is it possible to do this with 3v instead of 5v?

It may be. Different colours of LEDs have different forward voltage drops. Try the LED in your button on 3V with a 47R resistor in line and see if it lights.

 

Oh, the circuit in post #17 will have the LED on all the time.

When you are testing the inputs, are you reading the status from the software?

 

There is one thing I don't understand for certain: The breakout box: Does that provide 2V and is looking merely for a closure of the switch? Or are you needing to provide a 2V source for the breakout box to see a closure?

If you're providing a source then either of the below should work. Unless someone can tell me why it wouldn't:


I'll work out the other scenario where the Bob (Break Out Box) is providing 2V and needs 2V on the sense circuit to assume the switch is closed.

 

Looking at your drawing and having deleted all the superfluous details pin 1 simply needs to be grounded to indicate a closure on the BOB. In this case the 1N4004 is not necessary. There should be no need to match polarity; and the diode will introduce an approximate voltage drop of about 0.6 to 0.7 volts. If you don't mind an external power source solely for the LED I have a solution. Give me a few min's to bang it out.

 

OK, I'm making some assumptions here, but Input 1 likely has a pull-up resistor internally. When Input 1 is pulled to ground BOB reads it as a closure. Hence, SW1 will pull Input 1 low. The LED is subsequently powered from an old cell phone charger outputting 5V probably at some current anywhere from 200 mA to 700 mA (typical, but may vary). D1 is a Schottky diode because it has a lower Vf than a regular diode. The diode prevents the 5V from the charger reaching input 1. If Input 1 is high at 2V then back feeding it with 5V might be harmful so the diode will be needed. Since I'm not versed in Schottky diodes I'll let someone else recommend a good candidate for this situation. Could just be a simple 1N914 signal diode for all I know. But not likely. Anyway, here's my latest iteration of a solution to your problem:

 

Schottky diodes are foreign to me. So looking around I found this but I don't know if I fully believe the claim. It claims a Vf of only 1mV. But don't just take my word for it:




Source: https://us.rs-online.com/diodes/rectifier-schottky-diodes/

 

Can you read the breakout box input status on your application software to see if it reads the switch correctly?
And if you have a spare USB socket, that could be used to supply the +5V. A couple of CR2032 batteries will not last long.
For a test, connect your multi meter, (on mA) between the GND and a breakout box input to measure the current an operated switch draws. If the current is low, a Germanium diode could be used as i has a lower forward voltage drop so the input will be closer to GND when on.

Another test could be to wire a variable resistor between the input and gnd, then measure the max voltage across it when the input is read as on. That wau it tells you how much headroom you have.

This button would typically only be lit for one hour once a week maximum, how long do you think a single CR2032 would last in this scenario?

 

This button would typically only be lit for one hour once a week maximum, how long do you think a single CR2032 would last in this scenario?

Good question. I had some old 2032's that I wanted to fully discharge. I don't know exactly what their power level was but I connected some superbright white LED's to them and let them burn 24/7. It took a couple weeks for them to visibly dim. So dividing that into hours and assuming two weeks (14 days) 14 x 24 = 336 days. Almost a whole year. But these were already depleted garage door opener fobs. Brand new - I have no idea how long they'll last. You COULD set up a test with a new CR2032 and a superbright LED and see how long it runs until the brightness is lacking. But you may find yourself waiting a month for it to go dead. But yeah, you can replace the 5V phone charger with a single coin cell battery.

They also make CR2450's which are a lot bigger and will last - obviously - longer.

 

Good question. I had some old 2032's that I wanted to fully discharge. I don't know exactly what their power level was but I connected some superbright white LED's to them and let them burn 24/7. It took a couple weeks for them to visibly dim. So dividing that into hours and assuming two weeks (14 days) 14 x 24 = 336 days. Almost a whole year. But these were already depleted garage door opener fobs. Brand new - I have no idea how long they'll last. You COULD set up a test with a new CR2032 and a superbright LED and see how long it runs until the brightness is lacking. But you may find yourself waiting a month for it to go dead. But yeah, you can replace the 5V phone charger with a single coin cell battery.

They also make CR2450's which are a lot bigger and will last - obviously - longer.

Thank you.

