EEVBlog Video 262 Please help

[rscott]

Hi guys

So i was really interested in video number 262
I thought it was a nice design and set out to build it in the simulator.

For the life of me i can't make it work.

Does anyone know why? Im using the exact same values and parts from the video.

Also please note, the P chanel has a threshold voltage of zero because its "ideal" for the sim.
Also the LED needs a forward current of .005.

Also i did think that maybe the switch in the sim is switching to fast or slow for the RC circuit to catch so i have tried many different size caps will still no luck

Also I've varied the supply voltage from 20-5 volts


Everything else is just like the video with no success

[joey120373]

I used that circuit in a project of mine as well, didn't even breadboard it, just slapped it on the board and hoped for the best  .

I suspected, as Dave eluded to, that i might have to play around with the values a bit, and i, He wasn't wrong.

I cant remember exactly what i did to make it work in my situation, but I do remember i had to increase C1 to 100-120uF, i think i also added a resistor (100K i think) from the base of Q3 to ground ( there by halfing the base voltage,  and helping ensure it stayed off when i wanted it off ). I also think i played around with R3 a bit. Its a  simple and neat circuit, but IMHO its not a plug and play circuit. It needs to be "tuned" for your application. There are other less analog, more digital circuits floating around the web that are less sensitive to component choices if you are looking for something that requires less fiddling. Might take some searching to find one that is a cheap to build as this one but i think you could get close. Adafruits soft power switch uses a fairly cheap ( $0.51 ) nand gate and a FET ( $0.45 ) is one example, may not be as cheep to build as this one but its still ~$1.00....

Joe

[rscott]

Hi joe, thanks very much for the reply. I appreciate your input

Im just curious how he get his working on the breadboard with those values.

I don't have the intuition to make adjustments to this in the correct manner.

[rscott]

anyone have any other good ideas?

I really need this thing to start working correctly
How is it possible to work in the video and not in the sim. Its clearly set up the exact same way

[Andy Watson]

Also please note, the P chanel has a threshold voltage of zero because its "ideal" for the sim.

So there is not much chance of the mosfet turning off!

Once the voltage on the load has dropped below the forward voltage of the LED (1.2V ?) there is no load on the circuit. Any leakage - no matter how small - must go through the base of the transistor Q3 - which in turn is amplified and pulls the gate of the mosfet further away from its off-state.

How is it possible to work in the video and not in the sim. Its clearly set up the exact same way

As you have discovered, because it requires mammoth amounts of computing power to make a completely accurate simulation, some corners are cut and some "ideal" components or conditions are assumed.

When you transfer this simulation to real components there will be other traps revealed by the non-ideal components, e.g. does the simulation account for button-bounce? 

Edit: Just noticed that R3 and the base of Q2 will allow take some current away - but only until the voltage drops to 0.6-0.7V - still enough to start Q3.

Another thought on simulation: If your are having problems with a simulation, break it down into small chunks and simulate these in isolation. Build the simulation from small bits that work and, importantly, you understand how they work.

[alsetalokin4017]

Well, I can't address the issue of why it doesn't work in the sim, but I've just breadboarded it and it works fine on my breadboard.

I used IRF9540n  mosfet, BC337-25 NPN transistors, a 47 uF capacitor and the 100k and 1M resistor values as in the video. An LED+220R for the "load", 5V supply voltage, and Bob's my Uncle. It works, first time and every time! I did notice that touching the Drain flange on the mosfet will turn the circuit ON (but not OFF). The switch works fine, and if the switch is held down the circuit oscillates as expected.