Oshpark Projects

Just did a search and the 70n02 still shows up but from a different seller. Until I use them up I’ll stick with the Lfpak56 since it fits more easily on a 105c than the vishay. I don’t do nearly as many mods as others so they should last awhile.

I have a question in my mind that when you guys are finding the low Ron FET, did you consider the low on Vgs of the FET is also a very important criteria for a single li-ion cell driver?

Microa, I'm comparing the RDS(on) at the lowest specified VGS. Is there something else I should be considering such as the "Gate Threshold Voltage"?

Most of the high power FETs show their Ron at 4.5Vgs and 10Vgs. I am not sure how many Vgs the attiny13A can provide to the FET but obviously under 4.5V. I doubt that the FET can be fully turned on. Low Vgs FET can be fully turned on at 2.5V or even 1.8V.

I’ve eyeballed the datasheets some more. See below for what I’m currently looking at.

That’s fine. Both 70N02 and 70N03 can go through 20A under 3Vgs.

I haven't caught up yet with the last 100(!) or so posts, but I tried the AOD510 and... the other one, both of them were flaky changing modes using the same known good gate&pulldown resistors as with the Vishay 70N02, so I didn't bother trying to get any current measurements for comparison.

Huh. Well that’s definitely the most valuable info at this moment.

I wonder where the hangup is. Come to think of it… can you precisely describe the flaky mode changes again?

It (both of them, really) acted the same as the early 70N02/17DD weirdness before the gate/pulldown resistors were added. It might just need different values there, but I didn't have time (and, I have like 150 of the 70N02s).

I figured: I just can’t remember precisely when/where you described the issue so I can’t just go read that.

Gate charge? Doesn’t it have to do with the mcu’s ability to supply the necessary current to the gate at the specified voltage.

Sometimes changes mode correctly, sometimes jumps two modes instead of one, sometimes hangs up, sometimes goes into some weird alien-Morse-code thing (dim-bright-BRIGHT!-off-BRIGHT!midlowbright-dim-off... etc.).

Chris are you able to post a short video showing the issue? It sounds like what I have but not sure. Also is this with a clicky or momentary setup that gives you issues (or it doesnt matter, it does it with both randomly)?

The biggest similar issue I’ve had is when I tried to use a stock FET I pulled out of an M6, it cleared up when I swapped it out for a 70N02 however I know the M6 FET’s work when left on their own board but controlled by a ATTiny13A on a piggy backed board, strange…

I had similar erratic behavior with both the irlm 2502 and lfpak 56 when using them without the gate resistor and above a certain current level. The gate resistor resolved this in both cases. It may be something to do with board geometry or who knows what since as far as I’ve heard it varies from board to board or FET to FET.

I tested the AOD510 & STD95N2LH5 on a 17DD v2.0, clicky switch (added to the side of my test battery box), single tame ICR 18650, test LED was triple XPG2 (parallel) on a CPU heatsink. I only tried them as drop-in replacements for the 70N02, so same 130/12k resistors, didn't try either FET without the resistors.

On paper, the AOD510 should be better at everything. Lower Rds(on), lower total gate charge, etc. Don't know why it wouldn't cooperate. I still have them, I may get back to it when I have some spare time.

Are you guys noticed how fast OSHPark has gotten these past few weeks? I don’t know it they were just going slow cause of makers fair and now that’s over or what happened but I’ve got shipping notices for boards I ordered just 8 days ago, used to it was 14+ days.

Doubt anyone from OSHPark is reading this thread or anything but if they happen to see this thanks for the hustle lately!

I created a driver for Roche F6 or Roche X3 if anyone would like it. 70N02 mosfet DD side switch. I would just prefer someone took it for a spin before I posting it in general.

Since 70N02’s aren’t current avaliable on eBay what’s the next best option from a US based seller?

See post #853
I suspect that nobody who needs to build drivers and has the capacity to determine what will work well has run into this yet. Many of us have some 70N02 stockpiled for now, so no need to buy more FETs and test them… comfychair went as far as to do a quick test and the driver failed to operate properly with other candidates.

Possibly adjusting the gate pull up/down resistors will solve the problem with AOD510, if so then it looks like a good choice.