BJT Matching?

Thread Starter

eng kryptonite

Joined Jan 25, 2010
17
What degree of matching in emitter area is required for the base-emitter voltages of two bipolar devices with identical collector currents to have a difference of less than 4.6 mV if we use VT = 25.85 mV? How to approach this? Its quite confusing for me. Thanks! : )
 

Ron H

Joined Apr 14, 2005
7,063
Here is how I would go about it.
For a given current, the current density is inversely proportional to emitter area. Think of the diode equation and current ratio.
 

Thread Starter

eng kryptonite

Joined Jan 25, 2010
17
Here is how I would go about it.
For a given current, the current density is inversely proportional to emitter area. Think of the diode equation and current ratio.
If all I have are the VT & the voltage .. how would I go with the diode equation and current ratio? Can you please show some figures so I can follow up or explain a little further? I still don't understand the question :confused:
 

Ron H

Joined Apr 14, 2005
7,063
Start with diode equations for the two transistor base-emitter junctions:
\( I_1=I_s*(e^{\frac{V_{be1}}{V_T}}-1)\)
\( I_2=I_s*(e^{\frac{V_{be2}}{V_T}}-1)\)
(Keep in mind what I said previously about emitter area and current density.)
For reasonable current levels, you can ignore the -1 term in each equation.
Now, take the natural log of the ratio \(\small \frac{I_1}{I_2}\). (I'm not going to do all the work for you).:)
You should come up with an equation that contains \(\small Vbe1-Vbe2\). Do you recognize what that difference represents? Plug it and \( \small V_T\) into your equation, then take the antilog. What answer do you get?
 

Thread Starter

eng kryptonite

Joined Jan 25, 2010
17
Thanks Ron .. The idea became much better now and I have came up with a solution thats attached. Nothing brilliant :D but now this thing makes sense to me. Sounds good to you?

what I got is 1.78%.
 

Attachments

Ron H

Joined Apr 14, 2005
7,063
My version of Word won't open your file.
Your answer is ln(I1/I2). You need I1/I2.

EDIT: ln(I1/I2)=0.178. Even if this were the correct answer, it is not equivalent to 1.78%.:confused:
 
Last edited:

Ron H

Joined Apr 14, 2005
7,063
\(\frac{I_{s1}}{I_{s2}}=e^{\frac{4.6}{25.85}}=e^{0.17795}\)

Are you aware that e = 2.71828..., which is the base for Naperian logarithms?

\( \frac {I_{s1}}{I_{s2}}=1.1948\)

Since \( I_s\) is proportional to emitter area, then the areas must match within ≈20%.
 

Thread Starter

eng kryptonite

Joined Jan 25, 2010
17
Ooops :D
I missed out the "e" part :)
for some reason under all this stress I calculated that as (10^x) ..
Thanks a lot Ron .. your help is appreciated once again.
 
Top