SPARK SP6 Review (5 x XM-L T6 | 6 x 18650 / 12 x CR123A)
- Thread starter turboBB
- Start date
Help Support Candle Power Flashlight Forum
rufus001
Enlightened
Great video! Spark have really done a good job with this light.
Thx! Here's the one covering heat testing w/out the use of a fan:
carl (and others) hope this addresses any concerns you have with heating issues. However, I must caveat as such:
This test was easier on the SP6 because:
- the SP6 was pointed straight up and there was no lens covering the heat allowing it to escape better (however keep in mind that air is actually a very poor medium for thermal exchange)
This test was much harder on the SP6 because:
- there wasn't a use of a fan and the reflector (which is actually a substantial chunk of aluminum) was not in place to further help draw heat away from the stalk (and as we all know, aluminum is a very good medium for thermal exchange)
Between the two, I'm leaning towards the latter but I'd like to hear your opinions on this.
Strangely, there wasn't any thermal compound on the base of the reflector. Personally, I feel this would've further aided heat dissapation. Of course, this has got my curiosity piqued so I'll do some add'l testing in the future to see what the heat on the reflector is without any compound and then again after Arctic MX-4 compound is added.
More testing to come! ;o)
Cheers,
Tim
carl (and others) hope this addresses any concerns you have with heating issues. However, I must caveat as such:
This test was easier on the SP6 because:
- the SP6 was pointed straight up and there was no lens covering the heat allowing it to escape better (however keep in mind that air is actually a very poor medium for thermal exchange)
This test was much harder on the SP6 because:
- there wasn't a use of a fan and the reflector (which is actually a substantial chunk of aluminum) was not in place to further help draw heat away from the stalk (and as we all know, aluminum is a very good medium for thermal exchange)
Between the two, I'm leaning towards the latter but I'd like to hear your opinions on this.
Strangely, there wasn't any thermal compound on the base of the reflector. Personally, I feel this would've further aided heat dissapation. Of course, this has got my curiosity piqued so I'll do some add'l testing in the future to see what the heat on the reflector is without any compound and then again after Arctic MX-4 compound is added.
More testing to come! ;o)
Cheers,
Tim
Last edited:
rufus001
Enlightened
I have a question but I don't know how to phrase it so you'll have to guess. :thinking: 
I think if the heat sink gets hot then that is good as it is meant to be drawing the heat away. But away from what? What part of the light causes flickering/failure due to overheating? And can that be measured to see how close it is getting to failure.
I think if the heat sink gets hot then that is good as it is meant to be drawing the heat away. But away from what? What part of the light causes flickering/failure due to overheating? And can that be measured to see how close it is getting to failure.Typically it's the circuit as LED's; and more specifically the XM-L's; have a max. junction temp of 150C/302F so theoretically a single XM-L can operate to that temperature before overheating. This is why LED manufacturers always suggest positioning the driver circuitry away from the LED's. Unfortunately I don't know enough about the specifics of the chips involved with each circuit and which specifcially is prone to heat failure so I'll defer to someone more knowledgeable to pipe in.
NorthernStar
Enlightened
A great review as always!
I like that the 18650´s batteries is included with the flaslight.If one could see some outdoors longdistance beamshots also it would be great!
I like that the 18650´s batteries is included with the flaslight.If one could see some outdoors longdistance beamshots also it would be great!
carl
Flashlight Enthusiast
turboBB - can't thank you enough for running that temp check. Thank you very much! The SP6 seems to run up to roughly 200F which is not enough to damage the LEDs - awesome light!
rufus001
Enlightened
Any idea of how the charger knows when to cut out? If it's based on an overall voltage then surely the batteries wouldn't get too much out of sync?
I'll graph the charging algo when I'm done with the bulk of the review but based on my initial testing, the carrier wires the batteries in series (6S1P) for a total v in the neighborhood of 25.2V (4.2V x 6). There is no balacing done so this is precisely why SPARK decided to include quality matched cells. Matched here means they are all of the same brand and labeled capacity (albeit I'm not certain if they actually conducted any internal resistance and real capacity testing). I believe charging is solely controlled by the adapter and not due to any circuity within the SP6 itself. When the charger (with a rated output of 25.2V @ 1A) is first plugged in, a small green LED will illuminate in the corner of the adapter. It will change to Red during charging and back to green when the batteries are fullly charged.
