tylerdurden
Flashlight Enthusiast
So if I'm reading this correctly, input current is a large contributor to the heating? If that is the case, then a downboy would be much less susceptible to this overheating than a badboy?
-
You must be a Supporting Member to participate in the Candle Power Forums Marketplace.
You can become a Supporting Member.
BB750 shutting down after 10-12secs?
- Thread starter hotbeam
- Start date
Help Support Candle Power Flashlight Forum
Input current is the dominant power heat generation in this particular configuration and for the internal Switch element.
For whatever reason, there is a beefier switch than the package can dissipate? Why Linear Technology picked these two combo's I'm not at all clear. Too bad it is not a mixed hybrid and used a FET for the switch element. That would be one awesome converter IC.
Yes, a stepdown is typically more efficient. With higher voltages, you need less current to get the drive the same load power.
For whatever reason, there is a beefier switch than the package can dissipate? Why Linear Technology picked these two combo's I'm not at all clear. Too bad it is not a mixed hybrid and used a FET for the switch element. That would be one awesome converter IC.
Yes, a stepdown is typically more efficient. With higher voltages, you need less current to get the drive the same load power.
shifty: The LS is AA'ed to the McModule and the BB750 circuit is potted with AA to the walls of the McModule so there should be enough thermal path for the heat to dissipate. /ubbthreads/images/graemlins/icon23.gif
ticey: I know ticey, this shouldn't be happening. I used the tip of the AA stirer to make sure (as much as practicable) that there are no air pockets in the AA. Yeah, I have the board with I haven't installed yet connected to power and a LS and powered it up. After a 10-15sec, the light would go off. I touched and smelt the copper heatsink and it is smokin' hot. Definately too hot to touch for more than 1 sec. I am at a loss too. /ubbthreads/images/graemlins/icon3.gif
doug: Yes Doug, a huge load of current is being pushed through the little LTNH IC and all the components on the circuit. As you say, in isolation, this circuit would not be a viable driver for a LS. The circuit is definately being maxed out. That's why I used AA as suggested by all.
ticey: "Too high a Vf?"... actually ticey, it is an X3T /ubbthreads/images/graemlins/smile.gif
dat2zip: Wayne, I heatsinked the top of the IC and the diode to promote a larger area for the heat to leave the circuit and move onto the AA onto the McModule. That would surely aid, rather than impede the heat transfer. /ubbthreads/images/graemlins/smile.gif Most people just pot the top of the circuit. I thought I would go for the 'overkill' option.
Also as mentioned above, the entire circuit has AA on it... right to the top of the McModule where the anode cap sits. I don't know what else I can do. Also the circuit that I haven't installed in the McModule also exhibit the same shutdown behaviour. That also has the heatsink on the IC and diode. I don't have a fan on whilst testing it so it did get very hot quickly. That shouldn't be happening, should it Wayne?
McGizmo: You must be doing the right thing to not experience the shutdown... and you have done quite a few too! I can't understand where I went wrong both in the uninstalled BB750 and the AA potted BB750 in the McModule! I am really at a loss /ubbthreads/images/graemlins/frown.gif. Oh, I am getting 1.8A at the tail with partially used ~90% 123s. Yeah the DB's would be more efficient and less heat would be loss.
Doug: "effect of lengthening this path"... actually Doug, my plan was to provide more area for heat from the chip to transfer to the AA. The top of the IC is only 4mm x 4mm. My copper heatsink adds at least 12-15 times more surface area for heat dissipation. Also the heatsink was put on the diode as well.
Maybe I just have to put this down to a learning experience /ubbthreads/images/graemlins/frown.gif
ticey: I know ticey, this shouldn't be happening. I used the tip of the AA stirer to make sure (as much as practicable) that there are no air pockets in the AA. Yeah, I have the board with I haven't installed yet connected to power and a LS and powered it up. After a 10-15sec, the light would go off. I touched and smelt the copper heatsink and it is smokin' hot. Definately too hot to touch for more than 1 sec. I am at a loss too. /ubbthreads/images/graemlins/icon3.gif
doug: Yes Doug, a huge load of current is being pushed through the little LTNH IC and all the components on the circuit. As you say, in isolation, this circuit would not be a viable driver for a LS. The circuit is definately being maxed out. That's why I used AA as suggested by all.
ticey: "Too high a Vf?"... actually ticey, it is an X3T /ubbthreads/images/graemlins/smile.gif
dat2zip: Wayne, I heatsinked the top of the IC and the diode to promote a larger area for the heat to leave the circuit and move onto the AA onto the McModule. That would surely aid, rather than impede the heat transfer. /ubbthreads/images/graemlins/smile.gif Most people just pot the top of the circuit. I thought I would go for the 'overkill' option.
