[On Sale] UVC LED Boost driver for sub 300nm UV LEDs typically around 8V up to 500mA

I got myself a 275nm 50mW LED and wanted to make a fairly simple driver for it
TI TPS61165 used

Used with Klaran WD

Quarz glass window for S2+ needs a bit grinding down

just UVC

17mm in UVA+UVC (17mm2 AMCs; 20,21,22mm up to 4 AMCs)

I have here shunts for different currents
1R0 200mA
0R51 392mA
0R4 500mA
0R33 606mA

Prices

17-21mm are available with BLF Firmware Bistro (tail clicky) or NarsilM/Anduril (momentary side switch)
11€ per driver just UVC
11.5€ per driver UVA+UVC

per loose BLF spring add 0.4€
per short spring 0.25€

springs can be soldered, but stiff long spring adds 1.5€ for shipping Outside EU; inside EU 2€
silicone cable can be added and soldered for 0.1-0.175€ per inch

Bistro, NarsilM, Anduril can be customized mode spacing adds 1-2€ depending on how many groups or ramping table you want changed

Shipping:
2€ envelope priority air mail shipping worldwide
4€ for optional tracking insurance up to 25€
Soldered long phosphor bronze springs are possible, but it adds about 2.5€ shipping costs
per full 100€ additional insurance 2€

1.3€ Germany
adds 2.8€ for signed tracking + insurance up to 25€

connection on drivers

interested UVA+UVC

You read my mind. I will probably order a couple of these for UVC.

Should be fun, 50mW UV-C leds are affordable nowadays. I’m in for one driver. Not that I have use for such a flashlight… :innocent:

Will this have any temperature regulation? Those emitters get pretty hot.

As any Bistro or Anduril drivers in the MCU is the temperature read out

those LEDs have DTP pad and fit DTP MCPCB
but at 7-8V and 500mA we are not talking of anywhere close to single XPL or so for the host but the emitters run relative hot so i would not overdrive them much

OK to speed things up about 2 weeks I got shipping changed to DHL even knowing that it costs 5$ more than normal shipping with DDP especially as everything is slowed down in China
And of course they charged me here almost 25€ taxes and fees

But Scientist donated 100$ to this project, so I spent it (would not with my own PCB orders)
So I added also designs with 20, 21 and 22mm to support more hosts, those can run the UVA LED up to 4 AMCs
I put also some other boards on the 100x100mm PCB, the top right is on Kit Lutara´s request a 30mm with up to 24 AMCs

Tested successfully with normal Bistro, nothing much that could go wrong with this relative simple boost chip

Some more driver tests

Electrical measurements:

Output stability
It keeps at 100% output current down to about 3.25V
at 3V it drops to 88% output current

Efficiency:

88% with full battery
Input: 4V*0,795A= 3,18W
Output 5,61V*0,5A=2,8W

79% with almost depleted battery
Input: 3V*1.04A= 3,12W
Output 5,58V*0,44A=2,45W

Output ripple
100% output → 176mV and 12% 68mV

Switching node first at 4V 2. at 3.3V

Thermal tests:

The driver is not connected to a host it constantly heats up constantly
with 500mA output
I made 3 measurements with some time in between and different input voltages

First picture was taken after 5 minutes on max. mode with 4V input voltage, 2. after 15 with 3.3V input voltage and last after 20 minutes

UVA+C has 2 different LEDs so it can’t use a standard Mode group in Bistro and Anduril

I thought of something like this for Bistro

MG 1: 10% UVA, 50% UVA, 100% UVA, 10% UVC, 50% UVC 100% UVC, 10%UVA+C, 50% A+C, 100% A+C

MG 2: 1% UVA, 25% UVA, 100% UVA, 1% UVC, 25% UVC, 100% UVC, 1%UVA+C, 25% A+C, 100% A+C

MG 3: 1% UVA, 25% UVA, 100% UVA, 10% UVC, 50% UVC, 100% UVC, 1%UVA+C, 25% A+C, 100% A+C

MG 4: 1% UVA, 25% UVA, 100% UVA, 10% UVC, 100% UVC, 1%A+10%C, 10%A+100%C,100% A+C

MG 5: 10% UVA, 100% UVA, 10% UVC, 100% UVC, 10% A+C, 100% A+C

MG 6: 50% UVA, 100% UVA, 50% UVC, 100% UVC, 50% A+C, 100% A+C

MG 7: 100% UVA, 100% UVC, 100% UVA+C

MG 8: 100% UVA, 100% UVC, 10% UVA+100% UVC

MG 9: 100% UVA, 100% UVC, 100% UVA+C

MG 10: 10% UVA, 100% UVA, 100% UVC, 10% UVA+100% UVC

MG 11: 1% UVC, 10% UVC, 100% UVC

MG 12: 10% UVC, 50% UVC, 100% UVC

MG 13: 1% UVC, 10% UVC, 50% UVC, 100% UVC

MG 14: 1% UVA, 10% UVA, 100% UVA

MG 15: 10% UVA, 50% UVA, 100% UVA

MG1 6: 1% UVA, 10% UVA, 50% UVA, 100% UVA

Muggle: 1% UVA, 10% UVA, 25%UVA, 10% UVC, 10%UVA+10%UVC

Suggestions for more mode groups are welcome

I am not sure if someone can program a strobe that is slowly switching between UV-A and UV-C like 2Hz or so

