I'll need to investigate this further. I rebuilt the circuit and I am getting the same results. Point is, I would not release a circuit like that unless I personally tested it and made sure that it worked. I'll get back to you soon.
Thank you very much for your insight. I believe you have hit the nail on the head with this problem. I am going to have to think about this one.
Beau,
You rebuilt the circuit and are getting the same results as when you built it in back in June 2009 (and I presume it worked then), or the same results that I am getting currently? Thank you so much for your help. I also just discovered the other thread regarding a very similar topic which I will read:
Try it with four diodes in the emitter circuit, instead of two, or else a 2.4V zener. I think the key is that the switch threshold has to be higher than the brownout threshold. OTOH, that may not give you enough drive for the pMOSFET, depending on the input voltage.
I will update the schematic later. apparently the regulator that I originally built the circuit with either did not have any 10uF capacitors or they were much smaller values.
The simple fix is to replace the 330 Ohm resistor with a 47k resistor. Additionally you need to add a 0.1uF capacitor from the Base of the NPN transistor to Ground.
To rebuild the circuit, the Regulator that I used is a ua78m33 with a 10uF capacitor on the output to Ground. The MOSFET that I used was a IRF4905
Note: For the auto shut off to work, you need to have the USB programming cable disconnected or else there is ground loop formed that allows enough parasitic power to turn things on.
@Beau, I've been following this thread for a while, since it peaked my interest in shut-down via my remote control. I got these values to operate, and the only annoyance is when I turn on my overhead florescent lights. They give off a brief 'spike' of light that gets processed as 'infrared' @ 38KHz. But, I'm not sure that is all that bad either? Kind of neat have it come on with the flip of a switch!
On the down side ... you cannot read the 'switch' (which isn't there anyway) as the 'IR' input provides info regarding shut down. There is no resistor between the 'OFF' pin and the 1uF cap (which is a no-no), however the idea is to get as much charge into the shut off cap before it dies and the propeller has a 40ma limiter in each pin, so this worked out nicely! This circuit is not able to reach all your original 'goals' ... but, it was just what I was looking for! Thank you.
Other notes: I changed the 100K to 10K, to protect the IR input pin and still operate the MOSFET. The IR receiver is 'open-collector' output.
But as I can see most of designs on forum lack that ones.
IF I say that way ---> Every IC need an decoupling capacitor on its V-Rail ----> V-Reg are an IC to ---> and have 2 V-Rails
@Beau, I've been following this thread for a while, since it peaked my interest in shut-down via my remote control. I got these values to operate, and the only annoyance is when I turn on my overhead florescent lights. They give off a brief 'spike' of light that gets processed as 'infrared' @ 38KHz. But, I'm not sure that is all that bad either? Kind of neat have it come on with the flip of a switch!
On the down side ... you cannot read the 'switch' (which isn't there anyway) as the 'IR' input provides info regarding shut down. There is no resistor between the 'OFF' pin and the 1uF cap (which is a no-no), however the idea is to get as much charge into the shut off cap before it dies and the propeller has a 40ma limiter in each pin, so this worked out nicely! This circuit is not able to reach all your original 'goals' ... but, it was just what I was looking for! Thank you.
Other notes: I changed the 100K to 10K, to protect the IR input pin and still operate the MOSFET. The IR receiver is 'open-collector' output.
But as I can see most of designs on forum lack that ones.
If you mean 'I'm sloppy' ... your right.
Do you mean the IR receiver should have one? Or, the ones I used are too small? Hey, I was just glad to under-stand where Beau was going with the 2 doides in the emitter!
And now I'm pondering perhaps replacing them with a color LED (I heard different colors have different forward drops) of the correct forward drop? But alas, w/11K I doubt I'll see it! Perhaps 1.5K and 15 ohm in series instead of 10K and 1K, which would restore Beau's original ratio of 100K to 1K, and see if the propeller can take it? With tiny amounts of current the 100K makes the MOSFET to 'touchy'.
But, whatever I do ... I must keep a value > 1.65 (or whatever voltage a solid '1' is?) on the collector of the NPN (with 2 diodes in emitter) ... then the IR receiver can (gnd - open - gnd - ...) Open comes in as '1' and the 38KHz signal grounds like Beau's switch did. Therefore, when you press 'any' key on the IR remote ... it turns on!
You are correct -- I missed that one (IR) it gives 5 decoupling capacitors ---> 100nF
Look on attached pic.
It is not that You are 'sloppy' ---> But as I see most of people on forum think as if them have 1uF capacitor and biger them have already one decouplig capacitor.
BUT that is not case ---> As in filters that in V-Reg constructions that parallel capacitors have diferent function.
100nF are decoupling one , 1uF and bigger are as some sort Batery that shall smoth curent on its Input and Ouput!
Do you mean the IR receiver should have one? Or, the ones I used are too small? Hey, I was just glad to under-stand where Beau was going with the 2 doides in the emitter!
And now I'm pondering perhaps replacing them with a color LED (I heard different colors have different forward drops) of the correct forward drop? But alas, w/11K I doubt I'll see it! Perhaps 1.5K and 15 ohm in series instead of 10K and 1K, which would restore Beau's original ratio of 100K to 1K, and see if the propeller can take it? With tiny amounts of current the 100K makes the MOSFET to 'touchy'.
But, whatever I do ... I must keep a value > 1.65 (or whatever voltage a solid '1' is?) on the collector of the NPN (with 2 diodes in emitter) ... then the IR receiver can (gnd - open - gnd - ...) Open comes in as '1' and the 38KHz signal grounds like Beau's switch did. Therefore, when you press 'any' key on the IR remote ... it turns on!
