using a bjt in the linear region

Thread Starter

Gibson486

Joined Jul 20, 2012
355
When I use a BJT in the linear region, can current be more than in the saturation region? Or is max current in the saturation region?
 

studiot

Joined Nov 9, 2007
4,998
The correct name for what you have called the linear region is the active region. It is not truly linear over the whole range of the active region but only linear over a very limited sub range.

As the collector current increases the transistor passes from the active region to the saturated region at which point the current does not increase.

So the maximum active current can equal, but never exceed the saturation current.

Does this help?
 

w2aew

Joined Jan 3, 2012
219
The correct name for what you have called the linear region is the active region. It is not truly linear over the whole range of the active region but only linear over a very limited sub range.

As the collector current increases the transistor passes from the active region to the saturated region at which point the current does not increase.

So the maximum active current can equal, but never exceed the saturation current.

Does this help?
Hmmm - that's really not true. Saturation occurs when the collector voltage drops to the point that the base-collector junction gets turned on. This can happen at very low collector currents when using a large resistor in the collector for example.

Similarly, with little or no resistance in the collector, such as with an emitter follower, the collector current can be quite high and the transistor will be no where near saturation.
 

studiot

Joined Nov 9, 2007
4,998
This can happen at very low collector currents when using a large resistor in the collector for example.
OK so how would you then drive more current through the transistor, without changing the power supply?
 

w2aew

Joined Jan 3, 2012
219
OK so how would you then drive more current through the transistor, without changing the power supply?
It's all in the biasing. Picture a circuit where the base is biased to 1.7V, and there is a 1K resistor connected from emitter to ground. There is approximately 1V across the 1K resistor (1.7 -Vbe), so there is about 1mA of emitter current, thus pretty close to 1mA of collector current. If the collector was connected to a 5V supply via a 1K resistor, then the voltage at the collector would be 4V (5V - 1mA*1K). If that resistor was changed to 4.7K, then the collector voltage would "want" to be 0.3V, but instead the transistor saturates when the b-c junction turns on at about 1 or 1.2V or so, thus collector current would be limited by the supply and collector and emitter resistors - to a little less than 1mA.

Same circuit, but this time make the collector resistor zero ohms. Reduce emitter resistor to 100 ohms. Now, the collector current is 10mA, and the transistor isn't saturated.

Thus, the collector current range in the active region can easily exceed the collector current in saturation - it all depends on the biasing.
 

Thread Starter

Gibson486

Joined Jul 20, 2012
355
The reason I ask is because I am trying to control a solenoid with a BJT transistor. The issue is that the voltage drop on Vce prevents me from fully opening it. As a result, I thought that I could put it in active region and have the current go above what it would be in saturation.
 

w2aew

Joined Jan 3, 2012
219
The reason I ask is because I am trying to control a solenoid with a BJT transistor. The issue is that the voltage drop on Vce prevents me from fully opening it. As a result, I thought that I could put it in active region and have the current go above what it would be in saturation.
It's difficult to say what is happening in your circuit without seeing a schematic - i.e. what is connected to the base and emitter - these things may be limiting the available current. If you just need it to act as a switch, a MOSFET might be a better choice...
 

Thread Starter

Gibson486

Joined Jul 20, 2012
355
It's difficult to say what is happening in your circuit without seeing a schematic - i.e. what is connected to the base and emitter - these things may be limiting the available current. If you just need it to act as a switch, a MOSFET might be a better choice...
I need it to modulate...perhaps I just modeled it wrong. I have no resistors in the emitter or collector and I probably need to add it. Basically, the valve is in the collector and the base has a resistor divider to set the current a voltage (currently pots at the moment). Judging from what you wrote, I probably did it all wrong....

The original intent was to control with PWM, but I can not get a response until I get to a 40% duty cycle.
 

Attachments

studiot

Joined Nov 9, 2007
4,998
Now I understand your point.

Yes if you change the circuit configuration the currents will be different.

And yes it is possible to find one configuration where the saturation current is less than the active current in another.

This follows from the characteristic curves of a transistor.
 

w2aew

Joined Jan 3, 2012
219
I need it to modulate...perhaps I just modeled it wrong. I have no resistors in the emitter or collector and I probably need to add it. Basically, the valve is in the collector and the base has a resistor divider to set the current a voltage (currently pots at the moment). Judging from what you wrote, I probably did it all wrong....

The original intent was to control with PWM, but I can not get a response until I get to a 40% duty cycle.
You mentioned earlier that this is a solenoid, and based upon your statement here, I'm assuming it is a solenoid valve of some sort? Most solenoids are designed to be either "on" or "off", and not linearly controlled to something in the middle. Since you mentioned PWM, it sounds like you are trying to take "analog" control over the valve/solenoid position. I think you'll find this difficult to do unless the device is designed to be driven this way.