All, updated runtime chart on Max and some material added to Design & Features section.
Cheers,
Tim
All, updated runtime chart on Max and some material added to Design & Features section.
Cheers,
Tim
rufus001
Enlightened
So if I use 6 identical high quality batteries I should be fine to simply use the charger? Don't worry, I'll treat the answer as "All care. No responsibility." 
I would use the internal (unbalanced) charging system only when in a bind.
When you use protected cells theoretically you should be fine, however I feel the individual cell protection should be a last line of defense an not used on a regular basis.
I have a lithium battery pack for a rc-car where one of the cell exhibits a significantly higher self-discharge rate.
Using a balanced charger the pack works absolutely fine, however if I were to charge it unblanced results would most likely be catastrophic.
While this is not 100% applicable here (no individual cell protection, cheaper quality) it highly discourages myself from handling any lithium rechargeables carelessly.
Although it is probably already careless enough to be still using this particular pack...
When you use protected cells theoretically you should be fine, however I feel the individual cell protection should be a last line of defense an not used on a regular basis.
I have a lithium battery pack for a rc-car where one of the cell exhibits a significantly higher self-discharge rate.
Using a balanced charger the pack works absolutely fine, however if I were to charge it unblanced results would most likely be catastrophic.
While this is not 100% applicable here (no individual cell protection, cheaper quality) it highly discourages myself from handling any lithium rechargeables carelessly.
Although it is probably already careless enough to be still using this particular pack...
rufus001
Enlightened
So what does the balancer actually do? I thought the problem was it would shut off charging before all batteries were fully charged if there was no balancer. But from the previous post it appears the balancer alters the current to each battery depending on how discharged it is? Therefore preventing to much current going to a deeply discharged battery resulting in 
?
?Balance charging (a feature thus far found only hobby chargers) is able to monitor the voltage of each cell and manages the charge of each individual battery that is charged in series through the means of balancing leads. What it does is to ensure that no individual cell is overcharged by actually discharging any that exceeds the preset limit (and this can vary depending on the chemistry of the batteries in question). It's a bit heavy to go over in a reply so just do a search for balance charging and you can read up on it.
For the record, there is NO balance charging on the SPARK SP6. The charger plugs to the end of the light and charges all 6 cells in series. As such, there is a potential for each cell to be mismatched. The easiest way to understand this is to think in terms of resting voltage not matching between all cells; ie: you can end up with 3 cells at 4.19 and 3 cells at 4.21 as a simple example OR you can end up with any combination of under/over charge for each cell as long as it equals to 25.2V (or whatever is the terminating voltage that the charger is programmed at - I don't know yet since I haven't graphed the charging algo yet). This typically happens as cells age and their internal resistance builds up thus the cells with higher resistance takes longer to charge thus staying lower in voltage while those with lower resistance charge more quickly and thus are prone to overcharging. It's a bit ironic that the one bad cell will actually be safe from overdischarging in this scenario while the healthy cells will take a hit.
So let's say that of 6 cells, 5 are identical in IR and 1 is a bad cell with pretty high IR. Now imagine a scenario where by the time the 5 healthy cells reach 4.2v each (for total of 21v) the bad cell is just at 3.9v thus representing a total of 24.9v. Now let's say that the charger has been programmed to terminate at 25.2v exactly, so the charger at this point will continue to charge all cells since it's programmed for just that and isn't aware of what each cell is at. To make things easy, let's pretend that it'll continue to charge all 6 cells equally to reach 25.2v so that means each cell will get an additonal .05v at which point you now have 5 cells at 4.25v and that one bad cell at 3.95v for a total of 25.2v. Again, as far as the charger is concerned, it's done its job and will terminate charging regardless of what each cell may be at. While 4.25 is overcharged, it's actually still within specs for many cells (which typically state 4.2 +/- .05V). However, now imagine that there were actually 5 bad cells and a single good cell, you can imagine how fast that single cell would fill up and be overcharged badly leading to very undersirable outcomes! This is the potential danger of charging batteries in series without a balancer.
What the use of protected cells will do is that each cell features a protection circuit that will trip in the event the cell is over-charged/discharged. Since each cell features this circuit, you basically have six times the safety redundancy in the event one of these cells has a faulty/bad circuit.
