BC547 hfe

[davidbear]

Still a newbe, but working on it. I wanted to use a BC547 NPN transistor to power a blue LED. Did the calculations and arrived at the circuit below.

To my surprise, I measure the base current (~800 uA) and the collector current (14 mA) and found an Hfe of ~18. I then got out two other BC547s and placed them in the circuit and found the same thing. Am I mistreating my transistors? Am I making a bad calculation? These are cheep transistors, so am I expecting too much? Any thoughts would be appreciated.

DavidBear

The data sheet is https://www.electro-tech-online.com/custompdfs/2009/01/BC547.pdf

 

[Nigel Goodwin]

Your completely bizarre circuit is messing up any readings as it doesn't work properly.

What are you actually trying to do?.

 

[davidbear]

I'm sorry my circuit is bizzare, it seemed like a good idea at the time. Where there is a switch, I actually have a counter, which does not have the current to drive the blue led. I am just trying to turn the LED on and off at something like maximum brilliance. I had anticipated an hfe of 200, and was quite disapointed with the low light level of the LED. Perhaps you would be so kind as to suggest a better switching circuit? If you can clear up any of my confusion, it will certainly be appreciated.

DavidBear

 

[Nigel Goodwin]

Assuming the red 'blob' is the LED?, move it to the collector instead - that will make the circuit work correctly.

 

[eng1]

I'm sorry my circuit is bizzare, it seemed like a good idea at the time. Where there is a switch, I actually have a counter, which does not have the current to drive the blue led. I am just trying to turn the LED on and off at something like maximum brilliance. I had anticipated an hfe of 200, and was quite disapointed with the low light level of the LED.

In order to drive a transistor into saturation, you have to force a base current greater than Icsat/beta by a factor (say 10 for example).
This current gain is called forced-beta because you set it at will. If it's 18 in your circuit, then it's ok.
Of course you have to move the LED as suggested first.
The 100 ohm resistor could be too high. A blue LED has a voltage drop of 3.6V @ 20 mA typically.
I'd use a 56 ohm resistor with a 4.8V voltage source.
You can check/recalculate the base reistor.

 

[davidbear]

Yes the blob is the LED. It was shown that way by the simulation (Circuit Simulator Applet) software I was using to check my circuit. I remain confused as to why the LED needs to be connected to the collector rather than the emitter, but when I hook the circuit up with the LED on the collector, I measure a hfe of 186, well within my expectations. Thanks for the help.

 

[Boncuk]

... and remove the diode between the switch and the base resistor.

 

[steveB]

I agree with the feedback that everyone is giving the OP, but I was wondering something. Isn't he correct to wonder why the measured current gain is only 18?

That transistor is supposed to have a much higher gain, obviously. Also, it is clear that the gain will be much lower once in saturation. However, based on the voltage readings in his figure, the transistor is just barely entering the saturation region (i.e. Vbc=4.4 mV, Vce = 0.64 V).

I would still expect higher gain under the conditions he quoted.

Can anyone explain this?

 

[Nigel Goodwin]

I agree with the feedback that everyone is giving the OP, but I was wondering something. Isn't he correct to wonder why the measured current gain is only 18?

That transistor is supposed to have a much higher gain, obviously. Also, it is clear that the gain will be much lower once in saturation. However, based on the voltage readings in his figure, the transistor is just barely entering the saturation region (i.e. Vbc=4.4 mV, Vce = 0.64 V).

I would still expect higher gain under the conditions he quoted.

Can anyone explain this?

Placing the LED in the emitter and the resistor in the collector is so incredibly bizarre?, it would be bad enough putting BOTH in the emitter. But doing it this way gives negative feedback as well, drastically reducing current gain. While the LED in the emitter isn't a resistor, it will still give negative feedback in association with the collector resistor.

 

[steveB]

Placing the LED in the emitter and the resistor in the collector is so incredibly bizarre?, it would be bad enough putting BOTH in the emitter. But doing it this way gives negative feedback as well, drastically reducing current gain. While the LED in the emitter isn't a resistor, it will still give negative feedback in association with the collector resistor.

Yes, I understand what you are saying. However, my question is much more fundamental. I'm simply noting the quoted conditions for the transistor as follows:

Ic=14.21 mA
Ib=789.69 uA
Vce=638.91 mV

If I compare this to Fig. 1 in the data sheet for the BC547, it doesn't seem to agree. If I look at Vce and Ic on the graph, it seems to predict that Ib would be about 10 times less.

So, while I agree with the recommended circuit changes that people are suggesting, I'm just as baffled as the OP as to why the current gain is so low in his circuit.