Also as mentioned above, the entire circuit has AA on it... right to the top of the McModule where the anode cap sits. I don't know what else I can do. Also the circuit that I haven't installed in the McModule also exhibit the same shutdown behaviour. That also has the heatsink on the IC and diode. I don't have a fan on whilst testing it so it did get very hot quickly. That shouldn't be happening, should it Wayne?
McGizmo: You must be doing the right thing to not experience the shutdown... and you have done quite a few too! I can't understand where I went wrong both in the uninstalled BB750 and the AA potted BB750 in the McModule! I am really at a loss /ubbthreads/images/graemlins/frown.gif. Oh, I am getting 1.8A at the tail with partially used ~90% 123s. Yeah the DB's would be more efficient and less heat would be loss.
Doug: "effect of lengthening this path"... actually Doug, my plan was to provide more area for heat from the chip to transfer to the AA. The top of the IC is only 4mm x 4mm. My copper heatsink adds at least 12-15 times more surface area for heat dissipation. Also the heatsink was put on the diode as well.
Maybe I just have to put this down to a learning experience /ubbthreads/images/graemlins/frown.gif
jtice
Flashaholic
I think you would have had to learn something, to put it down as a learning experience HB. /ubbthreads/images/graemlins/frown.gif
This makes no sence, and should run fine, and even though they will run hot, it should run with a X3T and 2x123.
You deffinately have us all stumped on this one.
It seems you did everything correct from what I can tell.
I would say its a bad converter, IF you didnt have TWO doing this! /ubbthreads/images/graemlins/ooo.gif
/ubbthreads/images/graemlins/frown.gif /ubbthreads/images/graemlins/icon3.gif
This makes no sence, and should run fine, and even though they will run hot, it should run with a X3T and 2x123.
You deffinately have us all stumped on this one.
It seems you did everything correct from what I can tell.
I would say its a bad converter, IF you didnt have TWO doing this! /ubbthreads/images/graemlins/ooo.gif
/ubbthreads/images/graemlins/frown.gif /ubbthreads/images/graemlins/icon3.gif
Let me add a data point that might help? I just took a McLux-PR head that has a X3T and BB750 and powered it up on my bench supply. At 4.14 Vin, it was drawing 1.5 amps. I didn't run it long enough to see if it would ultimately shut down but did give it a minute and a half and it was stable in current draw for the full time. Now I don't know what the voltage drop would be on a couple 123's but would expect it not to be as low as the 4.14 that I had on the bench supply.
I think your 1.8 amps coupled with quick shut down is indicative of something wrong with the circuit in situ. As to what that might be, I am clueless! /ubbthreads/images/graemlins/frown.gif
I think your 1.8 amps coupled with quick shut down is indicative of something wrong with the circuit in situ. As to what that might be, I am clueless! /ubbthreads/images/graemlins/frown.gif
I think it is how you heatsink the board to the MCModule, I don't think it's transfering the heat fast enought.
McGizmo: can you run yours down to 2.5v to see if it can still stay in regulation with out shuting down. Meaning can you run one 18650 li-ion cell.
McGizmo: can you run yours down to 2.5v to see if it can still stay in regulation with out shuting down. Meaning can you run one 18650 li-ion cell.
tvodrd
*Flashaholic* ,
I've only built-up 3 BB750's- 2 in 6V/5W's and the 3rd is a .20" thick sandwich with an R2H attached. All 3 have the IC and diode AA epoxied against grounded copper- housing or copper "emitter board" in the case of the R2H. I run the latter in my 1X CR2 light which I watched the outer case hit 190 degF in 1 hour. A fresh CR2 seems to be capable of 6-7 Amps into a dead short, but seems to current limit Iin to the module. I have run both the 5W's all the way down. I presume I have just been lucky- never seen any flickering.
