DIY Driver Nanjg Style/AMC7135 using Z8F082A MCU

Made a new programming interface/connector. Instead of soldering pins on the driver and using clips, I use this one.

Much faster to hook it up! This is my wired Asgard($35 BT?) version!

I've had this thread bookmarked for a long time now. Been wanting to read and learn from it. This is a great endeavor you have undertaken. It appears that you have had to push forward a lot of the way on your own. Not only are you showing how to make a similar driver, but also the process to getting custom fab circuit boards, etc. Thank you for this great effort and contribution to BLF.

I wish I had more time, because this looks like a great way to customize a light.

Waiting to get my hands on a few more of these next week. I have plenty of hosts waiting to put them in.
Have you tried to increase the PWM frequency any higher? I can hear PWM whine on the non-100% modes. I’m thinking 21kHz should be doable with the internal 5.53MHz oscillator. I’ve used that frequency before on this part and you end up with a byte for the PWM register, so 255 levels of adjustment (0.3% to 100%). Just a thought. I think QLite drivers are around 17kHz.

I know how it feels to do research(?) with minimal feedback. Glad you are keeping this up. Press on. :beer:

Thanks Darren! I’ll revisit this on Monday when I can use a scope to verify. Good input!

@Rufus - Yup! For your turn signal, I didn’t use a PWM so it’s not an issue. Also I used the WDT and Stop feature of the MCU to
save power when it’s OFF. :beer:

Yes, for sure keep pushing on! Got my parts a few days ago, so I need to get on it and build up a module with this driver. Really can't wait to try it out

That’s nice. I’m sure he’ll be doing more coasting than braking so the additional off time won’t hurt.

Even though I can’t hear the whining, I changed it from 4.5KHz to 21kHz. Changed also the # of groups to 3 and have re-arranged the groupings.

I like this one better! :beer:

Hi Tivo,

Sent you a PM…

Ohaya sent me one East-092 for testing and was able to reproduce the problem on getting stuck to HI mode.
Did a quick experiment:

1. De-soldered and lifted Pin#1 of East-092 FET (06N06):
2. Using one of my spare Zilog MCU driver with the same code as above:

  • Connect Vbat+ to East-092 Vbat+
  • Connect Vbat - to East-092 Vbat -
  • Connect MCU PWM pin to East-092 Pin#1

3. Power up and can cycle to different modes and groups as shown in the table above.
Moonlight is working fine. With Bench PS: 4.1V, direct connection(no switch),
I’m getting 4.4-4.6A. Not getting hot with PCB floating in the air. Should be better
when in a pill or flashlight.

I might switch to this format as soldering several 7135s is a pain!


This might be interesting - a new source for a DD driver like the old EAST-092, but with more “modern” modes and mode groups?

Maybe you should start a new thread on this development?

Sharing the Design files and Programming Procedure.

TZF082A LED Driver_AMC7135_Rev0.04 PDF
in Word file with links to parts.


Updated Build Documentation to Rev 1.0 in the OP. Cheers! :beer:

Very very nice!

Cool. Ordered 9 boards. Thank you Tivo. The programing contact points look slightly different. Will the same cable from your other driver project work for this board to?

HI 4Wheelr!

It was too late when we realized that the pins and sequence were not the same. In case we do another PCB designs we will make the programming pins layout the same or standard so that the last one is re-usable. The programming pins for this 8 AMC7135 pads is documented in the PDF/word file. Here’s a pic:


Thank you. I better label my 2 cables clearly.

This no longer in "beta" ready to go?

Thinking of getting some of these too!

Yes WarHawk. I replaced all my P60 and C8 driver using this.
You can also use this as a controller for slave boards like this.

Cheers! :beer:

With updated codes in the OP and here: TZF082A_LED_Driver_AMC7135_1.1