-
You must be a Supporting Member to participate in the Candle Power Forums Marketplace.
You can become a Supporting Member.
GD1000 flashing
- Thread starter K-T
- Start date
Help Support Candle Power Flashlight Forum
If you are testing it with a power supply it could be the supply is getting overloaded and then shuts down and then comes back to try again causing the driver to flash.
Long wires and or small gauge wire could also cause the driver to flash.
The driver is going to try to pull around 1A. Your supply or batteries need to be able to supply this amount of current and have a low internal impedance.
Start up currents can easily exceed the nominal operating and this usually freaks power supplies out.
To supply 1A to a driver you should have a 2A or 3A power supply as peak currents could be 2X or more.
One trick on power supplies is to power up live. What I mean is to turn on the power supply so that the power supply output capacitors are fully charged and then use an external switch to apply power to the converter board. The charged power supply capacitors are usually enough to kick start the converter into regulation at which time the demand on the power supply is a lot less.
Wayne
Long wires and or small gauge wire could also cause the driver to flash.
The driver is going to try to pull around 1A. Your supply or batteries need to be able to supply this amount of current and have a low internal impedance.
Start up currents can easily exceed the nominal operating and this usually freaks power supplies out.
To supply 1A to a driver you should have a 2A or 3A power supply as peak currents could be 2X or more.
One trick on power supplies is to power up live. What I mean is to turn on the power supply so that the power supply output capacitors are fully charged and then use an external switch to apply power to the converter board. The charged power supply capacitors are usually enough to kick start the converter into regulation at which time the demand on the power supply is a lot less.
Wayne
Hello,
I just had this happen with an Aleph LE I was building - my first one. I put it all togther, it lit up, but after a few seconds it would start flashing. I had not run any power through it until it was wired into the can with the LED and everything and ran it on a 16340. With a CR123 it would not flash but it ran at a much reduced output. I have since further damaged the LE
and it was potted anyways so there is no fixing it, but I am curious if there is something to do differently next time. Thanks for your help!
I just had this happen with an Aleph LE I was building - my first one. I put it all togther, it lit up, but after a few seconds it would start flashing. I had not run any power through it until it was wired into the can with the LED and everything and ran it on a 16340. With a CR123 it would not flash but it ran at a much reduced output. I have since further damaged the LE
and it was potted anyways so there is no fixing it, but I am curious if there is something to do differently next time. Thanks for your help!The GD converter driver IC is a unique beast. Internally it has four switches allowing it to run a true synchronous buck or synchronous boost. I don't recall all the variations of non functional behavior. I can only guess at this point in time. One might be heat related and the converter shutting down and then after it cools starting up again causing the light to turn on/off at the cooling/heating cycle. If this is the situation then one needs to provide a better thermal management for the converter.
Another situation may arise from low input voltage at the converter board. The lower the input voltage is the harder the converter has to work to regulated the power to the LED. usually the worst offender is the switch followed by wiring and losses in aluminum if the return path is through the battery tube. There could also be losses in threads if the electrical path has to cross from battery tube to tailcap. All these may be minor, but, can add up to a significant loss.
If you can mock up the bench setup and measure the voltage directly at the converter board and at the battery terminals you can measure the loss across the complete electrical path. The Kroll switch for example if not greased I have seen a resistance as high as 2 ohms. Properly greased contact resistance came way down.
Hope this helps,
Wayne
Another situation may arise from low input voltage at the converter board. The lower the input voltage is the harder the converter has to work to regulated the power to the LED. usually the worst offender is the switch followed by wiring and losses in aluminum if the return path is through the battery tube. There could also be losses in threads if the electrical path has to cross from battery tube to tailcap. All these may be minor, but, can add up to a significant loss.
If you can mock up the bench setup and measure the voltage directly at the converter board and at the battery terminals you can measure the loss across the complete electrical path. The Kroll switch for example if not greased I have seen a resistance as high as 2 ohms. Properly greased contact resistance came way down.
