USB Syncroized 5V@10A power supply POL. Now Taking Orders!!! - Page 7

Beehphy · 06-02-2008, 03:10 PM

Latest project update:

I have built and partially tested the new revision of the ACPPSv1.

Better, stronger, faster… but surprisingly not bigger.

Rev 1.03 is now rated at 10A output! (previous revisions were 7.5A). The output is steadier and quieter. The operating temperature range has been extended to -40c to 65c (-40 to 85c for less then 5A output).

I may be making another production run in the near future. Contact me if you are interested.

Beehphy · 07-11-2008, 04:24 PM

Bump!

FYI,
Parts and PCB's have been ordered to construct 15 units. It will still be a few weeks before they are ready to ship.

Thank you everyone for your patience.
Jesse

FockMasterDan · 07-11-2008, 04:40 PM

The V2 sounds very interesting for me, as I'm looking to get regulated 12 as well as 5. You mentioned increasing the efficiency of the V2, and by that do you mean it will decrease input or increase output? I'm just not sure if it would have enough juice to be everything that I want, but it sounds like it would be plenty.

My issue is because I'm building a laptop based setup, I don't have a nice PS to run extra hardware off of.

Beehphy · 07-12-2008, 11:26 AM

Quote: Originally Posted by FockMasterDan

The V2 sounds very interesting for me, as I'm looking to get regulated 12 as well as 5. You mentioned increasing the efficiency of the V2, and by that do you mean it will decrease input or increase output? I'm just not sure if it would have enough juice to be everything that I want, but it sounds like it would be plenty.

My issue is because I'm building a laptop based setup, I don't have a nice PS to run extra hardware off of.

My big thing with efficiency is because of heat. I have designed a ~50W power supply with no fan. Without active cooling the PS needs to transfer that heat away by natural means. Unfortunately natural convection only moves so much heat. If you create more heat than can be pulled away, eventually the magic smoke makes its way out.

The ACPPSv1 has ~93% efficiency, meaning 93% of the power entering the PS is leaving through the output as electricity, the other 7% is being wasted. Some of it is coming out as light from the LED, but primarily the waste is comming out as heat. So, when the ACPPS is opperating at 50W, 3.5W is turning to heat and needs to be removed.

With the ACPPSv2 opperating at 85% efficiency and outputing 50W, there is 15% or 7.5W of heat being generated. Unfortunately 7.5W is edging on too much heat to be pulled away by natural convection. And if I can't find a way to get it over 90% I'll have to find a way to cool it.

So to answer your question, with v2 I'm looking to reduce the input required for the same output and at the same time reduce the heat generated at full output.

The ACPPSv2 will be a 50W power supply, just like v1, but that power will be split between the two outputs. 25W to the 5V output and 25W to the 12/18/19V output.

Power = Volts x Amps
Output 1: 25W = 5V x 5A
Output 2: 25W = 12V x 2.08A or 18V x 1.39A or 19V x 1.32A

So, figure out what components your system will include (GPS, OBD2, CD-r, Audigy etc.) and add up all the power requirments to find out if my supply will suit your purpose. If you have any problems doing that contact me. It would be a great tutorial to add to this thread.

fredsub · 07-12-2008, 08:23 PM

congrats on your successfull project,
i'm going to build my own tho, because I can.I need 15V@6Amp

can i ask what controller your using?

BTW, for the 5V, if you use a sychronous switcher seems you could get efficiency even higher, maybe 95% ?
12V would still need to be a separate buck/boost circuit tho.....

Beehphy · 07-13-2008, 10:02 AM

Quote: Originally Posted by fredsub

congrats on your successfull project,
i'm going to build my own tho, because I can.I need 15V@6Amp

can i ask what controller your using?

The controller I'm using for the v1 dosen't do buck boost, flyback, or SEPIC, like you'll need to get 15V. But its a TI sync buck. In the v2 i use an LM3488 in a sepic configuration. but that chip cant handle an output as bigs as your looking for.

I would look at national and TI (maybe Maxim/Dallas) for controller chips. they seem to have more bang for the buck.

Quote: Originally Posted by fredsub

BTW, for the 5V, if you use a sychronous switcher seems you could get efficiency even higher, maybe 95% ?
12V would still need to be a separate buck/boost circuit tho.....

r1.03 runs at 93.5%, while there are lights and relays in my circuit stealing power. That efficiency is for the entire unit, wires in to wires out. The 95-6% claimed by most controllers is for the switcher all by its self with no support circuitry.

I have gotten the design up to 94.5% (whole unit) in some tests with certain configurations. unfortunately the extra 1% of efficiency comes at a high cost. it added almost 10% to the cost of parts for each unit. I made a comprimise to keep the cost down.

You dont have a problem tuning for peak efficiency, a 1 off power supply does not have to be production friendly.

syberkowboy · 08-04-2008, 04:35 PM

Hey Beehphy is the aux output auto selected or is there some jumper inside. I'm putting a 12v LCD on the roof and want a regulated PS to keep it clean. I really like the unit and am hoping to be able to use if for: DVD drive in dash, hub, ?external HDD?, and maybe the LCD. Haven't done the wattage calcs yet but am still in the plan stages for my second carputer.

Beehphy · 08-11-2008, 12:18 PM

Quote: Originally Posted by syberkowboy

Hey Beehphy is the aux output auto selected or is there some jumper inside.

Sorry bud, You have a v1 which only has 5V regulated. The 12V output, is just the 12V input switched to that pin when the ACPPS is turned on. Any fluctuations of the cars power system will be present on that 12V output pin.

The Prototype ACPPSv2 has a jumper that selects the output voltage.

On to other topics.

The next run of ACPPSv1 r1.03 is almost ready to start shipping (See the stack of circuit boards below). There have been some improvements and refinments since the last run that make the current models better in many ways. The most recent improvement is the Finish on the case, It is now completely poweder coated, for improved durability.

I had to machine 200 end caps (seen in the pic) to make the best use of the $200 minimum order at the powder coater. It took me ~80 hours to machine all of the end caps, by hand on my minimill. I spun the handles of the mill aproximately 80,000 times over the two weeks while I was building them. I had to jimmy rig a power drill to help me with some of it (one particular step of machining took ~40,000 turns of the handle all by itself).

And after all that work, I've decided I'm upgrading my mill to NC/CNC control. It will cost ~$1200 to do all the upgrades I want, but It will be well worth it. The end caps will be more accurate and take less time to cut. But, beyond making production of ACPPS easier, the CNC mod will make all my projects more refined.

Also, this production run, the circuit boards have silk screens and solder masks. The previous run, the PCB's had exposed tinned copper traces. With a solder masked PCB it is more difficult for a piece of debris to short things together if it gets inside.

The new design (r1.03) still includes the expanded temperature range and smaller output ripple voltage as discussed in previous posts.

I should (hope) be sending out bills for the preorders later this week. If you are interested in purchasing an ACPPS, now is the time to make your reservation.
Jesse

PS. Sorry the pics are crap, they were taken with my phone.

Beehphy · 08-14-2008, 02:11 PM

ACPPSv1 is now for sale again.

There's a (new) good pics in the first thread and the manual has been updated.