An LM311 is a normal power single comparator that has over 8 times the output current of a comparator in a low power LM339 or LM393.
It drives a TIP31 which is a medium power power transistor.
You didn't say how much current your relay coil uses but this circuit should provide plenty of current and there are no extra comparators to disable.

Attached Images

fuel pump driver.PNG (29.1 KB, 22 views)

__________________
Uncle $crooge

 

The LM311 is only capable of sinking 50mA. His relay will draw at least 133mA (90 ohm coil). He will have to use a transistor to drive the relay. A 2N3904 should work nicely (200mA continuous current). Again, going back to your original circuit, use the non-inverting input for input signal. You can use the LM311, like audioguru suggested. This will save real estate on your design.
You will want to add the things that were suggested earlier such as the snubber, flyback diode, voltage divider and pull-up resistor on the output of the LM311 (since it is open-collector). You may want to put a diode in series with your transistor base as well since the output of the LM311 will not saturate to 0V. This will require the output to be 1.4-1.5V before the transistor turns on. Otherwise, depending on things like the output current and temperature, may cause the transistor to go into conduction prematurely. Try using a 2KO resistor for the pull-up to keep the output current low. This will help in decreasing the saturation point

 

rezer...I have done some modifications to the circuit you drew. I added the NPN as recommended. Not specifically yours mentioned below.

2 questions, is the flyback resistor R4 in my drawing? It is now at 240 ohm, is that the one you recommend 2K ohm?

"The "diode" in series with the base".......did I draw it correctly here? If so, what is a recommended value for that? Same as the other one in paralell with the coil?

Otherwise, I think I have added everything as you said, and as given. Looks like I am ready to etch a board here if this new design works.

Stu

Attached Images

New Circuit Design 2 post.JPG (31.6 KB, 19 views)

 

Yes, R4 is the resistor I was referring to as 2KO. Again, yes, the diode is drawn correctly. You want to connect R4 to the anode side of the diode. A 1N914 (1N4148) diode will work. Use what is marked on your schematic for the flyback diode across the coil of the relay, 1N4001.
I am assuming by your statement "...if this new design works." to mean that you will test it out on a breadboard first.

 

If anyone of you are interested, this is how I drew it up for my final breadboarding. How it works, is during cruising, and under a certain RPM, the onboard ECU sends a 2.5V signal to the fuel pump ECU. That in turn then outputs a 9V power supply to the FP. As you can see, through the SPDT relay, the car is operating in this matter. You will also see a SPST relay on the left that is always supplying the fuel pump enhancer a 12V input. (armed)

When the car goes into higher RPM range (4K), the ECU sends a 5.0v signal to the FP ECU, and that in turn bumps the voltage up to 12V. Now the 3rd function is the pump enhancer. This is not quite activated as yet, but it is fully "armed". So when the Hobbs switch is activated (5.5K rpms), the enhancer (BAP) jumps the voltage to the fuel pump to 18V.....Increasing the fuel flow by quite a bit. The comparator is set to see this voltage change, and sends the signal to the coil to activate the relay and switch power to the BAP enhancer.

My concern about this whole set up was not having the enhancer unit kick from either (9v or even 12v) all the way to 18V in one switch. I wanted to do it in "steps"

Lets hope it works~!...Thanks to everyone, and I will report back after testing.

Stu

Attached Images

BAP Wiring-Circuit.JPG (150.7 KB, 19 views)

 

OK, might need a bit more help here. I totally benched tested the circuit board and it worked OK. I have exactly 3.3v as the referance. I inputed approx 2.7v and it did not trigger. I then applied about 4.3 V and it triggered. My problem now is that when it is all installed in my car, the coil is "hot", or "on". If all is right, the coil should not be triggering until it gets its 5V signal. Here is one possibility. I measured the input referance voltage in my car, and it was not 2.5V at idle, it was 1.9V....In further testing, I am getting a reading 7.3V on the TIP31 Collector to Coil (Pin 85). Now if the output of the TIP31 "is" triggered, shouldnt that be 12-13V? Why am I getting 7.3V.........and why am I getting "any" voltage at all if it isnt supposed to be triggered until the car gets to 5V +/-?? What could I be missing? What is the typical voltage for an automotive 12V relay at the coil in order to trigger it? Would 7.3V be enough to turn the relay on.
Maybe I have a bad relay.........cause it kind of hums a little.

Thanks for the help.

