FREE DIY open source BOOST driver!!! Tested & working!!

[benmwv]

How would you like free boost driver designs?! :drool:

Though I can't control anyone's actions, I would appreciate it if anyone wishing to sell this circuit would get permission from rhd or me first.

This is a project that I have been working on with RHD for a pretty long time and it is finally ready!

Here it is(R6 R9):
(Schematic)

(Board)

(Don't be fooled by those "design pictures" this thing is real! Rhd is going to post some nice pics on the first reply to this post)

Rhd has dubbed it "The Ben Boost" and I think its a pretty good name

So this is a boost driver based on the lm3410. Single li-ion or 2x/3x alkaline/nimh ONLY. It is meant to drive diodes with a Vf greater than the input voltage, like 445 and 405 diodes. It has been tested at a little under 1.2A on a 445nm diode and worked great, it may go a bit higher but I wouldn't expect anything over 1.3A. These drivers can also be paralleled to get higher currents like 1.8A. Rhd has a build using two of these drivers in parallel, see it in the next post!

The driver can boost to quite high voltages (28v IIRC) , so you could run two or three reds in series with this driver if you wanted. But remember, with a higher output voltage your maximum possible output current will decrease.

Current is calculated by: I (in Amps) = 0.19 / Rset (in Ohms)

Here is a table of values:
R1,R2,Current (ma)
1R0,xxx,190
R75,xxx,253
1R0,1R0,380
1R0,R75,442
R75,R75,507
1R0,R50,570
R75,R50,633
R50,R50,760
R33,R75,830
R30,R75,888
R33,R50,955
R50,R30,1013
R33,R33,1151
R33,R30,1210
R30,R30,1267

(Each of the resistors above you can get on digikey in 1/3W 0805 in 1 part quantity for $0.35 from the Susumu manufacturer)

This driver is 9.5mm x 13.8mm.

Now here's kicker: total part cost is about FIVE bucks! That means you could make two of these and parrallel them to 1.8A for about $10 (compared to $50-60 to do the same thing with flexdrives :eg

The release version of this driver has reverse polarity protection which isn't in the version in Rhd's pics. We added reverse polarity protection because the first three working prototypes of this driver died by shorting, mechanical damage, and reverse polarity so we fixed what we could. I assure you all of this was done on purpose as testing to see what extremes the driver can handle and they weren't mistakes at all :na:

BIG Update!

I feel confident the "Benboost Mini" is ready. This is a smaller, and much better version of the benboost. The LD+/- cap has been completely removed, now there a 2x 22uf caps from LD+ to GND. It uses the LLP 3x3 version of the lm3410 which has a thermal pad for heat transfer. There is a pad on the bottom of the board directly beneath the lm3410 where a heatsink can be soldered in. With just a small cube of copper on this pad and no other heatsink the driver must be run very hard to cause it to overheat. Since the heatsinking is so much better we can now utilize the full switch current of the driver without overheating. These drivers can handle 1.7A from 4.2v (full battery) in down to at least 4v, and most can go lower than that. For example, I have one with a max switch current of 2.6A (200ma below the chips average of 2.8A) and it can do 1.7A down to around 3.9V. An average 2.8A switch current chip should be able to go down to 3.8v or 3.7v. I feel confident saying all can provide 1.4A into a 445 diode over the entire discharge of a li-ion, many should be able to do 1.5a or 1.6A.

You guys should definitely try these out, they are WAY better than the original! Parts list below in the changelog, files at the bottom of this post.

More info, including oscope tests in the is post:
http://laserpointerforums.com/f67/f...iver-tested-working-71433-47.html#post1104257

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Feeling generous?
Do you like this driver? Has it helped you?
Make everyone jealous by getting your name on the supporters list!
I've got a couple new designs in my head that could use some funding



BenBoost Supporters: Me, RHD, Mohrenberg, Anthot, Gillza, Blord, Wee40811
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Warning: I am recommending everyone to use 22uf caps, and leave off C3. The cap was causing some oscillations in the output. It will be taken out on future design updates. With these changes it is safe to use this on sensitive diodes such as 12x.

