Help with Driver for red laser

Well, I'm intending to build a red laser pointer, and I need some help, most of all with the driver.

My LD will most likely be from a 22x DVDRW drive,

And I'm not sure what host to use... I've found three that I kinda like: sku.3405, sku.1442, sku.46
What are your recommendations? (To me 3405 looks very sturdy, but 46 is alot cheaper)
I'm intending to run the laser on common batteries like AA and AAAs, so if there are other suggestions for host, do bear that in mind.

For the driver, I'm assuming I'm taking one of those hosts above, and will be running on 3 AAA batteries. I want to use rechargable batteries, so the output available would be 3.6V for me to play around with.
The thing is, what is the voltage requirement for a red LD? Anyone can cite a typical range?

If it's 2.9V like I heard around here and there, then DDL's driver would not be able to work since that takes around at least 2V off. If that's the case, then would a simple driver like here http://www.allaboutcircuits.com/vol_6/chpt_5/15.html
be good enough?
Or, as another possibility, would an LM317 used as a voltage regulator hooked up to the base of a MOSFET or BJT be a good current regulator?

[edit]Sorry to have to revive a thread a bit on the old side, but I was thinking it'd help others who are building drivers. The content has not been edited so that anybody can follow what I went through to hopefully understand everything here. This driver is a semi-low voltage driver and would work rather well for blu-rays. Currently (refer to my last post) it outputs 166ma at max with a power source at 4.5v. Higher voltage will of course give better current.[/edit]

For your application i think the best driver to use would be a MicroFlex Driver from Dr Lava or the Groove driver that he sells also. It allows your battery to be used for longer period because of its low voltage dropout. But than again for a red diode,you dont need much to power them. A Rkcstr driver will also work.

Only if your building a BR will the volatge be higher and you will need to use special battery if your using a Rkcstr driver. But for the driver from DrLava you dont need them.



Tommy

Hmm... so how many volts does a typical red LD take?
I probably wanna try build my own driver too, hopefully.

I'll be taking this in stages... first trying to build a reasonable driver, then get the LD and try to take it out before finally getting the host to put everything together...

I am building a 300mw red. I was told that with a rckstr driver that you have to have a higher voltage than 3 normal AAA's which only gets 4.5 volts or right around there. Best choice if wanting to stay that size is use these batteries 10440. Use two of them and a dummy battery to get the required voltage to run the rckstr.
Other drivers do take lower voltages also.

:Edited to add.
Look up the schematics for the DDL Driver. Pretty easy to make.

Hmm, so the rckstr driver is out.... but so is the DDL, cos that one require's at least 2V overhead if not 3V.

But my main concern now is: What is a typical voltage for a red LD? If it's rather little, then there won't be much of a problem with the driver selection.


Oh and of course, anyone to recommend which of the hosts, sku.3405, sku.1442, sku.46, is best?

Ok, erm so, referencing from Viracocha's design (http://www.laserpointerforums.com/forums/YaBB.pl?num=1222777857) , I decided to come up with a small driver for my red LD too....

I'm using an LM317 as a voltage regulator to control an N-channel MOSFET. LM317 alone as a current regulator has too high a dropout voltage as compared to a transistor. Apparently BJT (normal transistors) have higher dropout voltage compared to MOSFETs, so I'm using one here.

R1 : Around 10k, 1/8W, for C1 to bleed (~2s for complete discharge), while not being too small (only takes 0.36mA while running)
R2: 0-2k. This is what adjusts the LM317's voltage output.
R3: 1k, 1/8W Minimise program current (if 1k's too much, might use 240 1/8W)

C1: 47uF? Main cap for stablising current.
C2: 1-10uF, reduce ripple, if any
C3: 1uF, reduce ripple, if any

D1: 1N5401, handles up to 3A, which will be gd in protecting in event of reverse polarity.
D2: 1N4001, handles 1A, still good. Not really necessary, but protecting the LD is a better safe than sorry thing.

Q1: Not sure about part no. I heard NTE2980's good...

[edit]R3 is too much at 1k and would make the regulating inaccurate. A 240 ohm resistor is better. For Q1, an NTE2980 might not be enough, you need a really sensitve one. I tried the IRLZ44 which turned out to be pretty ok. If you can find a more sensitive one than that, that'll be gd. To know how sensitive a transistor is, check out it's characteristics graph in the datasheet.

If turn on delay is needed, make C2 bigger. E.g. 47uF[/edit]

So yea, anyone can tell me if this is worth a try?
And, is a potential divider needed for the MOSFET? The minimum voltage that the LM317 is capable of providing is 1.25V, and I dunno if that is too much for the transistor. Anyone can clarify?

Judging from the specs on the NTE2980 and my limited knowledge, the output from the LM317 is about right.
http://www.alldatasheet.com/datasheet-pdf/pdf/10186/NTE/NTE2980.html says it has a gate threshold voltage min of 1v and max of 2v (max of 10v before damage) so if your adjuster can vary the range between those ouputs it should be ok. If not then you could add a simple 1.25:1 voltage divider which should bring the 1.25v output to 1v and reduce the upper level of the output so the pot gives a greater range.

Also, the LDs i've read about (from StoneTec) give voltage ranges from 1.9v-2.3v (IR) and 2.5v-3v (red) so with a normal transistor, the voltage drop might even be benificial.