Attached Images

Relay Trigger Circuit.JPG (61.2 KB, 14 views)

 

Actually, I have a quesion that "rezer" mentioned above. He states to use the "non-inverted" as the "input". What actually are you considering as the "input". Is this the "referance" signal of 3.3V.....Or the signal that changes from the 2V to 5V for the trigger.?
Pin #2 of the LM311 is called the "non-inverted" pin. This is where I have the "trigger" signal coming in.

Also, I did not read through the post, you state that I should switch the 240 ohm to a 2K ohm to reduce current. I did not do that. Is this a possible cause? Are my 2 pins reversed for Pin #2 and Pin #3?

Stu

 

The non-inverting input is pin 2. It looks like, based on your earlier schematic, that is what you are using. This is the signal input that changes from 2-5V. I would fix you last posted schematic to show the +/- inputs of the comparator do reduce the confusion. The TIP31 will turn on at 1.8V and the LM311 has a Vsat of .75V, so you should be fine there. It can sink 50mA. With a 240 ohm, that puts it at 43mA. Again, it should be fine. What is the output of the LM311 when installed in your vehicle?

__________________
"Remember, you're special.....just like everyone else."

 

Oops! I calculated your current sink off of the Vsat of the LP339, not the LM311. Also, your alternator will run at around 14V, not 12V. Isink is 55.2mA. You are over-driving the LM311. Replace the 220Ω with a 470Ω.


First replace the resistor. If it doesn't fix the problem, here are some further questions.
Are you measureing 7.3V across the coil or at the collector to ground? If your reference voltage drops to 1.9V, what is the trigger voltage at? What is your voltage swing at the reference from idle to normal speeds (you can measure this by leaving the car in neutral and revving the engine)? Refer to the relay spec sheet and it will show what the Vlatch is. This sounds a little low for the relay to latch, but it may be on the threshold, which is where the "hum" comes from. But the main issue is your relay should be either full on or off. The transistor should not be operating in the active region.

__________________
"Remember, you're special.....just like everyone else."

 

This statement confuses me. At 2.0V (+/-), I DO NOT want the relay coil energized. I want it OFF so it is using "Input A" for the fuel pump supply voltage. I want the TIP31 to turn "ON" and energise the coil when the input signal jumps to 5V (+/-). This might be where we are confused on the (+) (-) non-inverted thing.


Also, the above "latest" schematic shows a 2K ohm as R3. That is an error. That is where I am using the 240 ohm. BUT....as you recommend, I will switch that tonight to a 470 ohm.

Also, my reading of 7.3V was from collector to ground.

The "Trigger" voltage will be somewhere in the mid to low 4V area. Won't know that until I go out on the road.

Lastly, the specs on the relay are 12VDC Nominal. 7.8V Operate Voltage. 1.2V Release Voltage. Coil 90 Ohms

 

Sorry for the confusion. The 1.8V would be at the comparator output to the transistor, not the trigger.
What voltage is your alternator running at? If at some point the relay gets energized, when it goes to shut it off, it may rattle as it approaches the holding voltage of the relay. The Vhold will be lower than Vlatch.

__________________
"Remember, you're special.....just like everyone else."

 

Should I change the 240 ohm to 470 ohm, or 2K ohm ? This is the first thing I will do tonight.

Stu

 

Change it to 470 ohm. This will lower your sink current to an acceptable level.

__________________
"Remember, you're special.....just like everyone else."

 

OK, well now things still arent working. I apply 4.8V to the trigger the stock FP and the stock pump kicks up to 12V at Idle. (Simulating). So I know it is getting the proper trigger voltage. The problem is the relay doesnt trigger when I apply the 4.8V now. Voltage's are identical accross the coil Pins 85 and 86 read 12V to ground. I would think that the voltage at the collector should be near zero, then when the signal comes out of the LM311, it grounds the coil out and energizes. So... it could be the diode went bad, or the TIP31 is bad? I don't think it can be the ref voltage/trigger voltage because I shouldnt see 12v at the collector if it is not energized? hmmmm kinda lost now. I can simulate the relay energizing by applying a 9V batter accross the coil. I did not try grounding the collector, thinking that might not be good idea.

So what is this telling me if I am reading 12v on both sides of the coil to ground? Is this a symptom of bad electronic peices or more of the circuit is not operating properly?.

Stu

 

Actually, maybe I have the inputs backwards. It keeps getting switched around. So I need to be clear on which pin is the "ref" (3.3v) and which pin gets the dual voltage signal. First drawing had one way, and your latest had it reversed. Maybe this is the possible problem?

I have it now, Pin #3 called the "Invert" is getting the 3.3v ref. V. Pin #2 called "non-Invert" is getting the dual input voltages (2.5 to 5.0V) Not sure which one needs to be the (+) or (-) if it really matters.

Stu