How many IR receivers can I wire in parallel?

If you pair a red 50mW 5mW laser with a TSOP4838 you get more than 100 feet range.

CAUTION when using Lasers !

Maybe look into lenses

image

What does the data sheet for the receiver say about maximum current related to output voltage? That ought to give some idea about the one and common pull up resistor.

If you used the TSOP38238 I would estimate

internal 30K
"absolute max" current of 5 ma.
3.3v

Assume 4 ma usable

R = 3/i = 3.3 / 0.004 = 825 ohms

30,000 / 828 ~ 36 So I would guess 25 to 30 would be a reliable "fan out"

If you needed more, you could simple add a Mosfet to each and / or a multiple gate logic chip.

Only the bare spec ... 5mA
But you have to respect the output voltage spec as well. It's stated as 100mV maximum at 0.5mA output current. When you apply that condition, the maximum number of allowable units in parallel comes down somewhat.

Parallell connecting true open collector npn outputs ought to allow for quite a few outputs but any internal pull up must be removed. Only one common pullup should be used. A kind of bus.
What do You think?
I haven't read the spec for precise knowledge about numbers....

I don't see any harm in having multiple pull up resistors. The expectation is that inactive receivers (ones that didn't see any signal) won't have any effect since it's still a wired "or" arrangement.

The only risk I can think of, is that the manufacturer has never considered this use case, and so the block diagram presented might omit some detail that would kybosh the whole thing.

If I was making this professionally, I would contact the manufacturer, explain my application and ask for advice.

However, V/I curves could be obtained experimentally to gain enough confidence. When you do it that way, the results only apply to the batch that you tested.

Btw: IR Modules-Electronic Goldmine
(Cheap ir modules)

@aarg
I know You know ohms law. I see the effect of many distributed pull up resistors connected in parallell will end up in a to low resistor value between Vcc and the common output rail, overloading any sensor output trying go low. Stripping the sensor from all pull ups and applying one well calculated resistor could reach the maximum number of sensors.

The resistor is internal, it can't be removed. Let's see... 5V-100mV = 4.9V. 4.9V/0.5mA = 9800 ohms. 30k/9800 ~= 3. So you can parallel 3 units if you want to respect the spec sheet. Now I'm glad I did the math...

I'm thinking that 100mV is already a ballpark safe figure for a reliable digital LOW signal. I think we all know that VceSAT increases with Ic.

Bastard design.... I didn't pick up that.

A germanium diode can be used as an OR gate keying to GND.

How many is "many"?

Germanium? Are they still available? Joking! Why not Schottky?

100 feet is alright I guess. I was hoping for more around 300-500.

Also I did look at lenses and they are very expensive.

Oh yes, I have 2,000

:nerd_face:

I said more than 100 feet, haven’t try fo a maximum.

I'm in the middle of building a combat robot with Arduino electronics and it's been a cake walk compared to what I think this project is going to be.

Are You pushing for making private business here? Joking!

I'll have to test myself then!

If I got rid of them, I wouldn’t be able to sleep, security blanket.
:sob:

2 Likes