NUBM0G

[RedCowboy]

At a 25A draw that Molicel cell will sag a lot, maybe down to 2.25V or less if not fully charged. It's ok for making a nickle chrome wire glow red hot in a vape gadget, but it can't hold up like that for very long maintaining a high current load.

The longer battery may have twice the capacity ( mAh ) but half the idle voltage of 2 half length cells in series and boosting will demand twice the current plus what it takes to overcome the additional battery voltage sag under a high drain load. You're almost always going to get more run time from 2 half length cells in series bucking than 1 full length cell boosting and often when boosting to 4+ amps your wire dia. and switch contact resistance can become an issue from what I have seen. Actually I get twice the run time with 3 cells in series than 2 cells in series just driving a nubm44 @ 4.5A, remember we recharge when the li cells get down to 3.5V without any load, but I have discharged them down to 2.75V unloaded, however that's fully discharged.

 

[Giannis_TDM]

Looks totally sick! Agreed we need at least one commercially available host/kit that is setup for cylindrical lens correction. Many have done so here but as far as I'm aware none have been produced at quantities for sale.


That's why we need compact high amp boost drivers. Lazeerer is MIA and @Giannis_TDM is working extremely hard making excellent drivers for us! But having a reliable 5A+ boost driver will mostly eliminate the need for multiple small cells (16340/18350 etc).

The highest amp boost I've found is the 22mm Convoy 6v 8A driver. They also have a 17mm 6v 5A boost driver. I'd love to get setup to test how suitable they'd be for driving LD's. Start-up ripple, Vpp etc.

The chemistry advancements made also means we have the Molicel P28B 18650 with class leading perfomance of 25A continuous, and for those "lower" powered applications the Vapcell F38 3800mAh 18650 with 10A continous. (F38 18650 has almost double the capacity density at 79.2mAh/g compared to 41.6mAh/g for the 26350).

If you have an extra of those drivers I will happily test em for you for free, just send em over.

 

[dubious speculation]

At a 25A draw that Molicel cell will sag a lot, maybe down to 2.25V or less if not fully charged.

At 30A the P28B when fully charged will sag to 3.5v after 25s and has a capacity of 2400mah down to 2.8V at 30A. Bench test.

Let's use the Nubm44 and a 5A boost as an example.

The Vf of the NUBM44 at 5A is 4.84v:

NUBM44 power draw (W) = 4.84 x 5 = 24.2W

Now the best boost drivers are somewhere around 95% efficiency, but we know that's not usually the case, plus there will be additional resistance in the wires, springs and switches. We'll even assume a worst case voltage drop for the P28B at 3.5v. Even using an average efficiency of 75% we get:

Power draw from driver (W) = 24.2/0.75 = 32.27W

Amp draw from cell (0.7v drop) (A) = 32.27/3.5 = 9.22A

Less than 10A! Even some 14500 and 18350's can sustain 10A. Now let's use the more accurate Vf drop of the P28B at 10A of 3.9v:

Amp draw from cell (0.3v drop) = 32.27/3.9 = 8.27A

So again, the picture is much better than you're depicting. I will agree with you and say that it's easier to find reputable high amp buck drivers. The same cannot be said for high amp boost drivers and that's another reason I'm advocating for them so much.

A single cell is less to worry about and has less potential issues from cells in series being mismatched and discharging at different rates, causing imbalances and uneven voltage drops. Plus the energy density gains for 18650's are significant, almost double compared to the best 18350.

You're almost always going to get more run time from 2 half length cells in series

Let's put the numbers to the test and compare runtimes of the buck and boost setups:

2S 18350 amp draw (A) = 32.27W / 7.8V = 4.14A

18350 runtime (Vapcell A11) = 0.9Ah / 4.41A x 60 = 12.25 minutes

18650 runtime (P28B) = 2.6Ah / 8.27A x 60 = 18.86 minutes

Not double but that's 54% longer runtime! With the high capacity Vapcell F38 your runtime would be double the best performing 18350 at 24.3 minutes (Keep in mind all calculations here use the real world measured mAh and Vf drop of each cell at the respective current draw we're using).

(All Ah/Vf numbers used for calculations are taken from Mooch test results)

If you have an extra of those drivers I will happily test em for you for free, just send em over.

That would be excellent @Giannis_TDM ! @loreadarkshade said he will be testing that driver I linked within a couple of weeks. If not I'll send over a few drivers for you to keep and test out

 

[kecked]

Going back to the heat sink conversation, has anyone used indium foil on these diodes to get better transfer? This might make a real difference. No power supply hit either.

 

[wolfram]

try some gold leaf foil.

 

[loreadarkshade]

@loreadarkshade[/USER] said he will be testing that driver I linked within a couple of weeks. If not I'll send over a few drivers for you to keep and test out

Tested here for anyone following this thread