Undiagnosable circuit problem

[dilmah]

Hey guys,

I'm having a little trouble building Daedal's LD driver circuit. The first time i built it, it didn't work at all (probably because i wired the load the wrong way around haha), so i rebuilt the entire thing with new parts. Now, the loads do light up, and i get the following measurements (with the pot at lowest resistance):

5.89 v across the 4xAA batteries
1.5 v between v in and v out pins of LM317T
0.06 v between v out and v adj pin of LM317T
0.06 v across the two resistors in parallel
0.00 v across the pins of the pot
4.28 v across load
4.28 v across the two blobs of solder that connect a heap of things
4.28 v across the pins of the 1N4004 (shouldn't this be like 0.7 volt drop or something?)

The above results were taken after the results displayed in the photos below, where the batteries had drained slightly.

The resistance of the two 10 ohm resisters in parallel was measured at 5.8 ohms.

Ive searched for a short somewhere, but i cannot find one. I've also replaced the LM317T (but i replaced it with an older one) and got exactly the same results.

Close up pics of the circuit are at:

http://dilmah.googlepages.com/home

If anyone can see why this is happening, or have any suggestions. or have had a similar problem, i'd love to hear from you!

dil

p.s I also want to sincerely thank daedal, he's examined my circuit and helped me out via PM, and i only found out now that he had the flu..

 

[Gazoo]

The cathode of the silicon diode is marked with a band. To see the flow of current you can refer to the following link:

http://www.geocities.com/mistertippy/components/diodes.html

You should always be seeing 1.25 volts between the output and adjust of the LM317. The best and easiest to understand source I have found for using the LM317 as a current regulator is here:

http://users.pandora.be/davshomepage/current-source.htm

You might want to refer to the above source and recheck your connections. Also it has been my experience that anything less than 6 volts input will cause the regulator to begin dropping voltage.

 

[chimo]

Does anything get hot when you power it up? I would guess you have a short somewhere (likely between Vout & Adj).

 

[o_DEATH_ANG3L_o]

Yes you should recheck ALL your connection. and as Gazoo said, if your input is below 6 volts, the LM317 is not going to act efficient. Next time try a bread board socket. they're great for testing circuits without soldering.

 

[dilmah]

thanks for the replies guys!

gazoo i read through those websites, and the diode thus appears to be correctly wired.
nothing gets hot when i power it up, but i dont keep it on for long, its a pushbutton switch haha i'd have to hold it down...

I've looked long and hard, but i cannot see any possible short between those two pins. But im a little worried about the multiple connections i have directed at one glob of solder. for example, in my circuit, the anode of the 1N4004, the negative terminal of the capacitor, the cathode of the LED and the negative wire from the battery are all directed to one glob of solder with no particular order. Daedal hasn't commented about this wiring method, so i assume its fine, but i just want to confirm that i can actually do this?


thanks again guys

dil

 

[chimo]

Power it up for at least a minute of continuous running. Then feel the components to see if any of them are hot.

Instead of visually checking for shorts, measure the resistance between the legs with a DMM.

 

[dilmah]

GENIUS

i never would have thought of that

thats insane

check for shorts with a DMM

ill go do that now!

 

[dilmah]

ok i measured the resistances

0.08 ohms if i connect the terminals of the DMM together
0.08 ohms across the pot (turned down to lowest resistance)
5.9 ohms between the Vout and Adj pins of the LM317T
5.9 ohms between Vout and glob of solder connecting a heap of positive things
5.9 ohms between the two 10 ohm resistors in parallel.


...that means there is no short right..?

WHAT THE

edit: does that mean both my LM317T's are broken?

 

[chimo]

Instead of those two series connected LEDs as a dummy load, try a 5 to 8 ohm resistor and power it up again.

 

[Daedal]

The solder you have there dil is fine

The resistance test... should have told you about that one... ;D

I really was very stumped when I saw the problem and found no solution to it. But I think something is not measuring right. I mean... given the voltage drop it would seem like the LM is working right, but then the readings do not confirm this at all :-?

And the biggest problem I am having is why the Vdrop across the Vout and Adj lines is so low :-/

The Vdrop across the 1N400x is also weird... I'm not sure, but you might have more than one dead component. Have you tried switching the diode... even though, that still doesn't explain the readings that you are getting :-?

--DDL

 

[dilmah]

WOAH

WOAH

I CONNECTED THE 8 OHM RESISTOR LOAD (two 3.9 ohm in series)

I GOT 1.25 V

DUDE

YOUR A LIFESAVER CHIMO

but why didnt it work with the LED's?

does that mean its safe to put in the Laser Diode?

 

[dilmah]

whoops didnt see Daedals reply

hmm it seems to be working now, but do you think i still have dead components? ill just go remeasure with the new load..

still stumped as to why the LED load did not work.. Chimo is there something unusual about LED loads that made you suggest using resistors, or was it a wild guess?

 

[dilmah]

ok new measurements:

1.25 volts across resistors in parrallel
1.25 across vout and adj pins
1.97 across load
5.83 across power source
1.99 across 1N4004
1.99 across the two huge globs of solder

hmm that means the 1N4004 is broken right

 

[chimo]

Arg, I had a whole post with the explaination but my router disconnected!

Here's a shorter explaination.

The 1n4004 should be fine - it's in reverse bias so it should read the load voltage.

The 2 red LEDs in series had a Vf that was too high for the LM317 to maintain regulation. A better dummy load (if you want to use LEDs) would be 3 or 4 red LEDs in parallel. Good luck!

Paul

 

[dilmah]

haha thanks dude

lawl damn i just replaced the 1n4004 haha

hmm but shouldnt i be getting like 3 volts out the other end (i.e across the load)? and would it be safe for me to discharge the cap and attach the LD now? also, to discharge the cap, we just connect the positive and negative terminals of the cap together for a few seconds right?

edit: p.s whats Vf? voltage...final?

 

[chimo]

250mA across 8 ohms = drumroll......... 2 volts
Your measured voltage = drumroll........ 1.97 volts

I'd say it's working.

Disconnect the power and the cap will self-discharge through the resistor.

haha thanks dude

lawl damn i just replaced the 1n4004 haha

hmm but shouldnt i be getting like 3 volts out the other end (i.e across the load)? and would it be safe for me to discharge the cap and attach the LD now? also, to discharge the cap, we just connect the positive and negative terminals of the cap together for a few seconds right?