# Parallel voltage regulator

Discussion in 'General Electronics Chat' started by tim90, Jan 27, 2016.

1. ### tim90 Thread Starter New Member

Jan 15, 2016
20
0
Hi,to everybody, i am using this circuit in the picture

The main problem is related to the two resistors R3 and R4 (to balance the two outputs), because since the maximum current is 15A, i have a voltage drop across the resistor (15A*0.01OHM), so the load regulation will be reduced. I already tried to put the feedback resistor (r1 and r2) after r4 but nothing change. Do you have some idea on how to solve the problem?

2. ### blocco a spirale AAC Fanatic!

Jun 18, 2008
1,461
370
That's pretty much from the datasheet. The voltage drop at 15A, due to the resistors, is only 0.075V so do you really have a problem that requires fixing?

3. ### tim90 Thread Starter New Member

Jan 15, 2016
20
0
no, the voltage drop is 0.15 v (15*0.01), i know also because i simulated it and yes, 015v is to much for my purpose

4. ### blocco a spirale AAC Fanatic!

Jun 18, 2008
1,461
370
But you aren't drawing 15A through each resistor they are effectively in parallel so it's 7.5 x 0.01 or 15 x 0.005 = 0.075V

Evanguy likes this.

May 20, 2015
673
194
See

6. ### hp1729 Well-Known Member

Nov 23, 2015
2,098
232
Those resistors low ohm resistors could be replaced by wire in the wire harness, assuming you have about two feet of #18 wire in a harness.

7. ### ScottWang Moderator

Aug 23, 2012
4,932
777
I haven't use LT1803, but if I want to use then I won't use its rating current as 7.5A to the real output, maybe the largest is 5A or less for each chip, so if I want to get 15A then it will be needs 3 piece, it is depends on how hot of the chip, if the chip is too hot then maybe use 4 piece.

8. ### blocco a spirale AAC Fanatic!

Jun 18, 2008
1,461
370
This won't help as it doesn't eliminate the resistors, it just replaces them with a different type of resistor i.e. copper cable.

I agree with ScottWang about using more regulators rather than operating them at the max 7.5A current, this will also further reduce the troublesome output resistance.

9. ### dl324 Distinguished Member

Mar 30, 2015
3,378
651

It seems that you require relatively high currents, but you're paralleling 3 terminal regulators and not using remote sensing to compensate for voltage drops in the conductors.

10. ### tim90 Thread Starter New Member

Jan 15, 2016
20
0
The current varies for 0.1A to 15A and i need a load regulation of some ten of millivolts. What do you think if i built a circuit like this:

is it enough precise?

11. ### Bordodynov Active Member

May 20, 2015
673
194
See:

cmartinez likes this.
12. ### crutschow Expert

Mar 14, 2008
13,501
3,375
Bordodynov, I simulated you circuit in LTspice but the output oscillated due to high loop gain from M1.
Perhaps there's a difference in our TL431 models.
To stabilize it I changed R1 to 1kΩ and add a 10μF capacitor between the gate and drain of M1.

Also the output capacitance need not be that high.
10μF to 100μF should be sufficient.

13. ### Bordodynov Active Member

May 20, 2015
673
194
I apologize for the misinformation.
I agree with you. Yes, I took this model, but it was necessary to take a smaller step. I was surprised that I did not need the frequency response correction. I modified my stabilizer.
The output capacitor I ranged 10 uF - 10,000 uF.
Now I checked my stabilizer with different models TL431.

Last edited: Jan 29, 2016
14. ### AnalogKid Distinguished Member

Aug 1, 2013
4,697
1,297
No, the voltage drop is 0.075 V because only half the current if going through each resistor. I pointed this out on the other forum where you posted this question.

And nothing speaks to inexperience so much as taking a simulation result over the advice of a seasoned, professional designer. Blocco is correct; you, and your computer, are wrong.

ak

15. ### tim90 Thread Starter New Member

Jan 15, 2016
20
0
AnalogKid, It is not necessary to insult. I was sure about the voltage droop, and I have simply supposed that Blocco didn't read each values correctly (it can happens) but when he claims the same thing again, I understood that i was wrong. As you can see from my posts I always write " is it right?" " what do you think?" and other stuff like this because i am not presumptuous, i know the limits of my knowledge and i know that there are many people with an incredible knowledge like Blocco. Anyway, sorry Blocco.

Whent I connect together the ground at the end of the load and the ground on R3 (feedback resistor),since we are speaking about tens of amps,also the parassitic resistance of ground will create a preoblem. Below R3 we will not have anymore 0V but a voltage equal to the current through the load times the parassitic resistance. Is it right? If yes, how can I solve this problem?

16. ### ISB123 Well-Known Member

May 21, 2014
1,240
531
You have to trust a man with the mustache.

17. ### tim90 Thread Starter New Member

Jan 15, 2016
20
0
ahhahaah Right!!!