op-amp for my power supply?

[Sajjadkhan]

op-amp for my power supply,current sense?

Hi! i am making my own lab power supply 0-30V and 3-4 amps.

First thing i am stuck in is the current sense.

I know how to sense current but the problem is i want to use presion 0.1 ohm resistor. measured current resolution would me 10mA. so 10mA x 0.1 = 1mA.
previously i was using lm358 i also have lm158 but there offset voltage does not go below 5mV. so you guys know any LM series of op-amps which has 1mV of offset even at higher temperature. max ambient temperature here in summer climbs to 40 degree C.
Another option is to use 0.5 ohm resistor to get the required offset voltage but naaaa i don't wana generate extra heat + keeping the resistor size small and i want that R.sense to PCB mount, not on any heat sink, it will look ugly and add more weight.Thanks.

*note: LT and MAX series are not in my country, AD series might be.

 

[goldsmith]

Dear Sajjadkhan
Hi
Don't confuse about that current sensor resistor . you can amplify the voltage across it with Av= 50 with your current fed op amp simply .
Thus your precision will increase .
Best Wishes
Goldsmith

 

[alexan_e]

10mA x 0.1 = 1mA.

You mean 1mV

Don't confuse about that current sensor resistor . you can amplify the voltage across it with Av= 50 with your current fed op amp simply .
Thus your precision will increase .

The input offset is amplified too so the gain doesn't change the problem

 

[goldsmith]

Dear Alex
Hi
my mean was with differential amp with op amp , i used this way more than hundreds ! and successfully .
Best Wishes
Goldsmith

 

[alexan_e]

Are you saying that in a differential amplifier the input offset will not be amplified too?

 

[goldsmith]

It was my problem , that why when i used that arrangement , the circuit worked fine , because i expected that the offset destroy , my circuit .
Would you like that i attach my circuit , here ( what worked fine ) ?
Best Regards
Goldsmith

 

[alexan_e]

You mean that you have a opamp that has 5mV input offset , you have a 1mV per 10mA input from the shunt resistor and the output can give you a better accuracy than 50mA (which equals the input offset of 5mV)?
Yes I would be interested to see that.

 

[goldsmith]

Hi again , as you can see , i designed that circuit with an awful op amps that called 741 ! you can control your out put current from some uA ( instead of v5=0 and up to each ampere that you required .)

View attachment circuit for alexan.pdf
View attachment circuit 2 for alexan.pdf
Respectfully
Goldsmith

 

[alexan_e]

I don't see how your differential amplifier with a gain of 50 can give a better resolution that the opamp input offset voltage.
The input offset voltage can be anywhere from 1mV to 5mV (LM741) which will introduce the result error but you expect to get an accuracy of say 1mV in the input?

So if the input is 1mV and the opamp has 1mV offset you will get the same result as when you have 1mV input and 5mV offset?

 

[goldsmith]

How about this ?



Best Regards
Goldsmith

 

[alexan_e]

Is the simulation model you have a static constant offset so you can use such tricks but in a real opamp the offset will be within a range and you can't even be sure which input is higher than the other, it can be the positive or negative.

Doing a calibration each time you change the opamp will improve the accuracy but there will still be a drift for the offset which will introduce an error while the circuit operates.

 

[goldsmith]

A magic thing is that i used this circuit in practice and it worked fine , and if you want i can take a picture from working of my circuit .
Best Wishes
Goldsmith

 

[alexan_e]

I'm not saying that it doesn't work, I'm saying that you can't have a accuracy better that 50mA which is what the original poster requests.
You can only achieve it if you use an opamp with a lower input offset which is what he asked for.

 

[goldsmith]

Again i say , it worked with high precision . and i don't know why , but did you test it in practice until now ?
Anywhere , if the problem is op amp selection , you can use a comparator like LM311 or .....
Respectfully
Goldsmith

 

[Sajjadkhan]

@ alexan_e yes i ment 1mV.

obviously i would agree alexan_e if your op-amp can not detect below 5mV than how would it amplify 1mV. op-amp would be blind in this case. your circuit might be working for say 50mA and you might not have noticed it.

 

[goldsmith]

Hi again
If your problem , is really , with offset , you can improve it with a simple potentiometer .
Best Wishes
Goldsmith

 

[Sajjadkhan]

and what is the affect of temperature on input offset voltage, there is no graph in any op-amp datasheet telling this. all values are @ 25 degree C.

---------- Post added at 16:24 ---------- Previous post was at 16:21 ----------

ok suppose i use a wire wound 10 turn pot. how much is it reliable vs temperature? as you know even a little drift will cause a big error!

 

[goldsmith]

How much precision do you need exactly , and why this high precision ? is this important really ? ( my mean is the effect of temperature on offset voltage ? are you making spacecraft project ? haha !
If you tell me a bit more about thing that you want to do , perhaps i can help you at better way .
Good luck
Goldsmith :-D

 

[Sajjadkhan]

well i passed out my university like a year ago finall i am selcted for a job and i have 4 months free. so i thought to make my own lab power supply which should be reliable and accurate up to mili amps of current yet powerful. for example if i have to limit current in mili amps then it should first sense current accurately in miliamps. if iuse 10 bit ADC then 1024/100 = 10.24 which means i can measure upto 10mA of current. thats why

 

[goldsmith]

No problem , i think with an standard comparator you'll be able to do that , or with better opamps e.g LM358 or LF356 or LM118 or the other types . or with an ADC and a microcontroller ! or with a hall effect sensor .
Wish you the best
Goldsmith