In search of 22A 4V buck driver.

[Alaskan]

I have the parts to build a HH laser which draws 40 amps

 

[RedCowboy]

Are you going to pump a crystal or do you have a whole diode block?

Maybe something else?

I have an idea for repurposing everything up to the phosphor wheel, but just how far with a large home made BE it could work is yet to be seen.

Is this build under way?

Can you give a hint?

 

[Nutball]

If only analog and digital circuit design was a common 4 year high school class, so we'd all be experts at designing things like power supplies. I sure wish I knew enough to do that.

Anyway, I'm sure it is very possible to have a non heatsunk buck converter smaller than the ebay one capable of over 25A for 2 minutes. (based on the results of a not very refined design I was working on a couple years ago)

 

[Alaskan]

Are you going to pump a crystal or do you have a whole diode block?

Maybe something else?

I have an idea for repurposing everything up to the phosphor wheel, but just how far with a large home made BE it could work is yet to be seen.

Is this build under way?

Can you give a hint?

I really think the idea of paralleling several small laser diode drivers is the easiest way to get your 22 amps.

RCB, who is that question for? If for me, it's just a 40W 808 nm FAP with a fiber collimator on the end, easy to do with a 2 volt lead acid battery, one of those Cyclon's. I have a 25 amp hour one which runs down in 10-15 minutes with that FAP.

 

[diachi]

I really think the idea of paralleling several small laser diode drivers is the easiest way to get your 22 amps.

RCB, who is that question for? If for me, it's just a 40W 808 nm FAP with a fiber collimator on the end, easy to do with a 2 volt lead acid battery, one of those Cyclon's. I have a 25 amp hour one which runs down in 10-15 minutes with that FAP.

You still have >70W waste heat whatever way you go about it, unless you go for a switching supply which, properly designed, could bring that number down to <10W.

That's a whole lot of heat to get rid of in a small space.

We need maximum dimensions from OP and more details about the project.

 

[Alaskan]

There is no waste heat in the driver because for my FAP800 with the Cyclon 2.1 VDC lead acid battery there is no driver, the battery drains down too quickly to over current the FAP array when a big heat sink is put under it. I know, unconventional and completely against the rules, but I tested this over and over and never had an over current situation.

 

[DashApple]

Are the blackbuck M8 able to be paralleled ?

 

[AaronT]

Well #### I guess I have to let the cat out of the bag... A bit.

22A at 4V as that is the "max" power recommended for the single diode I have.

Yes, lots of amps, lots of sag, lots of heat, almost no runtime...

The 26650s I have are rated at 35 amps/ea...

For the driver "small" is entirely relative...

I had not considered running drivers in parallel, that may be the simplest way to do this... Could even "fan" them around the diode head for "style" points...

If I hit a brick wall I'll drop down to a 15 watt diode...



My last "bad" idea involved helping make a friends 180 watt CO2 laser from a used cutting table "portable"... Bout as portable as an AT4 anti tank rocket... Aim... laze, POOF!!!

 

[RedCowboy]

AT4 that happens to also be 84mm, those Swedes are too funny.
Chinese however are often liars, the continuous and surge ratings on batteries are......optimistic, they will put out a lot for a short burst, but even if they can stand to dole out 14 amps the voltage sag will be under the bucking drivers minimum is short order, you can give your 2 x 26650 batteries a test with a pair of 1 ohm resistors in parallel, use big ones as they will get hot, and also put a volt meter across the batteries while you load test them and see how long until they drop under 6.5-6.8v the SXD's minimum, or the minimum of the bucking drivers you will be using.

 

[Radim]

I'm looking for a 22A 4V driver. (2 26650s as power source)

A small driver with the ability to use an external adjustable pot (as blackbuck M8) would be great.

Any designs for such a driver would also be useful.

22 amps and 4 V DC? 88 W of power? Well, physics is going to limit you a lot here, thick cables, not really small driver/power source. As advised above - I would think about use of more buck converters in parallel. However they are not going to be really small for such currents. I would also add some capacitors extra to balance output and input here. Do not forget proper heatsinking. Maybe if you look for robotic parts for arduino/raspberry they should have some high current adjustable buck DC DC converters for powering motors - still I would pay attention to balancing them well to prevent overload in case they are not balanced well and one delivers no amps due to sufficient voltage, but others deliver too much current. Maybe this solution is too complicated and not reliable.

What about building some custom one? You just should need sufficient and fast switching MOSFETs or maybe some small fast switching thyristors, some high current fast switching schottky diode and inductance with thick wire (or more wires winded in paralel). Also caps should be of high capacity (electrolytic) and thick leads (and for high frequency operation). You can base it on some IC (+ this IC powering circuitry) used in other regulator, just change the power carrying components with correct specs (do not forget to match IC driving circuitry as well). All components specs however should be calculated and fitting IC switching characteristics to have smooth and stable output. And one more thing - look for high switching frequency IC (like 150+ kHz), than you can use smaller power components than with lower frequencies.

Edit: Consider adding sufficient fuse on input to protect it from too much current. SMD fuses should be small enough for your purposes.

 

[Pman]

You guys are way over my head but at least you have some cool vids
All I want is a cheap driver to power some PT121 Phlatlights (or at least one). Messing around I've blasted a single green at around 40 amps and man that thing was nasty bright!
Have one of those Jayrob PT54 mag builds and a 1000lm of green in the eyes is quite blinding.

Also, why are you all up so early? I couldn't sleep as my wife usually gets up at 4:20 to get ready to drive a school bus.

 

[Radim]

In my time zone it is almost noon now.

 

[RedCowboy]

The SST-90 LED runs at 18A CW so a flash light driver for that chip could get you close.

Maybe you could run 3 of these in parallel : https://www.ebay.com/itm/3-Mode-Led...640971?hash=item4887a8f00b:g:q78AAOSwaNBUhxdV


I still haven't played with these programmable drivers: https://www.ebay.com/itm/0-10-Amp-c...hash=item2eed2125b6:m:mWt3rUraLwLP7GsWYvJ0TdA

 

[Pman]

Hmmmm that driver is cheap. Could try one on a PT54.
If your not worried about longevity on the phlatlights they will handle stupid amounts of current. Don't have more than a 10A power supply to really push one but have done some series/parallel testing with Samsung 25r cells momentarily where the voltage sag hits the top of their rated range and I think the highest I saw was 43A. Don't tell anyone I did this though..... didn't run more than a few seconds as I was a bit nervous although I will say the batteries had no indication of distress for such a short duration.

 

[Hemlock_Mike]

Texas Inst. has a line of high current power supplies.
Part numbers begin with PTH08*****
Might be worth a look.
I'm running one at 40 Amps / 2.8V (as I recall) with an input
of 14 volts using 20 amp Li Ion cells. Runs cool.
HM

 

[AaronT]

PTH05020WAZ Non-Isolated PoL Module DC DC Converter 1 Output 0.8 ~ 3.6 V 22A

DC12V/24V 8V~36V Step-Down to DC 5V 30A 150W

Due to the size of those parts I'm thinking a 3 Cell Maglite host and FEILONG 32650s

Feilong 32650 Cells TESTED!