Hi Tom-E!
Yes it can support an e-switch and I have an experimental code done before but have not finish it yet.
And yes, I was using an I/O pin to wake up the MCU.
The MCU has 8K code. The current code is around 2K but I have a lot of un-used codes in there. Not optimized yet for size.
Low Batt and Temp monitoring can also be supported.
Since my objective is to share, hopefully those interested can learn and improve the code, create their own groupings and modes and start sharing. One can post/share the hex file and guys here can try it on their light.
Once you learn the basics, you can easily update the code using the programmer and the batch file.
Inputs are welcome.
Thanks! I have cleaned up the source code to removed some of my groups which are not needed.
I now officially changed it to Tivo532-092A_LED_Driver_1.0. The zip file now includes the ZDS II-Z8 Encore 5.0.0 installation software, pdf and docx files.
Ok, I'm in. This seems to be the best driver out there that's real. It sure has a great feature of linear regulation and direct drive with nice sets of modes. I know it's a little early - of course would be nice to see others first try it out, but think I'll go slow at first - try it out and see. Checked around at pricing - there's some investment involved, but I don't have to order the qty you indicated for the pricier parts for now. Z8's in qty 1 at DigiKey are only $2.05 instead of $1.88 at qty 25 - no big deal. Figure I'll order parts to build 3-5 up first. I've programmed the Z80 back in the day when it was hot, but didn't know Zylog was still around.
I go through Nanjg's like crazy, and thinking this will sub pretty well, most of the time.
For max output builds, I really want to see what it will do in C8's and HD 2010's on copper pills. I got a Nanjg at 4.9A with a de-domed XP-G2/Noctigon reflowed onto the pill and can only get about 4.65A out of it with a Powerizer cell. I'm hoping this driver in "Turbo" mode will beat that by having lower resistance than the 14 7135's. De-domed XP-G2's on copper seem to be the worse case I'm seeing with 7135 based drivers.
My house is all apple. Could I build some up and send them to one of you folks that do windows to flash. I was contemplating the purchase of a windows laptop just to program the driver. After all as a flashaholic it makes sense. But my wife was kind enough to point out that buying a computer just to program a light is rather wasteful. Damn I hate when she is right.
I think I might have a compaq that has ’98 on it in storage. With any luck it will work and the wifi card in it will work. It’s been like 6 years since I’ve opened it up. Forgot about it until you said windows 98.
TiVo - I noticed there's no battery monitoring, or temp monitoring. Did you test these features out? Do you plan on adding support? I could look at porting what's in luxdrv for battery monitoring but need the I/O info. The 10 msec interrupt looks good.
Also, can you provide DigiKey refs/links for the Diode and resistors? Think I got the cap figured out but not sure of the diode. I want to get all the parts from DigiKey in one order, including the USB Smart Cable. If not, I'll get someone to help me at work.
I ordered 6 of the boards for now.
Also, it's very easy to edit the OP (Opening Post) - would be a good idea to have all your latest links and info in the OP, rather than having to scroll through this thread. I'm thinking this thread is gonna get rather long... (maybe from me alone...)
Just submitted the order. Here is what it looks like. Hope it's all good, think it should be. I'm only planning on building up qty 5 boards for now. The ribbin cable I should be able to get rigged up at work, with some help.
The MCU has a Low Voltage Detect feature and code can be added if needed. It’s just that I don’t think I need that feature. When my light is dim or acting strange(my indicators! ), I’ll just pop in a fresh battery. Got a lot from laptop pulls.
The MCU has also a Temp sensor and code can also be added. And yes we can create a code with these features later.
Cheers!
Just realized (my bad) I should have ordered more MOSFET's, because for a hot rod setup, would want at least 4 on the board. We figured a 0.17v drop for just 1 MOSFET - maybe you guys discussed this already.
Wait, what? Could you expand on that a little for those that have a lot to learn. I thought as is designed was “hot rod”. What am I not understanding about the number of mosfets and hot rod. I would like to run @ 9 amps allowing battery, host and thermal limits- on direct drive. Will I be able to do so with the current design?
Whoa there - 9A on a direct drive 1 cell light? Dunno if that's possible... The really nice thing about this design is you can use one MOSFET part, but it will add a fairly significant voltage drop, but if you parallel in more MOSFET's, the voltage drop will be reduced. So his design has foot prints to populate 4 MOSFET's, reducing the voltage drop from 0.17 to 0.004v. In addition, you could piggyback more MOSFET's for an even lower voltage drop -- this is all great stuff. Dunno if 9A would be possible for an XM-L2 on copper though. If you look at the charts (somewhere), the voltage requirement is quite high at high amps.
Tom, nope we have not discussed the FET voltage drop yet.
I did a quick test with 2 DMG6968U soldered on the board here’s what I’m getting:
Voltage across GND and Vbat(LED+) = 3.87V
Voltage across LED- and Vbat(LED+) = 3.81V
Looks like we have 0.06V drop across the two FETs for these particular configuration/setup.
From my initial observation, two DMG6968U is good enough for one XM-L. No significant improvement adding FET. The FET is rated at 5A in a typical situation(or ideal). But as we all know, contact resistances on wires and switches, Battery and Temp contributes to what actual current you can get. Cheers:
want to give you a man hug!
), I’ll just pop in a fresh battery. Got a lot from laptop pulls.