Calculators with Support of the 82240 IR-Printer

[Christoph Giesselink]

(08-07-2014 06:33 PM)Martin Hepperle Wrote:  If you only sent 7 pulses (which is perfectly within the HP specs of 6-8 pulses) you gained some time for the pause (one 32kHz cycle), which may have helped the TSOP to detect the pause. The HP 48 and 28 seem to send 8 pulses for each burst, resulting in a shorter pause.
Currently I am playing with the 1133, maybe I should try some 1833 instead, if I can get some (I have one 1838, but thought it would be better to move closer to 33 kHz).

I had a look at the CLARKE External Reference Specification and I can confirm that the 1LT8 Clarke chip inside the HP48SX use 8 pulses. The situation at the 1LR2 Lewis chip inside the HP28S and the 1LP2 Centipede chip inside the HP28C is a little bit different, here you're enable a channel sending 16us pulses to the IR transmitter LED. The number of pulses depends on the time the STL pin is set to high by the CPU. This is inaccurate.

For more information you may have a look at the HP48 I/O Technical Interfacing Guide PDF. There's also a schematic of the HP48 IR receiver.

The TSOP1833 is out of production for many years now. I had a look at my personal REDEYE receivers, they all use the TSOP1833 chip. None of them use the later TSOP4133 chip which need one more pulse pause time than the TSOP1833. So I have no long time experience with the TSOP4133 in connection with the HP48 or HP28S.

The attached file show my smallest REDEYE receiver with a RS232C interface.