Shop OBEX P1 Docs P2 Docs Learn Events
IR Receivers: Hoarding Time! — Parallax Forums

IR Receivers: Hoarding Time!

ercoerco Posts: 20,254
edited 2015-03-30 11:26 in General Discussion
I've been searching for a while, and finally found this jewel at Mouser: Vishay's TSOP4038, a special little IR receiver that is "continuous signal compatible". The data sheet says it will even detect an unmodulated 38 kHz signal like the old Sharp 40KHz tin cube sensors would. You're probably yawning already, but believe me, they are RARE. And very useful for IR beacon tracking. Most IR modules can't detect a constant 38 kHz signal for more than a second or two (including the ones that come with Boebot, unfortunately). Try it and see for yourself...

Anyway, I'm stocking up for a variety of reasons. I'm doing some beacon experiments in ROBOT magazine; they're cheap at 78 cents each; and finally, they may be going away soon, the datasheet says "not for new designs".
http://www.mouser.com/ProductDetail/Vishay/TSOP4038/?qs=%2fha2pyFaduj0SkmO2XfIrLlK1mhOELebdZ0wsDRU2VQ%3d

Datasheet at http://www.vishay.com/docs/81926/tsop4038.pdf

You haven't heard the last of me on beacon tracking, BTW. :)
«13

Comments

  • jmgjmg Posts: 15,148
    edited 2012-09-04 17:42
    err, so why not modulate the beacon ?
  • Ron CzapalaRon Czapala Posts: 2,418
    edited 2012-09-04 17:47
    erco,
    The datasheet says it can also handle 38 kHz bursts. Does this mean it could be used in place of detector Parallax sells?
  • ercoerco Posts: 20,254
    edited 2012-09-04 19:41
    jmg wrote: »
    err, so why not modulate the beacon ?

    Precisely! A 556 dual timer makes a perfect cheap & easy beacon, half generating the 38 kHz carrier and the other half modulating it at 800-1000 hz. But most IR receivers filter out any regular modulation like that. If you want to, you can add a micro to a beacon and pulse out Sony codes to almost any IR receiver, but that's more complicated & expensive, and it takes longer to decode. Simpler & faster is better in my book. With this receiver, you could have dozens of beacons all over your house, each modulated at its own unique frequency so even a BS2-based robot can tell them apart in 10 milliseconds.

    This particular receiver is the rare bird that can also detect an unmodulated 38 khz signal, so if you want to simplify further, your beacon could just be a 555 timer. Those of course, all look identical to the receiver.
  • ercoerco Posts: 20,254
    edited 2012-09-04 19:43
    Ron Czapala wrote: »
    erco,
    The datasheet says it can also handle 38 kHz bursts. Does this mean it could be used in place of detector Parallax sells?

    Yes, it should function to receive standard Sony IR codes and such. But it may be more susceptible to noise and sunlight since it lacks some of the filtering that other receivers have. Will advise when I get 'em in.
  • GordonMcCombGordonMcComb Posts: 3,366
    edited 2012-09-05 14:25
    Excellent choice. I wouldn't worry about Vishay's declaring them EOL, as it doesn't even look like the replacement mentioned in the data sheet is available. And besides, Mouser have about 5K left, and I don't think even you can horde *that* many!

    For general remote control applications, I think the TSOP38238 might get the nod, and it's applicable to the Sony's as well.

    -- Gordon
  • ercoerco Posts: 20,254
    edited 2012-09-05 15:12
    I've been testing various modules and most don't work in this "continuous signal" application. One which does is half-commonly available: Radio Shack's 276-640 ($4.19, ouch) comes in 2 flavors. The one pictured on their website, with a silver metal bracket attached, works fine. But in the same package, same stock number, they also sell an all-black unit (no bracket) which does NOT work. I have a feeling they are changing to the non-working unit!

    http://www.radioshack.com/product/index.jsp?productId=2049727 Yep, that's my review: BUSTED!
  • W9GFOW9GFO Posts: 4,010
    edited 2012-09-05 16:44
    I ordered a few of the TSOP4838s, other than using less current what other advantages do the TSOP38238s have?[h=1][/h]
  • GordonMcCombGordonMcComb Posts: 3,366
    edited 2012-09-05 17:12
    Not a whole hecka lot of difference as far as I can see, and it takes careful comparison of the data sheets to find them. One critical difference between the 4838 and the 38238 appears to be current consumption: 0.85 mA for the former vs 0.35 mA (typical) for the latter. I use up mA's just blinking, so a difference this small doesn't mean much to me. I usually get what's cheapest, as long as it can read the Sony codes.

    -- Gordon
  • ercoerco Posts: 20,254
    edited 2012-09-08 15:47
    I tested the TSOP4838s, they don't work for my continuous signal app. I'm amassing quite a collection of IR receivers in my tests. Lots of Waitrony units work, but those aren't as commonly available as Vishay und der Telefunken...
  • ercoerco Posts: 20,254
    edited 2012-09-13 15:36
    Wow, my 4038s came in and they are surprisingly sensitive, showing at least twice the sensing range of other "continuous signal" sensors I have tested. It is a fixed gain sensor, as opposed to most other IR sensors with AGC: http://www.eebeat.com/?p=949 I'll have some beacon tracking videos up shortly. I may need a bigger house to fully test the limits of this sensor.

