ground opening a path w/out using a relay?

Okay here's my diagram, i need to replace the relay in the diagram with something else... i'm not sure what though? nothing inside the box i've drawn can be modified. A is a constant ground that reads resistance. B is a switched ground that switches somewhat rapidly (so the relay is not the best choice). supply voltage is 12 volts. A needs to read ~4.2kOhms when B is closed and 420 Ohms when B is open.

**broken link removed**

what do you think? is a solid state relay a reliable choice? or is there another way to do it using different components? just for curiosity sake im more interested in a way of doing it without using any kind of relay. thanks for your help

Edit: heres more detail my car has a solenoid that opens vacuum pressure to a diaphragm that controls a butterfly valve in the factory air box silencer. theres a sensor on the valve that reports to the engine computer weather or not the valve is open. i have an aftermarket intake so i removed all of the components related to the factory intake but now it throws a check engine light because that sensor is not there. "B" is the ground for the solenoid, switched on or off inside the ECU and "A" checks resistance to determine weather the valve is open or closed. so i need to use B to some how change the resistance on A. i made a prototype that worked using a relay but the problem is it clicked every time stepped on the gas or let off. annoying.

What operates the "relay"? It seems that like it needs some sort of signal to tell your circuit if the resistance to ground should be 420 or 4200 Ohms?

Perhaps you should describe what you're trying to do, instead of how you're trying to do it.


Regards,

Torben

MikeMl said:

What operates the "relay"? It seems that like it needs some sort of signal to tell your circuit if the resistance to ground should be 420 or 4200 Ohms?

Click to expand...

the relay is ground triggered via the switch inside the box i've drawn. a should read 420ohms until the 466ohm resistor is activated. according to my calculations, both resistors in parallel should read ~4.2k at ground A

my car has a solenoid that opens vacuum pressure to a diaphragm that controls a butterfly valve in the factory air box silencer. theres a sensor on the valve that reports to the engine computer weather or not the valve is open. i have an aftermarket intake so i removed all of the components related to the factory intake but now it throws a check engine light because that sensor is not there. "B" is the ground for the solenoid, switched on or off inside the ECU and "A" checks resistance to determine weather the valve is open or closed. so i need to use B to some how change the resistance on A. i made a prototype that worked using a relay but the problem is it clicked every time stepped on the gas or let off. annoying.

any ideas? im really curious to see what will need to be done to make this work i havent worked with electronics like this since i was a kid and its rekindling my passion

Replace your relay with a P-channel MOSFET. Source to +12V, Drain to the right end of the 466Ω, Gate to the grounding switch, and a 10KΩ resistor between Gate and Source....might work. But, since this going into an automotive environment I can't be sure.

Ken

See if this works for you: It creates a resistance to ground. Is that what you wanted?

thanks for the responses! but i dont know the values of resistors 1-5... and what does the capacitor do?

ill try KMoffett's suggestion.

so i put what kmoffett said into the simulator and it works flawlessly i am very happy about that! now, what are your concerns about the automotive environment? this circut will be enclosed in a plastic project box and tucked into the passenger's kick panel. or are you worried about varrying voltage and interference? i also included a diode, do you think its necessary? scratch that, i remember reading that transistors work like diodes anyway.

**broken link removed**

now, i want to use this as an opportunity to learn as much as possible so i have a few questions... which im about to go research anyways but it's easier for me to retain information when i receive it directly through a situation that i can relate to.

1. why a mosfet and not a normal pnp or npn transistor?
2. is a p channel mosfet similar to a pnp and a n channel similar to a npn? how do they differ?
3. i believe the current will be less than 50ma through this circuit (i will go double check soon) does this change anything?

There is something horribly wrong with your calculations. If you start with 420Ω resistor, and then you switch a 466Ω resistor in parallel with it, the total resistance becomes (420*466)/(420+466) = 195720/886 = 221Ω, not 4200Ω. Is that what you want?

The values in my sim were shown with the K in place of the decimal point, i.e. 4K7 = 4.7K = 4700Ω. My circuit powered the transistor from a tie to the 12V battery which has nothing to do with terminal B where the resistance to ground is measured. The capacitor was to suppress a spike in the output resistance.

math was my worst subject... haha well thank you for pointing that out. and okay, that makes sense. although, resistance is measured at terminal a, not b. voltage/resistance is not important at b just for the record. can you please explain to me why/how this spike would occur? or at least point me to an article explaining it so i know for the future? also, can you tell me what it's called when you have two resistors in parallel so i can look up the algorithm.

i've been working with the information KM gave me in the simulator and this is what i've come up with so far. i'm using a mosfet like he said but i also tried it with a pnp and it worked too. im not sure which to use yet or why. in my physical model i'm using the mosfet for now and theres a picture below with the information on the back of the mosfet box. i've taken this to mean that the voltage at the base and source should be no more than 4.5v should i run 4.5 at the base to fully saturate it or is my logic completely off? in the diagram i have it at 3v to be safe, dont wanna max it out unless i have to. what are your thoughts on the diagram? and how, after i recalculate the value of resistance that needs to be introduced to the circuit, do i calculate what my 2 resistor values will need to be so that the output is 4.2k AND the voltage at the source pin of the mosfet is at or below 4.5v woooh. thats a lot to ask but i really appreciate your help. i'm learning so much
**broken link removed**
**broken link removed**

I am still confused about how terminal A on your ECU measures resistance? Does it look for resistance connected between terminal A and the 12V (battery) supply?

Are you trying to say that terminal A is always at or very close to 0V?

And just to clarify; at terminal B of the ECU, there is a switch inside which either connects to ground or is open. This is an output from the ECU?

When the switch at B is closed, the resistance from +12V to A should be 4200Ω?
When the switch at B is open, the resistance from +12V to A should be 420Ω?

The formula to compute resistors in parallel: Rt = (R1*R2)/(R1+R2). Resistors in series: Rt = R1 + R2

question 1. yes, for example, if i were to use my ohm meter i would probe at the 12v src and terminal a, the reading i get should be the reading the ecu sees.
2. if i were to check voltage at terminal a with a multimeter it would always read 0v.
3. correct.
4. correct correct
5. thank you very much so how do i calculate what ohm resistors i need if i know rt? its been a while since algebra...

Ok, here is a no transistor solution. Look at the red trace in the simulation. I'm using LTSpice to display the effective resistance between A and Bat. Note that the resistance is 417Ω with switch B open and 4213Ω with switch B closed.

Note that I consider the battery voltage to be 14.2V with the engine running...

impressive! thank you. can you upload that file so i can experiment with it? if you still have it that is

Here is the LTSpice Draft99.asc file for this circuit. LTSpice uses text for its schematic files.