Driver Brainstorming Puzzle / identification / issue resolving

Flashlight Modding and DIY Parts
4

“What were you using for an ON/OFF switch for your test?”

No switch for test man… I used clamp meter and thick gauge wires for amp readings. My test is very accurate regarding that. I am 100% sure in that.

“How were you connected to the cell and the ”load”?“

By thick gauge wire. By classic clamp meter amps reading way

“What was your ”load” device and how was it connected?“

Load device? XPG2S42B DD, on DTP MCPCB, this crappy driver with carobronze bypassed springs, Samsung INR 30Q…

“The SS602 appears to be an N-FET although i couldn’t find an exact datasheet with that marking code.”

Maybe or maybe not? We need to confirm that

“What’s that little solder bridge above the black wire—was the trace cut or broken?”
Nothing… Just nothing… This is one of the batch of 500 drivers (yes you heard that right) :wink:

“Can you trace or determine which pins of the etched chip are being used and which are not connected to anything?”
Everything is connected and working… This is factory made driver. I just tried to modd this crappy driver for more current draw. But that ain’t happening. :person_facepalming:

Thanks :beer:

5

PWM you say? :laughing:

“Remove and bypass that (i hope its a) schottky diode.”

Any other way except loosing bat reverse protection? That is must feature…

“Connect led + through hole to the springs top”
Not possible without drilling and damaging components. I am using 20AWG wire. Even 22AWG won’t go…

“Remove that capacitor from + and connect to mosfets output and minus”
Hmm… Do you maybe know what does that capacitor do? :beer:

6

i was thinking it might be PWM also, do you have a scope that you could take a look at FET pin 4 or etched pin 5 to see it switching?

The FET has the same pinouts as Infineon BSC067N06LG in the PG-TDSON-8 package. You could check for a forward diode drop from pins 1-3 to 5-8, pin 4 is the gate. Yours is obviously a different vendor but same function.

Is the lower capacitor connected to the top of the 303 resistor, i can’t see if there is a trace?

On the etched chip, pin 1 has a trace to the upper capacitor and to the reverse protection diode?
Pins 2, 3, 4 have no connect?
Pin 5 connects to the 303 and to pin 4 of the FET? Pins 6,7 no connect? Pin 8 connects to cell (-)?

But some pin must be connected to the top of the 104 resistor?

We might be able to identify the etched chip from the pinouts. Or you could remove it and see the traces underneath and any markings on the bottom of the chip.

7

“I was thinking it might be PWM also, do you have a scope that you could take a look at FET pin 4 or etched pin 5 to see it switching?”

Friend I don’t even know what and where is Fet pin 4 and etched pin 5… You have to show me that on pics :slight_smile:

Where is it?

“The FET has the same pinouts as Infineon BSC067N06LG in the PG-TDSON-8 package. You could check for a forward diode drop from pins 1-3 to 5-8, pin 4 is the gate. Yours is obviously a different vendor but same function.”

I can’t do that… Why? I don’t know how… :person_facepalming: As I said I am good in modding stuff if I have components but I don’t have any electro-technical knowledge to… I am just regular Joe with soldering skills… I could perform test but only with tools and if someone explains technique.

“Is the lower capacitor connected to the top of the 303 resistor, i can’t see if there is a trace?”

I don’t know what you mean here… You will have to see by picture… Plus vias leads to capacitor and capacitor vias leads to 303 resistor which leads to one of MCU pins and then to FET PIN.

“On the etched chip, pin 1 has a trace to the upper capacitor and to the black wire? Pins 2, 3, 4 have no connect?”
Pin 5 connects to the 303 and to pin 4 of the FET? Pins 6,7 no connect? Pin 8 connects to cell (-)?

I don’t know where is pin 1,2,3,4… Pin 5 you say? What I know is that 303 is connected to that mcu pin and mcu pin leads to pin on SS602 tranzistor?

“But some pin must be connected to the top of the 104 resistor?”
Pin connected on top of 104 resistor… I don’t understand…

“We might be able to identify the etched chip from the pinouts. Or you could remove it and see the traces underneath and any markings on the bottom of the chip.”

I uploaded pictures… Do you see anything from pictures?