    They may have another advantage over other sensors, the ability to swag distance, per http://www.vishay.com/company/press/releases/2010/100628irsensors :

    Many other applications require a reflective sensor that detects not only presence but also proximity by measuring the strength or weakness of the reflected signal. Instead of a fixed detection threshold, analog information from the sensor is needed. This is possible with the TSOP4038, TSOP58038, and TSOP58P38 IR proximity sensors. The length of the sensor’s output pulse in response to the emitter signal varies in proportion to the amount of light reflected from the object being detected. For near objects, the output pulse approaches 100 % of the emitted pulse, for far objects the output pulse becomes shorter.
  • PublisonPublison Posts: 12,366
    edited 2014-08-05 09:19
    Can't believe this thread is two years old already!

    Anyway, I just went to order some more TSOP4038's from Mouser, and stock is zero and obsolete. I am contacting the companies that have them in stock, and I wanted to see if any one wants to do a group buy?

    I am waiting on a stock quote from one company right now.
  • ercoerco Posts: 20,254
    edited 2014-08-05 09:34
    What! Are you kidding me? First the PDB and now this?

    You know me, I'm down for a hundred. Where from, Jim?
  • PublisonPublison Posts: 12,366
    edited 2014-08-05 10:13
    erco wrote: »
    What! Are you kidding me? First the PDB and now this?

    You know me, I'm down for a hundred. Where from, Jim?

    I'm trying Solarbotics first:

    https://solarbotics.com/product/TSOP4038/

    Allieded has 35 that can ship in 8-10 days:

    http://www.alliedelec.com/search/productdetail.aspx?SKU=R1084633&utm_source=FindChips&utm_medium=buyNow

    Farnell Element12 Asia has 3222 in stock.
  • Ron CzapalaRon Czapala Posts: 2,418
    edited 2014-08-05 11:37
    Supposedly 2400 available at Bonase Electronics http://www.hkinventory.com/p/d/Tsop4038.htm
  • PublisonPublison Posts: 12,366
    edited 2014-08-05 11:51
    Ron Czapala wrote: »
    Supposedly 2400 available at Bonase Electronics http://www.hkinventory.com/p/d/Tsop4038.htm

    Thanks Ron,

    Worth checking into.
  • __red____red__ Posts: 470
    edited 2014-08-05 12:01
    WARNING:

    These IR receivers caused a product launch failure for me. They switched the component from a 5V component to a 3.3 -> 5V component without changing the part number. It cost me a serious 4-figure loss of money and reputational damage.

    Here's the response I got from Vishay:
    Hello Red,

    Allow me to answer your complaint since I am the one responsible for the decision. When the IC was changed from 5V only to 3-5V, this was an extension of the data sheet. Old applications should work correctly with the new parts. All customers of record were informed of the change. In our business, the majority of our customers do not want to see any change in part number or any restriction in data sheet parameters but extensions are usually OK. So there is a huge economic argument for not changing the part number.

    The unfortunate thing in your case is that your handler appears to have sold you 10 year old parts. All of our official dealers and distributors have stock plans which prevent this sort of thing and generally we do not guarantee any stock that is older than 2 years. It may be worth investigating whether your fabricator may be held liable for your damages.

    Best regards

    Be very careful buying stock of Vishay parts.

    The published datasheet says 3.3v -> 5v, the parts you get sold may be 5V parts.
  • PublisonPublison Posts: 12,366
    edited 2014-08-05 12:05
    __red__ wrote: »
    WARNING:

    These IR receivers caused a product launch failure for me. They switched the component from a 5V component to a 3.3 -> 5V component without changing the part number. It cost me a serious 4-figure loss of money and reputational damage.

    Here's the response I got from Vishay:



    Be very careful buying stock of Vishay parts.

    The published datasheet says 3.3v -> 5v, the parts you get sold may be 5V parts.

    Thanks for that update.
  • ercoerco Posts: 20,254
    edited 2014-08-05 12:28
    Sorry to hear about your experience, red. Good to know that. To date, I'm a 5V sensor man, but it's definitely good to know exactly what parts you're getting.

    So the parts you received were 5V parts and would not work in your 3.3V application. Did your supplier ever come clean?
  • ercoerco Posts: 20,254
    edited 2014-08-05 13:45
    GordonMcComb wrote: »
    And besides, Mouser have about 5K left, and I don't think even you can horde *that* many!

    Somebody did, Gordon! :)
  • PublisonPublison Posts: 12,366
    edited 2014-08-05 14:04
    __red__ wrote: »
    WARNING:

    These IR receivers caused a product launch failure for me. They switched the component from a 5V component to a 3.3 -> 5V component without changing the part number. It cost me a serious 4-figure loss of money and reputational damage.

