adding offset to opamp

Thread Starter

kubeek

Joined Sep 20, 2005
5,794
In the attached diagram you can see a small votlage source being added to the inputs of the amplifier and comparator. How would you realize such an offset in a real circuit?

I thought about a Vbe multiplier, but those can only have more drop than Vbe, not less. I am replicating this ideal diode controller with discrete parts, so in the real circuit (normal differential amplifier) I solved it by having slightly different voltage dividers for each input, but that makes the offset change with Vcc. What would you use to achive this small offset of inputs?lm5050.PNG
 

OBW0549

Joined Mar 2, 2015
3,566
In the attached diagram you can see a small votlage source being added to the inputs of the amplifier and comparator. How would you realize such an offset in a real circuit?
It's important to realize that when you see a diagram like this, the details are drawn to illustrate the function of the circuit, not to show how it's actually constructed. Those little voltage sources may not even exist as distinct physical entities; more likely, the comparators themselves are designed with a deliberately unbalanced input differential stage (e.g., the two transistors in the input diffamp have different emitter areas) that yields the same result.

The effect is as if a millivolt voltage source were placed in series with one of the inputs of an ideal comparator, but the actual implementation is something altogether different.
 

Thread Starter

kubeek

Joined Sep 20, 2005
5,794
Yes I understand that. My question is more pointed to how would you realize such an offset in a discrete circuit with normal comparator that has offset <1mV? With the differential amp I can see more solutions, but not with the comparator.
 

OBW0549

Joined Mar 2, 2015
3,566
My question is more pointed to how would you realize such an offset in a discrete circuit with normal comparator that has offset <1mV?
Most practical way, I think, would be to put a small resistor in series with the comparator input, and connect a constant-current source or sink to "pull" the comparator input either up or down, as needed. For example, a 1 kΩ resistor and a 30 μA current source will result in a 30 mV offset. Offhand, I can't think of any other solution that wouldn't result in a lot of head-banging.
 

ifixit

Joined Nov 20, 2008
652
You could use a matched transistor pair in 1 package (eg 2SA1873). Run a different base current in each until the difference in the base voltage is 30mV. This difference should remain constant over a reasonable temperature range. If biased from Vin through resistors then both base currents will change together and tend to keep the difference the same. This is easy inside an IC, but somewhat unpredictable on a circuit board.

Just an idea
Ifixit
 

wayneh

Joined Sep 9, 2010
17,496
In the attached diagram ....
Your post got me interested. I had looked into building an active rectifier some years ago but it was a little too complicated for me at the time. Your ideal diode got me thinking that things have probably advanced, and sure enough they have.

So far I've found a Linear IC that I'd like to try:
http://www.linear.com/product/LT4320

This looks like it would be a ideal for a hobbyist-level project to build a high-efficiency rectifier. Great for windmill generators, for instance.
 

Thread Starter

kubeek

Joined Sep 20, 2005
5,794
When I am back at work on monday I will post my discrete solution, which theoretically could be also used as a rectifier.
 
Top