I’ll leave commenting on this to Pilotdog68, DEL, Tom and TA for now.
.
.
.
.
And well let’s be honest and mention the not so good too.
Got an email from engineering department
First their English is hard to understand and I am sure our English is even harder for them to grasp
1 They mention “noise” due to a component
2 they do not seem to understand how Narsil works.
Considering their office is not at the same location Barry, thus marketing and middle management works they probably cannot view BLF easy.
@1 we asked for more input here.
@2 the English text with Narsil manual is emailed and Barry asked to translate it. Engineering reacted with gratitude.
@2 we have been working on a flowchart to explain Narsil (well “we” meaning Tom E and Joechina as graphic designer)
Hopefully this chart can be made complete to the point it can be mailed to TF soon.
Well, as for a “completely new” board design, I agree that it’s wise to leave it up to those you mentioned. AFAIK, nobody has done any real testing of the TA SRK board yet, but both TA and Tom have been testing the other TA drivers already. Also, another thing I think I remember TA saying about his SRK board is that it won’t need a zener to do multi-series-cell arrangements, because of the way he designed for the Batt- of the first cell to connect on the outer GND ring and the Batt+ of that cell to connect to the first quadrant in the middle, for 1S power to the MCU. The remaining cells each connect to one another and to the remaining quadrants so that the board can be arranged in any of three ways (1S4P, 2S2P, 4S1P).
Edit: Oh yeah, his board arrangement will require that the cells don’t turn against the driver or tail PCB. They must load straight in somehow to stay in perfect alignment with the proper contacts.
I'm confused - the TA SRK board I read about in his OP did not mention 4P, only 2S and 4S? Something I missed? All we care about is 4P - other configs are nice to have.
Sadly I think the time for changes to the drive board are past unless of course TF rejects the design we already submitted.
The largest change I would like to see to the drive is a 3rd channel added with a bank of 7135’s, the voltage divider removed and in it’s place an indicator LED.
The 4S driver has had no testing but should work in theory, although trying to use diodes to act as polarity protection is not really an option as all the current would have to flow through them. I doubt they would be able to even handle that at all and they would cause a lot of resistance.
Sadly polarity protection in a 4P setup such as this is simply not real feasible. This is part of why I would have much preferred to see the Q8 use a 4S setup. Not only would you get more performance but it would make polarity protection a lot less of an issue. Too late for that though.
In that design it is indeed setup for 4s or 2s2p, although allowing it to be used with 4p as well would not be that hard but since the stock driver handles 4P just fine, I figure it should focus on series setups.
How about using two (0.5mm thick 1mm wide) insulating rings spaced 12~13mm apart glued to the battery side of the driver, it will mean button tops only or a dab of solder
When the battery is inserted correctly the button top will contact the driver, if upside down the insulating rings will stop it touching.
Added benefit, no added resistance
I’m not sure I would trust the thin wrapper to prevent a short on the - side with the driver board. Any slight bulge of the battery could create a contact.
But you’re making me think I could simply rewrap my cells with a small plastic ring on the - side. Something 1mm thick or so should make it much safer.
That sounds right that there aren’t any diodes, that would fit, with such low forward voltage. An improvement on steel springs as fuses would be bronze or bypassed springs and fuses with lower resistance and lower melting point, such as (thick) lead tin or solder. Maybe the connections on one board or the other could be done in such a way that the current has to cross a solder filled gap. That is, the path through the solder would be longer than just two pieces of copper soldered together. Or maybe such a thing is available as a fuse.
A nice thing about 4P is that we can use mixed cell types.
A while back on a modded Shocker with fully bypassed springs, someone had a short in the carrier and it simply fried a trace on the carrier - next weak point. Probably the same would happen on the BLF Q8 spring tail plate. Those traces make nice fuses...
That is comforting. Copper makes a better fuse than steel, I think, because its melting point is lower. If we don’t think it is safe enough, we could put a narrow place in each trace.
Hey add me to this list. One I dont own a monster level flashlight, I need one and you guys always get some awesome specs into these things (working on catching up on whats going on with the build)
Can I still jump in to get this beauty? I have one of those cheap end SRK copies but im not so happy with its quality. This might come in handy now that days are getting shorter here in Finland
We discussed this back around post 1510 Sharpie was going to look into doing some calculations and figure out how to design fusing traces for specific currents. Unfortunately his posts seem to only live on where he was quoted and I don’t know that he ever finished educating himself or posted his results.
TexasAce, you seem to be turning out boards with DipTrace like an… Ace, can you look into using traces as fuses?
Hhmm, can't recall where we left off as well, but the intention was narrowed traces to act as fuses. These can be easily repaired as well by adding a jumper wire across the break, hoping it's the point of weakness that fails. Potential issue is lose of amps of course, but think it was determined if it was only a short distance, it would make little difference. djozz knows more bout this though...
Yeah, I think Sharpie said that a very short distance of narrow trace wouldn’t contribute much to resistance in the circuit during normal use, yet would act as a fuse in times of overcurrent. It would be nice if fuses were to be made into a test-PCB and someone with a good test set-up (djozz?) would test and show what size traces blow at what current level.