 

[davidbear]

Dear All,

Thanks for the suggestions. It appears that when I use the circuit below, my measured readings are now in agreement with a transistor with an hfe of ~220. I am confused as to why having the LED on the collector is that different than having it on the Emitter. I am not understanding the "negative feedback" comment, nor do I see why putting a resistor in the circuit is bizarre, I was just trying to limit over current in case the hfe was too high. My confusion doubtlessly arises from a deeper misunderstanding of transistors in general, but for now I am willing to accept my ignorance and keep moving. I'll just remember to place my switchable loads on the collector side. If anyone has a web reference that clarifies this issue, it will be appreciated. Also, if anyone knows of a free simple schematic drawing program that I can use, my LEDs will no longer look like red "blobs."

DavidBear

 

[Hero999]

You still need a series resistor, use 220R for a red LED run at 5V.

 

[kchriste]

I'm simply noting the quoted conditions for the transistor as follows:
Ic=14.21 mA
Ib=789.69 uA
Vce=638.91 mV
If I compare this to Fig. 1 in the data sheet for the BC547, it doesn't seem to agree. If I look at Vce and Ic on the graph, it seems to predict that Ib would be about 10 times less.
So, while I agree with the recommended circuit changes that people are suggesting, I'm just as baffled as the OP as to why the current gain is so low in his circuit.

The reason is because the transistor is in saturation (Vce=638.91mV) so the extra base current just passes out the emitter. Assuming a beta of 200 and a VLED of 2.6V: If you raised the 4.8V (on the collector resistor only) to 25V you'd see apx 160ma in the LED and then smoke and fire.
With real parts the current would be lower because transistors exhibit lower beta at higher collector currents.

 

[davidbear]

Hero,

Actually, I don't think need the 220 Ω resistor (BTW the LED is blue), since the transistor is limiting the current to 30 mA when on. Would it be better design to put the series resistor with the LED and remove the base resistor? I was thinking that with the current design that I would be giving the circuit a higher input impedance, thus limiting the power requirements of the circuit.

Since I am not an engineer, I really don't know about best practices.

DavidBear

P.S. - The 30 mA is a real number from a real circuit (as are the 4.8V [I was anticipating a voltage of 5V]and 138 uA base current) using the blue LED. Simulation suggests that this circuit would create a current of 38 mA with an input voltage of 6V and 12 mA with a voltage of 3.1V.

 

[audioguru]

Can't you paste the real symbol of an LED and its color into your schematic drawing software? Cant you make the schematic a positive with a white background instead of a negative with a black background?
Can't you remove the dots on the wires?

 

[steveB]

The reason is because the transistor is in saturation (Vce=638.91mV) so the extra base current just passes out the emitter. Assuming a beta of 200 and a VLED of 2.6V: If you raised the 4.8V (on the collector resistor only) to 25V you'd see apx 160ma in the LED and then smoke and fire.
With real parts the current would be lower because transistors exhibit lower beta at higher collector currents.

Trust me, I understand what you are saying, but somehow I'm not making myself clear. The characteristic curves shown in the data sheet are representative of the transistor performance. Fig. 1 shows the Ic vs. Vce plots (for various Ib) including saturation effects. The point I mentioned (Vce, Ic and Ib), which is quoted from the OP, does not match the curve. In fact, it is very different.

Basically, I'm curious if anyone can explain why. The numbers presented don't add up. Although the transistor is in saturation, it has only just barely entered that region. Thus, I would expect higher current gain. The curves confirm this fact. The only thought that crosses my mind is that perhaps the OP took DC measurements with a meter (i.e. without an O-scope), and in fact had oscillations present in the circuit.

Anyway, I guess it doesn't matter now since the OP seems to have things working. I was just curious.

 

[Pommie]

You are relying on the gain of the transistor to limit the LED current which is not a good idea as the gain can vary greatly from one device to the next. Reduce the base resistor to around 1k and insert a resistor in series with the LED to obtain the correct current.

Mike.

 

[Pommie]

Basically, I'm curious if anyone can explain why. The numbers presented don't add up. Although the transistor is in saturation, it has only just barely entered that region. Thus, I would expect higher current gain. The curves confirm this fact. The only thought that crosses my mind is that perhaps the OP took DC measurements with a meter (i.e. without an O-scope), and in fact had oscillations present in the circuit.

Think of what would happen if the collector was open circuit, a base current will still flow and so the gain would be zero. In the OPs circuit the collector current reached the maximum it could and the only way to get the transistor back on the curve would be to increase the collector voltage (as pointed out by kcriste).

Mike.

 

[audioguru]

Post a video of the LED blowing up because the hFE of the transistor is higher than normal.

 

[steveB]

.... In the OPs circuit the collector current reached the maximum it could and the only way to get the transistor back on the curve would be to increase the collector voltage (as pointed out by kcriste).

Mike.

This doesn't make sense. The transistor always operates somewhere on it's characteristic curve such as shown in Fig. 1 of the data sheet. According to the characteristic curve, Vce should be lower when current gain is 18. For example, Vce of maybe 0.3 V could match up with Ic=14 mA and Ib=790 uA.

Please look carefully at Fig. 1 in the data sheet before trying to answer this question. That region of the curve is a little cramped, you must look closely to see what I'm saying. This is not a newbie question.