Using a DC/DC converter to power the laser

[IgorT]

This is great!

I just recieved an email from CoilCraft, that they shipped the second sample order (the one for the 3410), so i asked my girlfriend to check her email, and she also got a reply, that they sent her samples for "her student project"...

I made this order on her name, while my first sample order was on hold.. Later they shipped my first order, now the second, and the funny thing is, that the email she recieved says, that her samples were shipped on 20st and 21st..


I believe this means they shipped them twice. Probably because the samples cart malfunctioned on the first try (didn't send a confirmation email) so she made the order again...

Soon i'll be supplied with all the coils i might need.. Well, at least for a while...


Next i'll try the LM3402 circuit, now that i learned how to draw and etch them...

 

[IgorT]

I don't know, a smd schottky diode from some SMPS, it seemed to have a low forward voltage when tested with my multimeter.

Oh, that's the weird square thing, right? What forward voltage did you measure?

Maybe you should try the 5817.. Could give you better efficiency..

all in all it hasn't been as hard as i expected, soldering a chip with no pins...

I should gather up the currage and give it a go.. But this lack of any pins, on a board without a solder stop mask scares me.. That, and the fact, that i killed one already..


I only have three left (should have ordered 5) and four of the YSD, and i still don't know how these are supposed to be used.. Oh well, i'll just try them and see..

 

[woop]

my multimeter measured 250mv but that was only at a few milliamps. so who knows. I guess i could measure the voltage across it in circuit.
no shops in my area have the 5817, the best i can find is the IN5822 with a 525mV drop at 1A

well if it doesn't go well first time, you can simply put it back in the oven to desolder the chip. I haven't killed one yet this way. it helps to put a bit of solder paste on the board so you know if it has reached melting temp yet. and the max temp should be about 200C in the oven
how else where you planning to solder the LLP?

 

[phenol]

I only have three left (should have ordered 5) and four of the YSD, and i still don't know how these are supposed to be used.. Oh well, i'll just try them and see..

it should be the same, just pick components with larger values, for instance 15-22uH instead of 4.7uH, 4.7-10uF instead of 2.2uF, the input filter cap could stay 10uF, the output one - at least 10uF. you wont have to use different resistors. eventhough the switching freq is lower, the slew rate is the same, which calls for the same spikes as with the 1.6MHz version. when probing with the scope the voltage across the inductors, the ringing on the rising edge had frequency i excess of 200MHz. this thing radiates copious amount of RFI,thereby blocking any radio up to VHF within meter or two

 

[woop]

when probing with the scope the voltage across the inductors, the ringing on the rising edge had frequency i excess of 200MHz. this thing radiates copious amount of RFI,thereby blocking any radio up to VHF within meter or two

wow. maybe we should put a faraday cage on the thing ;D
have you had much success getting rid of spikes? will a few ceramic caps do the job?
I wish i had a scope

 

[IgorT]

my multimeter measured 250mv but that was only at a few milliamps. so who knows. I guess i could measure the voltage across it in circuit.
no shops in my area have the 5817, the best i can find is the IN5822 with a 525mV drop at 1A

If you want, i can send you some.. I have a bunch and they're almost for free. But the 5822 could work just as well.

well if it doesn't go well first time, you can simply put it back in the oven to desolder the chip. I haven't killed one yet this way. it helps to put a bit of solder paste on the board so you know if it has reached melting temp yet. and the max temp should be about 200C in the oven
how else where you planning to solder the LLP?

On a hot plate, like in that video..

How long can you safelly leave this in the oven? And how much solder paste would you recommend putting on?

I guess i'm spoiled by the boards always being proffesionally made with solder stop mask and all...

 

[IgorT]

it should be the same, just pick components with larger values, for instance 15-22uH instead of 4.7uH, 4.7-10uF instead of 2.2uF, the input filter cap could stay 10uF, the output one - at least 10uF. you wont have to use different resistors. eventhough the switching freq is lower, the slew rate is the same, which calls for the same spikes as with the 1.6MHz version.

Thanks for the info!

That's what i'll do..

when probing with the scope the voltage across the inductors, the ringing on the rising edge had frequency i excess of 200MHz. this thing radiates copious amount of RFI,thereby blocking any radio up to VHF within meter or two

Are your coils shielded? Did you already recieve the correct ones from CoilCraft?

Or does that have nothing to do with this?

 

[woop]

How long can you safelly leave this in the oven? And how much solder paste would you recommend putting on?

it only takes 2-3 minutes, i wasn't timing. just wait till the paste melts then take it out
use only a small amount, only whats required to make contact. i used a toothpic to apply paste only to the pads, avoiding getting it between the pads
i used a mini toaster oven, so i can see the board clearly through the glass

 

[IgorT]

How long can you safelly leave this in the oven? And how much solder paste would you recommend putting on?

it only takes 2-3 minutes, i wasn't timing. just wait till the paste melts then take it out
use only a small amount, only whats required to make contact. i used a toothpic to apply paste only to the pads, avoiding getting it between the pads
i used a mini toaster oven, so i can see the board clearly through the glass

Thanks!

I really appreciate all the information you guys give me.. As i said, this is all new to me.. Never etched my own boards before. And with SMD, we always had the complicated ones soldered in the factory. (only costs 30% more and saves a lot of time)


Well, if i decide to use an oven, i have a microwave, that can be used as a convection oven.. If i set it to convection and 200°C would that be ok?

Does your oven have convection or just heating from above and below?


And the IC doesn't mind any of this? The datasheet specifies very short reflow times - 15 seconds at 220°C convection...

