Problems with toslink receiver

[Zipdox]

I'm figuring out how to use toslink to connect my interrupter to my SSTC and the transmitting end works fine but I'm having trouble with the receiving end.

Sometimes the receiver is normally-high and other times it's normally-low. I have no idea if there's any sort of consistency or if I'm doing something wrong. I found a toslink product guide online that says the receiver contains an inverter but that doesn't help at all. Anyone have any experience with toslink?

[Hydron]

TOSLINK is designed for a "DC balanced" manchester-encoded bitstream. Using it in an interrupter application is not guaranteed to work, as your signal will be far from what they are designed to handle. This is the main reason why people typically use more expensive industrial fibre optics parts.

[Zipdox]

TOSLINK is designed for a "DC balanced" manchester-encoded bitstream. Using it in an interrupter application is not guaranteed to work, as your signal will be far from what they are designed to handle. This is the main reason why people typically use more expensive industrial fibre optics parts.

But that still doesn't explain the seamingly random switching between high and low. As km sitting here typing this the output is a witching from time to time without any optical input.

[Zipdox]

Also worthy of notice is that I was completely unable to find a datasheet for the toslink receiver. On the back it says DO-017A.

[davekni]

It wouldn't surprise me if the receiver had an AC-coupled stage followed by a schmitt-trigger.  The schmitt-trigger could stick in either state without signal.

If you can find any, there are DC-coupled raw detectors for 1mm-core fiber, IF-D91 with a photodiode and IF-D92 with a phototransistor.  They clamp onto the jacketed fiber, not a TOSLINK connector.  Those are what I use mostly.

Another option that I've also used is to buy TOSLINK couplers.  Drill out one side to 3mm and insert a 3mm photodiode or phototransistor.

[Zipdox]

It wouldn't surprise me if the receiver had an AC-coupled stage followed by a schmitt-trigger.  The schmitt-trigger could stick in either state without signal.

If you can find any, there are DC-coupled raw detectors for 1mm-core fiber, IF-D91 with a photodiode and IF-D92 with a phototransistor.  They clamp onto the jacketed fiber, not a TOSLINK connector.  Those are what I use mostly.

Another option that I've also used is to buy TOSLINK couplers.  Drill out one side to 3mm and insert a 3mm photodiode or phototransistor.

Where can I get phototrannies for cheap? Can I maybe saw an optocoupler in half?

[dexter]

With some extra components the toslink receiver can be made to work no matter how long pulses you want to send.

on interrupter side add an extra 555 to send a train of high frequency pulses instead of a flat ON

on receiver side add an integrating circuit to recreate the flat ON signal

[Zipdox]

With some extra components the toslink receiver can be made to work no matter how long pulses you want to send.

on interrupter side add an extra 555 to send a train of high frequency pulses instead of a flat ON

on receiver side add an integrating circuit to recreate the flat ON signal

I've just molested 3 photocouplers from a dead PSU and managed to separate the LED from the photodiode on two of them. The inside was made from milky white epoxy so I had to do some careful sanding to allow more light through but I managed to fit it in a toslink casing and it works wonderfully. I'll use it to pull a pin in a schmitt trigger high along with a 10k pulldown and it should work without a problem.

[davekni]

Glad to read that you have a solution - clever idea.  If some day you want to purchase phototransistors, EBay is a reasonable option.  Search for "3mm phototransistor" and pick ones that have clear housings, not the dark gray or black ones.  Those dark cases are IR-pass filters for use with IR LEDs, so won't respond well to red light.  The clear ones will respond well to red even if labeled for 940nm or other IR wavelength.