Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

an anolog CURRENT divider

Status
Not open for further replies.
T

tufan

Junior Member level 3
Joined
Oct 3, 2004
Messages
26
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
228
I am currently studying on an anolog CURRENT
divider, since I need to calculate the division :( of sum
of and difference of currents as:
(I1-I2)/(I1+I2) in an analog current domain by using BJT or MOS and without using voltage domain.So I need of a discerete component
availabe to do only division.If it is available in such an IC .Can you tell me about it?number and where can I find...and how can I divide with 4-5 transistor using :(
 

CURRENT DIVIDER WITH TRANSISTORS

HI ...I NEED TO DO ABOVE CALCULATION BY USING TRANSISTOR NOT OP-AMPS...SO IN FACT I NEED AN ANALOG CURRENT DIVIDER CIRCUIT.ANYONE CAN TELL ME SUCH A CIRCUIT SCHEMATIC... :?
 

Dpeneds on your application and specs, you may try current mirror.
 

    T

    tufan

    Points: 2
    Helpful Answer Positive Rating
THANKS BUT mirror lonely work to divide? DO YOU HAVE ANY CIRCUIT?
I TRID CURRENT MIRROR BUT I FAILED...PLEASE EXPLAIN IN DETAIL..

:roll:
 

SO ADD SOME INFO:

this is for a read-out circuit of a position sensing detector(psd),TWO CURRENTS ARE fotocurrents which occurs when laser beam comes. so I try to measure laser beams location on the psd taking difference and normalizing by dividing sum...since output of psd is in current domain I dont want to use op-amps that needs voltage conversion I simply want to do it with transistors so use less components and cause less noise..
 

You got to use opamps. Otherwise, it would be difficult. By the way, if, one day, you figure out a circuit to do this without an opamp or circuits similar to opamp, please let me know.
 

    T

    tufan

    Points: 2
    Helpful Answer Positive Rating
What are the values of the currents and how big do you expect the difference to be?
 

    T

    tufan

    Points: 2
    Helpful Answer Positive Rating
What do you mean current divider? By analogy with voltage divider it should be I/k, where I is current and k is some coefficient. In this case current mirror is the most simple and acceptable solution. But you want to divide not by coefficient. You want to divide current by current. What do you expect to get?
 

    T

    tufan

    Points: 2
    Helpful Answer Positive Rating
By dividing current to current I normalize the output therefore make less depend the illumunation level of the laser beam.

well;I hve another question I send you all the circuits for differencing and summing now can you help me please finding this circuit in discreate IC builded by any semiconductor company..


thanks ......
 

you can refer to
C.Toumazou,F.J.Lidgey & D.G.Haigh

Analog circuit design -- current mode.
 

    T

    tufan

    Points: 2
    Helpful Answer Positive Rating
Both your bipolar circuits are wrong. They never will work as you expected. If you want to use bipolar you need to read some book about bipolar device physics.
For subtractor an adder you can use cobination of current mirrors like in your CMOS examle. You can use both bipolar or CMOS mirrors. In your example both left and right parts (mirrors) are subtractor (or differencing) circuit. If you connect only drains of PMOS (or NMOS) transistors it will be adder. About divider I have no idea. But your circuit is not a divider and not a multiplier.
 

    T

    tufan

    Points: 2
    Helpful Answer Positive Rating
Here's a way to do it: the translinear principle of gilbert. But forget about this if you can't have matched transistors, as in discretes. Another way is to use a chip from analog:

AD539

But these are not cheap!
 

    T

    tufan

    Points: 2
    Helpful Answer Positive Rating
I don't think an opamp is necessary if you are planning to implement this
on-chip. The current multiplier/divider can be implemented by using MOS
transistors in the subthreshold region. Of course, you'll have to scale the
currents to something around 1uA, to keep the transistor aspect ratios
within acceptable levels (depends on your size constraint of course).
I'm assuming you are planning to multiply/divide the DC currents?
 

I try off-chip and dc current and currents are 2mAs
 

This idea used in some PFC controolers
 

Previous circuit is really nice idea!

Please also tell me what bandwidth you expect (or rise time/fall time/pulse width).
 

    T

    tufan

    Points: 2
    Helpful Answer Positive Rating
In PFC controoler this circuit works on frequency about 100Hz.
I never modell it to get bandwidth.
You can do it with a proper model of transitors
 

    T

    tufan

    Points: 2
    Helpful Answer Positive Rating
Hi friends ,
the pfc contro0ller circuit is realy nice but now I have a problem about the output transistor is not in linear region therefore I can not get any output.does it realy works?because simulations fail
 

How accurate do you need in your current divider output? if you need to be the range of highly accurate arithmetic operation (may be 0.1%? ) I think you must use some kind of opamp in feedback config.
 

I advise the following paper:

Klaas Bult and Govert Gleelen, An Inherently Linear and Compact MOST-Only Current Division Technique IEEE Journal of Solid-State Circuits, vol. 27 (12), Dec. 1992.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top