Design an adjustable voltage switch

[egil]

Eric,

Thanks for your warm welcome, and I apologize for not yet knowing how to accurately describe the voltage regulator in my truck. Here is my understanding of the regulator: (1) It takes the 14v input from the alternator. (2) It turns the 14v output on and off in order to produce an average output of 5v.

Here is another description I've found:

"The CVR's purpose is to provide an average voltage to give consistent readings with voltages ranging from as little as 6V up to 16V. The CVR is in fact a early mechanical 'logic' circuit and is one of the first steps to the computer controled cars of today. Today it is called pulse width modulation, but what ever you call it the basic concept is by varing the ratio of 'on' time to 'off' time or % dwell to produce a target "average" voltage."

Does this explain how the CVR in my truck works? Many thanks.

Charles

 

[ericgibbs]

Eric,

Thanks for your warm welcome, and I apologize for not yet knowing how to accurately describe the voltage regulator in my truck. Here is my understanding of the regulator: (1) It takes the 14v input from the alternator. (2) It turns the 14v output on and off in order to produce an average output of 5v.

Here is another description I've found:

"The CVR's purpose is to provide an average voltage to give consistent readings with voltages ranging from as little as 6V up to 16V. The CVR is in fact a early mechanical 'logic' circuit and is one of the first steps to the computer controled cars of today. Today it is called pulse width modulation, but what ever you call it the basic concept is by varing the ratio of 'on' time to 'off' time or % dwell to produce a target "average" voltage."

Does this explain how the CVR in my truck works? Many thanks.

Charles

hi,
The PWM describes the method used OK.
It would be possible to 'rectify' and smooth this PWM signal to give a 'dc' voltage.
This dc voltage would be connected to a comparator circuit [type posted earlier], which would be set to switch at a predetermined voltage level.

ie: low fuel level warning.

EDIT: I should ask what level of electronics expertise and equipment do you have to build this project.??

 

[egil]

My expertise is none. My desire to add the indicator light has inspired me to learn, with the ultimate goal of figuring out how to design and build one. But I also realize that I will have much to learn about before I can ultimately tackle this. So I've been reading the newbie FAQ links to start. I'm open to any and all suggestions about books, sites, beginner projects so that I can learn more. Thanks, again.

Charles

 

[ericgibbs]

My expertise is none. My desire to add the indicator light has inspired me to learn, with the ultimate goal of figuring out how to design and build one. But I also realize that I will have much to learn about before I can ultimately tackle this. So I've been reading the newbie FAQ links to start. I'm open to any and all suggestions about books, sites, beginner projects so that I can learn more. Thanks, again.

Charles

Hi,

I will help and I am sure others will also be keen to help you get started.

Some links:
Digital Electronics Tutorial

Electronics Tutorial



Let us know when you are ready to build and we will help with diagrams.

 

[stuhagen]

Charles,

Welcome...you will find this forum VERY helpfull. Wonderfull people here. I might suggest starting a new thread with your request, and "link" this thread to it. This one is getting a bit confusing with all the other people's interst.

I have finished my project. I have also made it somewhat universal in usage. I will post the schematic and circuit board layout here in a few days. I have made sepatate "pads" for the relay so it can be used as a switch "on-off" as well as supplying a voltage source. It would just be a matter of adding a "jumper" wire to the circuit board. I also added a LED so you would know if the circuit is on or off. Next step is a box, wiring, and a Bourns 10 turn pot.

Stu

PS: A lot of cars, most Japanese and GMC use PWM. My understanding is it can be read as a flat DC voltage.

 

[RogerTango]

Thinking outside the box, you can use a Picaxe chip to sample the voltage, and have it either close or open a circuit (via MOSFET?) at a programmable value.

Andrew

 

[427mustang]

One more silly question. In the circuit diagram, what does the SOT below R3 signify?

 

[stuhagen]

One more silly question. In the circuit diagram, what does the SOT below R3 signify?

The SOT is for the Adjustment of the "dead band" Depending on the usage of this circuit. Dead band is a feedback loop for sampling and making the "return" state a bit less. I asked for this because I didnt want to "osciallate" the circuit at the trigger point. I think Eric determined that a .2v deadband was all I was going to get out of this circuit. But that is good enough since that represents a few psi on the MAP sensor.

BTW, I have tested the circuit, got all of the Gerber files, and am close to get proto's made. If your interested, LMK. It can work an any MAP sensor, and all GM 3 bar and 2 bar sensors. Or any 0-5v sensors. There all real linear.
For me, 10psi = 4V and 17psi = 4.8v...so a high precision 10 turn pot is really needed. My test uses two full 360 degree rotations for only .8v range. Pretty cool~!

Stu

 

[stuhagen]

Final Circuit

Here is "my" final circuit. I added a few things, like a LED so I know the relay is on or off. I added a "jumper" because my initial usage is for supplying a 12v supply to a device. If I do this, then I just wire the jumper. If I want to use this circuit as a "on-off" or a "off-on" switch, then I would not use this jumper. Think of it as a typical SPDT relay that can be N/C or N/O.

Hope that people can use this. It was derived from a public forum, so I am posting my results. I have installed this into a 1"x3"x4" hobby box, with a Bounes 5K-10 turn pot sticking out. It works great on the bench........but it has been too cold to install in my car and go test. I am sure it will work fine.

Thanks to Eric and members for there feedback. (PS I never grounded out pins 5&6, but I may when my proto's come in)

Stu

 

[Roff]

Here is "my" final circuit. I added a few things, like a LED so I know the relay is on or off. I added a "jumper" because my initial usage is for supplying a 12v supply to a device. If I do this, then I just wire the jumper. If I want to use this circuit as a "on-off" or a "off-on" switch, then I would not use this jumper. Think of it as a typical SPDT relay that can be N/C or N/O.