    Here's the response I got from Vishay:



    Be very careful buying stock of Vishay parts.

    The published datasheet says 3.3v -> 5v, the parts you get sold may be 5V parts.

    So are you still sitting on these parts? You may be able to sell a lot here.

    Jim
  • ercoerco Posts: 20,254
    edited 2014-08-05 14:22
    My name is erco and I'm a hoarder. :)

    I just bought more TSOP4038s than I need from Solarbotics. Please don't tell the wifey!

    If you want some, PM me. Our friendly Canadian neighbors gouge considerably on shipping and handling on small quantities.

    Jim? Whit? Duane? Mike? Martin? Rich? I'm counting on you boys to bail me out! :)
  • jmgjmg Posts: 15,148
    edited 2014-08-05 14:52
    Publison wrote: »
    Anyway, I just went to order some more TSOP4038's from Mouser, and stock is zero and obsolete.

    The latest data sheet says this

    Not for New Design - Alternative Available: New TSSP4038 (#82458)

    and that replacement TSSP4038 looks to be more widely available. at Element 14, Future & Digikey


    Of course Murphy's law says that continual design (25m) has a different pin out from the NEC AGC pulsed one I just chose...(TSOP2438, 45m) and the other pin version (TSOP4438) is less available...
  • LevLev Posts: 182
    edited 2014-08-05 15:07
    Thanks for offering! I can use 10.

    Rich
  • Ron CzapalaRon Czapala Posts: 2,418
    edited 2014-08-05 15:29
    jmg wrote: »
    The latest data sheet says this

    Not for New Design - Alternative Available: New TSSP4038 (#82458)

    and that replacement TSSP4038 looks to be more widely available. at Element 14, Future & Digikey


    Of course Murphy's law says that continual design (25m) has a different pin out from the NEC AGC pulsed one I just chose...(TSOP2438, 45m) and the other pin version (TSOP4438) is less available...

    Thanks for that info. The spec sheets look pretty similar.
  • __red____red__ Posts: 470
    edited 2014-08-05 16:07
    erco wrote: »
    So the parts you received were 5V parts and would not work in your 3.3V application. Did your supplier ever come clean?

    Supplier blamed the vendor. Vendor blamed the supplier. I nearly lose my shirt over it.

    I won't use either again.
  • __red____red__ Posts: 470
    edited 2014-08-05 16:13
    It's a real shame because we built an IR mesh network for text messaging between conference attendees and the basic IR functionality simply didn't work because 3.3v wasn't enough to power the part. The lesson to learn is to always source your production components from the same source as your prototypes. After some long discussions with others on the subject, I'm not the first or last to be burned by tomfoolery like this.

    I'm not bitter towards IR and have moved to RF.

    I can now get bi-directional RF up to 300kbps between devices (P2P or broadcast) with a potential range of a half-mile for about the same price as RF. It just needed a driver written... and that's pretty close to done :-)

    Sorry for the slight topic detour.
  • jmgjmg Posts: 15,148
    edited 2014-08-05 17:41
    __red__ wrote: »
    Supplier blamed the vendor. Vendor blamed the supplier. I nearly lose my shirt over it.

    I won't use either again.

    The moron that created a situation where one part number can apply to two distinct and incompatible items should have been sacked.
  • ercoerco Posts: 20,254
    edited 2014-08-05 17:45
    Jim & Rich, thanks for your interest. Will advise after I get the parts. First thing I will test is how they work at 3 volts. :)
  • W9GFOW9GFO Posts: 4,010
    edited 2014-08-05 19:15
    erco wrote: »
    I just bought more TSOP4038s than I need from Solarbotics. Please don't tell the wifey!

    If you want some, PM me. Our friendly Canadian neighbors gouge considerably on shipping and handling on small quantities.

    Jim? Whit? Duane? Mike? Martin? Rich? I'm counting on you boys to bail me out! :)

    How many do you need to get rid of?
  • ercoerco Posts: 20,254
    edited 2014-08-05 20:15
    W9GFO wrote: »
    How many do you need to get rid of?

    'Bout a hundred. I bought way too many to get free ship. Might take a while to sell, but I may eventually use 'em all up myself. I'm doing some contract work now and there is amazing demand for localization, beacon tracking, and return to charger apps.

    These are rare "continuous signal compatible" types, but they also work great for standard IR control, like Sony codes etc . And as I mentioned previously:

    These have another advantage over other sensors, the ability to swag distance, per http://www.vishay.com/company/press/releases/2010/100628irsensors :

    Many other applications require a reflective sensor that detects not only presence but also proximity by measuring the strength or weakness of the reflected signal. Instead of a fixed detection threshold, analog information from the sensor is needed. This is possible with the TSOP4038, TSOP58038, and TSOP58P38 IR proximity sensors. The length of the sensor’s output pulse in response to the emitter signal varies in proportion to the amount of light reflected from the object being detected. For near objects, the output pulse approaches 100 % of the emitted pulse, for far objects the output pulse becomes shorter.
Sign In or Register to comment.