HP-48G IR Test Failure

[aj04062]

Can anyone say what the typical root cause might be? Would it be the LEDs?

If so, are the pair from the HP-38s the same?

I can't get a unit I bought to send or receive.

Thanks!

[BobVA]

It's been awhile since I've done this, but as I recall you need to place a piece of white paper (or similar reflective surface) in front of the IR port (about an inch or so away) for this to work properly. Did you try that? (Or am I thinking of another calculator?)

Regards,
Bob

Edited: 6 July 2011, 5:52 a.m.

[aj04062]

I did try that and it didn't work.

I swapped both LEDs with ones from a 38g parts machine I had. No luck. I then tried to print with the unit and it did print, so I know the send is working. When I try to send an object to another hp-48, no luck. This doesn't make sense to me.

[Thomas Chrapkiewicz]

A simple check to see if the transmitting LED is working is by viewing it (on the LCD display) with a pocket camera (or smartphone camera), as the image sensor in the camera is IR sensitive and you can 'see' the LED illuminating with IR when it is transmitting.

TomC

[Paul Berger (Canada)]

I tried the test on my 48G and 48GX and the test passes even if the IR is pointing off into space, on the GX it will fail if the card cover is not in place, as warned in the manual. You are likely able to print as the print communication would be one way, however when you are trying to transfer to another calculator there would likely be some handshaking between then which would fail if the receive circuit is bad. My 48S has this problem it prints but the IR test fails and it will not communicate with other calculators.

[aj04062]

Do you think this is a failure in the LED or in the rest of the circuitry?

[Paul Berger (Canada)]

I have not looked into it too much, the 48S is kind of a junker so I was not too concerned, however with what little looking I did it seems that the photo-transistor does detect IR pulses which would lead me to believe it is OK and the problem is somewhere beyond that. Schematics for the 48 can be found at http://www.hpcalc.org/hp48/docs/misc/.

[Paul Berger (Canada)]

Well I fixed the IR on my 48S it appears that there was an intermittent connection between the second transistor and the input pin on the processor, it started working while I as probing around so I replaced that connection with a piece of wire wrap wire.

While inside the 48S I noticed that the processor is not a Clarke as most documentation lists as the saturn version used in the 48S/SX but is in fact a Yorke with the speed pin tied low so that it runs at 2MHz. I did see an article that claimed that early production runs of Yorke did not yield a lot of chips that would run reliably at 4MHZ perhaps these found there way into late production 48S/SX. The serial number of my S suggests it was built week 46 of 1992.