My LM3410x based Boost converters: Spark driver! PART 1

[billg519]

"especially when you do not use testing load like me :banghead:"

I think that you have enough 1N4007 diodes and probably a 1 ohm resistor, so make that test load and use it. Save the LD's for proven working drivers.

 

[ReNNo]

It's my fault.
Last time when I and Eudaimonium exchanged packages he sent me a lot of 1N4007 diodes as I asked him.

So he is probably out of silicon diodes.

 

[lasersbee]

"especially when you do not use testing load like me :banghead:"

I think that you have enough 1N4007 diodes and probably a 1 ohm resistor, so make that test load and use it. Save the LD's for proven working drivers.

Good Idea (no... the best idea)... use 6 1n400X dieds with a 1 Ohn
resistor as a test load....
If you need some 1 Ohm resistors see here.....

http://laserpointerforums.com/f64/free-1-ohm-test-load-resistors-42409.html


Jerry

 

[Eudaimonium]

Guys, I know how to build a test load!

And I do have a lot of 1n400x rectifiers, ReNNo, nothing here is your fault OK?

Bill, I do have them and will test it.
The reason I used 3.5 Vdrop load is because simply I had it assembled at the time and had some trouble finding some more rectifiers (which were in a paper bag somewhere else, instead of Bills packages...),
And my impatience got the best of me.

Next time I won't repeat the same mistake guys, I cannot afford to.
When you see the part2 , you know that the driver is successfuly prototyped and fully operational.

Thanks for all your support here!

 

[EzPk0970]

Sounds awesome Eudaimonium ! Looks like a great new driver. Hopefully the price for each when it's all said and done will be much more affordable then others out there.

 

[Eudaimonium]

Sounds awesome Eudaimonium ! Looks like a great new driver. Hopefully the price for each when it's all said and done will be much more affordable then others out there.

Well thoughts of selling them did cross my mind yeah, however, there are couple of things to consider:
1st: it is simply homemade, as opposed flex and other which are made in factories (correct me if I'm wrong)
2nd: I am still not quite, what'dya call, famous here, guys wanting a boost converer will choose already tried and commonly used drivers, then something new from unknown member.
3rd: I am in a dorm, 85 km from my home, I can only get something done through the weeekend. So it would be a very slow manufacture rate (I think, considering the 'demand' it is more than enough :crackup

If I would sell them, they would aprox. come off a 10 or 15 $ a piece maybe... but that is still a long-shot guess.

Also there is a mayor setback: the diode must in any possible combination be insulated from the case, as there are no common points with diode and source.

I just hope to get the working and use the in my builds since they are simple to build, yet effective little drivers. Can power anything from red to bluray (hopefully, with good coil).
Also, there will be no potentiometer in my design what so ever. Only a fixed resistance that determines the current (2.0 ohm for 100 mA , PHR, or 0.47 Ohm for ~400 mA for red).

 

[Flaminpyro]

Good work 1+ for you buddy !

 

[Eudaimonium]

Good work 1+ for you buddy !

:beer:
Thank you FlaminPryro
BTW since you are already here...
I searched some lm3410s on this forum by Jerryies request.
I found this dude , 'wannaburn' (with that facepalm avatar), who is making and selling lm3410 based boost converters, and apparently there was some sort of fuss because he is, kind of, usign your name on them 'Pyro' ?
Could also apply to Styropyro I guess.
I also noticed that there is a couple of guys here who do not like this guy...
Care to shine some light on the subject for me?

He was also selling those for ... 40$ or something :crackup:

 

[lasersbee]

You will learn in time that if you want to sell something (anything) that
there are many unforeseen things to consider...

1) the actual price of parts including Duty and Taxes
2) the Printed Circuit Board costs including Artwork and design costs
3) the equipment needed to assemble and test your product
4) your Time and Labor
5) yotal Shipping costs to you and from you
6) your Profit... (no-one works for free)

Only when you take all these basic things in consideration can
you predict a price for a product... IMO

IIRC... the name on those PCBs you found was Pyrotronic...
I think that is also his name on eBay...

Jerry

 

[Eudaimonium]

Laserbee, don't worry I can set a price right, it is not the first time I would sell something.

I do not value my time and effort, only price of parts and shipping, plus small profit for me.
Profit takes care of all my effort and time.
I do not have any kind of equipement except my DMM and my pair of eyes.

The coils used on this thread were 2$ a piece. I do not know the price of LM3410. All other misc components were practically obtained for free. Either from Bills packages or from DVD writer PCBs.

So price could get low pretty good.

But we have the biggest two problems:
Competition (DrLava),
and I am, already mentioned, UNABLE to recieve any kind of paypal funds.

If I wish to sell something online, there are two choices. Someone kan either do a bank tranfer... pretty expensive.

Or... I could ask, Bill for example, to make a paypal account for me. No funds on it what so ever. Now, I need budget in cash to make this happen: I would use my bufget to build stuff and sell it. Payment would be to that paypal account. Until I have enough money on that paypal, I would use my budget.
When enough money is accumulated in that account, I can use that to order parts, with continued income. Now, my budget will never be returned in cash anyhow. I would not earn money, but I could sell stuff online without real cash payments.