Alright! Thanks!
I will try to build the driver and find out if it's gd.

Test load is the typical 1 ohm resistor plus 4 diodes right?

based on http://stonetek.org/shop/index.php?main_page=product_info&cPath=71&products_id=223

red = 3 diodes and blue = 6

I went downtown to get the parts and stuff. They didn't have NTE2981s (these would be better than the 2980s), or the NTE transistors at all actually, so I got myself the IRF510 (couldn't think of anything else at that time)

Boy are MOSFETs expensive.... the IRF510 cost be $3 each here (about US$2) and the NTEs are apparently more expensive. But they didn't carry it so I wouldn't know how much more it'd have cost. I really hope I don't fry them during soldering when I get them.

Well, the voltage regulator part of the circuit's great. Works exactly as planned. The LM317 takes away about 2V, although the max voltage I achieved with it was 2.8V, for a 3.6V battery source, but then I was compromising on the voltage regulation property, as when I added in an extra battery, the voltage changed.

So now I'm checking up on other n-channel low turn on voltage MOSFETs. I'll post more results here.

Shops have large overheads so will always charge more for items, especially small relativly cheap stuff like components.

Buying online will get you exactly what you want, will cost less and the item is delivered to your door so you save on travel costs!

It's the future! ;D

tw15t3r,

The IRF510 was the first mosfet I used (for the same reason...). I also ran into the same problem. The logic level mosfets solved it...

Just look for IRL instead of IRF on your part numbers...

cheers,
kernelpanic

p.s. - I've got some logic-level mosfets (N-channel) surface mount), I probably won't use them all. I don't know how much shipping to you would be but if you cover it, I'll send you a couple.

Well the main reason I don't buy stuff online is because I live in Singapore, and most online stores are all overseas, so it's not a good idea to only buy one or two parts and pay so much for shipping as what most online stores charge...

Anyway, yea Kernel, you're right! I went to get an IRLZ34, and it works well, but I think it's not good enough. My load currently is 3 1N4001 diodes and a 1ohm resistor (that seems to be slightly more than 1 ohm) but anyway, my max current is only 120mA (with 3x 1.5V batteries), which is kinda lousy since I'm hoping to use rechargables which only give 1.2V each, giving about 3.7V in total.

Currently I'm doing some checks to see if the transistor is already saturated at max current setting. If it is, then I would probably have to get an even "better" transistor. Based onthe graph of the output characteristics given in the datasheet for each transistor model, I think the IRLZ44 may do the trick. If not, I'm totally lost.

And if the transistor isn't saturated at max voltage I'm giving, then that means that the voltage regulator is taking too much of a voltage away from the gate, and that'll mean I'd need to change the LM317 to some other kind of regulator that either can provide a variable voltage with less dropout, or use a preset voltage with a voltage divider, then feed it to the gate.

Currently, that is what is going through my mind.

Already spent S$15 (about US$10) on transistors already (I kept buying them in pairs in case I blew one)
I hope I don't have to waste too much on this driver.

Oh, and also, the 1k ohm resistor for R3 is too much. It makes the voltage regulating irregular. I tested it by adding one battery and taking it away while measuring the voltage between the output and ground. 240 ohms is better, so I'm sticking to convention. That'll mean of course that R2 will have to change from a 2k pot to a 500 ohm pot.

Well, so that's about it.

tw15t3r,

You may just be running into the dropout of the LM317. At the low current used to drive the mosfet, there is probably 1.5-2.0V dropout. Meaning that with a 4.5V supply you can give 2.5-3.0V max to the gate of the mosfet. This may not be enough to open the gate all the way... (Even if it worked for 4.5V, it definitely won't work for 3.6V as the regulator dropout will likely put the gate voltage below the threshold needed to open ).

This however doesn't rule out using a voltage divider or Zener reference instead of the LM317. I actually used the mosfet so that I could eliminate the LM317 precisely because of the dropout. But I do like the idea of using the regulator to give a voltage reference. I'm actually thinking of replacing the voltage divider in my driver circuit with the LM317 for the 'Labby' hi-power version of the driver.

cheers,
kernelpanic

I did some tests: There's almost 1V across the transistor at the maximum voltage the regulator can output, which means that the transistor is not saturated yet. However, I don't think I can get much better a regulator, so I will try another transistor. (wasted $15 on transistors in that case... US eq of 10)

This transistor I'm looking at is the IRLZ44 instead of the IRLZ34 which I just tried. According to the characteristics in the data sheet, it looks like a very small voltage is required to give a large current. The graph shows that only 2.25V is required at the gate to give a current of 2A (YES! 2A! Not 1, but 2A!). The current voltage regulator I have can give 2.8V max, which is more than required, so it's looking good. Hopefully this will be the last transistor that I need to try to get a good low dropout driver circuit. And..... hopefully the shops carry this type of transistors too.

[edit]Oops! Didn't realise you posted while I was posting. The LM317 as a voltage regulator is kinda useful, because it doesn't consume much current, and apparently, according to my measurements, it's dropout in voltage regulation is on the low side of just under 2V. And, I've found what looks like a very low threshold MOSFET that looks promising! Wish me luck![/edit]

Indeed, Good Luck!

kernelpanic