Fade on, Fade off Interior lights

This should be a relatively simple circuit to design/build. I used to do stuff like this back in school but forgot just about everything..lol

I will look through my old notes so if nobody else can post a suggestion know that I am on the case.

I am pretty sure this would work. I think...... What do you guys think?

Now another thing, is I would like it to stay on for approximately 20-30 secs when I get out.

I will have to look into it, not sure if that circuit will work or not, might need a resistor, and also will depend on current draw ( the sum of how many lights for the interior, or if only for one light the current draw of that) to determine length of time for fade. If there is a wattage rating for the bulb(s) let me know and I can use ohms law to determine current. Most likely only 3-5a each light.

As far as the lights staying on when you get out, most  newer vehicles have a delay as standard equipment. What kind of vehicle do you have?  You can use an off the shelf timer circuit and relay to do the delay but I need more info about your vehicle. IE: Is it a positive or negative switch for interior lights?

What you can also do is buy a few caps of various mfd's and use a 12v light bulb similar to the one in your vehicle and a 12v power supply or battery. Set up the circuit and see what happens when you turn the light on and off. Cheap and easy way to find out if it works.

Personally, I think you'd either need to go with a transistor controlled by an RC circuit, or use PWM to control the light.. both ways have good and bad points.

The circuit as drawn won't work the way you want it to.. you're on the right track with thinking about using caps and resistors, but you don't have a resistor in the circuit.  Do a search on RC circuits.. Resistance Capacitance. PWM is Pulse Width Modulated.

https://webphysics.davidson.edu/physlet_resources/bu_semester2/c11_RC.html

https://www.rwc.uc.edu/koehler/biophys/4g.html

Here's a page with more application specific information..

https://www.discovercircuits.com/L/lite-dimmer.htm

Personally, if you're interested in electronics then this one is _easy_.. an RC circuit driving a transistor is a pretty basic circuit - but you'll likely want to play with it on the bench.. Read through the links I provided.. if you don't have any interests in electronics, then find someone to buy a module from..<grin>

Any questions.. feel free to ask.

Jim

Good links Mad Scientists, your name suits you. You seem to always know the whacky stuff.. Those links should help with the "fade" circuit. I knew there was a resistor involved but it's been a while since I have done this stuff. And after seeing those formulas again I really don't miss it..lol

As far as the 20-30 second shut down delay I still think a timer/relay is the way to go.( adj from 0 to 60 or 90 seconds I think) DEI has one and it's inexpensive but may be others around too. Just to simplify things. Using a diode to prevent back feed we can trigger the timer with the appropriate polarity when the doors are closed and have it send the approriate polarity for the dome light to remain on and adjust for the time desired before it "fades" out with the RC circuit.

Keep us posted.

Thanks everyone, this is helping quite a bit.

"What kind of vehicle do you have?".................1986 Ford Ranger

"Is it a positive or negative switch for interior lights?"....................Positive (I never checked it) But I know how to reverse polarity with a relay, so I'm capable of making it work for this circuit if need be.

"Most likely only 3-5a each light"..................I am not sure, can I determine it with a DMM?

"The circuit as drawn won't work the way you want it to.. you're on the right track with thinking about using caps and resistors, but you don't have a resistor in the circuit."..........................Yeah, I know this circuit is not going to stay on for any period of time, it is just meant to fade on, then fade off. Would I need a resistor in there? Which of Ohm's or Kirchoff's laws do I need to determine the size of the resistor?

Sorry, I have so many questions, but I'd rather hear the right answer! I also have the Basic Installers Studyguide, I have read it twice but I am still unsure how to apply some of the laws to this application. If anyone can actually post one full equasion with a brief description, I could go by the example to determine my own. It would be greatly appreciated. BTW I liked those sites Jim, now I have them added to my favorites list.

Again, thank you for any, and all info provided!

"Which of Ohm's or Kirchoff's laws do I need to determine the size of the resistor?"

Sorry, let me correct myself..... 

Which of Ohm's law is needed to specify the correct resistor?

Which laws do I use to determine the amount of time that the light stays on? Kirchoff's?

I'm guessing that the diagram posted above would allow the light to fade off (the filament would be the bleed resistor that would allow the stored current of the cap to discharge).  However, it won't fade on, and it won't delay at all.

You can get time delay relays that would integrate with what is posted above to give you a delay, then a fade out.  I personally don't think the fade on is that necesarry.  I like instant light! 

I found out that the only bulb that it will control's specs are:

volts: 12

amps: 0.833

watts: 10

"However, it won't fade on, and it won't delay at all."

It wont? Even though the capacitor will be taking in some of the charge?

