Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Help with transistor relay switch 3

Status
Not open for further replies.

MitchSF

Computer
Apr 29, 2005
17
I need a circuit for a simple transistor switch for an automotive hobby application, designed so that when the base of a switching transistor is grounded, a relay that is connected to +12v on one side, and the collector of the transistor (I guess), on the other side, will be closed. The emitter will be connected to ground.

One catch is that the base will be connected to a wire that sends varying resistances, from about 500 ohms, to 2k ohms, and I don't want that functionality interfered with, so I guess the input to the transistor switch (base) should have a high impedance, maybe 50k ohms or so.

Any help, including other ideas, will be greatly appreciated.

Thanks in advance,

Mitch
 
Replies continue below

Recommended for you

I'll try this later today. I want to be clear on a couple things - the zener goes from ground to one input, and a 5-47k resistor goes from the same input to 12v.

The sense input goes through a 47k resistor. I still don't see why the relay won't close if this sense input is grounded through a 200 ohm resistor, because that will provide a voltage differential. I only want it to close on a direct ground of the sense input. But I'll try. Sorry for the beginner questions, and thanks for the help.

Mitch
 
MichSt:
The 5 V may be too much. You need a reference voltage significantly less than what you get with any other switch.

This circuit doesn't eliminate the horn turning on at
power on/off.


<nbucska@pcperipherals DOT com> subj: eng-tips
read FAQ240-1032
 
What 200 ohm resistor! You are combining too many solutions into one circuit.
 
Here are the resistor values that are grounded from each switch:

220 ohms
600
1300
1600
2800
direct ground (horn)

I can change the 220 and 600 ohm resistors to some higher value. The Peripheral Electronics module trains itself to associate resistor values that it reads with infrared commands, so the values are not critical.

If you don't think this will work, I'll just do it with the Basic Stamp, but I'd rather not have to do all that work.
Thanks,
Mitch


 
Sorry, these things go on so long and with other posts the origional question gets lost. The basic comparator circuit will work. You may have to put a voltage divider pot on the zener to get the appropriate trip voltage. Is any current applied to these resistors to create a voltage that you monitor or will you have to supply that too?
 
Hi, no problem. Thanks again for the help.

There is a very small current on the input to the Peripheral Electronics module, a few milliamps, at around 10 volts, so I don't have to supply the voltage. The switch assembly simply switches each resistor to ground, and the horn switch grounds the wire completely.

Will the pot go between the zener and the input, or will I replace the 47k resistor from the input to 12v, with a pot? Remember, I'm not an EE like you guys, so please keep it simple.

Also, should I change the 220 and 600 ohm resistors to higher values?

Thanks,

Mitch


 
The lowest resistance before the short is 220 ohms. That 10V could be a fixed current source or just a resistance. If you figure about 2ma for a start, your reference voltage should be about 0.44 volts. You have to get below that so 0.2V would be a good value to start with. From the 4.7V zener, have a 10K resistor going to a 470 ohm to ground for a voltage divider. That would give you 0.21V which should be low enough. Much below 0.2V, I would be concerned about noise. The 47K goes to the switch sense and it would be a good idea to have a .1uf disk capacitor at the input of the op amp to prevent RF from false triggering the relay.
 
I'll try this later today or tomorrow. I can always change the 220 to something higher, if it doesn't work.
Thanks again,
Mitch
 
It worked! I calibrated it on the bench with a 150 ohm resistor on the input using the 10k variable pot, and it worked fine in the car. Steering wheel switches worked perfectly, and the 200 ohm switch did not close the horn relay.

Now the last problem - the horn honks when the ignition is started, and also when it's turned off, because the switch does ground the + ignition line at a point when the key is rotated. After playing around for a little while, I also found that if the circuit's + is not connected at all, the horn is on constantly when the car is started. So I think I need a switch that will not allow the relay to close if there is no +12v.

I'm considering using an NPN transistor with the emitter connected to ground, and the collector to the relay (the side that will be grounded to close the relay). The base will go to the output of the gates through a resistor (not sure what value, but probably not critical). If it's done this way, I assume that only one of the gates is necessary because the transistor will handle the current, but since they're all already connected, I'll leave the circuit the way it is.

Is there a better way, like using one of the gates to do this on the input, instead of a transistor on the output? If so, please let me know how.

Thanks,
Mitch
 
On second thought, I think it would be a PNP, with emitter to the relay wire, and collector to ground.

Mitch
 
As I said in a previous post, the problem is the 10V is switched off when the vehicle is being started. If you move the relay power to the switched 12V buss there won't be a problem. To put any time delay in would require you to find that switched power anyway. You could try a 47K from the 10V of the sample line (not at the op amp pin)to your 12V. That would add another 0.03ma and mightlift tthe voltage high enough so the horn doesn't blow. This added current would not effect the sense currents that much.
 
I can't change the 12v connection to the horn relay. Even with the added transistor, the same thing happens because the circuit is still powered with 12v through the horn relay. The easiest way at this point, I think, is to add a small 12v relay on the output, to isolate the relay coil. I'd rather use something that's solid state, but I'm sure this will work.

Thanks again,
Mitch
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor