JLSE
1
- Joined
- Dec 13, 2007
- Messages
- 3,580
- Points
- 0
LOL, I disappear for a while and there’s just a feeding frenzy going on! I’m flattered guys, you really didn’t have to.
Though as I read through these pages I notice IgorT that you have some facts wrong.
Firstly,
-“Unfortunatelly they told me the drivers were for personal use - cos they didn't trust FlexDrives... I was never told the drivers would be used in lasers for sale or that the drivers themselves”
I was looking into a boost driver rather than buying them from Dr Lava, that part is true. I was looking into the LM2577 (IIRC) and asked you for advice. I had also asked you about another maxim chip and so on. You constantly compared the 3410 to Lavas drive and finally sold me on it as well. The last time you claimed that all you had said about the flexdrive was based on what I told you and my experience? LOL, you call me a liar?
You suggested to me after many emails to look into the 3410, which you always referred to as ‘your driver’. In a later PM you said that you would help me so long as I didn’t sell the handhelds I made with the 3410 on LPF. Something to the effect that ‘guys like Jake21 were slowing you up by selling cheap. Obviously you weren’t interested in any further competition… Understood, and respected.
Nowhere did you say to me that I couldn’t sell the drivers themselves, nor sell them in anything anywhere. There was no such agreement…
-“What happened in the end was, that the driver became a DIRECT copy of Woop's driver (anyone remember Woop?) - plans posted and burried deep in the Experiments and Modifications section of this forum...for sale..”
“Hmm, in that case, it can't possibly be a direct copy of Woop's driver, can it?!? Maybe i just got confused by the layout being so identical to Woops... Yeah, that must be it! (ok, ok, parts were moved around slightly, but in general it's still pretty much identical)”
Incorrect, while yes I did use woops diagram initially, I changed it completely when I switched to the sot-23 5pin. Approx a year ago...
The layout since I started using the 5pin over the LLP has changed several times. As you refer to your driver as “my driver” I also refer to my version as my driver.
I also have had the help of an EE help test this board in different versions, and still to this day. I have also since acquired a DS O-Scope and continue to test and look to make improvements.
As for any hype about the IC, you were the one who constantly boasted about how superior it was, yet you downplay it in the forum? Wasn’t it you who said “stability and capability far superior to any FlexDrive”? Last time I checked, the only other boost driver for 405nm out there is the flexdrive, I don’t make any direct claims against it though.
-“Why? Cos he didn't know how to convert a schematic into a PCB by himself... And then he went on to lie he developed it himself (and that it's the best driver ever, with zero ripple and other fantastic but unspecific claims). Also, he lied to me about his intended use for the driver, while asking for help while making it...”
When did I lie? Do me a favor since you are so adamant about this whole topic... Dig up the PMs that I “lie about the intended use” All you ever said was “you shouldn’t sell these without first testing”. I never said I was going to sell the boards on their own initially, simply because my bigger need was for my lasers. If I have extra why not sell them?
You call me a liar, lol, you are telling this forum lies about what we had talked about, what I have said and more.
I had made a mistake in the ad when I wrote it by putting “no ripple” when it should have read ‘no harmful’ which was put in the other ad for the same driver. I had not corrected this on the one that sold, but had made the correction to the other.
I will be sure to remove “far superior” and add “stability and capability far superior to any FlexDrive” as you once so succinctly put it. Perhaps this will make you feel better?
Almost forgot, the crack about the inductors im using? lol, are your fingers bleeding from all that scraping? I have not once ordered a sample other than IC’s from maxim, once. I use coilcraft only, from the beginning till now. I can send you a pic of the reels if it will cheer you up. I remember you posting on how you were getting samples under you GF’s name, neighbors dog… etc… lol, why would you make a crack about samples?
And finally to Dave… I gather the pointer you are poking fun at doesn’t meet your requirements of what would be classed as a laser? I still talk to many of my customers who have bought higher priced items. Anyone who buys from me gets 1 year of free repair at no additional cost. So if it blows (which it wont), the customer does not get left hanging. I know that what you’d like the people who read your post believe that.. But that’s just an inaccurate claim. I have had enough returns in 2 years, I can no longer count them on both hands. But ive stuck to my words and always take care of the customer.
Its funny how all these things are being said, yet none have had any business with me. That’s a lot of assumption, but hey if it works for ya..
To help refresh your memory.... Found those PM's for you.