I will update the schematic later. apparently the regulator that I originally built the circuit with either did not have any 10uF capacitors or they were much smaller values.
The simple fix is to replace the 330 Ohm resistor with a 47k resistor. Additionally you need to add a 0.1uF capacitor from the Base of the NPN transistor to Ground.
To rebuild the circuit, the Regulator that I used is a ua78m33 with a 10uF capacitor on the output to Ground. The MOSFET that I used was a IRF4905
Note: For the auto shut off to work, you need to have the USB programming cable disconnected or else there is ground loop formed that allows enough parasitic power to turn things on.
Dear Beau,
After modifying the circuit as you stated, it behaves very differently.
The push button now must be held down until the prop starts up and holds its pin High. After that, as soon as the prop makes it's pin an input to read the state of the button, the power shuts down.
I have used the attached code to test this circuit and determine at which point the circuit shuts off.
I tried lowering the 47K Ohm resistor to 10K, but that reverted the operation back to what is was before, causing the brownout to reset and not allow the power to shutdown.
The 0.1 cap didn't seem to make any difference.
Let me know if I should give up and go with one of those fancy IC V regs that have all the extra pins.
I'm sorry your having so much trouble with this circuit. What does your schematic look like? What other devices are connected? Power management needs to be applied everywhere in a circuit like this and can be very tricky. You run a real risk of creating a ground loops that can cause unwanted behavior of not turning on or off properly.
Comments
Thank you very much for your insight. I believe you have hit the nail on the head with this problem. I am going to have to think about this one.
Beau,
You rebuilt the circuit and are getting the same results as when you built it in back in June 2009 (and I presume it worked then), or the same results that I am getting currently? Thank you so much for your help. I also just discovered the other thread regarding a very similar topic which I will read:
http://forums.parallax.com/showthread.php?138694-Using-Propeller-to-control-it-s-power-source-Beau-Schwabe-Challenge!/page2
Thanks,
Marcus
Try it with four diodes in the emitter circuit, instead of two, or else a 2.4V zener. I think the key is that the switch threshold has to be higher than the brownout threshold. OTOH, that may not give you enough drive for the pMOSFET, depending on the input voltage.
-Phil
The simple fix is to replace the 330 Ohm resistor with a 47k resistor. Additionally you need to add a 0.1uF capacitor from the Base of the NPN transistor to Ground.
To rebuild the circuit, the Regulator that I used is a ua78m33 with a 10uF capacitor on the output to Ground. The MOSFET that I used was a IRF4905
Note: For the auto shut off to work, you need to have the USB programming cable disconnected or else there is ground loop formed that allows enough parasitic power to turn things on.
On the down side ... you cannot read the 'switch' (which isn't there anyway) as the 'IR' input provides info regarding shut down. There is no resistor between the 'OFF' pin and the 1uF cap (which is a no-no), however the idea is to get as much charge into the shut off cap before it dies and the propeller has a 40ma limiter in each pin, so this worked out nicely! This circuit is not able to reach all your original 'goals' ... but, it was just what I was looking for! Thank you.
Other notes: I changed the 100K to 10K, to protect the IR input pin and still operate the MOSFET. The IR receiver is 'open-collector' output.
On Yours SCH -- I see 4 caps to little.
But as I can see most of designs on forum lack that ones.
IF I say that way ---> Every IC need an decoupling capacitor on its V-Rail ----> V-Reg are an IC to ---> and have 2 V-Rails
If you mean 'I'm sloppy' ... your right.
Do you mean the IR receiver should have one? Or, the ones I used are too small? Hey, I was just glad to under-stand where Beau was going with the 2 doides in the emitter!
And now I'm pondering perhaps replacing them with a color LED (I heard different colors have different forward drops) of the correct forward drop? But alas, w/11K I doubt I'll see it! Perhaps 1.5K and 15 ohm in series instead of 10K and 1K, which would restore Beau's original ratio of 100K to 1K, and see if the propeller can take it? With tiny amounts of current the 100K makes the MOSFET to 'touchy'.
But, whatever I do ... I must keep a value > 1.65 (or whatever voltage a solid '1' is?) on the collector of the NPN (with 2 diodes in emitter) ... then the IR receiver can (gnd - open - gnd - ...) Open comes in as '1' and the 38KHz signal grounds like Beau's switch did. Therefore, when you press 'any' key on the IR remote ... it turns on!
... Tim
You are correct -- I missed that one (IR) it gives 5 decoupling capacitors ---> 100nF
Look on attached pic.
It is not that You are 'sloppy' ---> But as I see most of people on forum think as if them have 1uF capacitor and biger them have already one decouplig capacitor.
BUT that is not case ---> As in filters that in V-Reg constructions that parallel capacitors have diferent function.
100nF are decoupling one , 1uF and bigger are as some sort Batery that shall smoth curent on its Input and Ouput!
Dear Beau,
After modifying the circuit as you stated, it behaves very differently.
The push button now must be held down until the prop starts up and holds its pin High. After that, as soon as the prop makes it's pin an input to read the state of the button, the power shuts down.
I have used the attached code to test this circuit and determine at which point the circuit shuts off.
I tried lowering the 47K Ohm resistor to 10K, but that reverted the operation back to what is was before, causing the brownout to reset and not allow the power to shutdown.
The 0.1 cap didn't seem to make any difference.
Let me know if I should give up and go with one of those fancy IC V regs that have all the extra pins.
Thanks again,
Marcus
PowerTest.spin
I'm sorry your having so much trouble with this circuit. What does your schematic look like? What other devices are connected? Power management needs to be applied everywhere in a circuit like this and can be very tricky. You run a real risk of creating a ground loops that can cause unwanted behavior of not turning on or off properly.