Larry
Larry
Toneloe,
At 2.04 Vin the driver was drawing 1.78 amps
At 2.5V 1.76 amps
At 3V 1.73 amps
AT 4V 1.5 amps
These are the numbers I read off my bench supply. I was geting constant light at these voltages but something tells me that the current to the LED was not 750 mA in all cases! Not once did the current get up to 1.8 amps?!?!? I didn't hold these levels for more than 10 to 20 seconds as it was ackward to do so. I don't know if there would be a thermal shutdown eventually. Certainly there was no quick shutdown like Hotbeam is experiencing. I spoke with Wayne a bit on the phone tonight on this issue. I asked if a short across the sense resistor could cause this behavior and he thought that might be a posibility but wasn't sure.
At 2.04 Vin the driver was drawing 1.78 amps
At 2.5V 1.76 amps
At 3V 1.73 amps
AT 4V 1.5 amps
These are the numbers I read off my bench supply. I was geting constant light at these voltages but something tells me that the current to the LED was not 750 mA in all cases! Not once did the current get up to 1.8 amps?!?!? I didn't hold these levels for more than 10 to 20 seconds as it was ackward to do so. I don't know if there would be a thermal shutdown eventually. Certainly there was no quick shutdown like Hotbeam is experiencing. I spoke with Wayne a bit on the phone tonight on this issue. I asked if a short across the sense resistor could cause this behavior and he thought that might be a posibility but wasn't sure.
Some observations. I tried the light again tonight and it stayed on for 26 secs before the BB750 shutdown the light. I measured the voltage of the 2 x 123s in the body and it read 5.67v.
I then got some fresh 123s and measured their voltage. They were 6.06v. When I turned on the light, the circuit only allowed it to stay on for 7 secs! With the 5.67v batteries (from last night), it stayed on for 26 secs before shutdown.
What can be concluded from this?
I then got some fresh 123s and measured their voltage. They were 6.06v. When I turned on the light, the circuit only allowed it to stay on for 7 secs! With the 5.67v batteries (from last night), it stayed on for 26 secs before shutdown.
What can be concluded from this?
AilSnail
Flashlight Enthusiast
I get a 1.2 or so amp draw from somewhat fresh batts to a bb750 powering an x3t.
After some minutes it starts to flicker and shifting from moon to full brightness irregularly.
After some minutes it starts to flicker and shifting from moon to full brightness irregularly.
Doug S
Flashlight Enthusiast
[ QUOTE ]
hotbeam said:
Some observations. I tried the light again tonight and it stayed on for 26 secs before the BB750 shutdown the light. I measured the voltage of the 2 x 123s in the body and it read 5.67v.
I then got some fresh 123s and measured their voltage. They were 6.06v. When I turned on the light, the circuit only allowed it to stay on for 7 secs! With the 5.67v batteries (from last night), it stayed on for 26 secs before shutdown.
What can be concluded from this?
[/ QUOTE ]
With it potted it will be hard to check but Don suggests a possibility above that is plausible. If there is a short across the sense resistor the IC will run in current limit regardless of input voltage. As I recall, the sense resistor input to the IC is via pins 2 and 3. There might be solder bridging these two pins.
hotbeam said:
Some observations. I tried the light again tonight and it stayed on for 26 secs before the BB750 shutdown the light. I measured the voltage of the 2 x 123s in the body and it read 5.67v.
I then got some fresh 123s and measured their voltage. They were 6.06v. When I turned on the light, the circuit only allowed it to stay on for 7 secs! With the 5.67v batteries (from last night), it stayed on for 26 secs before shutdown.
What can be concluded from this?
[/ QUOTE ]
With it potted it will be hard to check but Don suggests a possibility above that is plausible. If there is a short across the sense resistor the IC will run in current limit regardless of input voltage. As I recall, the sense resistor input to the IC is via pins 2 and 3. There might be solder bridging these two pins.
Hotbeam,
I haven't a clue what is wrong with your BadBoy(s) but something is certainly not right or normal! I suggest you salvage the X3T before you send too much current through it! This is assuming that the excess current is getting to the LED.
I haven't a clue what is wrong with your BadBoy(s) but something is certainly not right or normal! I suggest you salvage the X3T before you send too much current through it! This is assuming that the excess current is getting to the LED.