Hope this helps,
Wayne
- Joined
- Feb 2, 2006
- Messages
- 1,154
Looks like when a converter boards fails it's often the GD driver...
Either the driver cannot operate at the specified current rating (eg. only ~500mA on a GD1000), or the driver functions at the first instance then dies completely
It recently happened to me that two out of three GD1000s from a 2012 purchase have failed
Since I have had no way to test the boards in advance, there's no way to tell if a driver board is operative without fault until the board is sent out for the actual module build, which in my case can be years later
Have sent SS an inquiry last week regarding replacements but still didn't get a reply from them. The failed GD1000s were tested before they're being assembled, AFAIK.
Either the driver cannot operate at the specified current rating (eg. only ~500mA on a GD1000), or the driver functions at the first instance then dies completely
It recently happened to me that two out of three GD1000s from a 2012 purchase have failed
Since I have had no way to test the boards in advance, there's no way to tell if a driver board is operative without fault until the board is sent out for the actual module build, which in my case can be years later
Have sent SS an inquiry last week regarding replacements but still didn't get a reply from them. The failed GD1000s were tested before they're being assembled, AFAIK.
- Joined
- Feb 2, 2006
- Messages
- 1,154
Had the very same experience just recently when enlisting a local mfyer to build a custom XPG2/GD1000 module for the Nuwai Q3 (TM-303X). When powered on (with an AW protected RCR123a) the light output flickers a bit at first then stabilizes, only to flash within some 25-30 seconds while the flashlight body begins to warm up rapidly. Again, no flashing when powered up with a primary CR123a, but output has since halved.I just had this happen with an Aleph LE I was building - my first one. I put it all togther, it lit up, but after a few seconds it would start flashing. I had not run any power through it until it was wired into the can with the LED and everything and ran it on a 16340. With a CR123 it would not flash but it ran at a much reduced output. I have since further damaged the LE
Have since swapped the module into another Nuwai light body, this time the TM-301x (2x123a but driven with 1x17670). Same result.
Strange though, another backup GD1000 module built a few years ago (with the very same custom heatsink) but with an SSCP4 installed, does not exhibit such flashing problem whatsoever. Light powers up with decent output, declining a bit upon continuous on and a little body warming up, but that's it -- no flashing.
Can someone suggest what exactly is going on? This is the third GD1000 in a row that appears to be faulty......
- Joined
- Feb 2, 2006
- Messages
- 1,154
A little update on my GD1000 situation. The mfyer responsible for my custom Nuwai module build has determined that the driver (that flashes) is indeed faulty and has since replaced it with another one. Now, without any flickering or flashing, the module outputs at full level (in this case, "freaking bright") with one 16340 or CR123a, and even though the latter outputs a bit less than the former, it's not as bad as before, which only yielded half the possible output.
Nevertheless, since the driver is very demanding in power, a small-capacity battery like the CR/RCR123a cannot retain flat regulation, and output drops gradually (albeit slowly). In the case of a protected RCR the healthy driver will drain the cell to the point where its protective circuit cuts out the power, which is normal.
Unfortunately during this custom module building exercise (which involved this Nuwai module as well as another one or two MagAA type), a total of three GD1000s have been found faulty, which the mfyer finds that very annoying.
Definitely the failure rate of the GD1000 is much higher than, way, the NG1000. And it's more expensive too so to speak. SS please kindly improve the overall QC of the GD1000!
Nevertheless, since the driver is very demanding in power, a small-capacity battery like the CR/RCR123a cannot retain flat regulation, and output drops gradually (albeit slowly). In the case of a protected RCR the healthy driver will drain the cell to the point where its protective circuit cuts out the power, which is normal.
Unfortunately during this custom module building exercise (which involved this Nuwai module as well as another one or two MagAA type), a total of three GD1000s have been found faulty, which the mfyer finds that very annoying.
Definitely the failure rate of the GD1000 is much higher than, way, the NG1000. And it's more expensive too so to speak. SS please kindly improve the overall QC of the GD1000!