Here is the parts list:
IC1:
LM3410 - $1.58: LM3410XMFE/NOPB | LM3410 Series 5.5 V 1.6 MHz Constant Current Boost and SEPIC LED Driver SOT-23-5 | NATIONAL SEMICONDUCTOR - Future Electronics
OR
LM3410 - $3.20: LM3410XMF/NOPB National Semiconductor | LM3410XMFCT-ND | DigiKey
Q1:
P-channel mosfet - $0.79: Digi-Key - SI3473CDV-T1-GE3CT-ND (Manufacturer - SI3473CDV-T1-GE3)
R3:
100k 0603 - $0.05: Digi-Key - P100KHCT-ND (Manufacturer - ERJ-3EKF1003V)
R1, R2:
Set Resistors (0805) - VARIES
L1:
Inductor - $0.66: Digi-Key - 587-2001-1-ND (Manufacturer - NR8040T100M)
C1, C2:
2x 22uF 0805 Caps - $0.77: C2012X5R1C226K TDK Corporation | 445-6797-1-ND | DigiKey
D1:
Shottky - $0.46: PMEG2020EH,115 NXP Semiconductors | 568-4118-1-ND | DigiKey

PCB - $0.31 : OSH Park ~ Welcome


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Changelog
R6 R9: Original posted design. Has reverse polarity protection, 0805 caps, No OVP.

R6 R9 Round v6: Originally posted design (0805 caps, no OVP) on a 16.5mm board to go right into most flashlight pills, with battery contacts on the bottom.

R6 R10: Replaced 0805 caps with 1206 caps (not as tall).
C1, C2:
2x 22uf 1206 Caps - $0.60: EMK316BJ226ML-T Taiyo Yuden | 587-1433-1-ND | DigiKey

R6 R11 OVP2: Added OVP (over-voltage protection) driver can now be powered up with no load connected. (still has 1206 caps)
D2:
12v Zener - $0.27: Digi-Key - 568-6279-1-ND (Manufacturer - BZX384-C12,115)
R4:
100 ohm 0603 - $0.02: Digi-Key - P100GCT-ND (Manufacturer - ERJ-3GEYJ101V)

Benboost Mini v5:
IC1:
LM3410 - $3.20: LM3410XSD/NOPB National Semiconductor | LM3410XSDCT-ND | DigiKey
Q1:
P-channel mosfet - $0.76: SI3493BDV-T1-E3 Vishay Siliconix | SI3493BDV-T1-E3CT-ND | DigiKey
R3:
100k 0603 - $0.10: ERJ-3EKF1003V Panasonic Electronic Components | P100KHCT-ND | DigiKey
R1, R2:
Set Resistors (0805) - VARIES
L1:
Inductor - $0.31: NR5040T4R7N Taiyo Yuden | 587-2373-1-ND | DigiKey
C1, C2, C3:
3x 22uF 1206 Caps - $0.60: EMK316BJ226ML-T Taiyo Yuden | 587-1433-1-ND | DigiKey
D1:
Shottky - $0.45: PMEG3030EP,115 NXP Semiconductors | 568-6761-1-ND | DigiKey
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

EAGLE board files are attached!!!!!!! [For EAGLE 6]

 

[rhd]

---- ON MY AGENDA ----
A few things have popped up via conversations with others, and feedback. I'm going to keep a little list of changes that I'd like to introduce. Small stuff:
------
1) Move C2 down so that it's not as close to the board edge. Otherwise, when placing the driver inside an Aixiz module, it risks contact with the module if a tall capacitor is used.
2) Trim about 0.5 mm off of the board width, to make placement inside an Aixiz module easier.
3) Choose a slightly smaller inductor pad. We're about 0.5mm larger on the pad spacing than we actually need for the inductors we're using.
4) Replace 0805 capacitors with 1206. This will allow for lower profile (less tall) capacitors, which will make it easier to heatsink the IC without fear of shorting the caps.
5)
6)
------

Alrighty, so the version of the PCB I have on hand is a few versions back (before we added the MOSFET reverse polarity protection, that missing feature cost me a 445!). So what you see here is a version or two old. Orders from Dorkbot take about a month to hit me here in Canada, so I'm still waiting on the newest version of the physical PCB to arrive.