Though as I read through these pages I notice IgorT that you have some facts wrong.
Firstly,
-“Unfortunatelly they told me the drivers were for personal use - cos they didn't trust FlexDrives... I was never told the drivers would be used in lasers for sale or that the drivers themselves”
I was looking into a boost driver rather than buying them from Dr Lava, that part is true. I was looking into the LM2577 (IIRC) and asked you for advice. I had also asked you about another maxim chip and so on. You constantly compared the 3410 to Lavas drive and finally sold me on it as well. The last time you claimed that all you had said about the flexdrive was based on what I told you and my experience? LOL, you call me a liar?
You suggested to me after many emails to look into the 3410, which you always referred to as ‘your driver’. In a later PM you said that you would help me so long as I didn’t sell the handhelds I made with the 3410 on LPF. Something to the effect that ‘guys like Jake21 were slowing you up by selling cheap. Obviously you weren’t interested in any further competition… Understood, and respected.
Nowhere did you say to me that I couldn’t sell the drivers themselves, nor sell them in anything anywhere. There was no such agreement…
-“What happened in the end was, that the driver became a DIRECT copy of Woop's driver (anyone remember Woop?) - plans posted and burried deep in the Experiments and Modifications section of this forum...for sale..”
“Hmm, in that case, it can't possibly be a direct copy of Woop's driver, can it?!? Maybe i just got confused by the layout being so identical to Woops... Yeah, that must be it! (ok, ok, parts were moved around slightly, but in general it's still pretty much identical)”
Incorrect, while yes I did use woops diagram initially, I changed it completely when I switched to the sot-23 5pin. Approx a year ago...
The layout since I started using the 5pin over the LLP has changed several times. As you refer to your driver as “my driver” I also refer to my version as my driver.
I also have had the help of an EE help test this board in different versions, and still to this day. I have also since acquired a DS O-Scope and continue to test and look to make improvements.
As for any hype about the IC, you were the one who constantly boasted about how superior it was, yet you downplay it in the forum? Wasn’t it you who said “stability and capability far superior to any FlexDrive”? Last time I checked, the only other boost driver for 405nm out there is the flexdrive, I don’t make any direct claims against it though.
-“Why? Cos he didn't know how to convert a schematic into a PCB by himself... And then he went on to lie he developed it himself (and that it's the best driver ever, with zero ripple and other fantastic but unspecific claims). Also, he lied to me about his intended use for the driver, while asking for help while making it...”
When did I lie? Do me a favor since you are so adamant about this whole topic... Dig up the PMs that I “lie about the intended use” All you ever said was “you shouldn’t sell these without first testing”. I never said I was going to sell the boards on their own initially, simply because my bigger need was for my lasers. If I have extra why not sell them?
You call me a liar, lol, you are telling this forum lies about what we had talked about, what I have said and more.
I had made a mistake in the ad when I wrote it by putting “no ripple” when it should have read ‘no harmful’ which was put in the other ad for the same driver. I had not corrected this on the one that sold, but had made the correction to the other.
I will be sure to remove “far superior” and add “stability and capability far superior to any FlexDrive” as you once so succinctly put it. Perhaps this will make you feel better?
Almost forgot, the crack about the inductors im using? lol, are your fingers bleeding from all that scraping? I have not once ordered a sample other than IC’s from maxim, once. I use coilcraft only, from the beginning till now. I can send you a pic of the reels if it will cheer you up. I remember you posting on how you were getting samples under you GF’s name, neighbors dog… etc… lol, why would you make a crack about samples?
And finally to Dave… I gather the pointer you are poking fun at doesn’t meet your requirements of what would be classed as a laser? I still talk to many of my customers who have bought higher priced items. Anyone who buys from me gets 1 year of free repair at no additional cost. So if it blows (which it wont), the customer does not get left hanging. I know that what you’d like the people who read your post believe that.. But that’s just an inaccurate claim. I have had enough returns in 2 years, I can no longer count them on both hands. But ive stuck to my words and always take care of the customer.
Its funny how all these things are being said, yet none have had any business with me. That’s a lot of assumption, but hey if it works for ya..
To help refresh your memory.... Found those PM's for you.
IgorT said:As i told you before, i will make a list of the exact parts with the exact sizes you would need to make a driver similar to mine. It would fit an AixiZ module. But my own driver is a tweaker single side driver, where i move some parts on the other side in a slightly "dirty" fashion..