This is why SPARK chose to pick 6 "matched" cells (again, note my caveat in previous reply) of the same type and brand since IR typically has less variation. However, while these Panasonic cells are indeed good quality, they are not protected cells (they lack a protection circuit). They do however feature a PTC thermistor (strangely the spec sheets directly on Panny's site doesn't mention this but I found this one that does) which provides a mechanical safety in the event of high current surge or temperatures.
Please read this great article on Battery University for a full explanation of the safety considerations when using Li-Ion cells (or just pop over to the Electronics sub-forum for many great detailed threads):
http://batteryuniversity.com/learn/article/lithium_ion_safety_concerns
In the end, since I'm typically risk averse, I usually like to use protected cells in my lights. However, I ALWAYS check the individual voltage of cells after charging (regardless protected or not) thus do not have any concerns with using unprotected cells as well especially when they are new since the likelihood of them being unbalanced is very low.
I have now run the Panny cells through two cycles and both time all cells were 4.20 (+/- .01) after charging regardless that their ending voltage were not all the same after runtime testing.
However, as they say, YMMV so you'll need to make your own judgement call on whether you feel comfortable/safe using non-protected cells as well as the included charger.
Cheers,
Tim
For the record, there is NO balance charging on the SPARK SP6. The charger plugs to the end of the light and charges all 6 cells in series. As such, there is a potential for each cell to be mismatched. The easiest way to understand this is to think in terms of resting voltage not matching between all cells; ie: you can end up with 3 cells at 4.19 and 3 cells at 4.21 as a simple example OR you can end up with any combination of under/over charge for each cell as long as it equals to 25.2V (or whatever is the terminating voltage that the charger is programmed at - I don't know yet since I haven't graphed the charging algo yet). This typically happens as cells age and their internal resistance builds up thus the cells with higher resistance takes longer to charge thus staying lower in voltage while those with lower resistance charge more quickly and thus are prone to overcharging. It's a bit ironic that the one bad cell will actually be safe from overdischarging in this scenario while the healthy cells will take a hit.
So let's say that of 6 cells, 5 are identical in IR and 1 is a bad cell with pretty high IR. Now imagine a scenario where by the time the 5 healthy cells reach 4.2v each (for total of 21v) the bad cell is just at 3.9v thus representing a total of 24.9v. Now let's say that the charger has been programmed to terminate at 25.2v exactly, so the charger at this point will continue to charge all cells since it's programmed for just that and isn't aware of what each cell is at. To make things easy, let's pretend that it'll continue to charge all 6 cells equally to reach 25.2v so that means each cell will get an additonal .05v at which point you now have 5 cells at 4.25v and that one bad cell at 3.95v for a total of 25.2v. Again, as far as the charger is concerned, it's done its job and will terminate charging regardless of what each cell may be at. While 4.25 is overcharged, it's actually still within specs for many cells (which typically state 4.2 +/- .05V). However, now imagine that there were actually 5 bad cells and a single good cell, you can imagine how fast that single cell would fill up and be overcharged badly leading to very undersirable outcomes! This is the potential danger of charging batteries in series without a balancer.
What the use of protected cells will do is that each cell features a protection circuit that will trip in the event the cell is over-charged/discharged. Since each cell features this circuit, you basically have six times the safety redundancy in the event one of these cells has a faulty/bad circuit.
This is why SPARK chose to pick 6 "matched" cells (again, note my caveat in previous reply) of the same type and brand since IR typically has less variation. However, while these Panasonic cells are indeed good quality, they are not protected cells (they lack a protection circuit). They do however feature a PTC thermistor (strangely the spec sheets directly on Panny's site doesn't mention this but I found this one that does) which provides a mechanical safety in the event of high current surge or temperatures.
Please read this great article on Battery University for a full explanation of the safety considerations when using Li-Ion cells (or just pop over to the Electronics sub-forum for many great detailed threads):
http://batteryuniversity.com/learn/article/lithium_ion_safety_concerns
In the end, since I'm typically risk averse, I usually like to use protected cells in my lights. However, I ALWAYS check the individual voltage of cells after charging (regardless protected or not) thus do not have any concerns with using unprotected cells as well especially when they are new since the likelihood of them being unbalanced is very low.
I have now run the Panny cells through two cycles and both time all cells were 4.20 (+/- .01) after charging regardless that their ending voltage were not all the same after runtime testing.
However, as they say, YMMV so you'll need to make your own judgement call on whether you feel comfortable/safe using non-protected cells as well as the included charger.
