Simple low current voltage doubling circuit needed

[Potato Cakes]

Hello, everyone,

I am in need of a low current (less than 10mA) DC-DC voltage doubling circuit with SMD parts to fit in a very small PCB footprint. I started out doing research to make my own but most everything I have found involves AC-DC. The small number of examples that I found which are actually DC-DC involve a MOSFET device that switches between capacitors in series, but no guidance on what voltage to feed the gate to create the switching effect. I do not have the time to figure this out and I also despise dealing with power in audio circuits which has become the biggest headache when trying to complete projects. Requirements for the circuit are as follows:

- Input voltage: 45-50VDC
- Input current: less than 10mA
- Output voltage: 90-100VDC
- Output current: 4-10mA
- Fewest parts possible for stable output voltage
- Circuit has to be tested and known to work as required

The last part is especially important. I do not have time to test and verify. A schematic with complete BOM including alternative part numbers is to be delivered upon payment.

Thanks!

Paul

 

[jp8]

Rethink your requirements: you want more output power than is available on the input side.

Jan

 

[gyraf]

If for phantom power, note that at 10mA you have almost no voltage..

/Jakob E.

 

[Potato Cakes]

Well, let's start with just getting voltage to double. The I think the current is actually less than 4mA. If someone is interested in taking this on please let me know.

 

[gyraf]

Important question - should it run on 48V phantom power? This is very different from "running on 48V"

 

[Potato Cakes]

It should run on phantom power.

 

[jp8]

It should run on phantom power.

Phantom power supply as available on a mic preamp input? I.e. through the 6.8k resistors? If so, mission impossible as @gyraf hinted to in post #3. If powered directly from the P48 supply, then reconsidder the amount of current available to the circuit to ease the circuit design.

And how much ripple is allowed on the output?

Jan

 

[JohnRoberts]

Hello, everyone,

I am in need of a low current (less than 10mA) DC-DC voltage doubling circuit with SMD parts to fit in a very small PCB footprint. I started out doing research to make my own but most everything I have found involves AC-DC. The small number of examples that I found which are actually DC-DC involve a MOSFET device that switches between capacitors in series, but no guidance on what voltage to feed the gate to create the switching effect. I do not have the time to figure this out and I also despise dealing with power in audio circuits which has become the biggest headache when trying to complete projects. Requirements for the circuit are as follows:

- Input voltage: 45-50VDC
- Input current: less than 10mA
- Output voltage: 90-100VDC
- Output current: 4-10mA
- Fewest parts possible for stable output voltage
- Circuit has to be tested and known to work as required

The last part is especially important. I do not have time to test and verify. A schematic with complete BOM including alternative part numbers is to be delivered upon payment.

Thanks!

Paul

10mA output at 100V is 1W of power. The maximum power you can pull from 48V phantom is around 170mW @ 24V

so a perfect (lossless) switcher will only get you around 1.7 mA @ 100V, actually less.

JR

 

[Potato Cakes]

Well, let's start with doubling the voltage even thought the output maybe slightly less than 1.7mA. I'm looking to see if someone who has power supply skills can take on this project and to give me a price. If it doesn't work in this particular instance then I can use if for other applications where the source current is higher.

 

[Newmarket]

The last part is especially important. I do not have time to test and verify. A schematic with complete BOM including alternative part numbers is to be delivered upon payment.

Alternative part numbers to what part numbers ? Do you mean you require a choice of components that can be substituted / used interchangeably ?
Do you have a ball park budget or maximum in mind. Quite a job if you require prototype build and test results.

 

[ccaudle]

let's start with doubling the voltage even thought the output maybe slightly less than 1.7mA

Doubling which voltage? The 48V is fed through 6.8k resistors, so as soon as you pull any current the voltage is no longer 48V.
If you double the realistic voltage you can get, you will be doubling around 24V to get 48V output. Is that what you want, or do you actually want to quadruple the voltage?
Or to state it a different way, do you actually want to double the voltage, or do you actually want 96V output? Because those two are not the same thing due to your misunderstanding of how phantom power works.

 

[Potato Cakes]

The idea behind an alternative part number is to be able to find components in case of components being out of stock. This mostly pertains to semiconductors. I can find different capacitors and resistors on my own if need be.

Further discussions regarding things like budget should be done via PM. I am trying to find someone who could take on this job and details should be discussed and amended between myself and potential hires.

Thanks!

Paul

 

[JohnRoberts]

Since a well executed design will never deliver more than 1.7mA @ 100V is that enough current?

To extract peak power from Phantom power you need to load down the two 6.81k resistors to 24V. Then you need to step up that 24V intermediate supply voltage up to 100V, not with a voltage doubler but a dc to dc switching "boost" convertor.

For only modest current output perhaps look at "charge pump" technology.

How well defined is your application... Is 1.7mA (or less) @ 100V really enough?

JR

 

[Potato Cakes]

This initial application is for the polarization of a rear microphone capsule.

 

[JohnRoberts]

OK that sounds like low current...

JR

 

[gyraf]

Perhaps everything would be easier if you clearly state what your aim is, what exactly you're looking for, under what limitations

Otherwise this will quickly turn into a XY-problem

/Jakob E.

 

[ccaudle]

for the polarization of a rear microphone capsule

That is not very closely related to the specifications you started with. I first thought maybe you were trying to power some kind of miniature vacuum tube or something, the output current requirement in the original spec is off by around 1000x for polarizing a microphone.

That is also a solved problem in every multipattern microphone. Is this a rear only cardioid mic, or is there an existing polarization circuit for the front diaphragm? Assuming the mic also has a front diaphragm, it might be better to tie in to the existing circuit to generate the additional voltage.
Is the front diaphragm polarized to -100V? Presumably if you need 100V, that is because the front diaphragm is polarized to -100V, and you want to be able to create a figure-8 pattern. Is that correct, or did the original spec miss the polarity requirement, and it actually needs to generate -100V?

 

[Potato Cakes]

As always, I greatly appreciate the responses and insights from those far more learned in circuit design than I will ever be. For the many inquiries that I have posted in the other forums looking for assistance with problems which I was solving on my own, I am forever grateful to the patience and insights generously shared with me. But for this particular project, I have come to the edge of my understanding and the chasm to get to the other side is currently too great to traverse on my own. So I posted here in hopes of hiring someone to free up my mental resources to focus on finishing current projects. So again, if anyone is interested in helping with getting this circuit designed please PM me.

Thanks!

Paul

 

[ccaudle]

hopes of hiring someone

Then I would recommend re-writing your request in terms of what you actually want to accomplish. As a working electrical engineer, I look at the first request and my impression is something like this person doesn't really know what is realistically needed, I don't have time for a project request which is physically impossible.
As currently stated you risk the most competent people skipping over your request as obviously impossible, and only finding someone who isn't skilled enough to catch that right away attempt to give you what you asked for (which is impossible) instead of what you need (which is probably just a CMOS inverter charge pump, like on Rog's OPIC pages).