I could also give you the transfer pics i made, with some hand soldering improovements - to make hand soldering of my regulator possible at all.
I just need to make a tiny change in the design. But it's NOT pot adjustable! As i told you, i hate pots, and prefer setting the current with resistors. I have put two jumpers into my design, that allow selecting a higher current by bridging two small gaps with a drop of solder.. I also have the last two flaws figured out, and could fix them over the weekend - when i'm done with the last lasers i have to ship out.
The only thing i care about is that you don't copy my laser. But i know that you wouldn't.. That is something i would only expect from Jake.
The driver i have is not a decent one, it's a perfect one. As i said, the efficiency is 84%, and it's not a force stabilized synchronous operating at the edge of it's abilities or beyond. The flexdrive can't work a load with a Vf under 2.7 or over 5.5V. Mine can do 0-27V, and is meant to be used as a diode driver. It's a constant current LED driver, and the high voltage ability is so that it could be used to drive a long chain of LEDs in series.
Some cheap LED driver ICs don't have constant current. They just limit the max switching current. This here is the real deal, and the only thing it requires is a filter cap on the diode itself, to reduce the current ripple to nothing..
The cost is $3 for the regulator (unless you buy 25, where it's $2.6 or so), the coil is $1.2-1.6 or so, the rest of the parts is dirt cheap. If you can fit an AixiZ module in your host, you could use it as a single sided board, with only a diode mounted through hole from the other side.. It's soldered directly on the diode, and then you solder another cap or two on top of the diode's legs..
The only "drawback" it has is that it is a simplified boost only version, which can only power PHR diodes or 6x diodes - it does not allow grounding of the diode, but blu(e) rays don't use the can pin, which is why this simplified driver is perfect!
I have another version, that is pot adjustable and requires a double sided circuit and more components, that allow the diode to be grounded, but in my oppinion (and the oppinion of one of my engineer friend, as well as my own engineer), that is a waste of money and time for a PHR. All the required resistors you need for the current are available, and with the two jumpers, you can select whatever you want (i can make it with two or three higher current options).
I also have a third version, which is the full buck/boost circuit, and again fits an aixiz module, but requires two coils and a lot more components. This could power a red from a single Li-Ion, much more reliably, than any synchronous, but for a blu ray, it's again pointless and overkill.
I have some more work to do, but i could finalize the last version over the weekend. I'm eliminating one microscopic (<1mm) resistor that is insane to solder in place, and at the same time reducing the chance of the related pads de-bonding when soldering as well as making one correction for the jumper pads, that are so small that they can also de-bond. The rest is perfect as it is.
If you can etch your own PCBs and solder legless ICs, you would be able to use it.
You asked me about ohm's law before.. Well, here you would have to use it.. Ohm's law is just the relationship between voltage and current and resistance. U = R x I..
In this case, a resistance sets the current. The regulator's feedback voltage is only 0.192V (317 has 1.25V!).
So.. If let's say you want 100mA.. U = R x I -> R = U/I
So U = 0.192V, I = 0.1A, R = 0.192 / 0.1 = 1.92
This means 1.92 Ohms give you 100mA.. I am using a 1.54 Ohm (ALL 1%) resistor, and it gives me 125mA +- 0.5mA, depending on the variances in the regulator Vf...
The driver keeps the current constant using ohms law.. At a certain current, the voltage drop over a known resistor is also known.
You put a resistor in. All the driver does is keep a constant voltage drop across that resistor.. So it measures the voltage drop, and adjusts it so that it is always 0.192V.
If the resistor is 1.92 Ohms, this means, the driver will always have to pump EXACTLY 100mA through that resistor in order to keep the voltage drop constant. This is how you get constant current.
If there is another aspect of Ohm's law you want to understand, just ask. But it's always the same thing, and once you get it, it's REAL simple.
IgorT said:I wrote quite a lengthy reply about it. I was sure you got it.. But it's not in my outbox. This is the fourth time this week. Yesterday i sent GooeyGus two PMs, one long, and the second very short, and he only got the second one.
I looked at the datasheet when you sent that, and i remember, that there are several versions. Some are preset to a certain voltage, one is adjustable.
The preset ones have a fixed high reference voltage, which is the same as what they are set to. They can only do that and nothing else. The adjustable one has a reference voltage of 1.25V, like the 317 for example.
