Shop OBEX P1 Docs P2 Docs Learn Events
PNA4602M discontinued — Parallax Forums

PNA4602M discontinued

tiger72tiger72 Posts: 6
edited 2011-04-30 11:38 in Accessories
I was wondering if anyone one knows of a replacement or substitute for the PNA4602M? Panasonic has discontinued the part. Digikey and Mouser have them as obsolete and don't give any idea on a replacement. Finding them has become difficult as most hobby electronic outlets have run out of stock

What is Parallax going to do? They list having 390 in stock. Once those are gone and all the kits that use that sensor are gone, what is the replacement? I know Sharp and Vishay make similar devices, but they reject continuous signals. Beside needing 38kHz, you have to turn the signal on and off for 600uS at a time. Plus the output of those devices does not stay low for long, you have to read the sensor quickly. Parallax would have to print new manuals and change all their code examples for these other sensors.

Anyone have any idea for a replacement?

Comments

  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2011-04-28 10:15
    There are plenty of other options:

    And those are just the ones DigiKey has in stock.

    -Phil
  • tiger72tiger72 Posts: 6
    edited 2011-04-28 14:02
    Not trying to be difficult, but did you read my whole post? The reason I ask is because your links are to the Vishay TSOP series of parts that I already mentioned are "not" direct replacements/substitutes. Again I will point out that if you don't read the data sheets you will not realize that the PNA4602M can handle a continuous 38kHz signal where as the TSOP parts cannot! The PNA4602M parts work with the way the Basic Stamp programs use them. You would have to rewrite all Basic Stamp programs to use the TSOP and Sharp GP1UX series of detectors. I have already tested a few of each Vishay TSOP and Sharp GP1UX parts and have not found one that accepts a continuous signal.

    That is why I am asking if there is a direct replacement for the PNA4602M. Does anyone know of a direct replacement or substitute? It needs to work just like the PNA sensor. If there is no source for a direct substitute, then I will have to rewrite and test code for a bunch of projects to use a different sensor. I can see this being an issue for companies that have based their products on this sensor. I guess since Panasonic has discontinued it there were not enough companies purchasing the part. It is not a huge issue, but it is annoying and time consuming. I have working code that I don't feel like rewriting. Parallax is going to have to go back through all their manuals and either reprint with updated code segments or print supplements informing user that the code has to change.

    So again anyone out there know of a substitute that functions the same as a PNA4602M?
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2011-04-28 14:27
    'Missed the part in your post about Vishay and Sharp -- sorry. However, every IR receiver I've ever used, at least, was incapable of responding to a continuous 38KHz input, and I never considered it to be a problem. In BoeBot obstacle detection, for example, continuous excitation is not used. But are you saying that the receiver output is not held low long enough, post-reception, for the Stamp to see it?

    -Phil
  • Mike GreenMike Green Posts: 23,101
    edited 2011-04-28 14:33
    The Stamps do not make use of a continuous modulated IR signal. If you look at the programs, the IR signal is generated by a FREQOUT statement which stops producing IR before the PNA4602M state is checked. The test for the PNA4602M state really has to be essentially the next statement so the test is done within a couple of hundred microseconds. I haven't looked at the Vishay detectors to see how quickly they raise their output level. It may be that a small capacitor is needed across the Stamp input to slow down the output change for a few hundred microseconds. That'll need some testing. I don't know what the Vishay output stage looks like. The PNA4602M is open collector with a pullup.
  • tiger72tiger72 Posts: 6
    edited 2011-04-28 19:35
    Phil and Mike,

    I do know in the Boe-Bot and the Sumobots you are not using continuous signal. You guys have part of the issue correct when you said the active low from the receiver does not stay low long enough. What happens is because the Basic Stamp executes slowly, the 38kHz signal stays on too long so the receiver thinks the signal is noise and changes from low back to high before the Stamp reads its input line.

    Let me try to explain it this way - I tested a Sumobot with a Sharp GP1UX51Q and a Vishay TSOP34438. They both have the same pin out as the PNA4602M and have similar functionality. The issue is they don't work in the circuit as designed with the Sumobot code. Randomly the Sumobot would register a detection, but not like when using the PNA sensor. Using an oscilloscope I checked to make sure the the Sharp and Vishay sensors were outputting an active low when detecting a signal. The do, but you are lucky if the low output lasts for a couple hundred microseconds. So this is not lasting long enough for the Stamp, in addition the 38kHz signal is lasting too long causing problems.
  • tiger72tiger72 Posts: 6
    edited 2011-04-28 19:47
    Forgot to mention on that last part what I mean is the Stamp is outputting the signal for too long causing the sensors to think the signal is noise.

    So anyway I tested the Sharp and Vishay sensors on an Atmel AVR - the ATmega168. I used the internal 8MHz clock and used Timer 1 to generate the 38kHz signal. Again this was for a Mini Sumo robot. Now the issue becomes the AVR reads the sensors output too fast. In a loop I turn on the signal for 1 sensor, read the sensor, trun the signal off, repeat for the other sensor, and send commands to the motor driver. This happens so fast that I lose detection during the sensors timeout period (this is in the data sheet). The timeout where it will not detect a signal can be anywhere from 1 to 3 milliseconds. Some of the sensors in these families can be even slower. So my sumo robot jerks around and loses the opponent often, where as when I was using the PNA sensor my bot smoothly tracked opponents.

    I'll admit some of the problem is because of my lax programming. With the PNA I am not worried about timing other than the 38kHz. Just turn on the IR LED with the signal, read the port, turn off the IR LED - done. Simple! Now with the other sensors I have to make sure that I only send a short burst (about 600uS) of signal, make sure I don't loop back to that sensor too soon, and continue the last motor commands until I can get an update from the sensor. This includes everthing I had to do for the PNA sensor. Just to get the smooth respone from a sensor - what a headache. That is why I used the PNA.

    Can you guys show me any working Basic Stamp code that responds to the Sharp or Vishay sensors correctly? Or in a similar fashion to the PNA4602?

    Thanks
    Tony
  • tiger72tiger72 Posts: 6
    edited 2011-04-28 19:55
    Mike,

    You'll have to read the output of the sensor before turning off the signal because the output returns high within 6 pulse periods max. You have from 25 microseconds to 150 microseconds to register the output of the sensor after you turn off the modulated IR. I have not tried this yet, but I will experiment with changing the code on a Boe-Bot or Sumobot - which ever I grab first, ha!

    Talk to you guys later,
    Tony
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2011-04-28 20:55
    tiger72,

    You've obviously analyzed this issue exhaustively. But I wouldn't worry too much. Parallax has extensive contacts in China, and I'm sure they'll find a suitable replacement for the Panasonic device.

    -Phil
  • tiger72tiger72 Posts: 6
    edited 2011-04-28 22:09
    Yeah, I am sure Parallax has already been looking into it. I was just hoping someone had already found a functional replacement. And after all my moaning and belly aching I know what really has to be done - fix my code so it doesn't matter which sensor I use. Then I am not stuck with a single source, and I can pick whatever is cost effective.

    I did like Mike's idea about putting a cap on the sensor output to keep it low a little longer. Might try that in addition to fixing the code.

    Thanks guys,
    Tony
  • ercoerco Posts: 20,255
    edited 2011-04-30 11:38
    Waitrony is a popular supplier with a few current modules that are continuous signal acceptable:

    http://www.waitrony.com/UploadFiles/20051131730796.pdf

    http://www.waitrony.com/UploadFiles/2005113171021829.pdf

    http://www.waitrony.com/eng/UploadFiles/20051110101911266.pdf
Sign In or Register to comment.