Adding and removing resistor component in circuit by using transistor......

Thread Starter

BigCircuit3302

Joined Sep 2, 2022
14
Hello everyone.....
I am asking for help with something I'm trying to do in a circuit.
I would like to know if it is possible to do what I'm about to describe with a transistor of some type???
I have a circuit which contains a resistor connected to a PIN of an IC chip. What I would like to do is replace this particular resistor with another resistor of a different value when the "switching transistor" receives a ground signal.....
When the ground signal is removed, the original value resistor will again be connected to the PIN of the IC chip.
I hope someone has a suggestion on how to accomplish this that's not too complicated.
Thanks in advance for any suggestions on how to accomplish this.
Have a good evening.
 

WBahn

Joined Mar 31, 2012
27,509
It should be doable using FETs, but a lot depends on the details.

What's on the other side of the resistors?

What is the value of the resistors and the range of voltages involved (on both sides of the resistors)?

Could you provide a partial schematic showing the two resistors and where you would like the connections made/broken (basically, where you would place the switches if you were using them)?

How good do the switches need to be (they will look like resistors themselves, so how low does their resistance need to be)?
 

Thread Starter

BigCircuit3302

Joined Sep 2, 2022
14
It should be doable using FETs, but a lot depends on the details.

What's on the other side of the resistors?

What is the value of the resistors and the range of voltages involved (on both sides of the resistors)?

Could you provide a partial schematic showing the two resistors and where you would like the connections made/broken (basically, where you would place the switches if you were using them)?

How good do the switches need to be (they will look like resistors themselves, so how low does their resistance need to be)?
Thanks for your reply.
I will get something to you.
 

MrChips

Joined Oct 2, 2009
27,196
If you already have the circuit with a resistor, post the circuit.
We want to see the value of the resistance and the circuit topology.
 

WBahn

Joined Mar 31, 2012
27,509
Since both R1 and R2 are tied to ground (or common), it should be pretty straightforward.

What are the values of R1 and R2.

What qualifies as "high speed"?

What qualifies as "simultaneous"?

You need to quantify these.
 

Thread Starter

BigCircuit3302

Joined Sep 2, 2022
14
Since both R1 and R2 are tied to ground (or common), it should be pretty straightforward.

What are the values of R1 and R2.

What qualifies as "high speed"?

What qualifies as "simultaneous"?

You need to quantify these.
Default values
R1 4.2K
R2 3.8K
Switched values
R1 20.5K
R2 17.0K
High speed = fast as possible
Simultaneous = R1 & R2 "switched" at same time
 

WBahn

Joined Mar 31, 2012
27,509
Default values
R1 4.2K
R2 3.8K
Switched values
R1 20.5K
R2 17.0K
High speed = fast as possible
Simultaneous = R1 & R2 "switched" at same time
For those kinds of values, a CD4066 analog switch is likely good enough -- depending on what is really meant by your speed specs.

"fast as possible" and "switched at the same time" are useless specs.

How much are you willing to spend to have them switch as "fast as possible" -- meaning that no one on the planet could make them switch any faster with current technology?

If it takes 1 ms for them to switch, is that too slow? If so, is 100 µs too slow? If so, is 10 µs too slow?

What is the slowest that they can switch and be be acceptable?

If one of them switches 1 ns after the other, is that close enough to "switched at the same time" to be acceptable or not?

What is the greatest difference between when one of them switches and when the other switches that can be tolerated?
 
Looking at your "Figure 1 Simplified Application" the dotted rectangle suggests a 20pin IC. I wonder if you could share with us the part number of that IC and what you are trying to do? The fact that R1 and R2 are shown as pots suggests that the speed and precise timing sequence of change is not critical. Both R1 and R2 are shown as connected via 28K resistors to the positive side of the battery so they are acting as potential dividers to feed voltages proportional to the battery voltage into pins 2 and 12 [Sense (Remote) and Supply Reg (Local)] respectively. But I don't see an obvious pattern between those two reference voltages?
 

Thread Starter

BigCircuit3302

Joined Sep 2, 2022
14
For those kinds of values, a CD4066 analog switch is likely good enough -- depending on what is really meant by your speed specs.

"fast as possible" and "switched at the same time" are useless specs.

How much are you willing to spend to have them switch as "fast as possible" -- meaning that no one on the planet could make them switch any faster with current technology?

If it takes 1 ms for them to switch, is that too slow? If so, is 100 µs too slow? If so, is 10 µs too slow?

What is the slowest that they can switch and be be acceptable?

If one of them switches 1 ns after the other, is that close enough to "switched at the same time" to be acceptable or not?

What is the greatest difference between when one of them switches and when the other switches that can be tolerated?
My acceptable time frame for "fast as possible" and "simultaneous switching" is 50 ms or less.
The maximum power supply voltage is 30 volts.
 

WBahn

Joined Mar 31, 2012
27,509
My acceptable time frame for "fast as possible" and "simultaneous switching" is 50 ms or less.
The maximum power supply voltage is 30 volts.
That's actually rather slow, so that helps out a lot.

How much switch resistance can you tolerate. As noted previously, the switches will look like resistors. For the CD4066, that resistance depends on the supply voltage. How will these be powered? From the battery in the schematic? What is the supply voltage? Is it the 30 V you mentioned here? If so, then that will need to be brought down. What is the lowest is might be?

For a supply voltage of 15 V, the on resistance is going to be ~150 Ω. For lower supply voltages it will go up. For a 5 V supply, it can get at high at 500 Ω. So what is acceptable? What is the tolerance of the resistors being switched (as well as those two 28 kΩ resistors above them)?