This means, that you can turn some parts around and get constant current, but 1.25V and a certain amount of power (depending on the current), would get lost on this resistor. The chip would always have to boost 1.25V higher, than what the diode needs - if you converted it to constant current.
If you use it as constant voltage with a driver after it, it gets even worse.
But all this is completelly pointless. If you want to make a boost driver, why not make one that is designed for constant current, and has a low enough reference voltage to achiev a high efficiency.
The thing is, a voltage booster creates a certain voltage across the output. What this voltage will be, depends on the sense resistor on the feedback pin. This resistor is in parallel with the output.
A constant current voltage booster will create not a certain voltage, but the voltage needed for a certain current to flow and adjust it, to keep it constant. It does so, by creating a constant voltage drop across the sense resistor, and this resistor is in series with the load.
Because it is in series, it always has to do output voltage + feedback voltage. The power lost on the sense resistor depends on the feedback voltage.
Regulators designed for constant current have a very low Vfb, just because of this. Since you want to make your own circuit anyway, make it right in one step, not in two, at a ridiculous efficiency. It'll be more work in two steps.
Look into the LM3410. It has a Vfb of 0.192V. This allows efficiency of almost 88%. And it does everything in one step. It's a constant current boost driver or a constant current buck/boost driver. It needs an input voltage between 2.5 and 5.5V, and always gives the load exactly what it needs.
The current is set with the feedback resistor, using Ohm's Law. U = R x I... U is feedback voltage. I is desired current. So what resistor you need for 100mA is calculated like this: R = U / I, R = 0.192 / 0.1, R =1.92 ohms.. This means 1.92 ohms will give you 100mA.
In the boost version it can drive PHRs, but the input voltage has to be UNDER the diode voltage at all times. A Li-Ion is perfect.
In the buck boost version it can drive any diode, but with reds, the diode has to be insulated from the battery. But since it can both boost and buck, it can work from any voltage between 2.5 and 5.5V, and give any diode whatever it needs.
With a special current sensing IC, it is also possible to ground the diode to the host, but means a more complicated driver. Then it also allows the use of a pot. It is not needed for blu diodes tho. Nor is the buck part...
IgorT said:The IC on the LavaDrive is pushed close to the limit, when powering a PHR diode.. But otherwise, it shouldn't have any problems. Even the PHR shouldn't actually be a problem. Still, i prefer to use a proper diode driver IC - one that is meant for this purpose, and which can also boost to 27V if necessary, so no laser diode is ever going to be a problem for it..
Anyway, i understand now what you wanted with the boost circuit.. But this regulator will hardly put out voltage high enough for you to put a linear driver after it. It would be better to rearrange the voltage setting resistors, and make it a lossy constant current circuit.
It could drive a red this way, but not a blu ray.. For a blu ray, you need a circuit designed to do constant current. Those have a VERY low feedback voltage.. Mine has a Vfb of 0.193V! I set the current with fractions of an Ohm.. 1.54 Ohms gives me 125mA for example..
This means very low losses on the Rsense, and an efficiency of >84%! Not 93% like the LavaDrive, but i trust mine more.... Probably ungrounded, but still. I know it, and what it can do, so i feel safe with it. The hardest part is making it.. I have to make it 13x9mm small! Before i had it in a form, that fit an AixiZ module, but my hosts don't fit an AixiZ module, so i had to cut a third away! One resistor between two legs is so small, that i don't know if i soldered it in place or inhaled it by mistake... ;D
The worst part is, that my regulator has no legs - just pads underneath.. There is a new version of it with legs, and i should switch to that one soon, cos i keep killing these, even tho i've been working with electronics for 10 years now...
Anyway, a combination of a boost circuit and a linear driver could work, if you can boost high enough.. The Maxim could maybe drive a red if you could boost to 5.5V with enough current behind it and rkcstr's driver after it.. But it could never drive a PHR.. For that, you would have to boost to 9V at least, again with a load of current behind.. Most of the Maxim IC's are incapable of the currents we need, so the voltage sags, and then you're nowhere.. I don't remember what this one can do tho..
But you might be better off using a single IC solution. It'll be SMD and small, but that's probably what you want anyway..
There are some synchronous ones sprouting all over the place, and there are several versions of the 3410 circuit available.. The first three in my DC/DC converter thread (Woop's, mine, Phenol's) are big, but Woop later made one that fits an AixiZ module..
