DC-DC converter glitch

Status
Not open for further replies.

zachtheterrible

Active Member
I built this circuit to drive a 1W LED and placed the whole assembly in a maglite. Everything works splendidly accept for one thing: If the flashlight gets jarred or dropped, the light will turn off, but the circuit still keeps on sucking current. I think there is a connection inside the flashlight that must get disconnected for a very short time, but still throw the circuit off. If you turn the flashlight off then back on, everything works fine. Sometimes though, when you turn the flashlight off then back on, the LED will come on for a half second then turn back off.

I tried using a 120 uf capacitor instead of the 22uf supply capacitor for C1 and this didn't make any difference. Space is a large concern so I can't really use any capacitor too big.

Any ideas to fix this? Thanks :lol:
 

Attachments

  • zxsc_400.gif
    31.8 KB · Views: 742
What is the value of R2?

If it is way too low, it could be 1/2 your problem, because it is using the power!

EDIT: the filter doesn't like the word that describes what vacuums do, that starts with S and rhymes with the swear word that starts with F.
 
Hiya Zac,
Eh mate while looking at the LED site you refered to in another thread I came across an application note on using a 12f675 pic to drive a 1-3 watt led. AN874 is the Ap note check it out mate as it looks like a good project.

Cheers Bryan
 
eh mstechca, the value for R2 is .82 ohms I believe. It is a low value current sense resistor.

eh mate :lol: , ill have to check that out when i get my bloody pic programmer to work. i think im going to just end up buying one :lol:
 
zachtheterrible said:
eh mate :lol: , ill have to check that out when i get my bloody pic programmer to work. i think im going to just end up buying one :lol:
Click to expand...

Hiya Zac,
Well if your going to buy a programmer check out sparkfun as they have a full range of programmers I lashed out and bought their ICD2 clone and I reckon I'll never look back. It's a bit pricey at $99 but it's a drop in replacement for the microchip one. Also they have much cheaper programmers too.

Just a Thought

Cheers Bryan
 
the value for R2 is .82 ohms I believe
Click to expand...
0.82 OHMS? :shock: :lol:
I think there is your problem right there.
I still think 82 ohms is a bit too low.

I have used IC's and 330 ohms even take the power out of the IC's and literally override normal circuit operation.

Nigel pointed out that he recommends a value between 10K and 100K as pull-up and pull-down resistors when dealing with IC's.

Of course that can be a ridiculous value to use here, because your LED won't shine.

Did you look at the specifications for the IC? pay special attention to anything talking about voltage and current.

I find it abnormal to connect an LED to a ridiculous low value resistor in series, and connect this in parallel with the supply, and connect the point between the LED and resistor to an INPUT pin of the IC.

I think you have done that.

Please don't forget about analog electronics, because they can affect digital electronics too.
 

You're missing the point of the circuit, the 0.82 ohm isn't a current LIMITING resistor, it's a current MONITORING resistor, so it needs to be low so as to not waste any more power than the minimum. Some circuits use far lower values, such as switchmode PSU's in many TV sets, I stock resistors from 0.1 ohm to 0.82 ohm (with 0.33 and 0.56 being the most popular) for repair purposes.
 
Yea, I got the values straight off the datasheet for the IC.

Bryan, I'm looking for the p16pro40 PIC programmer because it is supported by Nigel's software and covered in his tutorials. I found one at quasarelectronics, it was out of stock, but I think I'm gonna buy it right now. I've made two p16pro40's and I never got them to work. For $15 bucks I can have a guaranteed working one :lol:
 
MStechca,
Zach is using a special high-power LED that operates at hundreds of mA.
His circuit is a switch-mode to step up the battery voltage and has pulses of current into the LED of hundreds of mA.

Flashlights with the Luxeon LED and this circuit should be sold in Canadian Tire stores soon. Not the Noma ones for sale now using ordinary bright LEDs. I think Garrity will be making the high power ones.
Their few hundred mA looks as bright as an incandescent bulb that uses 1A or more. The incandescent bulb (not the battery) lasts about 10 hours. The LED lasts a lot longer.
 
Zach, If you don't have an excellent power supply to this circuit (example: battery), then I don't think it will last long (unless for some reason the LED's resistance is very high).
 
mstechca said:
Zach, If you don't have an excellent power supply to this circuit (example: battery), then I don't think it will last long (unless for some reason the LED's resistance is very high).
Click to expand...
Why? The IC supplies the LED with a regulated voltage and current, no matter what the battery does. An ordinary incandescent flashlight dims as the battery runs down. Not this regulated one.
 
He is mistaking the internal resistance of his worn-down battery as the LED's resistance. I recall a "smoke test" for batteries that he posted. I think his battery smoked more than its 1 ohm load! :lol:
 
audioguru said:
He is mistaking the internal resistance of his worn-down battery as the LED's resistance. I recall a "smoke test" for batteries that he posted. I think his battery smoked more than its 1 ohm load! :lol:
Click to expand...

Depends on the type of battery!, NiCd's have far greater capacity for self-destruction :lol:
 
I wonder what a shorted lithium battery would look like. I bet it would be spectacular!
 
MMMMMMMMM now you guy's have me thinking :twisted: I wonder if I thru a lithium ion battery down a hole I drilled in a rock I need removing say 3 foot deep then charged the battery to about 6 volts then short it out :twisted: what sort of result would I get. I'm thinking of a jet of flame back up the hole but If I added some petrol it should make a good show. Short of getting some tnt would mean a heap of paperwork and I do have about 2 dozen lithium ion batteries around and I do have a heap of rocks I need to either dig up or blowup so I can sow my hay paddock.

Cheers Bryan
 
I know I mentioned this once before, but a friend of mine's phone malfunctioned while charging and melted right into the floor :lol:. Probably a Li-ion battery.

Eh Audio, you said that they're going to be sold in tire stores??
I've already seen some flashlights with the luxeon in them at fry's electronics stores. None with any DC-DC converters in 'em though.

I'll send an email to zetex, the maker of my IC, and I'm sure they'll know exactly how to fix my problemo.
 
The Li-Ion battery will catch fire during overcharge, probably while you are pouring petrol on it! Ouch! It might burn your face off. :lol:
 
Status
Not open for further replies.

Similar threads

D
  • Question
Comms over DC power cable
Replies
8
Views
3K
dougy83
D
S
  • Question
isolated DC-DC convertor and mosfet driver
2
Replies
28
Views
5K
shortbus=
S
T
  • Question
Boost regulator
Replies
13
Views
4K
Tom120
T
S
Varying Speed AC Generator to Steady DC
2 3
Replies
48
Views
6K
Soul21
S
O
  • Question
DC motor not spinning with L293D motor shield
Replies
5
Views
2K
tepalia02
T
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…