from.

    I've tried to test my last remaining phototransistor that I desoldered from one of my pre-assembled PC boards but I don't know what I'm testing. I have put power to it and checked the amps with and without a light. Is this a sufficiently controlled test? The multi-meter never quite settles and I'm also not sure if this is the way to test a phototransistor anyway. Should I be using a resistor or some other type of test circuit as a jig? In a nutshell this is my original question. How do you test a photodiode?

    I assume I'll need similar light/dark current as the phototransistors I've already used. I'll be putting around 12VDC to the phototransistor in its end use.

    The secondary need is the wavelength specific aspect, where I want to exclude IR interference. I've checked Allied, Newark Element 14, Digikey, Vetco, Mouser and a few other vendnors. Manufactureres I've checked are Vishay, Honeywell, Sun, King-Brite, Lumex, Osram, Lite-On, Panasonic, Rohm, Sharp and a couple of others. I did not check Perkin Elmer. Hamamatsu PTs were optically filtered but they excluded the visible range, so no sensitivity below 600nm. They did show one photodiode in their data sheets that looked promising but I did not find any vendors for it via google. Their other photodiodes were way out of my price range, since I'm just using plastic phototransistors now.

    Thanks again for your comments.
     
  10. Johann

    Senior Member

    Nov 27, 2006
    190
    30
    Hi Tornado Bones,

    I would certainly include a resistor, say 1k, to do any testing with 12V just to prevent burn-out of the photo transistor under test. Not knowing anything more about the photo transistor, a simple comparison to a known working one would suffice, I believe. Once you've measured the current passed by the photo transistor for a given light level, you can compare that current to that obtained from the transistor under test (with the same light level of course).
     
  11. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,779
    1,103
    Have you contacted the board vendor to ask for the PT spec? Can you modify the board circuit to allow the use of a wider range of PTs?
     
  12. Tornado Bones

    Thread Starter New Member

    Nov 8, 2014
    14
    0
    I did contact the vendor, initially. He tells me the parts come with the boards from offshore and he doesn't have any specs. He offered to buy a large quantity for me but I've solved the problem for now. I just tested a list of different PTs with the same circuit and found that almost all of them worked. The ones that failed were too sensitive or the package dimensions were tricky to work with for my setup. Most of the PTs I tested prior to this thread failed to work for me. It must have just been bad luck and/or bad parts (some were from radio shack).

    What still confuses me is the collector emitter voltage and the emitter collector voltage. The parts I've settled on are TEPT5700 or TEPT5600. These have CEV and ECV of 6v and 1.5v, respectively. My meter showed 10vdc on the board where the PT goes. Without understanding better, I guessed I needed a higher voltage rating for the PT. Is the CEV more like a photovoltaic effect? Is that what it means? Am I using these parts above their voltage ratings? Will this cause failures or service life issues?

    BTW, the TEPT5700 and TEPT5600 both have peak sensitivity to 570nm. The package is a perfect fit and I've identified a mounting configuration in my sensor assemblies that is completely immune to ambient IR from sunlight.
     
  13. Tornado Bones

    Thread Starter New Member

    Nov 8, 2014
    14
    0
    I contacted the vendor first. He did not know the specs on the phototransistors. They come with the boards where he orders them in quantity offshore. He offered to order a large quantity of the phototransistors separately from his supplier but I've already solved the problem (I think).

    I found a part that seems to work well. It's peak sensitivity is 570nm and its package size fits perfectly into the 5mm hole that molded into my sensor assemblies. I've also identified a mounting configuration that seems to be totally immune to interference from ambient IR from sunlight. I posted the spectral curve below.

    The last remaining question I have is about voltage. The mfgr lists collector emitter breakdown voltage as 6v at 1mA. My meter reads 10vdc on the board where the PT goes. Am I exceeding the voltage rating of my PT? Will I have issues down the road, like failures or service life? Is there a workaround?

    I don't fully understand collector emitter and emitter collector voltage. Is this more like the photovoltaic effect?


    upload_2014-11-25_8-47-35.png
     
  14. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,006
    3,762
    I think they are referring to a reverse voltage situation. also, how do you have this breakdown voltage without having the datasheet?
     
  15. Brevor

    Active Member

    Apr 9, 2011
    279
    54
    You are correct, I have some of those TEPT 5600's and the max collector to emitter voltage is 6 Volts. It may be possible to modify the circuit to work on 5 volts. Can you post a schematic?
     
  16. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,779
    1,103
    VCE means the voltage between collector and emitter when the collector is more positive than the emitter. VEC means the voltage between collector and emitter when the emitter is more positive than the collector.
    If you have 10V at the collector and less than 4V at the emitter then you are exceeding the spec.
     
  17. Tornado Bones

    Thread Starter New Member

    Nov 8, 2014
    14
    0
    GopherT, sorry if my questions were unclear. I don't have specs or data sheet for the original parts that are part of the circuit I'm using. The boards come pre-manufactured with the phototransistor already mounted. I've tried a slough of phototransistors and at first, nothing I tried was working. I started with radio shack parts, which also had no data sheet. Now that I've found a couple of PT's that seem to work, I do have data sheets for those. Only problem is that the best parts I've tested have a lower voltage rating than my circuit is supplying to that component (if I understood Alec_T's comment). The schematic for my circuit is attached. I'd have to go back and check the components and read the color codes to give you all the values of each component.

    Can the 6v PT be made to work? For bench testing, I'm using a regulated power supply at 12vdc. In the field, it will need to sometimes run from automotive batteries that can exceed 13V. Will I need something else to control that power supply. I'm guessing it will be more likely to fry my phototransistors.
    Dark On Switch.jpg
     
  18. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,779
    1,103
    It would be helpful to know the resistor and pot values.
     
  19. Brevor

    Active Member

    Apr 9, 2011
    279
    54
    If you have a meter connect it across the phototransistor and expose the board to all expected light levels. If Vce stays below 6 Volts you should be OK.
     
  20. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,779
    1,103
    Adding a resistor Rx, as shown here, will reduce the voltage across the PT. The Rx value should be about 1.5 times R3 value to get the PT volts <6. You may need to twiddle the trimmer to adjust the light sensitivity.
    VoltsReducer.gif

    Edit: Ignore the ground symbol below C1, it's just to keep LTspice happy.
     
    Last edited: Nov 27, 2014
Loading...