Anduril Ramping table I am a bit unsure like 2 options extend the ramping table length by 101 values to 250 total set ramp ceiling to 250,

so it needs longer for a whole cycle and its easier to hit the value you want:

1- 50: 1-100% UVA,

51-100: stay at max UVA,

101-150: 1-100% UVC

151-200: stay at max UVC

201-250: 1-100%UV-A+C

250 is then double click from OFF

251 turbo set as user want it to react on double click from ON

Also open for suggestions here how to implement without a major reprogramming of Toykeeper

OK now going into testing Modegroups of the UVC in Bistro, it works perfectly out the Boost chip is linear down to PWM level 3

so no adjustent has to be done like with AMCs UVC only will be then MG12 by default

Moonlight is now set 0 and 0 for both channels, I will make here your personal preset if wanted
// UVC:
#define RAMP_PWM2 0,3,25,64,102,128,255,0, 0, 0, 0, 0, 0,25,64,128,255,3,25,255
// UVA:
#define RAMP_PWM1 0,0, 0, 0, 0, 0, 0,3,25,64,102,128,255,25,64,128,255,3, 3, 25

PROGMEM const uint8_t modegroups[] = {
9, 12, 13, 3, 6, 7, 14, 16, 17, 0, // 1: 10% UVA, 50% UVA, 100% UVA, 10% UVC, 50% UVC 100% UVC, 10%UVA+C, 50% A+C, 100% A+C
8, 10, 13, 2, 4, 7, 18, 15, 17, 0, // 2: 1% UVA, 25% UVA, 100% UVA, 1% UVC, 25% UVC, 100% UVC, 1%UVA+C, 25% A+C, 100% A+C
8, 10, 13, 3, 6, 7, 18, 15, 17, 0, // 3: 1% UVA, 25% UVA, 100% UVA, 10% UVC, 50% UVC, 100% UVC, 1%UVA+C, 25% A+C, 100% A+C
8, 10, 13, 3, 7, 19, 20, 17, 0, // 4: 1% UVA, 25% UVA, 100% UVA, 10% UVC, 100% UVC, 1%A+10%C, 10%A+100%C, 100% A+C
9, 13, 3, 7, 14, 17, 0, // 5: 10% UVA, 100% UVA, 10% UVC, 100% UVC, 10% A+C, 100% A+C
12, 13, 6, 7, 16, 17, 0, // 6: 50% UVA, 100% UVA, 50% UVC, 100% UVC, 50% A+C, 100% A+C
13, 7, 17, 0, // 7: 100% UVA, 100% UVC, 100% UVA+C
13, 7, 20, 0, // 8: 100% UVA, 100% UVC, 10% UVA+100% UVC
13, 7, 17, 0, // 9: 100% UVA, 100% UVC, 100% UVA+C
9, 13, 7, 20, 0, // 10: 10% UVA, 100% UVA, 100% UVC, 10% UVA+100% UVC
2, 3, 7, 0, // 11: 1% UVC, 10% UVC, 100% UVC
2, 6, 7, 0, // 12: 10% UVC, 50% UVC, 100% UVC
2, 3, 6, 7, 0, // 13: 1% UVC, 10% UVC, 50% UVC, 100% UVC
8, 9, 13, 0, // 14: 1% UVA, 10% UVA, 100% UVA
9, 12, 13, 0, // 15: 10% UVA, 50% UVA, 100% UVA
8, 9, 12, 13, 0, // 16: 1% UVA, 10% UVA, 50% UVA, 100% UVA
2, 0, // 17:
2, 0, // 18:
2, 0, // 19:
2, 0, // 20:
2, 0, // 21:
2, 0, // 22:
2, 0, // 23:
2, 0, // 24:
2, 0, // 25:
2, 0, // 26:
8, 9, 10, 3, 14, 0 // muggle mode
};
#endif

I appreciate your work on this driver Lexel, even though I’m too occupied with other stuff to provide useful input on the user interface. Whatever the outcome, I will be happy with it (ordered 1 UVA+C driver)

This is your driver preset on mode group 4

Ramping table
1- 50: 1-100% UVA,
51-100: stay at max UVA,
101-150: 1-100% UVC
151-200: stay at max UVC
201-250: 1-100%UV-A+C
250 is then double click from OFF

Turbo is here 10% UVA and 100% UVC so that the mW should be close to each other not UVA over shining UVC fluorescence