Opto coupler from mice

[mcs51mc]

I tried to re-use the optocouplers from an old mice in a slot car track as lap counter.
For the diode part no problem +5V with 680R should be ok .
But the transistor is a bit stubborn. What is the output type af that thing open collector, TTL, PNP, PNP, ... ...??? :cry:
Anyone used such things and have drawings to share with me?
There is no part number of any kind on it. Just the letters "EL" on top of it.

Another consideration:
I search the web for information and found several sites describing the functionalities of PS/2 and serial mices but no hardware . They talk about 2 square pulse signals 90° shifted to detect left and right movement of the encoder wheel. But how can one make two signals out of one diode and one transistor :? ?

Thanks a lot ...

 

[Nigel Goodwin]

I tried to re-use the optocouplers from an old mice in a slot car track as lap counter.
For the diode part no problem +5V with 680R should be ok .
But the transistor is a bit stubborn. What is the output type af that thing open collector, TTL, PNP, PNP, ... ...??? :cry:
Anyone used such things and have drawings to share with me?
There is no part number of any kind on it. Just the letters "EL" on top of it.

They are usually open-collector, you just need a pull-up resistor to 5V.

Another consideration:
I search the web for information and found several sites describing the functionalities of PS/2 and serial mices but no hardware . They talk about 2 square pulse signals 90° shifted to detect left and right movement of the encoder wheel. But how can one make two signals out of one diode and one transistor :? ?

They don't, they have two arranged as a quadrature encoder, have a look at https://www.electro-tech-online.com/custompdfs/2006/02/00519c.pdf which may be useful?.

 

[mcs51mc]

They are usually open-collector, you just need a pull-up resistor to 5V.

That was my first thought also but with a 2K2 resistor to +5V I have no change in output voltage when placing something between diode and transistor :cry:

They don't, they have two arranged as a quadrature encoder, have a look at http://ww1.microchip.com/downloads/en/appnotes/00519c.pdf which may be useful?.

What do you mean by "They don't" no use of quadrature signals or only one signal out of one transistor? I can assure you that I see only one component on each side of the encoder wheel. One diode and one transistor, so generatig quadrature signals with that looks impossible to me. But that quadrature thing is the least of my concern as long as it doesn't involve special components. Too bad there wasn't any drawing of the opto coupler in that application note :cry:

 

[Nigel Goodwin]

What do you mean by "They don't" no use of quadrature signals or only one signal out of one transistor? I can assure you that I see only one component on each side of the encoder wheel. One diode and one transistor, so generatig quadrature signals with that looks impossible to me. But that quadrature thing is the least of my concern as long as it doesn't involve special components. Too bad there wasn't any drawing of the opto coupler in that application note :cry:

Your mistake is assuming it's a 'transistor' at the receiver side, it's not a single transistor - it's probably two in a single encapsulation.

I've just taken a mouse to pieces, the sensors have three connections, the centre one goes to ground, the other two (the outputs of the two photo-transistors?) go directly to the chip. The other side of the slotted wheel is a single IR LED, which shines through on to the double receiver.

 

[mcs51mc]

You're right when you say that it's not a standard opto transistor, shame on me .
The center pin is connected to +5V (not ground like you wrote :shock: ) and the other two goes directly to a Zilog Z8631 704 chip.
The (preliminary) datasheet for that chip says that the four pins used by both opto-transistors have built in voltage dividers (25K pull-up / 7.5K pull-down), the input levels are designed for connection to the emitters of the opto-transistors and switch at a voltage level of 0.4 VDD. So far for the datasheet.
How do I interface these opto-transistors to a standard 8051 µP or HEF4093 NAND Schmitt trigger? Make a resistance divider 25K / 7.5K and connect output of opto-transistor and input of the 8051 or 4093 to the center point of the resistance divider? Is that the way to go?

Another consideration: Pins 11, 12 and 13 of the Zilog are used to drive the LED. When I place an oscilloscoop on these pins I see a 5kHz square pulse signal. Do I need the 5kHz because there is a demodulator in the opto-transistors or something like that. Any idea why they need the 5kHz??

Thanks a lot

You can find the datasheet for the Zilog at www.AllDataSheet.com

 

[Nigel Goodwin]

You're right when you say that it's not a standard opto transistor, shame on me .
The center pin is connected to +5V (not ground like you wrote :shock: ) and the other two goes directly to a Zilog Z8631 704 chip.
The (preliminary) datasheet for that chip says that the four pins used by both opto-transistors have built in voltage dividers (25K pull-up / 7.5K pull-down), the input levels are designed for connection to the emitters of the opto-transistors and switch at a voltage level of 0.4 VDD. So far for the datasheet.
How do I interface these opto-transistors to a standard 8051 µP or HEF4093 NAND Schmitt trigger? Make a resistance divider 25K / 7.5K and connect output of opto-transistor and input of the 8051 or 4093 to the center point of the resistance divider? Is that the way to go?

For a start I would try them with a resistor to ground, and see what results you get - once you know how they perform you can then start designing an interface.

Another consideration: Pins 11, 12 and 13 of the Zilog are used to drive the LED. When I place an oscilloscoop on these pins I see a 5kHz square pulse signal. Do I need the 5kHz because there is a demodulator in the opto-transistors or something like that. Any idea why they need the 5kHz??

No idea, I should try it without first, and see what happens?.

 

[mcs51mc]

There are two types of opto-units. Both have same dimensions but complete other functionality.

The first one I tried was in combination with a Logitec LSC507804P chip. The opto unit used with that chip has one connection to +5V, another to ground and one to the Logitec chip. How Logitec can determine in which direction the mice moves is still a mystery to me. Some Logitec engineers reading this, please come in... ... :wink:

The second one was in combination with the Zilog chip described earlier. Here indeed one connection to +5V and two to the Zilog chip. Replacing the Zilog chip with a 1K resisor to ground works fine.

 

[Nigel Goodwin]

The first one I tried was in combination with a Logitec LSC507804P chip. The opto unit used with that chip has one connection to +5V, another to ground and one to the Logitec chip. How Logitec can determine in which direction the mice moves is still a mystery to me. Some Logitec engineers reading this, please come in... ... :wink:

Try scoping the output to the chip, the processing might be done in the actual opto unit, and a serial data signal sent to the chip?. Compare it to the signals from the other mouse!.

 

[mcs51mc]

Try scoping the output to the chip, the processing might be done in the actual opto unit, and a serial data signal sent to the chip?. Compare it to the signals from the other mouse!.

My curiosity already forced my to put a scoop on the chip input and :shock: :?
With no movement of the mouse , the signal is constant low with execption of two short spikes at regular time interval. Moving the encoder wheel result in constant high signal with a multitude of spikes to ground.
Like you said it's kind of a code to the chip. No use at all for me; I just needed some optocouplers remember ... ...
I can use the opto-units from the mouse with the Zilog chip
Thanks for your guidance on this issue.