Basic Momentary 2-mode driver

[lonesouth]

Here is a first crack at a momentary two mode driver. The switch will activate a transistor to allow the second AMC7135 to parallel the first and allowing 700ma to the LED. when the switch is released, output drops back down to 350ma. Let me know what you guys think. I am not sure I have the LED hooked up correctly, but this first attempt is to attack the concept. Thanks!

admin note: I realize this is specifically referring to automotive applications, however, I think the concept can apply to more than just 12v systems and therefore have posted in the flashlight-electronics subforum.

 

[jeffosborne]

Hi Lonesouth! Your uncertainty about the LED hookup is well founded. Using the 7135, you want to attached the + side of the LED to the supply voltage (5v) and then the output(s) of the 7135 attach to the negative side of the LED. Also, I do not think you need the transistor. If it were me, I would give both 7135's the +5v and let the switch connect the second 7135's output to the LED. This assumes that the switch is good for 350ma, so maybe it is not. Also, connecting the base of the transistor directly to 12 volts is not good - it needs a resistor between. But anyway I think you have a good idea here, I like lights with various levels. What kind of LED are you going to drive?

Cheers,
Jeff

 

[lonesouth]

the idea is for brake lights that traditionally have two filaments in the bulb. The 12 volt line is actually what gets switched. The final circuit will have 2 12v sources, one for parking lights, one for brake lights. I figured I could either use a transistor, relay or two LM7805s. Figuring that the transistor would provide the least heat and the longest time between failures. The transistor listed claims to be good to 75v Vb and 6v Ve. I'm planning on running an XR-E P4, Q5 or R2. Need to get it installed behind the lense and see which one I need.

Thanks for the feedback!

 

[lonesouth]

Take 2: I added a diode to prevent back feeding the brake circuit and moved the transistor to reflect the actual opperation of the 7135

Edit: heat disappation in the LM7805 will be around 5w-6w so plan on mounting it in a 5w heatsink, such as this one.

 

[heavy_duty]

Hi Lonesouth

I'm not sure the circuit will work as connected. If you reverse the transistors' C and E connections, ( C to pin 3 of LM7805 and E to VDD of AMC7135, it will work as intended. Oh, and put a resistor in between the transistor B and the brake switch (anywhere from 100k to 500k should be fine).

 

[Mr Happy]

This is good as a "thought experiment" for learning about circuits, but you should not think about actually installing anything like this in a car. The brake lights in particular are a critical safety system and if you change them from what the manufacturer installed you will be violating various vehicle codes and rendering yourself liable to prosecution, fines and other sanctions.

Keep these experiments on your test bench and be safe!

 

[lonesouth]

Take 3:

I moved the transistor per heavy duty. Also, the transistor claims to be good to 75v on the base, so I'm not sure what use the resistor is on the signal in. Please do enlighten me on this.

Mr Happy, I certainly understand that this is an experiment and it in no way is going on my daily driver. These brakelights are what got me started down this path. I just can't stomach paying US$150 for something I can build for less than $10. I understand that my design is likely less efficient, but it should still be far more efficient than the incandecent bulb it replaces. My rig sees mostly trails and a few events, but rarely sees the road. It is a worthy disclaimer about homebrew use in daily traffic. Especialy with something like brake lights!

 

[heavy_duty]

The 75V you refer to is collector to base voltage which is quite different than 75V on the base...

In normal operation (when transistor is on) the emitter is about 0.7V below the base. So if you force B to 12V, E will want to be 11.3V. This is probably what will happen if you do not use a resistor. The transistor will behave like a diode (B-E junction) and force AMC7135 VDD to 11.3V (not good, ab max is 7V). Also, the transistor may pull a lot of current from the brake switch.

By inserting a resistor in series with the brake switch and the transistor Base (B), you limit the current flowing into the Base and also allow the voltage at E to be very close to that at C thus providing 5V to the AMC7135 VDD pin.

In order to keep the voltage at E very close to that at C, you need a collector to base current ratio of 10 or less. The VDD supply current of the AMC7135 is specified as 200uA max so you'll want to set the transistor base current >=20uA. Assuming 5V at E, 5.7V at B and 12V at the brake switch, this calculates to 315kohms or less. There shouldn't be a need to use a value lower than 50k.

 

[lonesouth]

Excellent explanation. I had completely misread the specs on the transistor and thought the spec was the max voltage for the pin. What you explained makes much more sense! Thanks!