I appreciate all your effort to help but adding an external power source unfortunately isn't an option, I would rather give up on lighting the button and use it unlit, I would loose a lot of value on my idea but it would still function. I know adding an external power source sounds like a plausible solution, and it is, even better than batteries since it wouldn't need replacement, but in this case, even though customers would appreciate it lighting up they just wouldn't accept the "hassle" for such a small simple button. They would dread another power cable, even a small phone charger.

Saltwater aquariums are massive, expensive, complex and rely on a delicate balance of chemicals to create a steady ecosystem that replicates the ocean to sustain fish and corals, anything out of parameter might not be able to be reversed and can cause a catastrophic result in a very short time. To make it thrive requires a huge amount of equipment and an enourmous amount of maintenance, cost and dedication. Any piece of cable, power socket, hardware that needs to be mounted or anything at all that has to be added to this complexity is almost unbearable unless its cutting down on the hassle.

Thats why I keep insisting on the batteries, I'm sorry.

 

Good question. I had some old 2032's that I wanted to fully discharge. I don't know exactly what their power level was but I connected some superbright white LED's to them and let them burn 24/7. It took a couple weeks for them to visibly dim. So dividing that into hours and assuming two weeks (14 days) 14 x 24 = 336 days. Almost a whole year. But these were already depleted garage door opener fobs. Brand new - I have no idea how long they'll last. You COULD set up a test with a new CR2032 and a superbright LED and see how long it runs until the brightness is lacking. But you may find yourself waiting a month for it to go dead. But yeah, you can replace the 5V phone charger with a single coin cell battery.

They also make CR2450's which are a lot bigger and will last - obviously - longer.

CR2450's sound great, I'm going to look for dimensions and add it in the CAD to see how it fits. Thanks for the tip.

 

Thats why I keep insisting on the batteries, I'm sorry.

It's your project. Don't apologize. You get to decide what goes into it and what doesn't.

 

OK, I'm making some assumptions here, but Input 1 likely has a pull-up resistor internally. When Input 1 is pulled to ground BOB reads it as a closure. Hence, SW1 will pull Input 1 low. The LED is subsequently powered from an old cell phone charger outputting 5V probably at some current anywhere from 200 mA to 700 mA (typical, but may vary). D1 is a Schottky diode because it has a lower Vf than a regular diode. The diode prevents the 5V from the charger reaching input 1. If Input 1 is high at 2V then back feeding it with 5V might be harmful so the diode will be needed. Since I'm not versed in Schottky diodes I'll let someone else recommend a good candidate for this situation. Could just be a simple 1N914 signal diode for all I know. But not likely. Anyway, here's my latest iteration of a solution to your problem:
View attachment 317466

So, if I switch to a 3v battery all I need is a 80 ohm resistor and a diode? If thats correct I just need to find the appropriate components, I would need help choosing the right ones.

These ok?:



https://www.digikey.com/en/products/detail/yageo/MFR200FBF73-80R/9136923

 

When I depleted my old used CR2032's I didn't use a resistor. I just let the LED's burn. And they never burned out. As I hear it - the 2032 has internal resistance and isn't capable of much current. Unless you want to control brightness you might not need a resistor. HOWEVER the LED inside the switch is touted as a 2.2Vf whereas my SuperBright White LED was 3Vf. So assuming you use a 3V battery (and for sake of argument we'll ignore the internal resistance of the battery) with a 2.2Vf LED you have 800mV (0.8V) to dissipate. ASSUMING you want 10mA. 800 ÷ 10 does in fact equal 80Ω. So, yes, 80Ω would be right. But the resistor you mention is 2 watt whereas you'd only need ( 3V x 0.01A = ) 30mW. A 1/8W resistor would be a whole lot smaller than a 2W resistor and is good to handle 125mW. Far above the 30mW you'd need.

As for the diode, the zener, I don't know if I trust that data sheet. It claims a 1mV forward voltage drop and I've never heard of such a low Vf. I'm hoping others will speak up and offer suggestions. I know they've already offered their opinions and I'm not one to say they are wrong; they collectively know a whole lot more than me. So I can't say for sure the 1N5407 would be the right choice. The two reasons for having it there is to prevent back-feeding voltage into BOB and in keeping the Vf as low as possible so as to not interfere with BOB's operation.