No, because the cap is in parallel with the light (resistor).  When power is applied the bulb will pull X amount of current and the cap will pull Y amount until fully charged.  The light will pull X amount regardless if the cap is there or not.

But, when power is removed the circuit changes (From a parallel circuit to a circuit in which the cap is the battery), and the cap becomes the power source.  As the voltage drains in the cap the light will dim.

I haven't tested this at all, but that is what I would guess would happen.  I may be wrong.

Time = RxC

C is the capacitance of your cap, R is the resistance of your bulb.  That should give you an approximation.

According to the information you provided the resistance of your bulb is ~14.4 ohms.  Your capacitance is 22000uF.

This would give you a time basis of 316800uS, or .316 seconds.  For a 1 second delay I would figure on using a 68000uF cap.

Okay, this is what I have come up with. The only problem I have is what resistor will go where the green symbol is, or even if I need one there. I believe I am going to need one that has a resistance of 14 Ohms so it will take 4 seconds to fully store, and to fully release. The plan is for it to have a delay of approximately 4 seconds after door shuts, then the relay will shut off. The 220,000 cap will then  fade until it is off in approximately 2 to 3 seconds.

What do you guys think? I know I will test it before I install it, but it looks good to me. If you guys have any suggestions, please let me know. And what is your input on the "Green Resistor"

Would the following circuit allow fade on as well? Logic tells me a cap is in a discharged state before the door is opened. Now once the door is opened a cap placed in parallel should initially shunt the circuit through the resistor  (hence no power will reach the light) and used with the light would allow a predetermined period of time before the cap charges and allows power to get to the light (with a fade I would assume determined by the "charge" rate.)

So when the door is closed and for whatever length of time decided via the timer cuts power and the discharge/rate will fade out the light. When the door is opened the above circuit should allow "fade" on if desired for whatever reason.

Am I on the right track in this thinking?

kymadan wrote:

---

Okay, this is what I have come up with. The only problem I have is what resistor will go where the green symbol is, or even if I need one there. I believe I am going to need one that has a resistance of 14 Ohms so it will take 4 seconds to fully store, and to fully release. The plan is for it to have a delay of approximately 4 seconds after door shuts, then the relay will shut off. The 220,000 cap will then  fade until it is off in approximately 2 to 3 seconds.

What do you guys think? I know I will test it before I install it, but it looks good to me. If you guys have any suggestions, please let me know. And what is your input on the "Green Resistor"

---

What I think will happen with this is:

Dome light signal will turn on, instantly energizing the relay, and almost instantly charging the cap (with no resistor).
As soon as the relay energizes the light will turn on, and the 220KuF cap will start to charge (light will not fade on)

Once the dome light signal is removed the 300KuF cap will keep the relay energized for a short period of time (but considerably longer then the light fade as the resistance will be about 5x greater).  Once this cap drains the relay will drop out, giving the 220KuF cap a chance to discharge in to the light, dimming it as it discharges.

I can't think of an 'easy' way to fade the light on with a cap, I would think a 555 timer or some other analog device (like mentioned in the links in the previous post) would be a better choice, but will obviously complicate everything.

Again, this is just my guess, I could be completely wrong.  Analog circuits really arn't my area of interest, I'm more of a digital guy!  :)

I can't help you with the green resistor value but regardless that circuit will not allow fade on. The relay is wired properly up to pin 87 which is connected directly to the light + after the cap/resistor network. The light will simply turn on via the relay instanly upon door opening. However if fade on is no longer a concern, this might work.

This circuit looks feasable to fade off but you stated earlier you also wanted a 20-30 second delay as well before fade out. This is where the timer/relay would come in handy. Once we have the RC fades straightened out (we will both need some more help with that) I can assist with a diagram to incorporate the timer delay if you desire.

BTW I am not refering to any of the diagrams posted when discussing (circuit), only about my thoery of using an additional cap/resistor placed in parallel with the light to allow fade on.

As far as fade off it looks like we are getting close, just need to know if the 20-30 second delay is still being considered before fade off. 

Well see, I decided against the 20-30 secs.... It would be a pain everytime you open the door it would stay on for that long. I decided approximately 6-7 secs is perfectly fine.

i know it will not fade on, I decided against that as well, it will come on immediately, but fade off. KPierson nailed it.

You lost me on your last post.  I think you are trying to slow the discharge rate to the coil of the relay.  However, like the filament in the lamp, the coil in the relay is your 'resistor'.  Once you select a relay you can measure the resistance of the relay and calculate the capacitance.  However, one thing to look out for is the drop out voltage of the coil.  Most 12vdc relays pull in around 12vdc and drop out around 6 (I'm guessing again).  So, once your cap discharges to 6vdc your relay will drop out (although it will continue to drain the cap). 