So the idea popped into my head to try paralleling the driver. The notion was that if it worked well, the circuit could be duplicated twice over, with balancing resistors added, on the same PCB. Step one was to try paralleling the single-driver PCBs together in a build. That's when the following SAIK build was born, using two boost drivers set to roughly 850mA each. You'll notice that the inductors on the two paralleled drivers are different than in the first photo. That's due to the fact that I've been testing a few different inductors, and a few of them work just fine. So I've continued to use the other inductors that have worked along the way, simply because they're here, on boards, and I don't want them to go to waste. But it's definitely the NR8040T100M that I'm liking most at the moment. Plus, those inductors in the paralleled build shot actually arrived damaged from Digikey, twice! They're too fragile.

Turns out - no balancing resistors were even necessary. The two drivers parallel just fine, and at 850mA each, there is great stability. Running from a single 22500, and using a G1 lens, I'm getting a peak of just above 2.3W from an H-series projector diode.

The best part, is that the version of the driver I used cost about $3.70 per unit (the MOSFET hadn't been added yet). Still, at under $5 per driver, you can have dual boost drive setup that parallels (pun intended) a flex drive setup that would cost $50+, for about $9. Even better - the prices Ben and I have quoted are not 1,000 QTY figures. They're single driver quantities! You could build ONE DRIVER (ignoring shipping costs) for the under-$5 we've teased in this thread

And, I know renders are silly since we have the real thing here, but, well, I love anti-aliased 3D graphics

 

[ARG]

Wow! This is incredible, at 5$ a piece to! Thanks for the amazing contribution to the community!

Edit: Not surprised, must spread more rep before giving to you again RHD

 

[LaZeRz]

You guys seriously are epic. I shall place an order of parts in due time

+1 to both of you's (if the system lets me)

I have been looking for a cheap boost alternative for a Saik build and here it is!

 

[Ghostchrome]

Nice job fellas. Can't wait to see more. You guys gonna start cranking these out?

 

[MarioMaster]

Glad to see others working on this project, the community has been in need of cheaper boost drivers for ages and I've had my hands tied up with LPM workings at Radiant.

 

[Tech_Junkie]

Very well done. Thank you to all involved.

I love when the hobby gets even more affordable.

 

[LaZeRz]

I'm going to be sending a PM to the guys over at DorkbotPDX to ask for a shipping quote to Aus.

This hobby is truly evolving. A boost driver that costs less than your traditional linear

 

[lasersbee]

How would you like free boost driver designs?! :drool:
A little disclaimer: You are free to make and use these drivers as you please, but you do not have the right to SELL these drivers without my or Rhd's permission.

This is a project that I have been working on with RHD for a pretty long time and it is finally ready!

Here it is:
(Don't be fooled by those "design pictures" this thing is real! Rhd is going to post some nice pics on the first reply to this post)

Rhd has dubbed it "The Ben Boost" and I think its a pretty good name

So this is a boost driver based on the lm3410. Single li-ion or 2x/3x alkaline/nimh ONLY. It is meant to drive diodes with a Vf greater than the input voltage, like 445 and 405 diodes. It has been tested at a little under 1.2A on a 445nm diode and worked great, it may go a bit higher but I wouldn't expect anything over 1.3A. These drivers can also be paralleled to get higher currents like 1.8A. Rhd currently has a build using two of these drivers in parallel and is going to make a thread about it, I'll pop in the link when he does.

The driver can boost to quite high voltages (28v IIRC) , so you could run two or three reds in series with this driver if you wanted. But remember, with a higher output voltage your maximum possible output current will decrease.

Current is calculated by: I (in Amps) = 0.19 / Rset (in Ohms)

This driver is 9.5mm x 13.8mm.