Doug S... There is definately no overflow of Artic Silver (AS) onto the legs of the chip. Now, all this talk is making me wonder whether there are a short on the diode. I *may* have put too much AS on top of the diode and that may be shorting it out. I can't tell now... for obvious reasons.
Don... You are right. The BB750 can be replaced but not the X3T. I should rip the guts out of the McModule and leave the X3T epoxied.
I think I will need to drill through the anode cap to clean the insides of the McModule. Any idea where I can get buy just the insulating anode cap and anode contact tab from? I don't think the Shoppe sells just those 2 little bits.
Don... You are right. The BB750 can be replaced but not the X3T. I should rip the guts out of the McModule and leave the X3T epoxied.
I think I will need to drill through the anode cap to clean the insides of the McModule. Any idea where I can get buy just the insulating anode cap and anode contact tab from? I don't think the Shoppe sells just those 2 little bits.
Doug S
Flashlight Enthusiast
[ QUOTE ]
hotbeam said:
Doug S... There is definately no overflow of Artic Silver (AS) onto the legs of the chip. Now, all this talk is making me wonder whether there are a short on the diode. I *may* have put too much AS on top of the diode and that may be shorting it out. I can't tell now... for obvious reasons.
[/ QUOTE ]
Hotbeam-
Arctic Silver is not sufficiently conductive to adversely affect this circuit no matter where it is. My comment about bridging the Isense pins applies to solder only. This would be a manufacturing defect. It seems improbable that you would have two different BBs with the same defect but I suppose it is not impossible.
hotbeam said:
Doug S... There is definately no overflow of Artic Silver (AS) onto the legs of the chip. Now, all this talk is making me wonder whether there are a short on the diode. I *may* have put too much AS on top of the diode and that may be shorting it out. I can't tell now... for obvious reasons.
[/ QUOTE ]
Hotbeam-
Arctic Silver is not sufficiently conductive to adversely affect this circuit no matter where it is. My comment about bridging the Isense pins applies to solder only. This would be a manufacturing defect. It seems improbable that you would have two different BBs with the same defect but I suppose it is not impossible.
HotBeam,
I am sure that Wayne or I can help you on those parts. If you are going to order something from the Shoppe, just include a note for a couple PM6 isolaters and anode rivets, N/C, per Don. I'll send some extras to Wayne in the next box that goes his way.
I missed the fact that you had ArcTic Silver on the PCB. I have trashed a few drivers that way until I learned to only use AS on the LED slug! Some of the shorts didn't happen until the board heated up and the silver particles likely were given the chance to finally allign and bridge the gap! /ubbthreads/images/graemlins/frown.gif Over a year ago, I had a completely custom build with one of Wayne's experimental dual level drivers (hall effect switch) I used a small rare earth magnet in a collar on the head to go from low beam to high beam. The light lasted all of about a minute before the AS destroyed the circuit. The thing was really sweet! We never had a chance to go back and visit that program.
AS + PCB = Disaster
I think this mystery has been solved.
EDIT: I see Doug S's post that came in while I was hunting and pecking. In the use of AS on PCB's I beg to differ with Doug. I think both AS epoxy and AS grease can effectively be put to use to destroy an otherwise functioning PCB. I don't know about you guys but much of the AS I have used has settled out in the syringe and it wouldn't surprise me if enough silver isn't in levels of concentration to take the AS from thermally conductive to electrically conductive! As I recall, even the manufacturer recommends AGAINST the use of AS in trace areas. I don't trust AS to be non electrically conductive and frankly I don't count on the electrically conductive epoxies for 100% performance.
I am sure that Wayne or I can help you on those parts. If you are going to order something from the Shoppe, just include a note for a couple PM6 isolaters and anode rivets, N/C, per Don. I'll send some extras to Wayne in the next box that goes his way.
I missed the fact that you had ArcTic Silver on the PCB. I have trashed a few drivers that way until I learned to only use AS on the LED slug! Some of the shorts didn't happen until the board heated up and the silver particles likely were given the chance to finally allign and bridge the gap! /ubbthreads/images/graemlins/frown.gif Over a year ago, I had a completely custom build with one of Wayne's experimental dual level drivers (hall effect switch) I used a small rare earth magnet in a collar on the head to go from low beam to high beam. The light lasted all of about a minute before the AS destroyed the circuit. The thing was really sweet! We never had a chance to go back and visit that program.