I would shoot for the 30 second delay, but I would throw in an extra relay that will kill the relay delay off part of the ciruit if the ignition is turned on.  That would be sweet.

The power rating of a resistor is simply that...a rating. Just like a lightbulb. That rating never changes BUT, if you exceed it's rating it will get quite hot and probably burn out. Put as with wire gauge, it will never hurt anything to go extra large, it can only help.

Ok next, I know you have simplified what you wish to do but now I am still curious since I put some thought into your initial plan.

Does anyone know if using a cap and resistor in parallel with the light allow the light to fade on. As I stated I base this theory on the princible of a cap. It will block DC. So it would seem logical that the discharged cap would shunt through the resistor, This shunting or short will not allow power to reach the bulb. As the capacitor charges the light should fade on. One the cap is fully charged it should have no other effect on the circuit or light.

This sound like it would work anyone?

Oh wait. To answer my own question I realize this won't work. With a resistor it will never see a short and the current would simply branch through the resistor and light.

Now if we put a cap by itself in parallel I imagine this would have no effect as well as it would charge instantly hence no fade.

Out of curiosity does anyone know a simple RC circuit that would case a light to fade on?

Kudos to you if you can figure this out haemphyst, but if it's possible, I know you will.

I am severly out of practice with circuit design and really haven't had much use for it the past several years until joining here. But I have a feeling over time it will come back ,however so slowly and with plenty of mistakes (re-learning/learning process) along the way.

I racked my brain for a while with the "fade-on" and after several failed attempts have given up. Other then some kind of IC or module I can't for the life of me determine how a simple RC can do this.

Thanks and good luck.

Still thinking about this...I can except failure, but dislike it.

There must be a way, although I don't know how simple, to build an electronically controlled variable resistor (potentiometer). It would need the input voltage to decrease the output resistance over time, until zero resistance (or very close to it) is accomplished and the light fades on.

The question remains...how to do design this using nothing but resistors, caps and perhaps diodes/relays. hmmm

I think I got it guys...... brb

Okay guys, this is what I come up with......

Now I have heard of the time delay relays, if these relays were .2 second delay then it would take 2 seconds for the full fade effect. The reason those resistors are in that order and size is because....

1. Total of 150 Ohm resistance, Use regular relay.  

2. Total of 135 Ohm resistance, Use a  .2 second delay relay

3. Total of 120 Ohm resistance, Use a .2 second delay relay

4. Total of 105 Ohm resistance, Use a .2 second delay relay

5. Total of 90 Ohm resistance, Use a .2 second delay relay

6. Total of 75 Ohm resistance, Use a.2 second delay relay

7. Total of 60 Ohm resistance, Use a .2 second delay relay

8. Total of 45 Ohm resistance, Use a .2 second delay relay

9. Total of 30 Ohm resistance, Use a .2 second delay relay

10. Total of 15 Ohm resistance, Use a .2 second delay relay

11. There is no resistor, it should allow light to be on full brightness. Use a .2 second delay relay

The regular dome light I have is 14 ohms, so this setup will have the affect of having 10 lights in a series circuit, and every .2 seconds 1 bulb is removed.... in turn making it brighter! The capacitor in for the sudden surges of the next relay turning on and to cause a fade effect to take placewhen it is shut off, it will last approximately 3 seconds .

Master 5 wrote

"Out of curiosity does anyone know a simple RC circuit that would case a light to fade on?"

This was the answer, but it is just not a good way to go about it.

hi,

for a pwm circuit device, this piece may work. https://www.vellemanusa.com/us/enu/product/view/?id=350605. now to control the input to allow a ramped voltage to trigger it, i'd have to check my notes.

mark

Could someone tell me where they are drawing up the circuits to post here?.. I see cardomain.com logos, but I haven't the time at this moment to go there and look around.

A simple RC circuit to control  an interior light would be one transistor, one resistor, and one capacitor... if someone clues me in on the drawing, I'll draw a picture. Basically the RC circuit feeds the base on the transistor.. the transistor controls the light.

Jim

I myself was kinda thinking something similar with the relays. Now all practicality issues aside this is all for fun and theory, not something I think anyone would actually use.

The reason I didn't bother trying to design the "fade" with an array of standard relays is that I think it would brighten in stages as each relay is energized, not really producing a smooth fade. The other factor is feedback from the relay coils, but I imagine diodes could correct that.

Now as far as transistors it's been a long time since I have messed with any. For the masters test all I needed to know regarding transistors is how to identify a "PNP" vs. "NPN". Easy enough. The main uses I have had for transistors in the past was when I needed to energize a relay coil using a device with too week an output. A resistor/transistor easily accomplishes this.

