Opto-Interrupter LED Forward Voltage

Your mistake was in using the word "error" in your report. You _should_ have said, "Within manufacturing tolerance." But, I suppose it's too late for that, and you'll have to write up the new spec, as Ian Sterling has recommended.

Also, you might have to review the existing designs, just to reassure everybody (including yourself) that it really _is_ a suitable sub. (I can't imagine it not being - 1.6 ~ 1.7V is well within tolerances for an LED.)

Sorry, Rich

Reply to
Rich Grise
Loading thread data ...

Recently an opto-interrupter we used went obsolete. The manufacturer suggested an alternate. I received samples and tested them and compared the specs and wrote a verification to approve them for our AML (Aproved Manufactures List) for this part number. Now the specs all match except some of the leakages and temperature specs are better on the new part and the forward voltage for the old one is 1.6v and the new one is 1.7v. I wrote for this one 'within margin of error'. The QC dept is calling me on it wanting me to define the margin of error. How do I define the margin of error on this spec? The 1.7v is at least defined at 10mA. The current the 1.6v was measured at is not defined. I'd hate to have to create a whole new part number and ECO all the products that use it to cut in the new part after using up our stocks of old parts. It would be alot of work for a minor difference in speccing.

Thanks

Paul C

Reply to
PaulCsouls

Meaningless question.

You need to determine if the new part will work with all your designs. It's quite possible (though not especially likely) that some poorly designed products might not quite supply enough current at 8% lower voltage.

I'd call this a fault in the original design (assuming that the required current is the same).

New part number, with a note about the change on the old one is probably the right way.

Reply to
Ian Stirling

It sounded better than 'close enough'.. How seriously do you a 'typical' value in a spec sheet with no max or min. Somebody grabbed a handful and did some testing. One guy's values rounded up, the other guys values rounded down.

Paul C

Reply to
PaulCsouls

I guess I'll have to got that route. I don't believe there is any real difference in forward voltage between the two parts. If I had to sort them into two piles by measuring the forward voltage there would be too much overlap to distinguish the two parts and no guarentee that the 1.7v (typical) would have the higher Max rating and the 1.6v (typical) would have the lower min (rating). But I can't prove it.

Paul C

Reply to
PaulCsouls

years ago we had a system of specifying devices using Company Spec numbers. Basically we took the suppliers data sheet and called it CSxxxx. If there were any particular parameters that could be relaxed, then this was amended in the CS spec. The we bought devices against the CS spec rather than manufacturer's data sheet. So if in your case, we could say that new devices were ok to our internal spec. Made extra work/time/cost, but seemed OK then. Too expensive now .... hth Neil

Reply to
neil

Unless specified otherwise, the Vf is forward voltage when driven by the nominal current for maximum CTR and this is usually 10mA. The range of current about the 10% band for maximum CTR is approximately 3:1, and probably in ratio 50:1 for a typical application, so you can see that this is not critical. Good design practice is to design for forward current at 1.25x nominal CTR,max to allow for LED output degradation with age. The 0.1V difference is figured into Vcc-Vf in most applications so that the error ratio of new to old current in the same circuit is -0.1/(5-1.6)=3% for a 5V circuit, and 1% for a 12V circuit. This error is one to two orders of magnitude less than the other tolerances driving the typical application circuit.

Reply to
Fred Bloggs

The 1.6v is an internal part spec. Our practice is to try second source our parts so we commonly loosen the specs to accommodate multiple parts on NEW part numbers. I could ECO the part number itself and do a 'where used' to track and verify that it doesn't effect any existing designs. I guess that is an option.

Paul C

Reply to
PaulCsouls

Thanks,

I'd bet the LEDs are all driven through 300 ohm resistors from 5 volts and the outputs are read from schmidt triggers. I guess it is possible that the LED could be being driven through some tiny resistor directly from a low voltage part and read directly by some gate where we might not get a legal low. I would think we would get failures with the 1.6v part and only increase the failure rate with the 1.7v part. I guess I'm going to need to check all the designs to see that no one's doing anything really stupid.

Paul C

Reply to
PaulCsouls

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.