 

[phenol]

oho! microwaves! just doublecheck the microwave portion is well switched OFF prior to starting the baking process. you would believe how much those microwaves like metal pieces and tiny little microscopic helpless LLP CMOS chips. once i baked a 8086 CPU in my microwave. its chip had a crunchy black finishing when the 5 secs cooking time had elapsed.

as regards coils and shileding and spikes, no, my coils are not shielded. im using toroids with initial relative permeability of about 700-1200. the field in them is supposed to stay confined in the core, but the core itself is a little bit conductive, the windings are not all that perfect, the freq is too high and all of these lead to the presence of some field near both inductors. parasitic inductance of PCB traces along with amperes of fast switching current is what generates those notorious spikes. thats when proper layout and component selection is crucial. as i dont think it is possible to completely eradicate noise on the board itself, it is probably more feasible to cope with it directly on the LD by soldering a 0806 or 1206 ceramic cap across its leads.

 

[IgorT]

oho! microwaves! just doublecheck the microwave portion is well switched OFF prior to starting the baking process. you would believe how much those microwaves like metal pieces and tiny little microscopic helpless LLP CMOS chips. once i baked a 8086 CPU in my microwave. its chip had a crunchy black finishing when the 5 secs cooking time had elapsed.

I did say, i'd only use convection, didn't i?


Did you turn on the microwaves by mistake? My microwave has a completelly separate convection, but it can be used combined...

BTW: I would believe, what microwaves can do.. Seen it myself, on purpose of course... The most fun was the ball of plasma..

it is probably more feasible to cope with it directly on the LD by soldering a 0806 or 1206 ceramic cap across its leads.

I did that, just like you recommended.. I used a 1uF, like you said in the PM. So is this really enough to make the output clean, or should i go for a higher value, or even multiple different values?


Who's gonna be the first to try if an LD can survive this? I have some flat chip 16x DVD LDs, which i can sacrifice.. I can also test it on IR LDs.. Don't want to just put on a good burner diode and risk losing it.


Anyway, i don't think i'll be using this circuit with Blue Rays... Those are at least 10 times as sensitive as the DVD LDs.. If you just look at them funny, they change their electrical characteristics and drop in output..

 

[chimo]

I'm curious, have you done any destructive testing to see if the paste re-flows under the ground pad of the chip? I have seen instances where the exposed or edge contacts will reflow, but the paste under the body doesn't due to it being a little more insulated than the other contacts.

Has anyone tried kapton tape as a solder mask?

How long can you safelly leave this in the oven? And how much solder paste would you recommend putting on?

it only takes 2-3 minutes, i wasn't timing. just wait till the paste melts then take it out
use only a small amount, only whats required to make contact. i used a toothpic to apply paste only to the pads, avoiding getting it between the pads
i used a mini toaster oven, so i can see the board clearly through the glass

 

[IgorT]

Has anyone tried kapton tape as a solder mask?

I heard of this once before, but forgot about it...

Can it be used in such small areas, as the ones between the IC pins and the central GND strip? That is my main worry....


Anyway, to make sure the middle under the IC reflows as well, i put solder blobs on the exposed "dog bone" ends on both sides of the IC and heat them with two soldering irons at the same time, while the whole thing is on a hot plate. Do you think this is ok?


I did "destructive testing", altho not on purpose, and the middle did reflow this way.

 

[phenol]

Did you turn on the microwaves by mistake? My microwave has a completelly separate convection, but it can be used combined...

BTW: I would believe, what microwaves can do.. Seen it myself, on purpose of course... The most fun was the ball of plasma..

i did it on purpose, just cudnt resist the temptation. i also put two forks like a dipole on the turntable. lovely sparks. i like the destructive microwave show they do n Brainiac on Discovery

I did that, just like you recommended.. I used a 1uF, like you said in the PM. So is this really enough to make the output clean, or should i go for a higher value, or even multiple different values?

1uf is adequate. since you have more laying around, you could also add a 10-100nF one in parallel with the 1uf even for rigidity sakes alone.

Who's gonna be the first to try if an LD can survive this? I have some flat chip 16x DVD LDs, which i can sacrifice.. I can also test it on IR LDs.. Don't want to just put on a good burner diode and risk losing it.

im intent on buyin my third LG soon. may try it with the IR one, but im going to make a bettr board first and also thy to mod it so that the LD cathode is grounded and the shunt is ''above'' it

 

[IgorT]

1uf is adequate. since you have more laying around, you could also add a 10-100nF one in parallel with the 1uf even for rigidity sakes alone.

If you think it would keep my LDs safer, if i add different values of these small SMD capacitors, i'm going to solder more on.. I found many different values in the DVD drives, so i would love to put them to good use..

im intent on buyin my third LG soon. may try it with the IR one, but im going to make a bettr board first and also thy to mod it so that the LD cathode is grounded and the shunt is ''above'' it

Well, i don't really care about my IR diodes, so i'll sacrifice them for testing.. I'll put on the same SMD capacitors, and see it they can take it. After that i'll try with the flat chip LD, and only if that makes it, will i attach a "canned" LD..

My first open can still didn't come, and when it does, i don't want to kill it because of current ripple..


So you're gonna try to mod it, so you can use a flashlight host? That's interesting.. You think it'll work? I'm sure you have some ideas already..

 

[woop]

under the chip seems to reflow well whenever i do it. i think the chip would actually be a hot spot, because it would absorb IR better than surrounding metal tracks.
IgorT, you have testing facilities don't you? when you get one working take it in and try adding caps to see what happens.
it should be possible to run a bluray with enough filtering. would it help to put another choke at the output?
only one way to find out.

I gotta get myself a new burner, any suggestions?