The combination of L1 and R3 forms a high-pass filter that allows signals above 1 MHz to pass. L1 is a standard noise-suppression choke. From this filter, the signal is fed to two inverters configured as amplifiers. The third and final inverter (IC1c) generates a logic-level signal. This 74HCU04 provides so much gain that there is a large risk of oscillation, particularly when the final stage is loaded with a 75-Ω coaxial cable. In case of problems (which will depend heavily on the construction), it may be beneficial to add a separate, decoupled buffer stage for the output, which will also allow the proper output impedance (75 Ω) to be maintained in order to prevent any reflections.
When building the circuit, make sure that the currents from IC1 do not flow through the ground path for T1. If necessary, use two separate ground planes and local decoupling. Furthermore, the circuit must be regarded as a high-frequency design, so it’s a good idea to provide the best possible screening between the input and the output. With the component values shown in the schematic, the range is around 1.2 metres without anything extra, which is not especially large. However, the range can easily be extended by using a small positive lens (as is commonly done with standard IRDA modules). In our experiments, we used an inexpensive magnifying glass, and once we got the photodiode positioned at the focus after a bit of adjustment.
IR–S/PDIF Receiver Circuit Diagram
We were able to achieve a range of 9 metres using the same transmitter (with a sampling frequency of 44.1 kHz). This does require the transmitter and receiver to be physically well aligned to each other. As you can see, a bit of experimenting certainly pays off here! It may also be possible to try other types of photodiode. The HDSL-5420 indicated in the schematic has a dome lens, but there is a similar model with a flat-top case (HDSL-5400). It has an acceptance angle of 110°, and with the same level of illumination, it generates nearly four times as much current.
The current consumption of the circuit is 43 mA with no signal and approximately 26 mA with a signal (fs = 44.1 kHz) That is rather high for battery operation, but it can handled quite readily using a pair of rechargeable NiMH cells. Incidentally, the circuit will also work at 4.5 V and even 3 V. If a logic-level output is needed, C3 at the output can be replaced by a jumper. Finally, there is one other thing worth mentioning. With the HSDL-5400 that we had to play with, the cathode marking (a dark-blue line on the side below one lead) was on the wrong side (!). So if you want to be sure that the diode is fitted properly, it’s a good idea to measure the DC voltage across R1, which should be practically zero.
Author: T. Giesberts
Copyright: Elektor Electronics
Copyright: Elektor Electronics
Post a Comment