circuit to split combined DRL/turn signal into separate circuits

[AMB_IV]

I have heard GM's charging system can deliver voltage > normal in certain charging conditions. **broken link removed** says the 2006 SRX has GM's Regulated Voltage Control and that under certain conditions the peak voltage can go as high as 15.5 volts.

 

[spec]

I fully understand that.
I just didn't see where that would occur in the circuit I posted. (555 pin above VCC)

Reference schematic of post #8
Assume the automobile battery voltage is 14V.
D2 will produce 13.4V on the 555 VCC
But the 555 trig input is connected to the automobile 14V line.
Thus Trig will be 600mV above VCC which will almost certainly blow the substrate diode, especially when you take any hash on the 14V supply line into account.

At initial turn on the situation is much worse: VCC= 0V and trig = 14V.

I don't think the circuit will exceed the 555 limit of 70°C (158°F) where it's located at the front of the car (unless it's parked facing the sun on a summer day in Death Valley).

You would be surprised how high and low the temperatures can get. There are standards that define theses temperature ranges and other automobile environmental conditions. I don't know if there is an automotive grade 555, but almost certainly as it is such a ubiquitous chip, but there are definitely automotive grade NMOSFETS and PMOSFETS. There is also automotive grade wire.

spec

 

[MikeMl]

Reference schematic of post #8
Assume the automobile battery voltage is 14V.
D2 will produce 13.4V on the 555 VCC
But the 555 trig input is connected to the automobile 14V line.
Thus Trig will be 600mV above VCC which will almost certainly blow the substrate diode, especially when you take any hash on the 14V supply line into account.

Since the substrate diode is Si, and the substrate diode is also Si, then this is unlikely to "blow" it.

At initial turn on the situation is much worse: VCC= 0V and trig = 14V...

Adding a 5K to 50K resistor between the 14V line connection dot and the Trig pin would take care of this...

 

[spec]

Since the substrate diode is Si, and the substrate diode is also Si, then this is unlikely to "blow" it.

Not true Mike. The diodes are completely different- the substrate diode is very delicate compared to a 1N4148 and 1N4001. Besides, taking the 555 pins outside the supply lines is not specified on the 555 data sheet. In addition any hash on the 14V line will instantly fry the substrate diode.

Adding a 5K to 50K resistor between the 14V line connection dot and the Trig pin would take care of this...

Once again, putting current through a substrate diode, especially the implied current, is not permitted. It is also unwise, and unnecessary.

spec

 

[spec]

I have heard GM's charging system can deliver voltage > normal in certain charging conditions. **broken link removed** says the 2006 SRX has GM's Regulated Voltage Control and that under certain conditions the peak voltage can go as high as 15.5 volts.

Hy AMB_IV

Automobile 12V supply lines are particularly nasty as far as electronic components are concerned; they are all blood and thunder.

There is a wide range of voltages 15.5V max, as you say for your Caddy, and down to 9V under cold cranking conditions. There can also be a load of hash on the 12V supply line due to the alternator, ignition and other items switching on and off. The injector coils are quite bad in this respect. Jump starting can also cause problems as can an external battery charger, especially the boost type.

That is all under normal operating conditions but the situation gets worse under fault conditions: bad earth connections, failed battery, or even battery being disconnected while the engine is running.

On top of all that there is high vibration, and extremes of temperature and humidity.

I have found that in practice you can get all sorts of odd failures with electronics fitted to automobiles. The first example of this was when I had the bright idea of fitting a 1000 uF capacitor across the supply line to a radio to reduce interference from the alternator. It worked fine... for a couple of days then one morning while driving to work the capacitor exploded.

So the best philosophy is keep it simple and do a belt and braces (suspenders) design.

spec

 

[spec]

Hy crtuschow,

I suggest the Infinion (International Reectifier) AUIRF4905 PMOSFET for the circuit of post #11. https://www.mouser.co.uk/ProductDet...TKZf7/IVqC62V4S//duFS%2b0oBuVCUtF9V3Iptmw6w==

It is automotive grade, comes in a TO220 case for easy handling and good power dissipation and has a medium gate drive with a minimum gate threshold of -2v.

spec

 

[crutschow]

Hy crtuschow,

I suggest the Infinion (International Reectifier) AUIRF4905 PMOSFET for the circuit of post #11. https://www.mouser.co.uk/ProductDetail/Infineon/AUIRF4905L/?qs=/ha2pyFadugmiTKZf7/IVqC62V4S//duFS%2b0oBuVCUtF9V3Iptmw6w==

It is automotive grade, comes in a TO220 case for easy handling and good power dissipation and has a medium gate drive with a minimum gate threshold of -2v.

spec

Sounds good to me.

 

[MikeMl]

Not true Mike. The diodes are completely different- the substrate diode is very delicate compared to a 1N4148 and 1N4001. Besides, taking the 555 pins outside the supply lines is not specified on the 555 data sheet. In addition any hash on the 14V line will instantly fry the substrate diode.

Once again, putting current through a substrate diode, especially the implied current, is not permitted. It is also unwise, and unnecessary.

Unlike you, I have designed, layed out, fabricated, and tested Integrated Circuits. I can tell you that the substrate diode on a Bipolar process (like the 555) will happily pass a transient of an Amp or so without damage. It is a Silicon diode, so has an IV curve similar to a 1n400X. The ultimate limit on injecting current into/out-of an input pin is fusing the bond wire between the package pin and the chip inside the epoxy.

 

[spec]

Unlike you, I have designed, layed out, fabricated, and tested Integrated Circuits. I can tell you that the substrate diode on a Bipolar process (like the 555) will happily pass a transient of an Amp or so without damage. It is a Silicon diode, so has an IV curve similar to a 1n400X. The ultimate limit on injecting current into/out-of an input pin is fusing the bond wire between the package pin and the chip inside the epoxy.

"Unlike you" I have designed and commissioned many units to go on automobiles and "unlike you" I do not believe in taking unnecessary risks and I can tell you that you are taking unnecessary risks in this case which will lead to failure in the field.

Besides which, regardless of what you and I have done or haven't done, the 555 specification does not say that you can forward bias the substrate diode with any current. As for forward biasing a 555 substrate diode with 1A, that is crazy. You also have no idea what the limitations are on the processes used for a particular 555, especially as they have been made since the 1970s by any number of manufacturers.

Why all this talk when a simple mod to the circuit would prevent the substrate diode from being turned on.

spec

PS: Designing, laying out, and testing integrated circuits sounds interesting. Can you tell us which integrated circuits you have designed?

 

[AMB_IV]

In one of the earlier posts Spec asked for current. Here it is:
3757 Amber incandescent bulb 2.08 amps
3157 single circuit 10 Amber Cree LED "bulb" 0.39 amps
3157 single circuit 72 Amber LED "bulb" 0.32 amps
3157 dual circuit 78 Amber LED and 42 White LED "bulb"
Amber circuit LED only 0.33 amps
White LED circuit only 0.24 amps.
3157 dual circuit = "switchback"

 

[AMB_IV]

for what it's worth, here are the bulbs I used in the measurement:

Left to right:
3757A dual filament bulb, only one filament used
3157 Miniature 50W CREE Amber LED (Also two shining forward under the glass lense)
3157 68-SMD Parking Amber Turn Signal Light Bulbs
3157 Dual Color Switchback White Amber LED Turn Signal Light Bulbs
Still hard to beat a filament bulb - only the "Cree" lamp seems to be as bright.

And this is what happens when you pump 2.08 amps at 14.10 volts through a 7W bulb:

I've got my mouser parts list made up, I'm gonna double check it and order some parts for testing. I'm favoring the circuit with the two mosfets and staying out of the 555 timer questions.
thanks again for everyone's help!!!

 

[AMB_IV]

Teaser shot:

Been busy on a vw and not enough spare time for this. And the next two weeks are spoken for. Maybe it'll be working by Halloween ?