Using jayrob's build but im having a problem

Ok so I used jayrob's build for the mxdl flashlight. I have it all built and modded into the casing. So the LED I tested with worked for some time, but then stopped working. So I measure the voltage and it goes crazy for a bit and finally settles at about 6V. Why is this happening and why is it not a 3V?

what type of driver? And, you're still testing this with the LED?

I'm using the ddl driver. The exact one that jayrob builds on the mxdl chip. Ya the led is dead, but I test the voltage where the wires connect to the led.

Sounds like a connection problem somewhere in the driver. There should not be any reading 'going crazy'. However, using a DDL driver, you will see very close to the supply voltage until the load is connected. Then, when the load is connected, you will see the voltage that the load is taking.

Double check things like polarity of the connections to the silicon diode and cap as well. And make sure that nothing is touching the board that is not supposed to! There are some points on the MXDL board that could touch the LM317 legs...

rkcstr makes an excellent test load for connecting to the driver to test current before connecting the LD. Remember to short the leads to discharge the cap after testing, and before you connect your LD!
Jay

Ok for the lm317 i connected the vin to the spot where the original resistor was on the the negative feed to the place where the bare wire went to the Original LED. The driver did not work. The vout is to the resistors as they should be, but then I souldered from where the original blue wire was connected to the vout also. Then my driver worked, but like I said it is giving me 6V and my test LED nolonger works.

Your going to have to start over with the driver build. Get a test load from rkcstr, or make one with 4 silicon diodes. Here's a diagram from rog8811 on a test load, and how to test current:
http://www.laserpointerforums.com/forums/YaBB.pl?num=1197651171#1

It is VERY important to test the driver before ever connecting your precious diode. Also, remember to discharge the cap after testing!

About the driver build...

The round circuit board from the MXDL has the stock resistor soldered in two spots. One of those spots is positive. The other spot is not positive or negative. Just an unused point on the board. You can use your DMM to test continuity from the battery contact point on the board, to the solder point.

* This positive solder point is where your Vin is connected on the LM317.
* The resistor goes between the other two legs of the LM317.
* The positive lead to the LD goes on the outside leg of the LM317.
* The negative lead to the LD goes from the negative point on the board, where you use a small wire to 'pinch' to the aluminum driver mount. (like the stock bare wire was)
* The silicon diode and cap get soldered between your negative and positive leads going to the LD. (check for polarity - there is a positive and a negative on these two components)



Check this picture and notice how the ground is. It is connected to the bare wire that is wrapped around the edge of the board. It gets 'pinched' in the aluminum driver mount.

Here's another picture:


Notice the small piece of electrical tape on the board (under the legs of the LM317), to protect from improper connections between the board and the LM317.
Jay

sounds good. I'll undo the vin and vout and then run the vin to the correct positive feed. The rest should be fine since I took a lot of time to make sure evrything was perfectly hooked up right.

Ok Now that I did what I said above I got the voltage to 4.98V. so Now what would the problem be?

Are you measuring voltage direct from the driver? Or, do you have a diode (or a test load) connected?

If you do not have a test load connected, you should see voltage that is very close to your supply voltage. When the diode, or test load is connected, you will measure the voltage of the load. (at the connection to the load)
Jay

I have an LED connected, but it died and will not light. Its just a 2.8V LED not a real LD yet. I Measure at where I have the LED connected. Is this my Problem, because I triple checked all my connections and made my corrections and they are all perfect and not touching? So is it because im testing at the dead LED?

The only way to know, is to test with a proper test load. You can use a dead LD too. If the dead LD still shows a dim light, it should serve as an accurate test load...
Jay

Would you predict that I have it correct though, because I took so much precaution this time and know it is hooked up right.

check the grounds of the host, thats my biggest problem i come across!


almost every time i have a problem its that.

chaosfourever said:

Would you predict that I have it correct though, because I took so much precaution this time and know it is hooked up right.

Click to expand...

You probably have it right, but there is no way to 'guess' this. You must have a test load. A test load is not hard to make. The silicon diodes cost pennies.

I recommend the very inexpensive test loads from rkcstr.

And yes, sk8 is right! The ground connection seems to be the part that people have problems with. But to me, it sounds like you have the ground. You just don't have a proper test load to simulate the LD...
Jay

first off i found that my n4001 was backwards so I corrected that. then i built the test load and attached it. Now I only have 0.65V so now what is wrong?

Study the diagram for the test load carefully. You probably have the positive and negative reversed...
Jay