12V or 24V from a 48V pack solution

completelycharged

Active member
Joined
Mar 7, 2018
Messages
1,083
Ok, this is not a sales pitch, been on the search for a cheap, good way to get 12V and 24V from my packs and stumbled upon some boards, which appear to be brilliant for powerwall use. I have seen various posts over time where 12V or 24V has been required and not a lot at a sensible price is available.

Having tried and blown a lot of cheap boards, some where the FET's can't even handle relatively small capacitors attached and looking for something that is not expensive and capable of some abuse. Think this board may be an absolute ideal bargain, due to it being sold as a spare/redundand part ? Ideal to feed 12V LED floodlights with.

They are sold by different sellers so have a look and check wider search for pricing and sources
https://www.aliexpress.com/item/4000089035197.html
Priced around 4 for a 30A converter, listed at $82 each on DigiKey (HPH-12/30-D48N-C) and 76 on the UK site !

image_htmdhd.jpg


Looking up the specification sheet for the board the features appear to be quite impressive and stunning for a 4 board
View attachment 1


With the isolation it could be possible to add 4 of them in series and get a 48V isolated (and current limited to 30A) output, which could make for a safe test connection that limits the maximum draw to 1.5kW. This could be where 4 of them end up for me when experimenting, however in series they could face a different set of issues.

Good Points
1. The board has isolated output, so add two in series to get 24V
2. Parallel them up to as many as you need
3. Automatic current limiting (see page 9 because it is not CC/CV so not a good choice to charge with)
4. Remote sense voltage correction (page 10) to prevent voltage drops on your supply cables
5. Seems to be able to cope with a lot of abuse
6. Can switch on/off remotely pulling one pin to the -ve line (but not when in series with common switch)
7. Input range 39V - 75V (to properly cope with full load regulation)
8. Over 94% efficient for 10A-30A load with 48Vin
9. MTBF of 1.4 million hours !

Bad Points
1. No load power is listed in the specifications as 150mA or 7.2W, which is quite high - the main big down side if only needing a few W !!!
2. No heatsink provided to use at full load for a long time but spec page 6 seems to say ok for 25A with no fan
3. =>10,000uF capacitor would appear as a short circuit

Note on the efficiency (see page 6 of the spec)the idle power is the equivalent of 2% at full load or 7W so if your only pulling 7W with an LED for example it would then appear to be just 50% efficient, however increase to 100W and that 7W is no longer as large a factor. The 36V efficiency is even better than the 94% at 48V.

I have 10 on order (4.73 each inc postage) and post a follow up once I have had a good try at abusing them, they do seem like an absolute bargain. lots of experiments...
 

Attachments

  • hph-12-30-d48.pdf
    1.1 MB · Views: 336
Wow! Nice find!
 
Since these are isolated DC-DC could they be used in part of an active balancing system?
 
Yes, they could be used for an active balancer with an additional step down buck, switch the output between cells and then have an active cell balancer with 30A capacity !

When I found them listed at the indicated price I ordered some before posting, lol, because I had been searching around for a good 48V to 12/24 step down and only just bought a Daygreen 20A unit, which are good for 24x7 because they dont have the 7W idle losses of these boards. For DIY buiders these are hopefully quite useful.

If your use has a reasonable loading level the efficiency and price are just seemingly a no brainer. My luck, they will turn up as DOA or way out of spec. I think I have already planned a use for about 8 of the boards !

For comparison, my 1kW inverter has an idle load roughly the same as two of these boards at about 14W.
 
Finally got around to having a look at the units after they arrived..... they are heavy and the PCB is multi layered way thicker than any standard board as the PCB has the coils embeded for the inductors. The two blocks on the boards are ferrite cores, which actually go through the PCB to make the PCB into the windings, which is a rather neat design. The heavier board makes soldering to the terminals quite interesting as the board with all the copper seems to act as one big heatsink so if you use a small soldering iron the board soaks up all the heat and you can't solder to them....

I have not powered any of them up yet and will post back with another update.

I did order some other 9.7V boards (2 in series for 19.4V) to be used as a generic supply for the NUC and laptop which both need 19V so that I can run them without having to have the inverter switched on for 220V. These have similar stanby vampire power drain but should overall be more efficient than multiple AC adaptors.

The issue with one of the boards was I had a moment of "where's wally" after finding a small surface mount capacitor in with the packaging and then trying to figure out where it had actually some from. The package had seemed to have taken a pass the ball momemnt en route and dropped or hit rather than just poor soldering so happy in one regard, not so it turned up bust. This was with one of the 9.7V units.
 
Finally got around to using these things properly. Bullet proof.

Appart from the 7W loss on idle they are great and unbeatable for the price. Attach anything to them.

The main difficulty is soldering to the pins because the whole board is like a heat sink so you can end up with solder joints that are very brittle if you dont have enough heat.

If someone finds some push connectors that would work with these thinks that would be ideal... very recomended for constand draw from 48V packs.
 
Back
Top