8

Just connect led+ with fat wire to sprigs top to see if any changes

9

Yes… I know what you mean… I could do that test when driver and MCPCB are out of the flashlight host and it would probably lower resistance a bit but this thing needs to be used in fully operational flashlight…

Ideally would be that hole in the middle as you said… But freaking China engineer dug the hole just for looks :laughing: Totally nonfunctional and redundant on this driver…But guy obviously like holes :laughing:

So if I do this and test show increased performance I will be unable to assemble light anyway right? :smiley: :cry: :person_facepalming:

I see other way… Drill through mod…

Check this idea and what do you think?


With my DIY Djozz FET driver same setup I had 4.5A current draw… With this one 3A :frowning: My DIY driver from above mentioned Djozz thread was far better than this crap…

And I really hoped that I will not have to mod anything from now on in my life but obviously I must mod everything I touch! :smiling_imp:

Freaking destiny…

10

yea do it, just make sure fat + contact

11

Please see this thread here comfychair-inspired quick+cheap+lazy single-sided 17DD FET-driver (poor man's nanjg92)

That is DIY FET…

How is possible that DIY is better than this? Thicker +(plus) vias or something? Does MCU regulates current flow or that resistors add additional resistance :laughing: ?

I don’t know… All I know my “new driver” sucks !

12

The etched chip is the one that has the marking code removed.

It may be a FET driver rather than an MCU, e.g.
Fet driver ic

Or it could be something as simple as a DC/DC voltage regulator, e.g.
Voltage Regulator

The drive voltage to the FET may be set low such that it is not being fully turned ON so it acts like a variable resistor rather than a low resistance switch. That would explain the limited current that you have experienced.

It seems that a person would need to determine how the circuit is designed to work before any meaningful changes could be made, but maybe that’s just me.

Good Luck to ya.

13

Thanks :+1:

But no luck :weary:

After wire drill through mode amperage raised a bit and that is all… Raised up to 3.6A…

This is probably not FET either…

Most probably this is ” DC/DC voltage regulator, e.g.
Voltage Regulator”

Or like Barkuti said:

” It is now pretty clear to me that what you got is just a fairly simple “resistor” driver, i.e. the driver is not regulated. The driver is using a power transistor for this task (the chip marked S6602).”

14

i kinda doubt there exists a bipolar transistor in that TDSON-8 package, but only you can determine whether it is FET or NPN—using the diode check function of your multimeter.

With the red probe on pin 4 then touch the black probe to pins 1, 2 or 3 and read the voltage drop; if about 0.6 then touch black probe to any of pins 5,6,7, or 8; if it reads about 0.6V, then it is a big power transistor NPN type.

If the above readings did not appear, then put the red probe on any of 1,2 or 3 and black on any of 5.6.7 or 8; if it reads about 0.5 to 0.7V, then that is the body diode reading of an N-channel FET.

If all this seems too complicated or outside of your abilities, then depending upon where you are located you could take your board to someone who can give you some professional electronics help if you really want to solve this. Or contact your vendor and have them identify the two chips.

15

I wonder what 104 resistor do? Its connected to minus and… ?

16

Thanks on help guys… I think this is resistor driver because it is producing a lot of heat… On that S6602 only 4 pins could be tested as pins from other side are invisible and soldered to minus.

Maybe this pic could inspire someone to make better driver than this (i think red traces are + and black traces minus)… So try to resolve puzzle :slight_smile: :+1:

17

Sure it have to be hot. It shorts +and - :smiley: And you have voltage drop on IC. Remove it.

18

Hmm… You mean to remove 104?

But flashlight works for more than 1 hour 30 minutes (see my test above)… Nothing exploded? Whole rig(battery, emitter, driver, switch) is fine after test…

Are you sure?

19

:smiley: :laughing: :+1:

I am going to pay a visit to my soldering iron now… Will report back…

20

nothing to lose anyway :smiley:

21

Yes… Of course… :+1:

BTW 303 looks fine to you?

22

Fine. Start test from 3V and increase ;))

23

I picked new driver from the batch and I did removed 104 from driver… short 20AWG wires, carobronze spring bypassed, XPG2S428 DD inside with DTP MCPCB,

I can only do test via clamp meter and fully charged Samsung INR30Q

And I did discharge test up to 3.8Voltage of mentioned battery.
so from Battery 4.2 to 3.8Voltage state it draw from 3.2- 2.6A of current (I did not measured minutes but it took a while to drop maybe 10 minutes or so).

It did run much cooler without 104
Reverse polarity protection works… I am impatient to see if battery overdischarge protection will kick.

So it is bit better without 104

Is there anything more that I could do or mod?

Could this be the cause of lower current draw or or S6602 is just plain resistor?

No one ever dealt with S6602?