What happened to the photo in post #105? I was going to compare the trace layout and component silk screen (they are not to the same scale in the PDFs) with the photo.

Ken

 

On the RoHS issue...this is only important if you are making and selling you detector commercially in countries where RoHS compliance is required. If reliability is your concern then no RoHS components or solder. The military and many medical products will not use lead-free assembly because of its potential long-term unreliability.

There should be no operational problem using the LM111 instead of the LM311...it has a much better temperature range.

Ken

 

why not try 4066

 

Chuwudi,
4066? How would you use that? Have you read this thread from the beginning?

payal.kellyyy,
Op amps are not good comparators. Texas Instruments has automotive-grade single comparators in SMD packages: http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=296-24702-1-ND]

Ken

 

Dear KMoffett,

Kindly let me know what is the main difference in functionality of LM111 and LM211 what I figured out from datasheet is the difference in working temperature range .Apart from temp range effect is there any other difference between LM111 and Lm211 in terms of electrical characteristic .

As you saw in the data sheet: "The LM211 is identical to the LM111, except that its performance is specified over a −25˚C to +85˚C temperature range instead of −55˚C to +125˚C. The LM311 has a temperature range of 0˚C to +70˚C."

Can you please explain in detail why this OpAMP NCV2904 as a comparator is not a good idea to implement .What all failures can be associated with the same .

"Op-amp [as a] voltage comparator" :http://en.wikipedia.org/wiki/Comparator#Op-amp_voltage_comparator

Ken

 

I see that Digikey has my original referenced IC as "NON-STOCK". Use this:http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=296-17493-1-ND

http://focus.ti.com/lit/ds/symlink/lm211-q1.pdf According to Figure 12 on page 10, just short pins 5 & 6 together.

ken

 

Offset balance is to correct for very small differences in the input currents. This is of no importance in your circuit. Yes, the emitter output should be grounded. The rest of the circuit should be the same.

Ken

 

It's not possible to predict "all" possible causes of failure. Even to predict the most likely causes of failure, you need to be able to describe as much as you can the circuit and it's operating environment.

Failure modes and operating limitations of all components.
Physical construction of the circuit and its attachments.
Interactions due to failures of, or anomalies in, external attached systems.
Operating environment, including temperature extremes, humidity, vibration, electrical...
Installation errors.
Operator errors.

And most important, what are the consequences of any failure of your device...none?...minor inconvenience?...major costs?...loss of life?

Ken

 

And mice, mice can chew things and cause failures.

 

And mice, mice can chew things and cause failures.

And insects!

Ken

 

Since I can't directly evaluate your design, installation, or environment, I can only make a guess. The most likely item to fail, long term is the 10uF electrolytic capacitor. These are prone to drying out, especially in high heat conditions.

Ken

 

Also, because of the large physical size, an electrolytic capacitor would be more prone to failure due to physical vibration. You might want to look at SMD tantalum capacitors like this: http://www.vatronics.com/tancap/smdtantalumcapacitors.pdf
And, maybe bump R1 up to a 1/2W

Ken

 

Dear KMoffett,
I am already using this tantalum capacitor in this project by using this capacitor is there still a chance of capacitor to get affected by vibration test?

A surface mount capacitor would be less prone to vibration failure than a radial or axial lead electrolytic.

Also let me know what all can be the reasons for failure of R1 and what is the purpose of increasing the wattage to 1/2W .

R1 will have to take the power dissipation for the clamping Zener in case of a "load dump" or spike. Since a little bigger might help and the cost is insignificant (unless you are going to make thousands of these) I suggested it.

Ken

 

Googled:unused opamp inputs

http://www0.fh-trier.de/~berres/Datenbl%E4tter/TEXAS/sloa067.pdf Section 4.2.1

http://www.maxim-ic.com/app-notes/index.mvp/id/1957

http://www.analog.com/static/imported-files/rarely_asked_questions/RAQ_unused_op-amp.pdf

Ken

 

I don't want to try to track back through the thread (137 posts) to find it. Please post the schematic with current component values.

Ken

 

In your attached pdf circuit there is a 5 sec. water-detected-indication during turn-on. After that, there is a near instantaneous water-detected-indication when water is detected. The water-detected-indication remains on during the presence of water. The water-detected-indication remains on for 5 sec. after the water is removed. I can see shortening the start-up delay from 5 to 3 seconds. But, why keep the 5 second delay after the water is gone? Just remove one of the 4.7uF capacitors and have a 2.5sec delay at both start-up and a 2.5 sec delay after water is removed.

Can explain how you see the operator using this device in the work-place.

Ken

 

5 sec delay after water is gone is the requirement of the customer what they want is just to reduce intial on time after circuit is detected to 3 secs from 5 secs.Cant we control the time delay after circuit is detected and after water presence is detected independently?

 

OK, this is an "in my mind" modification. I won't have time to test it until Monday.
R3, R4, C4, and Q2 (P-MOSFET) are added. C1/C3 are normally charged through R1/R2. These set the two 5 second delays in the previous circuit. With the new components, when power is first turned-on, Q2 is turned-on and puts R3 in parallel with R1/R2. This shortens the turn-on indication time. C4 charges through R4 in about 3 seconds, turning-off Q2 and removing R3 from the delay circuit. When water is sensed, and removed, you would have the 5 second delay before the indicator turns off.

Ken

 

R3 should be ~1MΩ. Since timing will depend on the spec's of Q2, C4 and R4 would have to be determined experimentally. I'll bench test it with a BS250 P-MOSFET.

Ken

 

Bench test gave me these values. R3=1MΩ, R4=1.2 MΩ, C4=1.0uF (C4 was a monolithic ceramic), and Q2=BS250. Delays may vary a little with component tolerances.

Ken