Do you have a single control signal that is HI (what voltage?) for one set of resistors and LO for the other? If so, can you provide both a HI signal for the set you want on and a LO signal for the set you want off (two signals that are always opposite each other)? If not, that can be done using an inverter of some flavor next to the 4066.
 

WBahn

Joined Mar 31, 2012
27,509
How do you communicate on this form through private messages?
You need to have a minimum number of posts before you can use the private message feature. I don't know what the admin's have it set at currently -- at one point it was ten posts.

But be forewarned, most of the members prefer that discussions take place in the forums since the big motivation is to share and gain knowledge all around.
 

Thread Starter

BigCircuit3302

Joined Sep 2, 2022
14
You need to have a minimum number of posts before you can use the private message feature. I don't know what the admin's have it set at currently -- at one point it was ten posts.

But be forewarned, most of the members prefer that discussions take place in the forums since the big motivation is to share and gain knowledge all around.
That's actually rather slow, so that helps out a lot.

How much switch resistance can you tolerate. As noted previously, the switches will look like resistors. For the CD4066, that resistance depends on the supply voltage. How will these be powered? From the battery in the schematic? What is the supply voltage? Is it the 30 V you mentioned here? If so, then that will need to be brought down. What is the lowest is might be?

For a supply voltage of 15 V, the on resistance is going to be ~150 Ω. For lower supply voltages it will go up. For a 5 V supply, it can get at high at 500 Ω. So what is acceptable? What is the tolerance of the resistors being switched (as well as those two 28 kΩ resistors above them)?

Do you have a single control signal that is HI (what voltage?) for one set of resistors and LO for the other? If so, can you provide both a HI signal for the set you want on and a LO signal for the set you want off (two signals that are always opposite each other)? If not, that can be done using an inverter of some flavor next to the 4066.
WBhan.......
Thank you for helping me with this project. I'm sure you can tell by my answers that I am not an engineer or highly technical. Just a person that attempts hobby circuits from time to time.
50 ms switching is just a guess...... A faster time would even be better.
I am not sure regarding your question about "switch resistance tolerance" ????
The supply voltage to the CD4066 would come from what's being produced by the alternator, which would vary depending upon what resistors are in circuit via the switching process. It would never be less than 12 v, but could be as high as 30 volts. The high voltage duration will never be more than 20 sec...... Most of the time the duration will be between 10 and 15 seconds. I would like to have one pair designated the "default pair"remain in circuit until a ground signal is sent to initiate the "switching process". When ground is removed, the default pair should be put back in circuit.
 

WBahn

Joined Mar 31, 2012
27,509
WBhan.......
Thank you for helping me with this project. I'm sure you can tell by my answers that I am not an engineer or highly technical. Just a person that attempts hobby circuits from time to time.
50 ms switching is just a guess...... A faster time would even be better.
I am not sure regarding your question about "switch resistance tolerance" ????
The supply voltage to the CD4066 would come from what's being produced by the alternator, which would vary depending upon what resistors are in circuit via the switching process. It would never be less than 12 v, but could be as high as 30 volts. The high voltage duration will never be more than 20 sec...... Most of the time the duration will be between 10 and 15 seconds. I would like to have one pair designated the "default pair"remain in circuit until a ground signal is sent to initiate the "switching process". When ground is removed, the default pair should be put back in circuit.
So the question of allowable switch resistance can be answered by asking something like:

What if, instead of:

Default values
R1 4.2 kΩ
R2 3.8 kΩ
Switched values
R1 20.5 kΩ
R2 17.0 kΩ

It turned out that if was:

Default values
R1 4.7 kΩ
R2 4.3 kΩ
Switched values
R1 21.0 kΩ
R2 17.5 kΩ

In other words, if the value of each resistor were increased by 500 Ω.

Would that be acceptable?
 

Thread Starter

BigCircuit3302

Joined Sep 2, 2022
14
So the question of allowable switch resistance can be answered by asking something like:

What if, instead of:

Default values
R1 4.2 kΩ
R2 3.8 kΩ
Switched values
R1 20.5 kΩ
R2 17.0 kΩ

It turned out that if was:

Default values
R1 4.7 kΩ
R2 4.3 kΩ
Switched values
R1 21.0 kΩ
R2 17.5 kΩ

In other words, if the value of each resistor were increased by 500 Ω.

Would that be acceptable?
Unfortunately, I don't believe that will work for this application. The value of the chosen resistor is what must be delivered to the IC pin ....... Is it possible to use a "J-FET" or some other device that will not add resistance to the IC pin connection? Also, I believe I saw in the spec sheet of the CD4066 18 was the high limit of of supply voltage?
 

WBahn

Joined Mar 31, 2012
27,509
Unfortunately, I don't believe that will work for this application. The value of the chosen resistor is what must be delivered to the IC pin ....... Is it possible to use a "J-FET" or some other device that will not add resistance to the IC pin connection? Also, I believe I saw in the spec sheet of the CD4066 18 was the high limit of of supply voltage?
So it won't work is the value of R1 is 4.21 kΩ?

There is no way to tell if any switching circuit will work unless you come up with a limit on how much resistance the switch can add.

We can deal with the max voltage limit using a zener diode (or other simple regulator).
 

Thread Starter

BigCircuit3302

Joined Sep 2, 2022
14
So it won't work is the value of R1 is 4.21 kΩ?

There is no way to tell if any switching circuit will work unless you come up with a limit on how much resistance the switch can add.

We can deal with the max voltage limit using a zener diode (or other simple regulator).
So question please.......
If each resistor value was reduced 500 ohms, would the value at the connection point be 500 ohms greater than selected resistor value?

46.5k = 47k at connection point using CD4066?
42.5K = 43K?
20.5K = 21K?
17K = 17.5K?
 
Top