# Foldback Current Limiter Question

Status
Not open for further replies.

#### p5taylor

##### New Member
The circuit attached is of a foldback current limiter, what determines the 20mA at the low end of the current limiting, and is there a formula to calculate this.

Kind Regards

Paul

#### Attachments

• 41.3 KB Views: 2,468

#### MikeMl

##### Well-Known Member
What is the emitter resistor of Q2 supposed to be?

I couldn't find either of the transistors in my LTSpice library, so I substituted ones that have similar specs, especially Hfe.

The circuit is very sensitive to the actual Hfe of Q1 (therefore, IMHO not a very good circuit).

First figure shows the current limiting as a function of the load resistance, R4 with a WAG for R3. Red trace is the voltage output. Green trace is the current through R4. Note that for load resistance < 18.6Ω, the output current is ≈ constant at 20mA. For load resistance > 24Ω, the output current is inversely proportional to R4, so no current limiting there. Between those values, the fold-back is happening.

Second figure shows the effect of changing R3 from 2K to 3K to 4K (order: left to right). Note that R3 seems to effect the onset of current limiting, but doesn't effect the folded-back value of 20mA very much.

Third fig. shows the folded-back current as a function of Load resistance R4 while R2 is varied from 50K to 80K in steps of 10K (order: Green, red, lt. blue and dk. blue, respectively). The f.b. current is a function of how much base current flows out of Q1.

This should give you something to work with.

#### Attachments

• 218.9 KB Views: 553
• 241.8 KB Views: 493
• 173.8 KB Views: 559

#### crutschow

##### Well-Known Member
Yes, that circuit depends upon the Hfe of Q1 to determine the current limit and is thus a very poor design. It's current limit will depend upon the particular transistor you happen to have, since Hfe varies widely from unit to unit.

#### p5taylor

##### New Member
Do you know of a better design that doesnt rely on the hfe thanks so far guys for you help.

#### ronsimpson

##### Well-Known Member
What voltage? What current? How much foldback?

#### crutschow

##### Well-Known Member
Here are some examples.

#### p5taylor

##### New Member
Hi Ron
As per original drawing 14v, 390ma down to 20mA

Thanks

#### unclejed613

##### Well-Known Member
you could get a similar effect using a constant current source as the limiter. the output voltage would begin decreasing when the current limit is reached, but the current will remain constant at the set-current of the CS. this is sometimes confused with foldback limiting, but it's only the voltage which folds. the required power dissipation for the current source is higher than a true foldback limiter.

#### p5taylor

##### New Member
unclejed613 that sounds interesting, i never thought of it that way to be honest so this would be a voltage foldback circuit? so in a current foldback the current decreases but do the voltage remain the same?

Kind Regards

#### p5taylor

##### New Member
Hi Dougy83 is there a circuit that will do that then?

#### schmitt trigger

##### Well-Known Member
Crustchow provided a good selection of web pages which show the many ways foldback current limiting can be implemented.

But as one of the app notes there warns, foldback limiting has some caveats:

" Linear foldback can have problems tripping current limit during start-up and returning to full load after a fault condition. These problems tend to lock up the Regulator in current limited state"

Therefore, if you implement a current foldback scheme, don't make it very aggressive...in other words, that the foldback is no less than 40 or 50% of the full rated current.

#### ericgibbs

##### Well-Known Member
Therefore, if you implement a current foldback scheme, don't make it very aggressive...in other words, that the foldback is no less than 40 or 50% of the full rated current.
I agree, this overhead rated current percentage is essential.

A poorly designed foldback circuit based on the rated load current can/will cause problems.

E

#### schmitt trigger

##### Well-Known Member
Because of the latchup problems, some people prefer "hiccup" current limiting schemes.

#### unclejed613

##### Well-Known Member
commercial vendors like Power-1 don't usually have the foldback percentage at an aggressive level either, usually something like 70-80% of the trip current (20-30% reduction). looking at a commercial supply schematic (one that uses standard off-the shelf components) would be very useful

some schematics and other information is here:
http://www.djerickson.com/p1hack/

Last edited: