I am designing a circuit that isolates the 5v logic/control circuit output with an opto-coupler (NTE3040). The 5v logic activates the LED in the chip and internally forward biases the base of the transistor, allowing 12v to flow on the other side; isolated, simple. Just keep the grounds separate in this case. But sometimes the chip WILL NOT OUTPUT.
Since we are prototyping, it takes a while for us to figure out that the brand new chip is garbage, and that's why the circuit doesn't work. Since I have to drive 50 miles to get the NTE3040's, I am curious if a 50% failure rate (2 out of 4) is normal and should i just get a bucket full and be prepared to throw half of them out? I haven't had any trouble with the other chips, but I was wondering if there was something they are sensitive to? or maybe WE are the ones destroying the chips faster than we can get them in an active circuit. Is there perhaps a manufacturer that might make a more reliable opto-coupler for the purpose?

 

I am designing a circuit that isolates the 5v logic/control circuit output with an opto-coupler (NTE3040). The 5v logic activates the LED in the chip and internally forward biases the base of the transistor, allowing 12v to flow on the other side; isolated, simple. Just keep the grounds separate in this case. But sometimes the chip WILL NOT OUTPUT.
Since we are prototyping, it takes a while for us to figure out that the brand new chip is garbage, and that's why the circuit doesn't work. Since I have to drive 50 miles to get the NTE3040's, I am curious if a 50% failure rate (2 out of 4) is normal and should i just get a bucket full and be prepared to throw half of them out? I haven't had any trouble with the other chips, but I was wondering if there was something they are sensitive to? or maybe WE are the ones destroying the chips faster than we can get them in an active circuit. Is there perhaps a manufacturer that might make a more reliable opto-coupler for the purpose?

Please share your schematic so we can see what's going on. High failure rates are unusual in my experience.

Show us your schematic and we'll have a better idea if you're doing something wrong or just terribly unlucky.

 

I have the circuit posted here: https://forum.allaboutcircuits.com/...ounter-still-skips-state.146742/#post-1253959

 

I have the circuit posted here: https://forum.allaboutcircuits.com/...ounter-still-skips-state.146742/#post-1253959

I don't see current limiting resistors for the LEDs. Don't know the output characteristics of the 7402, but that looks suspicious to me.

 

The input of the opto is an LED which needs current limiting.
If there's no input resistor for that, then that's likely what zapped them.

 

Now I'm wondering if it's the opposite problem... or maybe I'm just misreading this datasheet:

http://www.ti.com/lit/ds/sdls027/sdls027.pdf

It looks like maybe this thing can't source more than 1mA??? If so, it's possible the optos aren't damaged at all - they're simply being under-driven to a point where it's surprising any of them ever work.

Hopefully someone who's better at reading datasheets can help me make sense of this.

 

There are 500 ohm resistors between the NAND gate and the opto-couples limit the current to a maximum of 10mA if I have done my math correctly. Actually the datasheet shows a (1.18 to 1.5)Vfb for the opto-couple so that brings current to around 7.6mA, which the 7402 should be able to handle easily.
The ones that work were no problem, but the other ones never turned on. I was just wondering if there is a better manufacturer or if this device is inherently unreliable. I think I was just unlucky as ebeowulf17 pointed out.

 

The question that must be asked...

Where did you get them?

 

There are 500 ohm resistors between the NAND gate and the opto-couples limit the current to a maximum of 10mA if I have done my math correctly. Actually the datasheet shows a (1.18 to 1.5)Vfb for the opto-couple so that brings current to around 7.6mA, which the 7402 should be able to handle easily.
The ones that work were no problem, but the other ones never turned on. I was just wondering if there is a better manufacturer or if this device is inherently unreliable. I think I was just unlucky as ebeowulf17 pointed out.

It would be easier to troubleshoot and brainstorm if we had an accurate schematic! ***EDIT: Disregard my snarky comment about inaccurate schematics. I had overlooked the second schematic in the referenced thread.

What are you trying to do with the 12V output? How much current do you need through that side of the opto?

Depending on whether I look at the parts spec'd in the schematic or the NTE part number you mentioned earlier, I find transfer ratios ranging from 100% (7.6mA If = 7.6mA output) all the way down to 20% (7.6mA If ~1.5mA output.)

Also, maybe it's different with optos, but usually you'd put an NPN on the low side, not the high side. In other words you'd connect 12V supply -> load -> NPN -> ground. I'm not sure what kind of output to expect from an NPN opto doing high side switching. Again, I'm not sure, but I have serious doubts about this schematic as it currently sits.

Again though, the real question here is what you're connecting the 12V side of this circuit to.

 

Doh! Sorry, I was looking at the wrong schematic from that other thread. The second schematic is a much better representation of what you've described here!

Looking at the correct schematic, I see that the opto LEDs have been re-arranged, and it looks good as far as I can tell. The gates are sinking, not sourcing current, and the resistors seem fine.

That still leaves the question of how much current you expect to run through the opto 12V side. Powering a fan directly through that opto looks dubious at best. The current transfer ratio is way too low, and you're trying to use it as a high side switch.

I think your optos are/were fine, but you need something in between them and the fan to do the switching. Selection will depend on the size and current requirements of the fan. Flyback diodes are probably in order as well.

 

Ok, one last note and then I'll leave this alone until someone else replies, I promise:

I wasn't sure about high side vs low side v switching from the opto, so I did a little googling. Nothing rock solid, but I did find a few threads with some discussion which all seemed to indicate it was roughly equivalent, aside from maybe some differences in switching speeds:

https://electronics.stackexchange.com/questions/170839/optocoupler-high-side-vs-low-side-switch

https://electronics.stackexchange.c...pto-isolator-output-from-collector-vs-emitter

So, probably safe to ignore my comments on that potential issue. Now we're just down to motor current and the need for something acting as a buffer or amp between the opto and the motor.

 

The opto coupler is a logic device and not a power device so you certainly need a power control transistor to switch the fan power. So now you need to find a transistor that will carry the fan load with a gain high enough to switch on with the current that the opto isolator can provide.