Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

LPF Donation via Stripe | LPF Donation - Other Methods

Links below open in new window

ArcticMyst Security by Avery

Super Basic Adjustable driver

Joined
Nov 2, 2008
Messages
51
Points
0
I've been mucking around with LED's for quite a while and using LM317T regulators, recently i have been playing around with PCB etching and thought i'd try my hand at making a small driver that is stable, or at worst, will drop off voltage as the batteries die. The flexdrives are great, but the 'feature' of increasing the output as the batteries die just makes me think of blowing diodes if batteries aren't kept in top charged form or regularly replaced.

SO,

This is not a tutorial or such, just a little post to show what can be done with some simple parts and a little bit of time. I'm sure that some of the real electronics gurus will be able to justify/explain the need for all the filtering, polity protection circuits and stuff that the other great drivers have but for me, i just wanted something with no frills to fit inside a pocket pal sized host.

The basic board is just 1oz copper single sided, its a fairly thick FR4 construction so i'm hoping to find some thinner PCB at some stage to reduce the height.

I used ' ExpressPCB ' to design the board and custom made the pads for the SMD trimmer (not pictured, still on order). This is a free program that can be used to order boards directly, or as i have used it, to design and print. Its main limitation is no gerber output, but it works fine for me and is very easy to use (also can link to schematics if required for larger projects).

The pad for the LM317MDT voltage reg (in constant current mode) which will supply up to 500ma, was a standard package in the program already so i didn't need to create it specially.

I have it layed out so i can use 2x 0805 resistors (as pictured), or a single higher rated 1206 package SMD resistor. Using the jump wire (as pictured) the driver is set at a fixed value, in this case 2x 24ohms in parallel = 12 ohms = ~103ma (measured) draw. With an SMD pot in place where the white wire is the adjustment is quite good and you use the resistors to set the "max" allowed current so there can be no "mistakes" with turning a pot too far and having it jump in power and kill a diode.

As mentioned in the subject, this is a very simple, very basic driver whose design has served me well so far during testing and (although less sensitive) i've used similar design on LEDS for a long time with no issues.

It does not have input polarity protection, it does not have output polarity protection, or a cap (which i'm hoping to add without going double sided, or maybe just directly onto the diode, not sure yet). Some will notice there is actually no ground connection on this board at all. The soldering was done with solder paste and a heat gun for the SMD parts.

Like any linear regulator, it requires more volts in than a boost driver like the flexdrive and when the input voltage is well above the required minimum, the driver sheds it as heat, which can be an issue with such little copper available, but so far its OK for short periods up to around 5 minutes with 12V input it doesn't get crazy hot which is longer than i can imagine having a laser on.


Hope this gives people some ides for their own projects that keep things simple and adds that little extra edge of being customised and DIY.
 

Attachments

  • driver-v3-fixed.jpg
    driver-v3-fixed.jpg
    41.2 KB · Views: 166





Joined
Nov 2, 2008
Messages
51
Points
0
The schematic is basically the 'DDL driver' without the output cap and diode, details of that are here: http://www.laserpointerforums.com/forums/YaBB.pl?num=1185701612 . If you want the board layout file for ExpressPCB, i can post that too, just let me know.

I have some SMD 47uf tantalum caps but i doubt i'll be able to fit one on the board and keep it square and fit inside the heatsinks i made for the host. My first version was round, but impossible to cut out neatly, so i went back to square.

The regulator i'm using is the same as the DDL, just in a surface mount package and rated at 0.5A instead of 1.5A. Due to being on such a small copper pad though, i think that higher drive currents (haven't tested it with my open can yet) would cause it to heat up quite quickly with really no where for the heat to go so i'll have to play around with it a bit more.

I was planning to use a 50ohm pot instead of 100ohm for better resolution as i would rarely need to set a drive current below 30ma or so, but if you need to get down below 30ma, then you'd certainly need a 100ohm pot (or a large increase in the fixed resistor value which would heavily limit the max current).
 
Joined
Jan 20, 2008
Messages
1,724
Points
0
Nice work!

You'll need to provide more heatsinking eventually... maybe try a double sided board and add some througholes... You'll also want to add some caps, I've been told the output gets really unstable without them... Do you have a scope?
 
Joined
Nov 2, 2008
Messages
51
Points
0
Thanks. :)

Yeh i'm thinking i'll have to go double sided, but lining up the layers (for anything other than extra copper for dissipation) would be a nightmare for DIY'ing. I'll look into it though. It was actually running cooler than i thought at 100ma so that is a bonus. Driving an open can at 300ma will be a big test though. I'm Just waiting for some 10440's now for my host.

I don't have a scope so can't check the output to see if there are nasty spikes during power cycles etc, that would be the only 'downfall' as such. even if i can't fit a cap on the board (easily) i'm sure i can get a package small enough to solder across the terminals on the diode.
 
Joined
Jan 20, 2008
Messages
1,724
Points
0
Another thing to note is those 0805 resistors are what, 0.1W?

When you try to use that to drive higher powered red diodes you'll find yourself needing six or even 8 of them to handle the kind of currents you'll be running into...
 
Joined
Nov 24, 2007
Messages
65
Points
0
CoZZm0_AU said:
Thanks.  :)

Yeh i'm thinking i'll have to go double sided, but lining up the layers (for anything other than extra copper for dissipation) would be a nightmare for DIY'ing. I'll look into it though. It was actually running cooler than i thought at 100ma so that is a bonus. Driving an open can at 300ma will be a big test though. I'm Just waiting for some 10440's now for my host.

I don't have a scope so can't check the output to see if there are nasty spikes during power cycles etc, that would be the only 'downfall' as such. even if i can't fit a cap on the board (easily) i'm sure i can get a package small enough to solder across the terminals on the diode.

Without caps the voltage regulators can oscillate and get hot because of that. You would only be able to tell if that is happening with a scope though.
 

Benm

0
Joined
Aug 16, 2007
Messages
7,896
Points
113
The requirement for capactirors on voltage regulators like the LM317 remains a bit of a debate. The datasheets state they are required, but in practical application things often (but not always) work properly without.

These capacitors don't have to be big though - 10 or 100 nF capacitors of adequate voltage come in 0603 form factor, so even adding one to both in and output wouldnt increase circuit size that much... they are as large as the ones on the little circuit board pictured on the left.
 




Top