The lower the Rds(on) the more efficient the mosfet. Its as simple as I=V/R. We want more I, R being the Rds(on) and voltage being a li-ion(fixed 4.2v). R (resistance) ds being (drain to source) is the resistance in the mosfet when fully saturated (on). Just like a switch. If the switch is turn off very high resistance, no current can pass, when the switch is closed there is a resistance across the switch and a Vdrop but current can pass. The lesser the resistance across the switch (drain to source) the more current that can pass, ohms law.
I have no idea if there is other parts of these boards that are different. But a lower Rds(on) should give more current with less heat. I have some East-92 boards on the way with the lower Rds(on) mosfets off ebay. So I cannot comment on whether it makes a difference on these boards yet, but theoretically speaking, it should.
Did you mean that you have some EAST-092 drivers from Ebay, or MOSFETs from Ebay?
This is where moviles first introduced the driver with the mosfet. He had very good results using a ssr-90.
I believe the driver used in the mte sf-15 maybe almost a exact copy of this driver (east-92) from fasttech.
The driver on the left(40N06)measured 1.2A on high. The driver on the right(06N03)measured 4.6A on high at the LED. With a panasonic NCR18650B. Out of all the drivers I ordered (over 30) I received 6 drivers like the one on the left and the rest were like the driver on the right.
I still want to test all of them, in due time! I think I will order 1 or 2 more……… DOZEN! Hahahahaha! 
I don’t know why my pic turned. The top driver is the good one. The confusion was not intended. These drivers are confusing enough!
| Post | User | MOSFET | Amps | |
| 113 | comfychair | 935-4858NG | 2.45? | |
| 113 | comfychair | DTU40N06 | DC51U | .85/2.45? |
| 126 | viffer750 | DTU06N03 | DC50U | 5.3 |
| 144 | Hyprmtr | DTU06N03 | DC50U | 4.6 |
| 144 | Hyprmtr | DTU40N06 | DC51U | 1.2 |
I just ordered 40 more. :bigsmile:
Now that I got my last order of east92 drivers today in the mail and they are all with the supposedly good MOSFET, I couldnt resist!
Just curious… what are you going to do with all of them???
I will use a lot of them. I have been modding a few lights for coworkers and friends and family. I really like these little simple drivers. You cant beat the price either. As long as they all work properly.
Ken
Actually, your comment (“simple drivers”) brings up a question I’ve been wondering about. From looking at the MOSFET datasheets and the EAST-092 drivers, is it really that the MCU is just controlling the GATE pin of the MOSFET, and basically just opening/closing the MOSFET, and that’s all?
And for the modes, it just toggles the GATE pin in PWM timing (I guess the MCU has built in PWM timer)?
I mean, looking at the EAST-092 driver, there’re not even any components on the spring side, so all that’s there is on the non-spring side, which look like basically just the MCU and a (big) MOSFET, and some resistors and a couple of diodes and caps?
I just have a hard time understanding why it’s so hard to get a consistent supply of the ‘good’ drivers (FYI, I posted on the FT forum for ‘geek’: https://www.fasttech.com/forums/1127405/t/1056716/geek-east-092-can-fasttech-screen-for-the-good).
Swapping to supposedly 'good' MOSFET on the bad 092B drivers does not work. I tried two different parts with much lower Rds specs and no improvement. One was actually worse - .72A. The other I didn't even bother measuring, I could see the SOS was still far far brighter than the 'high' mode.
One of the 103 resistors needs to be swapped to an 010? Or both? Or is it in firmware, using PWM on 'high'? What do I jumper to where to override that, if that's what it's doing?
BAT+ to GATE? That should bypass the MCU, right?
Thanks for trying and posting, comfy.
I’m starting to think that it IS the firmware. I think that I mentioned that when I was trying to get a “good” HD2010 from Tmart awhile ago, there was mention that one of “the signs” of a bad HD2010 was that the strobe was much brighter than the high. It’d be nice if someone could try swapping the MCU from a “good” driver to a “bad” driver and then see if that makes the “bad” driver a “good” driver?
Either that, or ground to GATE (not sure if GATE high=>open MOSFET current path, or if GATE low=>close MOSFET current path).
If you do that, you’ll probably want to disconnect/cut the MCU-to-GATE path/trace?
Does anyone know if, in high mode, the original EAST-092 driver uses PWM (or not)?
I reinstalled the 092B's original FET. Voltage on the GATE is 1.83v on 'high', and 3.34v on SOS. (I realize this isn't an actual DC voltage, but just an average of the PWM signal) Anyway, jumping BAT+ to GATE does make it go DD, but of course with no modes. This isn't the same as jumping LED- to GND and completely bypassing the driver altogether, this is just commanding the FET fully on.
Current output, stock form: .89A
GATE jumped to BAT+: 2.67A (only MCU bypassed)
LED- jumped to GND: 2.67A (driver bypassed completely)
(yes I know these numbers are low, all I want to see is the driver output equal to the current with the driver bypassed, which is what I got)
So the 'bad' FETs are capable of doing direct drive like the original EAST-092s, IF they are given the right signal on the GATE.
p.s. MCU pin#5 is the PWM output
Hyprmtr-
Can you please do this test on one of your good true-confirmed-DD drivers: jump BAT+ (or LED+, they're the same) to the GATE on the FET, the left leg when viewed with the legs pointing down? I don't need a current measurement, just want to see if yours get brighter when that's jumped.
Can you tell if there’s PWM on high mode? Maybe that’s a redundant question (I think the answer is ‘yes, there’s PWM on high’, by implication of what you said), but wanted to confirm.
If there were no PWM on high there would be 3.34v (or more) at the gate, not the 1.83v I measured. I don't have any other way to test it and just doing it visually might not show anything if the frequency is high enough.
The reason I’m asking that is that PWM allows using a digital signal to “simulate” an analog voltage. So, if the driver does control the output in high mode using PWM (vs. just holding the GATE high), then the firmware might be using the wrong duty cycle for high mode, causing the GATE voltage to be lower than expected, and then causing the lower than DD current in high…
EDIT: I think we’re both saying almost the same thing, but just wanted to explain the reason for my question(s).
Yes, the MCU is sending the FET a signal with a duty cycle far short of being just 'on'. My DC voltage measurement is just an average of the PWM pulses. I have no other equipment to measure the actual PWM frequency.
Part number is hidden on the bottom of the chip, it's marked as 'T1H1K68CAE', whatever that is. Being a MCU that hasn't been reverse engineered yet this is something not fixable. Unless an ATTiny13 can be grafted on, but I haven't traced out the original MCU paths yet.
Removing the MCU to get the number killed the driver, 4 of the pads came off with the chip.