Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

LPF Donation via Stripe | LPF Donation - Other Methods

Links below open in new window

ArcticMyst Security by Avery

Driver for NUBM31 and for similar MDP chips sold on ebay/ali






Joined
Feb 12, 2016
Messages
589
Points
63
BTW today I have tried to use a 400W driver for driving the whole block of NUBM31 (output 40V, 6A) from 24V input.
The strange thing is that it could be set for such output with a test load (7 Ohm, 500W rated resistor) but once NUBM31 is attached to it it keeps blinking and immediately going out.

In contrary, the small 250W rated driver was more difficult to adjust for real 240W ouput but once this was done it worked. But heated up very fast, so needs a heatsink.

And in the driver for NUBM08 with R24 removed the PWM input works as on-off switch by applying 0.5V (not 5V as was shown on the ebay/ ali pictures).
 

additude

Member
Joined
Mar 3, 2023
Messages
59
Points
8
In the last months I have been searching to find decent drivers for NUBM31-38 MDPs and for this purpose I have studied some booster drivers from ebay. The similar looking pieces are sold on ali with different specs but I think they are all the same as hardware. After all this time I can say that I was mislead by those "purposed drivers for NUBM31" from techhood etc. and I have lost half a year thinking that these big things should be robust etc. compared to cheap and simple booster drivers from ebay but at the end those expensive big things were crap.

So I purchased a 1000W variable resistor and set it at 14 Ohms to use as test load in order not to risk burning NUBM31 array and tried several drivers with it: these were the small 250 and 400W rated booster drivers and the small driver for NUBM08 blocks found on ali.

Here are the comparative dimentions of all drivers one can use for NUBM31. I have not tested the last big one which different sellers promote both as 1200W and 1500W, but Smackitup has made a build with it recently...

At the result I was surprised to see that even the small 250W driver could do the job for 2 of the LD rows (42V at 3A) from 24V input very well and even from 12V input, however it was heating too much so I would not use it with 12 V input for long.

400W driver instead could do the job from 12V input without heating too much. But not every one - I received 3 of them: 2 worked well, one was defective but regarding the 6usd price...

Ali driver for NUBM08 can also be set to 42V, 3A and works from 24V input but 2 white ceramics resistors become hot pretty quickly. Looking at it makes me think that its board was originally designed for something different but later they soldered elements to the PCB to make NUBM08 driver in the same way as it must have been done with the quickly burning driver for NUBM31 I had so bad experience with...

In conclusion: a pair of these small drivers can be enough to drive NUBM31-38 blocks at 2s2p conection and the 400W driver may be even used alone if adjusted to 42V, 6A but I have not tried it yet...must wait for more of them to arrive.
Hello, I am reading this thread because I am searching for a driver board for the NUBM31 and I need to know why you selected 14 Ohms as the test load resistance output value for your testing.
Thanks.
 
Joined
Jul 10, 2015
Messages
9,913
Points
113
Been using these lately, good for up to 60V out so in the case of a nubm31 you would use 2 of them, one for 2 rows in series.
It has a pot to set the voltage limit and a pot to set the current limit.


SANY6755.JPG
 

additude

Member
Joined
Mar 3, 2023
Messages
59
Points
8
Thanks.
For 2 rows in series I'd need 41V @ 3A (or about 55v @ 4A "over driven"), correct?
 
Joined
Jul 10, 2015
Messages
9,913
Points
113
Eor your 2 rows of 5 ( 10 diodes ) you will draw 4 - 4.5 volts each depending on your current setting, my array is a nubm35 with 7 diodes per row and I'm only powering 1 row of 7 which takes about 31V @ 3.5A and I'm powering the driver with 4 x 18650 which do have to work pretty hard, probably about an 8 amp draw, more as they discharge/sag...... I should be running 6 x 18650 really but 4 do work.

Also the Chinese love to over rate things so you need to add a heat sink to the bottom which is easy as it's solid aluminum across the bottom, otherwise the thermal protection will shut it off after a couple of minutes.

For 10 diodes I would use at least 6 x 18650 in series, 8 would be better.
 

additude

Member
Joined
Mar 3, 2023
Messages
59
Points
8
OK, I am going to start off with feeding it with a 33v @ 10.5a CC power supply. I'm working off some Ohms law and the spec sheet and it drops 4.1v @ 3a. Because it's a diode, it should only take a few tenths of a volt to get it up to 4a - 4.5a.
 
Joined
Jul 10, 2015
Messages
9,913
Points
113
Yes the voltage will change very little as the current swings wide, this is why you must limit your current and it is also affected by heat, you can't depend on limiting only the voltage.

Yes I saw you said CC power supply, so it's capable of 10.5A but you need to limit the current to 3 - 4A max.
Remember the diodes are close together, so overdriving means a lot more heat in a small area.
 
Joined
Jul 10, 2015
Messages
9,913
Points
113
Count the watts, volts times amps equals watts.

If you want 4 amps at 54 volts out of your driver that's 216 watts.
The driver is 90 - 95% efficient ???
So if a graphene battery is 11V the draw from the driver will be about 20A
If 22V then about 10A so a pair of 2.5 watt hour 11V batteries in series will give you 15 minutes in theory.
If you add a 2nd driver and 2nd string of 12 diodes then 7.5 minutes and the draw is 20A.
 




Top