AS + PCB = Disaster
I think this mystery has been solved.
EDIT: I see Doug S's post that came in while I was hunting and pecking. In the use of AS on PCB's I beg to differ with Doug. I think both AS epoxy and AS grease can effectively be put to use to destroy an otherwise functioning PCB. I don't know about you guys but much of the AS I have used has settled out in the syringe and it wouldn't surprise me if enough silver isn't in levels of concentration to take the AS from thermally conductive to electrically conductive! As I recall, even the manufacturer recommends AGAINST the use of AS in trace areas. I don't trust AS to be non electrically conductive and frankly I don't count on the electrically conductive epoxies for 100% performance.
jtice
Flashaholic
I agree with Don on this one.
It is SILVER you know. /ubbthreads/images/graemlins/wink.gif I wouldnt think it would take much to foul up the works.
Maybe thats why it can run for a few seconds, cuzz it isnt a complete short, but enough to make something like a feedback loop, to quickly heatup the converter.
I dont use AS for anything, I dont trust it at all.
Heck AA scares me sometimes. hehe /ubbthreads/images/graemlins/wink.gif
It is SILVER you know. /ubbthreads/images/graemlins/wink.gif I wouldnt think it would take much to foul up the works.
Maybe thats why it can run for a few seconds, cuzz it isnt a complete short, but enough to make something like a feedback loop, to quickly heatup the converter.
I dont use AS for anything, I dont trust it at all.
Heck AA scares me sometimes. hehe /ubbthreads/images/graemlins/wink.gif
Doug S
Flashlight Enthusiast
[ QUOTE ]
McGizmo said:
EDIT: I see Doug S's post that came in while I was hunting and pecking. In the use of AS on PCB's I beg to differ with Doug. I think both AS epoxy and AS grease can effectively be put to use to destroy an otherwise functioning PCB. I don't know about you guys but much of the AS I have used has settled out in the syringe and it wouldn't surprise me if enough silver isn't in levels of concentration to take the AS from thermally conductive to electrically conductive! As I recall, even the manufacturer recommends AGAINST the use of AS in trace areas. I don't trust AS to be non electrically conductive and frankly I don't count on the electrically conductive epoxies for 100% performance.
[/ QUOTE ]
Well Don, we *do* differ on this one but I am always receptive to being persuaded by real world observations. My original statement is based upon the fact that the *stock* BB has no high impedence external control inputs that would be affected by the small leakage currents that would result from use of Arctic Silver assuming that it has the low conductivity they claim. You raise an interesting point with your observation about seperation of the components in the mix. Your Hall effect input to the Iadj pin is a different animal. I believe that your external circuit in this case had an impedence in the megohm range. In this case, small leakage currents could affect the circuit. I will be persuaded If you can report that you have seen a stock BB have its operation affected by gooping AS on its pins. I am standing by to stand corrected /ubbthreads/images/graemlins/grin.gif
McGizmo said:
EDIT: I see Doug S's post that came in while I was hunting and pecking. In the use of AS on PCB's I beg to differ with Doug. I think both AS epoxy and AS grease can effectively be put to use to destroy an otherwise functioning PCB. I don't know about you guys but much of the AS I have used has settled out in the syringe and it wouldn't surprise me if enough silver isn't in levels of concentration to take the AS from thermally conductive to electrically conductive! As I recall, even the manufacturer recommends AGAINST the use of AS in trace areas. I don't trust AS to be non electrically conductive and frankly I don't count on the electrically conductive epoxies for 100% performance.
[/ QUOTE ]
Well Don, we *do* differ on this one but I am always receptive to being persuaded by real world observations. My original statement is based upon the fact that the *stock* BB has no high impedence external control inputs that would be affected by the small leakage currents that would result from use of Arctic Silver assuming that it has the low conductivity they claim. You raise an interesting point with your observation about seperation of the components in the mix. Your Hall effect input to the Iadj pin is a different animal. I believe that your external circuit in this case had an impedence in the megohm range. In this case, small leakage currents could affect the circuit. I will be persuaded If you can report that you have seen a stock BB have its operation affected by gooping AS on its pins. I am standing by to stand corrected /ubbthreads/images/graemlins/grin.gif
