analogue adder subtractor

Status
Not open for further replies.

chatumal

Newbie level 5
Joined
Aug 5, 2013
Messages
8
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Visit site
Activity points
58
Are there ready made analogue adder/ subtractor ics?

if so can someone give me some ic numbers.
 

well, its for an adc and opamp circuits would definitely work. but I need a number of adders/subtractors and so i was hoping they would come in quad packages so i could save space. if you know of such ics please let me know.

chatumal
 

Many op amps are available in quad packages. If the ADC input is unipolar then you can use a rail-rail type op amp with a single supply voltage such as the MCP604 for 5V applications.
 

re there ready made analogue adder/ subtractor ics?

if so can someone give me some ic numbers.

Hi chatumal

of course you can easily do all kind of mathematics with op amps . for adding or subtraction or multiplication or division or integral or differential or ... .
Best Wishes
Goldsmith
 

................
of course you can easily do all kind of mathematics with op amps . for adding or subtraction or multiplication or division or integral or differential or ... .
Op amps can indeed perform a lot of functions (they are rather the Swiss Army knife of analog circuits) but they cannot do multiplication or division unless you have added non-linear circuits to help perform those functions.
 

Op amps can indeed perform a lot of functions (they are rather the Swiss Army knife of analog circuits) but they cannot do multiplication or division unless you have added non-linear circuits to help perform those functions.
Hi crutschow

Such a good example ( the knife ha ha ! ) i've never heard of that kind of knife and it is my first time to hear about that .
Indeed !
Certainly either multiplication or division are requiring the e^alpha*t function which can be achieved by a simple diode or transistor .

Best Regards
Goldsmith
 

........................
Certainly either multiplication or division are requiring the e^alpha*t function which can be achieved by a simple diode or transistor.
Yes, but that generally works only for a limited range of single quadrant signals. For a wide range of 4-quadrant signals you need a Gilbert Cell type multiplier.
 
Yes, but that generally works only for a limited range of single quadrant signals

Hi crutschow
indeed !

For a wide range of 4-quadrant signals you need a Gilbert Cell type multiplier.
I'm glad because i'm learning new things from you . in fact i have never designed any kind of gilbert cell . may you tell me more about it please ? i had a bit search regarding that in google and i have found something but i couldn't understand how it works as well , so i'll be grateful if you tell me more about it .
Thanks in advance+ Best Regards
Goldsmith
 

You don't normally design a Gilbert cell unless you are making an IC. The Gilbert cell is used in virtually all analog multiplier/divider chips, which is the usual way to obtain one. Analog Devices, for example, makes many different models covering a wide range of performance and cost.

Edit: Basically a Gilbert cell consists of two cross-coupled differential transistor pairs. The key to their operation is that the gm (gain) of a transistor is proportional to the current through it. This non-linearity is used to generate the multiplication (or division) function between the two differential inputs in the cell.
 
Last edited:
you can use ic 741 operational amplifier. it can work as adder/subtractor if inputs are applied to both inverting and non-inverting terminals
 

You don't normally design a Gilbert cell unless you are making an IC
Hi crutschow

May i ask you why unless we are about to design an IC ? is there any problem with designing them ?
Let me confess when i want design a multiplier or divider i prefer to use AD633 which was very good till now .

Best Regards
Goldsmith
 

May i ask you why unless we are about to design an IC ? is there any problem with designing them ?
Let me confess when i want design a multiplier or divider i prefer to use AD633 which was very good till now .

.................
Mainly it's that the cell is complex and requires close matching of the transistors to minimize offsets in it's operation which is difficult with discrete transistors but easy on an IC. Since Gilbert cell IC multipliers are readily available there is no good reason to try to design a discrete one. You certainly can build one, if you like, but the resulting circuit will likely operate poorly and only be useful to demonstrate it's operation, not for practical use.
 

Hi crutschow
I really appreciate your help because i'm enjoying with learning new things .
in fact i don't want to use it in my projects i just want see how it works ( learning purpose ) but when i need to use it certainly i will use an integrated circuit .
So may you guide me through some design lines for Gilbert cell please ?

Thank you so much
Best Regards
Goldsmith
 

You might look at this or this. I've never designed one so that's all I know about them.
Hi crutschow
Thank you so much for that . those links are really helpful ( especially the HFA3101 link ) i will try to find some ways to design one of them and if i could find a simple way i will share with you .

Thank you again
Best regards
Goldsmith
 

Status
Not open for further replies.

Similar threads

Cookies are required to use this site. You must accept them to continue using the site. Learn more…