I haven't posted my latest version tho, as it is not very user friendly.. Besides, i doubt anyone needs something that has to fit a half of the back part of an AixiZ module..
IgorT said:Wait, i'm a little confused.. You removed the diode, put on LEDs, to tinker safelly, and then the diode died?
Where did it die and which diode died..
I think you need to hook up your driver to those LEDs through a DMM. Make sure the LED dummy load has a Vf between 2.7 and 5.5V. Not more than 5.5V and not less than 2.7V. LavaDrive can't handle regulating the current outside this range.
Then you let it run and observe the current if it's changing.
So you are using the kit you made or the fully assembled tiny LavaDrives with a different pot?
It can't be the pot or the lower current. The only thing that could be wrong would be if the voltage on the PHR would be above 5.5V at that current.
I thought LavaDrive is a good driver. And Lava knows a hell of a lot more about electronics than me. I do not think he would sell a driver, that wouldn't work properly.
If you do the current test i mentioned, and the current doesn't change from what you set it to, then it could only be the ripple. But ripple at 70mA should not be dangerous to a PHR. And it's filtered anyway..
The only thing i can imagine to kill a diode is the current being unstable. But that could only come from a defective regulator. Or maybe oscillations in the circuit, if the cap that stabilizes it is not soldered properly, or broken. Tiny ceramic caps can break easily, with the slightest force. Like if you solder one side and then push the other side down to solder it, you can easily break the cap inside, and you wouldn't even know it. It could then result in an unstable circuit.
But in this case, it should be obvious in the current changing - do the dummy load -> DMM test..
P.S. And no, i'm afraid i don't have anything i could sell.. If i take time into account, my driver costs me more, than a LavaDrive would. But i can't squeeze a LavaDrive in my hosts... So i have to make everything myself.
The parts themselves cost a shit load of money. And i'm running out of inductors. The PCBs are easy to etch, but sometimes a tiny copper trace can be heated just a fraction of a second too long, and falls off the board.. Need to change the design yet again..
Also, i killed many regulators while soldering them by hand, simply because they don't have legs.. They have pads underneath, and nothing on the side, unlike the IC Lava is using..
The parts list is very short, but the sizes are important.. I'll look into it, to see if there is a simple solution - it has to be just right, otherwise you end up with parts you can not use.
In the meanwhile do that test and let me know what you find. Use one current setting, and let it run for 10 minutes, while observing the current. Then another setting. And make sure the output voltage at those currents is within the limits..
Oh, and did you by any chance ever power up the circuit WITHOUT a load? Think about it. It's an important question.
IgorT said:Oh man, i know what you mean..
Some people are just pissing me off.. This guy Jake21 for example.. He doesn't even know what he is doing!
But when i offered my lasers here for $185, he was still trying to sell his crap for $250! And that was an undegulated boost circuit driven laser, made with the LED abuser circuit he found in a flashlight, and that one day after he asked me how to put a pot in there!
I told him not to use that, but since i knew he would, i told him what to do.. The next day he was selling it, and then he tried to take away one of my buyers by offering them lower and lower prices untill they almost changed their mind!
Before that, when i didn't know he was selling lasers, he asked me to make him some drivers.. I said i would make them for him if he pays for the parts.. And then he sold the lasers he made with them!
And this is a person, who can't tell current from voltage! Never mind understand what a driver does.
It makes me sick! I try to do quality work, i spent two weeks designing my module and driver, and i spend a shit load of time to make it all just perfect, but because that idiot offering quickly thrown together crap for $130, i can't get more buyers here anymore.. And if the first lens attempt fails, i will then have to pay for the second one out of my own pocket!
Not to talk about some other people, who've been wishing me to fail from the start, even tho i was not trying to be their competition. I just wanted to cover my expenses and get out of this with some quality optics..
For a while i thought, certain people are acting this way cos they think i would keep the lenses for myself. So i started the GB feeler and told everyone the real price. But it became worse.
At this point, i regret not deciding to keep the lenses for myself. But if i have to make a second attempt, i just might do that still.
Otherwise, i know about working for yourself.. That's why i started my own company four years ago. But i have been doing the same thing for 10 years now - developing and designing custom electronic devices and building them..
Oh, if i have to go for aspherics in the end, i will not be able to do it without selling at least one more batch of the lasers.. Two would be better.. So any tips would be appreciated.
I won't have much more time to do this, so i'll try to prepare both batches now, it's taking too much time away from my company..
Last edited: