12V 3A high side switch with two input XOR logic

Thread Starter

Mogman

Joined Jan 20, 2022
21
I need help with a high side switch circuit (actually two a left and right) to switch a 2.5-3A incandescent lamp, it needs two inputs per circuit with XOR logic.
12-15V supply, rugged yet simple, open load tolerant.
What I am doing is I have a single filament stop/turn signal lamp that I need it to illuminate with the brake light switch (input 1) and to flash with a turn signal input (input2) whether or not the brake signal is present, I understand the light would invert flash with the other turn signal lamp when the brake is applied but that is OK.
The turn signal switch I have is not sophisticated enough to automatically isolate the turn signals and brake signals unlike most automotive turn signal switches.
I was thinking a simple TIP120 switch with a XOR TTL chip driving it but I suspect the TTL level is not enough to drive the TIP120 directly?
It has been decades since I have tried to design a circuit like this.
It would be built on proto board and potted.
Thanks
 

dl324

Joined Mar 30, 2015
16,845
I was thinking a simple TIP120 switch with a XOR TTL chip driving it but I suspect the TTL level is not enough to drive the TIP120 directly?
TIP120 is an NPN Darlington. For low side switching, it can be driven by TTL.

For high side switching, you need something like TIP125. You could drive that with open collector TTL. Otherwise, you'd need a discrete inverter stage to interface between 5V and 12V. In the latter case, you could use a PNP power transistor because the inverter could sink the required current to saturate the switch.

It would be helpful if you drew a block diagram.
 

Thread Starter

Mogman

Joined Jan 20, 2022
21
If a PNP switch needs a ground on the base to conduct then a XNOR input could possibly be used?
Maybe not that simple I do not think the TTL chip would "drive" the output low.
Maybe using the XOR and an additional transistor 'inverter"?
Or maybe a complementary pair darlington?
 

Attachments

dl324

Joined Mar 30, 2015
16,845
Is it legal in your locale to flash the brake light like that?
If a PNP switch needs a ground on the base to conduct then a XNOR input could possibly be used?
Not if the XNOR is TTL operating at 5V. You'd need a resistor on the base to 12V to turn the transistor off. TTL outputs can't be pulled up to 12V.
Maybe using the XOR and an additional transistor 'inverter"?
That's what I mentioned in my first post. The discrete inverter would need to be able to sink 10% of the lamp current and the TTL gate would need to be able to source 1% of the lamp current.
 

Thread Starter

Mogman

Joined Jan 20, 2022
21
Never mind does not look like it would work, a compound is like a Darlington only it has complementary transistors instead of the same transistors in it.
The problem is the emitter of the driving transistor is tied to the emitter of the driven transistor which seems to make it not desirable, trying to reduce the number of components, looks like it would make a good low side switch though. Might be hard to find also
Compound.JPG
 

Thread Starter

Mogman

Joined Jan 20, 2022
21
So I guess we are looking at a circuit like this, only R1 would be around 10K to be driven by the TTL chip?
How do I address the 12V inputs to the 7400 series XOR and how would it act if the load was removed (blown lamp)?
High side switch.gifSorry forgot attachment, I will work up a schematic based on this, then someone can point out all the mistakes
 
Last edited:

dl324

Joined Mar 30, 2015
16,845
a compound is like a Darlington only it has complementary transistors instead of the same transistors in it
You shouldn't make up your own terms. No one will understand what you're talking about.

The configuration you referenced is called a Sziklai pair after the man who is thought to have invented it.
 

dl324

Joined Mar 30, 2015
16,845
So I guess we are looking at a circuit like this, only R1 would be around 10K to be driven by the TTL chip?
How do I address the 12V inputs to the 7400 series XOR and how would it act if the load was removed (blown lamp)?
Don't see any schematics posted with an R1.

EDIT: post #9 was updated to include a schematic.
 
Last edited:

Papabravo

Joined Feb 24, 2006
21,159
Why not use an actual logic gate with a transistor buffer/level shifter? Sprague used to make such high voltage parts (e.g. UHP502) but they are long out of that business. Something like this. The counter and the clock block are just for simulation testing testing.
1660180978225.png
 
Last edited:

Thread Starter

Mogman

Joined Jan 20, 2022
21
Don't see any schematics posted with an R1.
Corrected the missing attachment, I am also sorry, I learned electronics a few years ago, we still used MC instead of MHz.
But it is/was a "correct" term.
"The compound/ Sziklai pair is a configuration of two bipolar transistors of opposite polarities, so will always consist of one NPN and one PNP transistor. The configuration is named after its Hungarian born inventor, George Sziklai. It is also sometimes known as a CFP (complementary feedback pair)."
I will build a schematic and submit.
 

eetech00

Joined Jun 8, 2013
3,859
So I guess we are looking at a circuit like this, only R1 would be around 10K to be driven by the TTL chip?
How do I address the 12V inputs to the 7400 series XOR and how would it act if the load was removed (blown lamp)?
View attachment 273613Sorry forgot attachment, I will work up a schematic based on this, then someone can point out all the mistakes
Might be better to use three mosfets, an NMOS (2N7000) to invert, an NMOS (2N7000) to drive a PMOS (FQP27P06) to drive the load.
If you don't need to invert then two mosfets, an NMOS (2N7000) to drive a PMOS (FQP27P06) to drive the load.
 
Last edited:

Papabravo

Joined Feb 24, 2006
21,159
Way too complex certainly do not need a clock.
No! No! No! I said the clock was just to run the simulation so you could see if the logic was correct. You only need R1, R2, Q1, U3
You are interested in the XNOR and the transistor inverter. 3mA of base current gets you 30 mA of collector current. That should be enough to drive a medium load.
 

Thread Starter

Mogman

Joined Jan 20, 2022
21
Here is what I have so far, I have a good stock of transistors and darlingtons, no mosfets.
There will of course be a 5V regulator circuit, I have no idea at this point the best way to handle the 12V inputs to the 7486
No! No! No! I said the clock was just to run the simulation so you could see if the logic was correct
OK
1st.JPG
 

Papabravo

Joined Feb 24, 2006
21,159
Here is what I have so far, I have a good stock of transistors and darlingtons, no mosfets.
There will of course be a 5V regulator circuit, I have no idea at this point the best way to handle the 12V inputs to the 7486

OK
View attachment 273616
You cannot run 12V inputs to a 74LS86. Don't use an 74LS86, use a CMOS gate, The 12V inputs are compatible with the CD4077B, and you only need a single transistor.

NOTE: Schematics without values are not much use.
 
Top