From my profits I can order myself a laser here and there. So , no cash would be going through my hands... only my laser collection would get bigger.

However, I SERIOUSLY doubt that I can actually make business like that, there would be no interest in homemade drivers, and I definetly cannot beat the O-like 405nm diode driver, 8$ plus shipping, about 10$ total.

I do not think any business in selling something in volume will take place, not in this year anyway.
I am fully aware of things, but one first has to make a name for himself before someone takes him seriously.

Like you say it Jerry:
Just my two cents.

 

[Asherz]

Nice driver, looking good

You do have something over o-like, the fact that shipping from you takes a lot less than shipping from china. Although the issue you have is quality control, things go wrong when everything is all hand done, and the cost of professional PCB creation etc gets expensive.

Talk to lava or rckstr and say they can use your design if you get a small cut of the profit

 

[Eudaimonium]

Nice driver, looking good

You do have something over o-like, the fact that shipping from you takes a lot less than shipping from china. Although the issue you have is quality control, things go wrong when everything is all hand done, and the cost of professional PCB creation etc gets expensive.

Talk to lava or rckstr and say they can use your design if you get a small cut of the profit

What profit?
We are taking too-long-shot fantasies here.

I am yet to finish the very first prototype!
It is very unlikely for me to make these in a volume. I can trade them to people who want them, for a diode or two or something like that.

Please hold your 'business ideas and observations' for the moment I say: It works! [video of PHR working off one 18650 here]

I can get professional PCBs made in my school no problem. However for the extra cost you do not get any extra functionality, it works exactly the same. And you will not even see the visual difference because it will be in a pretty host powering a pretty diode. You know what I mean?

 

[lasersbee]

Talk to lava or rckstr and say they can use your design if you get a small cut of the profit

If I'm not mistaken... the circuit design is the same one that is on the
Data Sheet for the LM3410... (unless Eudaimonium changed the circuit)
and belongs to the App Note design engineer...

@Eudaimonium...
you can get some prices on the LM3410 at DigiKey....
just enter the part number...

DigiKey Corp. | Electronic Components Distributor | United States Home Page


Jerry

 

[Eudaimonium]

Well my schematic is the same one as the first schematic in datasheet yes, although I did change the orientation, and components OFC (if LD is placed instead of LEDs, does that count? guess not) to the ones I had.

But I already said, I can have the professional lookin' board made in my school, but for the price difference it does not pay off in any way. The better looks will only be noticable for about 4 seconds, 'till you solder on your diode and close it up in a host.

So what is the point?

Anyhow... please guys, let's leave this profesional business stuff conversation for later.

I contacted IgorT...
He pointed out one mistake I made...
I need much higher capacitance at input and output, and ALSO need caps across LD, not only the one across LD + schottky.

It is very likely that I have enourmus ripple which killed my PHR.
He explained that I need at least 10 uF at input, 10 uF at output and at least double the output cap to the one across the diode.
I only have small 4.7 uF caps at the moment. It appears that I will have to stack two in parralel with another two on top of them to get near to 20 uF. Three dimensions FTW. Stacked capacitors will be used if I fail to find some more larger caps, saving some space.

I have just placed another free sample order with coicraft. I have picked two kinds of coils, two pieces a kind. They are about 4x4x2 milimeter in dimensions, with rated current around 1.5 A, can't remember correctly now, I know I picked those because they suited me best.
But, they are 4.7 uH inductance, not 10 uH.
IgorT explained he used 4.7 uH and around that with LM3410Y (520 and something KHz), while I have 1.6 MHz .

So they should be good.
Looking up some LM3410X on digikey now, thanks Jerry again

 

[lasersbee]

I'm having problems believing that the ripple killed your Laser Diode..
because the ripple is only on the max output line (looking at a scope)
and is normally only a few millivolts.

A Laser Diode is normally rated at twice the current/output in Pulsed mode...
and that is at a high frequency with a (usually) 50% duty cycle...


Jerry

 

[Eudaimonium]

I'm having problems believing that the ripple killed your Laser Diode..
because the ripple is only on the max output line (looking at a scope)
and is normally only a few millivolts.

A Laser Diode is normally rated at twice the current/output in Pulsed mode...
and that is at a high frequency with a (usually) 50% duty cycle...


Jerry

Well I am too, since the red diode survived and still kickin'.

But I used 3 or 4 times less capacitance then I should, and the coil cannot deliver such current at that Vdrop.

I noticed that when reading on my DMM the current on 5 V test load, the reading is acting like you are trying to measure AC with DMM set to DC. It continualy jumps around.

Meaning, it was VERY unstable...

However, nothing we do now is gonna bring back my PHR :cryyy:

On a brighter note: I have placed a free sample order for two kinds of coils on coilcraft, with an email, hope I have them soon. Also, would hooking up coils in parralel increase their current output?

And Jerry, I took a look at digikey... well the LM3410 I have is about 2.73$ I piece. But he shipping is 30$ flat, using USPS. I will have to find another distributor ...