Other then that I would be interested in learning how using a transistor can accomplish this "fade on" we are talking about. Is there a way anyone (mad scientisits, lol) can explain this. If its too complicated don't bother but I would think it not that difficult.

Thanks

Oh and as far as drawing my circuits I need some help also. I had a bunch of schematic symbols I collected over the years but when I need something different I will do a google search or do it by hand in paint program. usually plain black and whites and it is very limited.

Not very pretty but it works. However I would much prefer a way to have them neat and in color as I see others post. The way I have been doing it is very time consuming as I have to cut and paste each componenet and draw all the lines manually.

Thanks

Theory of operation:

When 12vdc is applied the resistor slowly charges the capacitor.  As the cap slowly charges the voltage at the resistor/capacitor junction slowly rises.  The 741 op amp is used for its super high impedance (doesn't put a load on the circuit).  The op amp is configured in a voltage follower configeration, meaning whatever voltage goes in comes out the output.  In between the output of the op amp and the feedback there is a transistor circuit that is designed to simply amplify the low current signal coming from the op amp.  With the transistors in this configuration they will amplify the current of the applied analog voltage.  Be careful and note that the emitters are drawn in a strange way (I mixed it up the first time I used that circuit).  Because the transistors are inside the feedback loop they have virtually no effect on the operation of the op amp.  What you have coming out of the op amp is a voltage that slowly rises to near battery level (it will never actually reach battery level though, one downside to the op amp) with enough current to drive your light.

So, in the end, I used an RC circuit to drive a transistorized buffer that will dim the light on.  Now, interfacing that with the rest of the circuit is going to be another challange.

Oh, and I simulated this in Electronics Workbench, so it should work.

I didn't put part numbers on the transistors either, the top one is an NPN and the bottom one is a PNP.  Depending on the current requirements you could go with a 2n3904/2n3906 (if under 200mA).  Over 200mA you'll have to find bigger transistors.

Impressive, might actually inspire me to get the parts and try it out. Thanks

As far as mspaint I could use some colors but the diagrams would still look like stick figures, just in color. For circuits like above this is fine but I kinda like the "cartoony" looking ones for more basic stuff. I am very artistic, but can't draw that well. Perhaps something to do with being a lefty.

Hey guys, do you think this would work okay? I  I know it will need to be adjusted here and there, but for the most part it looks good to me. Basically, what it is supposed to do is take approximately 6 seconds to fully fade on. (Thanks KPierson for the diagram!) Then when the domelight wire ceases to  power the circuit, it will drain the 300,000 uf cap and the 220,000 uf cap, and will fade off for approximately 7 seconds. Which in turn is the fade on-fade off circuit I have been trying to figure out.

Now as far as trying to get it to stay on for a couple seconds, it will do that as well approx. 4 seconds. Which will be fine. I have a switch inside the truck to keep it on longer if I need to.

Looks good to me. There will also be a slight v-drop through the diodes barrier voltage (if I remember correct it's .7 for ge and  .3 for si) but without actually building/testing this circuit I don't have the abilty to predict the time length/brightness loss with any great accuracy.

The parts are not expensive so you might want to build it and post the results. I am curious myself.

Good luck

I already have 1N4001 Diodes rated at 1A 50V, I believe this will work... but I am going to ask anyway to be sure, So will this work for my diode in my last schematic? My light is only rated at 0.844 amps.

master5 wrote:

---

Looks good to me. There will also be a slight v-drop through the diodes barrier voltage (if I remember correct it's .7 for ge and  .3 for si) but without actually building/testing this circuit I don't have the abilty to predict the time length/brightness loss with any great accuracy.

The parts are not expensive so you might want to build it and post the results. I am curious myself.

Good luck

---

Just wondering, but ir I put two diodes in paralell with each other, will that drop down to a .35 for ge... and .15 for si? Or will it double to 1.4 for ge, and .6 for si?

No, there is no way to decrease the v drop. If you wired 2 or more diodes in series the voltage drop will add. If you wire 2 identical diodes in parallel the v drop will remain the same. However, the current will divide up and the power handling will double, so if the current the diode needs to pass is exceeding it's rating, and you don't have a larger one, you can parallel them to get the same result. Otherwise the diode will burn up.

The concern about the v drop during forward bias is probably not that big a concern, as it is less then 1volt for a si, and less then 1/2 volt for a ge. However, since this circuit as some resistance in it at all times, the added v drop from the diode will make the bulb even dimmer. Decreasing the resistor value will help, but then you have more current. In that case make sure the diode(s) can handle it.

There are numerous versions of how to fade the interior lights all over the web, here's one of the simpler versions 

https://sound.westhost.com/appnotes/an004.htm