Cheers,
Tim
I honestly think the internal charger should not be used with these cells at all.
Basically there is not one protection system and 6 links that can break in the chain (7 if you count the charger).
I guess there a two possibilities that could happen:
1) Spark's idea is probably that if single cells get overcharged slightly, the system will balance itself naturally.
2) My scenario though: The lowest capacity cell of the six (the one which will reach over 4.2V first) will get overcharged.
Being overcharged a little every time, this cell will age faster and loose its capacity quicker than the rest and as a result will reach ever higher voltages with every charging cycle.
My first instinct is that Spark must have thought this through, I mean they even shrink-wrapped the cells as to say that this system requires no further intervention by the user.
However I think if you insist on using this charging system I think you should check the individual cell voltage after charging at least every couple of cycles (which is basically what a balancing system does every second).
rufus001
Enlightened
Thanks Tim!
The "interesting" thing is that it starts out at 500 lumens and then "slowly" ramps up to 3K lumens over a span of 2 minutes???
Unless I'm misunderstanding you here, this is not "interesting", it's a major flaw and a deal breaker. Forgive me, but at this price point, I expect an led flashlight to turn on at or very near it's full rated output when I press the switch, not 2 minutes later! I don't need to wait for my TN-31 or RRT-3 to warm up. I could buy a cheap HID light for a lot less if I wanted a delay in achieving full output. I'm surprised that no one else has commented on this. I think I'll be passing on this light.
Unless I'm misunderstanding you here, this is not "interesting", it's a major flaw and a deal breaker. Forgive me, but at this price point, I expect an led flashlight to turn on at or very near it's full rated output when I press the switch, not 2 minutes later! I don't need to wait for my TN-31 or RRT-3 to warm up. I could buy a cheap HID light for a lot less if I wanted a delay in achieving full output. I'm surprised that no one else has commented on this. I think I'll be passing on this light.
Last edited:
You're taking the statement out of context... If you look at the runtime chart, the ramp up is not linear over that span. It goes from 500lms to about 3000lms in 2 seconds and then ramps from there to about 3200.
SPARK has not touted this as a feature so I'm uncertain if it's a flaw or by design, in which case if it was the latter, I'm not surprised as a few of the high output multi-emitter lights feature this (e.g. XTAR S1, TN30...).
Besides most of us would never notice the ramp up (at least on the SP6) if it were not for test equipment to capture this. Let's think of it pragmatically, at night if your eyes were dark adapted, you'd actually appreciate a slow ramp up especially at these output levels.
I'll find out if this was in fact by design or a flaw on my sample, in the meanwhile, I'd like to hear back from current owners on whether or not they notice this ramp up.
Cheers,
Tim
SPARK has not touted this as a feature so I'm uncertain if it's a flaw or by design, in which case if it was the latter, I'm not surprised as a few of the high output multi-emitter lights feature this (e.g. XTAR S1, TN30...).
Besides most of us would never notice the ramp up (at least on the SP6) if it were not for test equipment to capture this. Let's think of it pragmatically, at night if your eyes were dark adapted, you'd actually appreciate a slow ramp up especially at these output levels.
I'll find out if this was in fact by design or a flaw on my sample, in the meanwhile, I'd like to hear back from current owners on whether or not they notice this ramp up.
Cheers,
Tim
warmurf
Enlightened
That ramp up time isn't a flaw and really isn't any issue. I am concerned thought about the unbalanced charging potential, however as someone said above, take them out every second charge and check them should work? So far I'm not seeing anything against buying this light, except the price. I may need to list some lights for sale.......
Well, the review was not conclusive in terms of ramp-up speed. It just mentions that it takes "2 minutes" to go from 500 lumen to 3000+ lumen.
And yes, 2 seconds would be tolerable (if not optimal), but it is still a flaw. 2 Minutes would be a disaster.
The instant on (allowing the use of intermediate light) is one of THE advantages LEDs have in very high output lights like these compared to HIDs (I own a HID that takes about 20s to get to full power and thats really annoying).
And yes, 2 seconds would be tolerable (if not optimal), but it is still a flaw. 2 Minutes would be a disaster.
The instant on (allowing the use of intermediate light) is one of THE advantages LEDs have in very high output lights like these compared to HIDs (I own a HID that takes about 20s to get to full power and thats really annoying).