Is the device designed to operate properly with simply a voltage placed across its coil, or does it need some current limiting? If it's the former, you might easily cook the transistor.

In general, a switching MOSFET will give you lower "on" resistance, and thus lower burden voltage in series with the coil when it is turned on.

PWM of an ordinary solenoid will generally not result in well controlled linear position.

So, more detail is needed again, I'm afraid. If your intent is some type of analog control over a mechanical device, then maybe a servo is a better choice.
 

Thread Starter

Gibson486

Joined Jul 20, 2012
355
You mentioned earlier that this is a solenoid, and based upon your statement here, I'm assuming it is a solenoid valve of some sort? Most solenoids are designed to be either "on" or "off", and not linearly controlled to something in the middle. Since you mentioned PWM, it sounds like you are trying to take "analog" control over the valve/solenoid position. I think you'll find this difficult to do unless the device is designed to be driven this way.

Is the device designed to operate properly with simply a voltage placed across its coil, or does it need some current limiting? If it's the former, you might easily cook the transistor.

In general, a switching MOSFET will give you lower "on" resistance, and thus lower burden voltage in series with the coil when it is turned on.

PWM of an ordinary solenoid will generally not result in well controlled linear position.

So, more detail is needed again, I'm afraid. If your intent is some type of analog control over a mechanical device, then maybe a servo is a better choice.
These are made for modulating control.

http://www.clippard.com/downloads/PDF_Documents/Product%20Data%20Sheets/EVP_Proportional_Valve_Data_Sheet.pdf

The original intent was PWM, but I could not get it come on until 40% duty cycle. Then, I tried adjusting current on the base to change current on the collector (non PWM). VCE prevents me from getting it fully open unless I compensate by raising V+. As the data sheet says, it takes some current to get the initial position going.
 

w2aew

Joined Jan 3, 2012
219
These are made for modulating control.

http://www.clippard.com/downloads/PDF_Documents/Product%20Data%20Sheets/EVP_Proportional_Valve_Data_Sheet.pdf

The original intent was PWM, but I could not get it come on until 40% duty cycle. Then, I tried adjusting current on the base to change current on the collector (non PWM). VCE prevents me from getting it fully open unless I compensate by raising V+. As the data sheet says, it takes some current to get the initial position going.
Controlling collector current via setting of the base current is not a great way to go. In the active region, it depends on the transistor's Hfe (Beta), which is not a very well controlled parameter. It will also vary a lot with temperature, and it goes out the window when the transistor saturates.

For PWM - a switching MOSFET will probably be less hassle - when driven on/off, it acts more like a voltage controlled switch than a BJT does. I assume you'll be relying on the internal inertia of the valve to filter the PWM pulses.

Another route would be a PWM (or other) controlled current source.
 

crutschow

Joined Mar 14, 2008
34,280
PWM is generally the preferred way to generate a varying voltage/current with a solenoid. You likely needed a 40% duty-cycle for a response because the solenoid requires at least 40% of the supply voltage to operate.

Make sure that you have enough base drive to fully saturate the transistor when it is ON. That means the base current should be at least 1/10th of the collector current. If the transistor is of sufficient size, then the ON voltage should be no more than a few tenths of a volt. For that you just need a proper value resistor in series with the base. You need no collector or emitter resistors.
 

Thread Starter

Gibson486

Joined Jul 20, 2012
355
PWM is generally the preferred way to generate a varying voltage/current with a solenoid. You likely needed a 40% duty-cycle for a response because the solenoid requires at least 40% of the supply voltage to operate.

Make sure that you have enough base drive to fully saturate the transistor when it is ON. That means the base current should be at least 1/10th of the collector current. If the transistor is of sufficient size, then the ON voltage should be no more than a few tenths of a volt. For that you just need a proper value resistor in series with the base. You need no collector or emitter resistors.
I like PWM because it is easy and cheap (no DAC needed). That said, I really need 100% of the span, which is why I am trying these quirky ideas to get more than 60%of the span that I currently get with PWM.
 

BobTPH

Joined Jun 5, 2013
8,804
I like PWM because it is easy and cheap (no DAC needed). That said, I really need 100% of the span, which is why I am trying these quirky ideas to get more than 60%of the span that I currently get with PWM.
This does not make any sense.

There are just as many intermediate numbers between 40 and 100% as there are between 0 and 100%. If the range of 40% to 100% does not give you enough resolution, you need to increase the resolution of the pulse width.

I.e. if you have 100 steps between 0 and 100% (i.e. each step is 1%) you need to make each step 0.6% to get 100 steps between 40% and 100%.

Bob
 
Top