 

When I depleted my old used CR2032's I didn't use a resistor. I just let the LED's burn. And they never burned out. As I hear it - the 2032 has internal resistance and isn't capable of much current. Unless you want to control brightness you might not need a resistor. HOWEVER the LED inside the switch is touted as a 2.2Vf whereas my SuperBright White LED was 3Vf. So assuming you use a 3V battery (and for sake of argument we'll ignore the internal resistance of the battery) with a 2.2Vf LED you have 800mV (0.8V) to dissipate. ASSUMING you want 10mA. 800 ÷ 10 does in fact equal 80Ω. So, yes, 80Ω would be right. But the resistor you mention is 2 watt whereas you'd only need ( 3V x 0.01A = ) 30mW. A 1/8W resistor would be a whole lot smaller than a 2W resistor and is good to handle 125mW. Far above the 30mW you'd need.

As for the diode, the zener, I don't know if I trust that data sheet. It claims a 1mV forward voltage drop and I've never heard of such a low Vf. I'm hoping others will speak up and offer suggestions. I know they've already offered their opinions and I'm not one to say they are wrong; they collectively know a whole lot more than me. So I can't say for sure the 1N5407 would be the right choice. The two reasons for having it there is to prevent back-feeding voltage into BOB and in keeping the Vf as low as possible so as to not interfere with BOB's operation.

Ditching the resistor sounds great, less cost, less labor and less trouble. I'm giving it a go, if something goes wrong I'll reconsider. I have to wrap up this project.

As for the diode, I'll add it, no problem. I looked into a couple of 1N5407s elsewhere and the forward voltage are different from one from the RS website you posted:

At Digikey I found one with a 1V Vf:



https://www.digikey.com/en/products/detail/nte-electronics-inc/1N5407/11643467

At Mouser I found one with a 1.2 Vf:



https://www.mouser.com/ProductDetail/Diotec-Semiconductor/1N5407?qs=OlC7AqGiEDnqapYcRODsQA==

Are one of these what I need? If not, If I get the specs of the diode I need I can try to look one up instead. If a Schottky diode is better I'd prefer it but if you're not versed on them imagine me!

Thank you for your honesty Tony, If anyone would like to pitch in and offer advice on the diode I need please dont be shy, this is the final piece in the puzzle.

Thanks.

 

1Vf might be too high. And for a Schottky diode I'd think that must be extremely high. You'd be better off with a small signal diode. I don't have time to look up different diodes right now. That's why I hope someone else can chime in and say something from a solid knowledge base.

 

1Vf might be too high. And for a Schottky diode I'd think that must be extremely high. You'd be better off with a small signal diode. I don't have time to look up different diodes right now. That's why I hope someone else can chime in and say something from a solid knowledge base.

I read this:

https://www.allaboutcircuits.com/te...chpt-3/introduction-to-diodes-and-rectifiers/

It made me understand what they are and how they work. Its like a check valve, we use that in aquariums a lot. I understand now that I need a very low Vf because it only pushes 2V through the diode and a high voltage drop could compromise the functionality of the BOB. 1V Vf would cut it in half!

I took a look at the datasheet of the 1mv diode and its actually 1V:



I'm looking for a diode with the smaller Vf possible, I'll let you know once I got it, thanks.

Question: Can I solder one of these PCB diodes like a "normal" diode?



They seem to be the ones with the lower Vf.

 

I found this diode with a 340mV Vf, it has the lowest Vf of the through hole mounting type diodes I saw. Lower than this only the surface mount types. I looked at dozens of datasheets from diodes at the RS website that claimed around 1mv and they all were 1V.



https://www.digikey.com/en/products/detail/smc-diode-solutions/95SQ015/8021508

 

Albertoinbox Great job of researching your components. You have far more drive than I do. A 0.3Vf sounds like a good choice. If the cost is not prohibitive then it's worth testing out on your BOB. Y'know what they say in Australia; "Bob's your uncle." Though I have no idea what that means.

 

Albertoinbox Great job of researching your components. You have far more drive than I do. A 0.3Vf sounds like a good choice. If the cost is not prohibitive then it's worth testing out on your BOB. Y'know what they say in Australia; "Bob's your uncle." Though I have no idea what that means.

Will do, thanks uncle Tony. And thank you to everyone else that lent me a hand. I'll get back to let you guys know if it works.

 

I would just like to share with all of you that it worked perfectly! Very, very happy.

I'm very proud of you, wish I could help somehow but guess I'll just have to pass it on to someone along the way.

Thank you ALL for your time, knowledge and willingness to help others without asking nothing in return!!!