High output current charge pumps?

[DuctTapeRiot]

I have been doing some searching but am seeming to get nowhere, and am going crosseyed from reading data sheets at the end of the work day.

Is it possible and practical to build a charge pump that:
- outputs 15v
- Can provide 400-500 mA output current. (looked through all the roadrage stuff, LT1054 puts out the highest current at 100mA)
- Takes a 9V input.  Also current consumption is key here, no point in even doing it if its going to pull like 2 full amps or something.

The idea is, for a pedal like a Radial Tonebone which takes 15v and claims to pull 400mA, to be able to build a little charge pump in a box to put between the 9v output of say a 1-spot or other power brick and the Tonebone.

Am I crazy?

[jubal81]

You're going to have to buy a 15v wall wart.

[Mike]

I think what you want is available, but expensive. After a quick search of Mouser, I think this would work.

-9-18V input
-15V output
-Current up to 533mA
-79% efficient (my math says 500mA will pull 633mA)
-50kHz switching
-DIP

Only $16.79. 

[DuctTapeRiot]

Thank you Mike, thats a great find (although they are actually $21, the $16 price is if you buy 500+).  So a bitimpractical in terms of cost and current comsumption, but definitely possible. Given a one spot puts out like 1700mA, it sounds like it could go on a daisy chain tap and still have lots of current left over.  Or it could definitely be used with one of the beefier power bricks like a Gig Rig Generator, a Yankee PS-M2 (which already outputs 15v) or a Power Pad II.  Hmmmmmmm

[Mike]

Heh... sorry. USD.

They have a bunch more, too, but you'll have to be a Mouser search master to find them, and then uncross your eyes to read the datasheets!

Mike

[stecykmi]

if you find one that "almost" works but is short on current capacity, one option might to run two of the devices in parallel two get (near) double max current. the LT054, and a couple of the other similar devices we often use, have provisions for parallel operation listed in the datasheet, it stands to reason other devices might as well.

that said, you'd have to run 5 of the LT054 in parallel to get the 500mA you're looking for, so the search for a device continues.

by the way, half an amp is a lot of current for an IC to handle, most power supplies with that requirement would probably use a solution with offboard MOSFETs. if you expanded your search (and are okay with a more complex solution), you might look at switching controller ICs.

[midwayfair]

That thing at mouser is also almost the size of a wall wart. It's 31mmx20mm!

[RobA]

There are several IC's that could get you the current you need. They'll be a bit more work to get tested and laid out, but you should be able to get a working solution with them.

The easiest might be the MC34063. It's like a charge pump, but inductor based.  say might be because I haven't actually breadboarded with it yet and I don't know if the noise level is going to be acceptable. TI's got an application note for it that describes how to use it http://www.ti.com/lit/an/slva252b/slva252b.pdf.

LM2735 and LM27313 are related boost and SEPIC converters that could do that job. I've played with these on breadboard and they do work. Layout and part selection (all SMD) is critical with these. But, there are very detailed spec sheets and app notes for them with specific instructions for layout.

There's another path that I've used that is really good for the clean power it delivers, but it won't quite get the power you want (350mA max). But, the idea is very simple and well described in the spec sheet and you may be able to achieve the current you need using an MCU or a timer (555) to get the clock signal and then a MOSFET power transistor to drive the switching. The complication with this one is finding a transformer that's got the Volt-time capabilities you need with the right primary:secondary ratio and is small enough to not take up too much room. The IC is the SN6501. It's the IC I used for the LiPo to 9V battery project I put up a while ago. For that one, there's a readily available transformer to do the job and it is tiny. The whole board including transformer, caps, voltage regulators, and connections ends up being about a quarter of the volume of a 9V battery.

I'd love to try to work out a higher power version of the SN6501 circuit but based on a clock and discrete components, but so far, my transformer finding skills have been too weak. If anyone can point me in the direction of a good transformer to use for this, I'll happily try to design a circuit.