a couple of current limiters

I was playing with a couple of simple overload-protection current limiters, and expected to need to add loop compensations. To my surprise, both were stable. They would of course usually not run in current limit; the opamps would normally be railed and the fets low-ohmic. So a little oscillation wouldn't be a show-stopper. But still.

Version 4 SHEET 1 2028 680 WIRE 400 -176 -112 -176 WIRE 560 -176 400 -176 WIRE 672 -176 560 -176 WIRE 1312 -176 672 -176 WIRE 144 -112 48 -112 WIRE 208 -112 144 -112 WIRE 400 -96 400 -176 WIRE 1312 -96 1312 -176 WIRE -112 -64 -112 -176 WIRE 48 -64 48 -112 WIRE 672 0 672 -176 WIRE 1312 32 1312 -16 WIRE 1312 32 1040 32 WIRE 48 48 48 16 WIRE 400 64 400 -16 WIRE 400 64 112 64 WIRE 1168 96 1120 96 WIRE 1200 96 1168 96 WIRE 400 112 400 64 WIRE 672 144 672 80 WIRE 208 160 208 -112 WIRE 1120 160 1120 96 WIRE -112 176 -112 16 WIRE -16 176 -112 176 WIRE 176 176 -16 176 WIRE 1040 176 1040 32 WIRE 1088 176 1040 176 WIRE 1312 176 1312 32 WIRE 304 192 240 192 WIRE 352 192 304 192 WIRE 1216 192 1152 192 WIRE 1264 192 1216 192 WIRE 112 208 112 64 WIRE 176 208 112 208 WIRE 960 208 928 208 WIRE 1088 208 960 208 WIRE 208 272 208 224 WIRE 1120 272 1120 224 WIRE 400 288 400 208 WIRE 480 288 400 288 WIRE 544 288 480 288 WIRE 400 320 400 288 WIRE 544 336 544 288 WIRE 1312 352 1312 272 WIRE 1392 352 1312 352 WIRE 1456 352 1392 352 WIRE 1312 384 1312 352 WIRE 1456 400 1456 352 WIRE 400 416 400 384 WIRE 544 464 544 416 WIRE 1312 480 1312 448 WIRE 400 528 400 496 WIRE 1456 528 1456 480 WIRE 1312 592 1312 560 FLAG 672 144 0 FLAG 208 272 0 FLAG 400 528 0 FLAG 48 48 0 FLAG 544 464 0 FLAG 560 -176 V+ FLAG 480 288 OUTX FLAG 304 192 GX FLAG 1120 272 0 FLAG 1312 592 0 FLAG 1456 528 0 FLAG 1392 352 OUTY FLAG 1216 192 GY FLAG 144 -112 +12 FLAG 1168 96 +12 FLAG -16 176 VREF FLAG 960 208 VREF SYMBOL voltage -112 -80 R0 WINDOW 0 57 36 Left 2 WINDOW 3 48 73 Left 2 SYMATTR InstName V1 SYMATTR Value 65m SYMBOL nmos 352 112 R0 WINDOW 0 121 8 Left 2 WINDOW 3 80 87 Left 2 SYMATTR InstName M1 SYMATTR Value BSC080N03MS SYMBOL Opamps\\UniversalOpamp2 208 192 R0 WINDOW 0 39 38 Left 2 SYMATTR InstName U1 SYMBOL res 384 -112 R0 WINDOW 0 64 45 Left 2 WINDOW 3 62 80 Left 2 SYMATTR InstName R1 SYMATTR Value 0.5 SYMBOL voltage 672 -16 R0 WINDOW 0 63 43 Left 2 WINDOW 3 70 81 Left 2 SYMATTR InstName V2 SYMATTR Value 6 SYMBOL cap 384 320 R0 WINDOW 0 -65 16 Left 2 WINDOW 3 -69 50 Left 2 SYMATTR InstName C1 SYMATTR Value 0.01 SYMBOL voltage 48 -80 R0 WINDOW 0 57 36 Left 2 WINDOW 3 59 69 Left 2 SYMATTR InstName V3 SYMATTR Value 12 SYMBOL res 528 320 R0 WINDOW 0 64 45 Left 2 WINDOW 3 65 78 Left 2 SYMATTR InstName R2 SYMATTR Value 70 SYMBOL res 384 400 R0 WINDOW 0 -67 40 Left 2 WINDOW 3 -69 75 Left 2 SYMATTR InstName R3 SYMATTR Value 5m SYMBOL Opamps\\UniversalOpamp2 1120 192 R0 WINDOW 0 40 35 Left 2 SYMATTR InstName U2 SYMBOL res 1296 -112 R0 WINDOW 0 -60 36 Left 2 WINDOW 3 -59 74 Left 2 SYMATTR InstName R4 SYMATTR Value 0.5 SYMBOL cap 1296 384 R0 WINDOW 0 -65 16 Left 2 WINDOW 3 -69 50 Left 2 SYMATTR InstName C2 SYMATTR Value 0.01 SYMBOL res 1440 384 R0 WINDOW 0 -49 38 Left 2 WINDOW 3 -45 74 Left 2 SYMATTR InstName R5 SYMATTR Value 70 SYMBOL res 1296 464 R0 WINDOW 0 -64 40 Left 2 WINDOW 3 -68 75 Left 2 SYMATTR InstName R6 SYMATTR Value 5m SYMBOL pmos 1264 272 M180 WINDOW 0 127 96 Left 2 WINDOW 3 94 13 Left 2 SYMATTR InstName M3 SYMATTR Value FDS4435A TEXT -36 455 Left 2 !.tran 0 1 50m uic TEXT -96 352 Left 2 ;Output Driver Current Limiters TEXT -24 400 Left 2 ;JL Oct 17, 2017 TEXT 456 160 Left 2 ;NFET TEXT 1376 216 Left 2 ;PFET

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin
Loading thread data ...

The second P-fet one would work off the 6V supply, which might be useful. How are you getting the 65mV reference? I guess some variation in the reference won't bother you too much.

George H.

Reply to
George Herold

Railing the op amps means they have lost control. You are requesting 65e-

3/0.5 = 0.13 A. The series string of 0.5 and 70 ohms can only supply 6/70.5 = 8.51e-2 A, so the op amps rail.

The 0.01 caps are actually 0.01 farad, or 10,000 uF. That is why the output take so long to hit saturation.

Change the caps to 0.01uF, and reduce the load resistors from 70 ohms to 39 ohms. Now you can see how the circuits work.

Reply to
Steve Wilson

My paragraph just above says that is the normal state of the circuit.

You are requesting 65e-

Yes. It's an overload protector, and I want the amps to be normally railed, and the fets to normally be milliohms of on resistance.

The huge cap is an easy way to make the output voltage ramp, from ground up, to scan the E:I behavior. The real load would be more like a few uF.

I can see how it works now. I was just surprised that both circuits, especially the nfet one, were stable first try.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin

In real life, I'd vary the +6 from maybe 2.5 to 6. This might be the current limiter for a pulse generator output stage, with programmable logic levels. That's speculative. One of my hobbies is designing pulse generator output stages; I don't have an immediate use for my new idea, but it would need a current limited supply.

If I use the pfet version, with Vhigh at 2.5 maybe, I'd need a negative supply for the opamp, to get a hunky gate drive and low Ron. The nfet circuit is mildly weird, but I can drive that gate from +12 or so, which I'd have.

The 65 mV could be divided down from a bandgap or a negative regulator that's referenced to the Vhigh rail. Or something simpler, like this

============+============ Vhigh rail +2.5 to +6 | | R | +-------------- | | | c gnd----b NPN e | | R | | | -5V

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin

On 10/17/2017 09:05 PM, John Larkin wrote: > I was playing with a couple of simple overload-protection current > limiters, and expected to need to add loop compensations. To my > surprise, both were stable. They would of course usually not run in > current limit; the opamps would normally be railed and the fets > low-ohmic. So a little oscillation wouldn't be a show-stopper. But > still.

Here's an extremely simple linear current limiter.

formatting link

One depletion-mode MOSFET, one resistor.

The main downside is that the sense resistor is rather large, so you waste some power in this. If you add an op amp, the loss can be much reduced.

I use this in an application where the voltage ranges from

120V down to 3V, and it works well over that range.

Design note:

formatting link

If you just want to turn something off on overcurrent, there are lots of ICs for that. Every USB port has one.

John Nagle

Reply to
John Nagle

Yeah, depletion fets are good, but they are not very precise and add a lot of resistance.

I want to current-limit a power supply that can range from 2.5 to 6 volts, and I want low voltage drop until it limits, and I want pretty good accuracy of the limit current. That has to be an active circuit with a small sense resistor.

There are a few chips that could do this, but the prices are crazy.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.