Okay. I don't have time today to try your circuit.
I'll be experimenting with them both when I get the
time, and commit them to my notebook. I have a variety
of mosfet "pulls" in my junk box that I would like to cycle
thru this schematic, evaluate different caps, and try higher voltages with it too.

I may come back to this thread with questions about what is going
on with both of these circuits. Veeery interesting! Thanks to you both,
your helps since I joined this forum is enormously valued, & has
gotten me over the hump of my timidity regarding my restart in
electronics, and the overhead of the learning curve in this hobby.

Two thumbs up, one each!

 

I have used the enhanced Miller Effect on a circuit I did 25 years ago, but I forgot about it on this particular day. It really is a very slick way to smooth out the transition in the Vgs to Id curve. You can use a 10 cent part to, "buy some time" as demonstrated by the drawing crutschow provided. I think engineers call this, "elegant". Dirt cheap and amazingly effective, with plenty of versatility for longer or shorter time constants.

 

What is the hi side voltage limitation of the crutschow circuit?
Is it limited only by my selection of mosfet?

 

What is the hi side voltage limitation of the crutschow circuit?
Is it limited only by my selection of mosfet?

The transistor is the primary limit, but the capacitors also must be rated for the maximum applied voltage (derate everything about 50% for good reliability).

 

I feel an urge to say that if your power supply voltage exceeds the Vgs limit of the mosfet, you will need a zener diode from the gate to the source of the mosfet in order to protect the gate. 10 volts to 16 volts should do the job. Similar to your second drawing in post #10 or my drawing in post #11. My preference is 10 volts for the zener diode because it allows a gate voltage generously more than the absolute minimum and very safely below the 20 volt limit we usually see on the datasheets. You (I) never know when some manufacturer will reduce the voltage limit to 15 volts, so 10 volts should be safe for any random mosfet you have laying around, even if you can't find the datasheet.

 

Right. Most bricks are 12v to 32v. Might also use this circuit to slowly bring up the lights too.

I might have landed myself a technician job at my company. I need to get some sleep so I can be fresh
for what tomorrow brings. I will goof around with your circuits next chance I get.

 

I hope this thread can be revived...

I have now experimented with #12's circuit, and crutschow's circuit.

I am leaning towards crutschow's circuit, as it seems to operate well,
and there isn't a power brick in my junk box that hasn't bent to the will
of this circuit. Note: I did add a 10v zener between the gate & ground,
per #12's advice.

PROBLEM:

Although crutschow's circuit slows the sudden current demand imposed
on the smps, defeating the over-current protection, it only works ONCE, sort of .

So, if the load is applied, the circuit tricks the smps, allowing the large
halogen bulbs to light. If I remove the load, and quickly reapply it, the
lamps will light again. I can do this over and over, no problem. However,
If I remove the load for more than about 5 seconds, the crutschow circuit
no longer works to slow the current down, and the limiter in the smps brick
trips.

However, if I connect the smps + crutschow circuit+halogen, and then
plug the smps brick into the wall, then it works, over an over again, fast or slow,
no problem.

I've tried placing a bleeder resistor across each capacitor, or a sand resistor
as a dummy load, and different combinations of caps, but nothing changes.
If I remove the load and try to reapply it after a pause of more than about 5
seconds, the over current protection in the smps trips.

?

 

Stop doing that?