Now here's kicker: total part cost is about FIVE bucks! That means you could make two of these and parrallel them to 1.8A for about $10 (compared to $50-60 to do the same thing with flexdrives :eg

The release version of this driver has reverse polarity protection which isn't in the version in Rhd's pics. We added reverse polarity protection because the first three working prototypes of this driver died by shorting, mechanical damage, and reverse polarity so we fixed what we could. I assure you all of this was done on purpose as testing to see what extremes the driver can handle and they weren't mistakes at all :na:

Here is the parts list:
IC1:
LM3410 - $1.58: LM3410XMFE/NOPB | LM3410 Series 5.5 V 1.6 MHz Constant Current Boost and SEPIC LED Driver SOT-23-5 | NATIONAL SEMICONDUCTOR - Future Electronics
Q1:
P-channel mosfet - $0.79: Digi-Key - SI3473CDV-T1-GE3CT-ND (Manufacturer - SI3473CDV-T1-GE3)
R3:
100k 0603 - $0.05: Digi-Key - P100KHCT-ND (Manufacturer - ERJ-3EKF1003V)
R1, R2:
Set Resistors (0805) - VARIES
L1:
Inductor - $0.66: Digi-Key - 587-2001-1-ND (Manufacturer - NR8040T100M)
C1, C2, C3:
3x 10uF Caps - $0.54: Digi-Key - 490-3347-1-ND (Manufacturer - GRM21BF51C106ZE15L)
D1:
Shottky - $0.47: Digi-Key - 568-4105-1-ND (Manufacturer - PMEG1020EH,115)

PCB - $0.31 : PCB Order | DorkbotPDX

TOTAL - $4.40 + SET Resisitors

EAGLE board files are attached!!!!!!!!!!!!!!!!!!!!!

Very Nice.... :gj:

The title says FREE DIY open source BOOST driver....
Did I miss it or did you not give a link to the schematics...

I would like to see the schematics of this open source
boost Driver...


Jerry

You can contact us at any time on our Website: J.BAUER Electronics

 

[Blord]

Great job !

Can someone buy 100 or 200 pieces of each and resale them as soldering kits ? That would make the driver much cheaper than everybody individually order the components.

 

[lasersbee]

Great job !

Can someone buy 100 or 200 pieces of each and resale them as soldering kits ? That would make the driver much cheaper than everybody individually order the components.

Since it was your idea.... we'll let you do it...:evil:


Jerry

You can contact us at any time on our Website: J.BAUER Electronics

 

[DTR]

Awesome contribution guys.:gj:

Do they need to be heatsinked @ 1A?

 

[rhd]

Blord: I think that's a cool idea. Better than taxing Dorkbot with a bunch of individual PCB orders.

Jerry: We didn't maintain an SCH file as we went, but I'm sure we can just put one together in the next little bit.

DTR: I doubt it will need heat sinking at 1A to 1.2A. I'll do a few builds today and find out.

 

[benmwv]

Thanks guys!

We don't plan on selling them, but we may find someone else to.

Very Nice.... :gj:

The title says FREE DIY open source BOOST driver....
Did I miss it or did you not give a link to the schematics...

I would like to see the schematics of this open source
boost Driver...


Jerry

Sorry, this board design has been changed and redone so many times there isn't a schematic file in sync with it.

I'll whip up a schematic later today :beer:

Awesome contribution guys.:gj:

Do they need to be heatsinked @ 1A?

I tested them at 1.16A and it needed heatsinking, so I would say they probably do.

I put 3 vias directy under the IC connecting the ground planes to help the heat dissipation, but since this IC doesn't have an exposed pad for heatsinking there isnt much you can do. If you put a tiny dab of thermal paste under the Ic before you solder it on it may help, but I don't know what current you can run them at without heatsinking.

Also worth noting that if you heatsink one of these drivers you should heatsink the Lm3410 AND the diode.

 

[tsteele93]

Mad props to both of you on this! Really a great contribution!

Great job !

Can someone buy 100 or 200 pieces of each and resale them as soldering kits ? That would make the driver much cheaper than everybody individually order the components.

You would need to get their permission first according to the first post...

 

[benmwv]

Jerry -- I added the schematic to the first post.