Hope that people can use this. It was derived from a public forum, so I am posting my results. I have installed this into a 1"x3"x4" hobby box, with a Bounes 5K-10 turn pot sticking out. It works great on the bench........but it has been too cold to install in my car and go test. I am sure it will work fine.

Thanks to Eric and members for there feedback. (PS I never grounded out pins 5&6, but I may when my proto's come in)
Stu

Your LED is pointing the wrong way.

 

[stuhagen]

Not that it isnt hard to fix, it works for me, but I always thought the resistor for a LED always went to ground. In other words, the short leg of an LED (-) is where I attached the resistor. The long leg is (+) and goes direct to 12V. Correct me If I am wrong, but isnt the 2 arrows pointing up to (+) ? Or is that top the (-) ?

Stu

 

[ericgibbs]

Not that it isnt hard to fix, it works for me, but I always thought the resistor for a LED always went to ground. In other words, the short leg of an LED (-) is where I attached the resistor. The long leg is (+) and goes direct to 12V. Correct me If I am wrong, but isnt the 2 arrows pointing up to (+) ? Or is that top the (-) ?

Stu

hi Stu,
The LED 'arrow' has to pointing in the direction of the 'conventional' current flow.
That is +V to the anode [triangle] and the -V to the cathode [bar], always include the LED current limit resistor.

What do the W3/4 terminals indicate, if its for the external pot it dosnt look correct.???

 

[stuhagen]

I corrected the circuit above. After looking at it, it really is hard to see what it does, so I added the component that this circuit is controlling. It's not really a Motor, but a VSV or Vacuum Switching Valve. When 12v is applied, it is "open", when 12v is removed, it is closed.

Stu

 

[427mustang]

The SOT is for the Adjustment of the "dead band" Depending on the usage of this circuit. Dead band is a feedback loop for sampling and making the "return" state a bit less. I asked for this because I didnt want to "osciallate" the circuit at the trigger point. I think Eric determined that a .2v deadband was all I was going to get out of this circuit. But that is good enough since that represents a few psi on the MAP sensor.

BTW, I have tested the circuit, got all of the Gerber files, and am close to get proto's made. If your interested, LMK. It can work an any MAP sensor, and all GM 3 bar and 2 bar sensors. Or any 0-5v sensors. There all real linear.
For me, 10psi = 4V and 17psi = 4.8v...so a high precision 10 turn pot is really needed. My test uses two full 360 degree rotations for only .8v range. Pretty cool~!

Stu

Stu

That's some good info, but it doesn't answer my question, which is 'what is an SOT?'

 

[ericgibbs]

Stu

That's some good info, but it doesn't answer my question, which is 'what is an SOT?'

Tom,
I have already replied to your email of this morning, regarding SOT.

SOT or AOT means Select On Test and Accept On Test.

It means that feedback resistor is chosen to SET the hysteresis [dead band] that suits your application
and its value is determined by the tests you do.

 

[427mustang]

Eric

Much thanks as always.

Tom

 

[stuhagen]

Circuit update

Well, thought I would throw this update out there and see if there are any thoughts for a fix. This circuit has been tried a lot lately. but there is a small issue with it. What appears to be going on is the lack of adjustment to a higher threshold. It has been acceptable to some but not to me.

The car sensor's voltage range, like most cars, is 0-5V. In this particular application, the maximum reachable sensor output voltage is about 4.8v Not sure why, just might be its maximum range for the sensor.

Here is the problem. In the real time test, the circuit cannot trigger on anything greater than about 4.4v. Once it gets past this voltage, the circuit will not trigger, even if the sensor voltage appears to be out putting a higher voltage. Lower the circuit fires as it should, or it reaches the circuits Vsrc succesfully. I need this circuit to go higher, or at least to 4.5v. Someone mentioned that the sensitivity of the car's computer, or the fact that there is a "tap" wire from the sensor to the cars ECU might be too much of a load for ECU and it cannot provide the max voltage output. They also mentioned that raising the Vsrc inline resistance to closer to 1meg from 10K might help to reduce the "load" taken from the 0-5v sensor output. I have attached a revised circuit so you dont have to dig back farther in this thread. If the sensor can reach 4.8v at it's peak, then why is the circuit failing to respond to anything 0ver 4.4v?

Stu

 

[Roff]

According to my calculations, the trip point will not go below 4.4V. Are you sure you are describing the problem correctly?
To make the minimum threshold lower, reduce the value of R8.

 

[stuhagen]

OK, well, maybe a better explanation. The origianl design had a 5.1Z and no resistor to ground on the R4 10K pot. Since my needs were vary sepcific I modified the circiut. All I wanted to adjust was from 4.0v to 4.7v So I installed a resistor to ground on the Pot leg to restrict the voltage from going below around 4.0v. No sense have a Pot with to much adj range. I also was having an issue with the 5.1z restricting the upper voltage range to under 4.8v. I wanted the pot to have the ability to turn in either direction fully and be from 4.0v to 5.0v, since that is what this circuit was all about.

So I can set the R4 to full counterclockwise and it will trigger once Vsrc gets to 4.0v with no problems. I can turn R4 Clockwise and adjust the trigger voltage as the Vsrc climbs linear. But, once i set R4 to 4.4+, the incoming Vsrc is not triggering the circuit. Now, an obvious test would be to have a meter on the Vsrc in real time as see what's going on. Well, I have done this. I can see the Vsrc coming in at a max of about 4.78v, but it is not triggering the circuit. So there is something going on in the sensitivity to my circuit where it is not able to go above this 4.4v to the IC or something.

Stu

 

[Roff]

You either have a bad part, or your schematic does not match your hardware. I'm guessing the latter.