I have 2 Ultrafire 504b lights that I bought a while back and have now got some time to "fix" them. The problem is that after a few seconds the lights will drop their output to around 5%, I can still see the LED glowing though. I am running 18650's in most of my lights (TF Flames) and they seem to be working fine with other lights. One of the lights is a XML the other is an XPG? Any advice?
Try the drop-in outside of the host by connecting a battery to it. That might help pinpoint the problem. If it still behaves that way, then it is something in the drop-in, otherwise in the host (or how the host and drop-in work together). If it is in the drop-in, try taking the reflector off and see if that helps. See if the LED is mounted properly to the pill.
Sounds like a temperature thing, but I wouldn't think it would trigger in just a few seconds.
When I use 2 CR123’s the light doesn’t act up as bad, it still happens just not as bad.
If you’re saying Both lights do the same thing I’m guessing it’s your battery…It would just seem too odd if two lights are both doing the same thing …What kind of amps is it pulling on your DMM ?.. Brted said test just the drop in but if you use the light without the tailcap that’s basicly what you’re doing …i’d claen everything and tighten and inspect then run a simple amperage test without the tailcap …what’s the voltage of your batteries ??
Since you didnt respond to brted’s suggestions 4 hrs after his post (you only mentioned 2cr123’s), im going to assume you didnt try them. But for us to try to pinpoint your problem, you need to give us some feedback.
i see you have a working 504B xml (signature). Take the other 2 dropins (the part with the reflector, springs and LED) and drop them in your working 504b. If they work, it might be ur tailcap causing problems. If they still act the same way, at least you know it has something to do with the drop-ins themselves. Perhaps the LED wasnt properly seated/soldered to the board.
You can also try in reverse by taking the known working drop-in and trying it in the incorrectly working bodies.
Okay I took some readings for you guys:
TF Flame 18650 #1: 3.97V
Tail cap reading on High: 0.92A
As others have said, try and isolate your problem. As far as I am aware it will be one of the following:
The torch body
This could include:
- Tail-switch has a loose/bad connection (if short the battery directly to the torch body and the problem goes away then this is probably your problem). I have had the contacts in a WF501B switch rust (inside the actual switch) which caused this.
- Drop-in has a lose connection, if the positive spring in not long enough or negative and positive are touching (short circuit)
- Bad connection between the torch tail cap and the main body, paint or anodising or non-conductive lube can do this.
The drop-in
- Bad connections: again the springs, wire soldering…
- Bad driver: driver not designed for 18650, some sort of heat protection or battery unable to provide current needed if you have a very inefficient driver.
- Shorted LED: wires on LED shorting to the reflector, happens if there is no isolating ring or it is worn though between reflector and LED.
The battery
- Low voltage (3.9V is Good so not that I think)
- Battery unable to deliver the current needed.
- Protection circuit cutting the power (shouldn’t be as it still goes)
- Battery wrapper is worn though and shorting to the torch tube (I have had it happen) so LED no go.
Hope this helps 
Hi, did you manage to isolate and solve the problem?
I had a similar problem with my 504B. It would change modes on me, typically dropping from the brightest mode to a lower power mode. I also thought it might be the battery but quickly confirmed that it wasn’t.
The difficulty with this problem is that it happens intermittently, not consistently. At first I also thought that it was the drop-in and the vendor very kindly sent me a second drop-in after I discussed with him. After switching the drop-in however, the problem still existed. I then found that the electrical connection between the switch in the tailcap and the threads in the tailcap was not very reliable. The threads in the tail-cap are dirty and don’t make good electrical connection with switch that it houses. If the electrical connection is intermittent, it could cause the electronics to detect a lower than needed voltage and thereby drop to a lower-power mode. In my case, I opened up the switch and cleaned out the threads using a cotton swab and some alcohol (I don’t know if this is the best way?) and for a while, my problem went away. However, it came back after a day or so. I cleaned it a second time, and a second time, it aparently went away and then came back again after some time.
The 504B and 501B have the same diameter thread (although they don’t screw down to the same depth, but you can screw it enough to make electrical connection), so I took the tailcap off a 501B that I have, and tried it on the 504B. It seemed to work without a problem, but I did not test it over several days, just a few minutes. Anyway, my conclusion is that the problem is still in the tailcap. I contacted the vendor again, but he hasn’t replied. I guess I don’t blame him since he already sent me a free drop-in replacement, but I don’t currnetly have working 504B.