DC-DC Converter

You mean something like this?

http://www.exhaustgas.com/ProductDetail.asp?ProductID=1815

Thanks for your reply,

In my project, My task is to convert voltage from a pyro material to a usable Voltage using Step up converter in pspice. i am done with getting the voltage from the pyro material, its about 0.2V to 0.8V around..with very current..milliamps...may be micro amps...Now i need a boost converter so that it hooks up the voltage to like..3V..so i am looking around for a good design of boost converter..i need your sujjestions and help..Thank you..

http://jonathankimball.com/pdf/low_input_boost_PESC.pdf

Boost converter design....

http://www.daycounter.com/LabBook/BoostConverter/Boost-Converter-Equations.phtml

You cannot get something for nothing. If your only power source is capable of less than a milliwatt of power (0.2V*2mA=0.4mW) then that's all the power available to you. So if you could could get a 100% efficient switching supply to work at these levels (not very likely) the most amount of current you could get at 3V will be hundreds of micro-amperes at best.

You could try a Joule Thief but I doubt it would work below 0.8V

Considering that most semiconductor junctions require about 0.5v to function then I think you will be struggling.
Maybe you can connect several of your devices to give you a more useable starting voltage. I'd suggest 2v is a reasonable workable minimum unless you are an expert electronics and semiconductor designer.
Del

the DC Dc convertors I have used have worked in th eother direction. I needed to guarentee 24VDC from a supply that varied from 28VDC to 36VDC. The convertor gave out exactly 24VDC regardless of the input volts.

Remember you cannot get out more energy than you put in. Normally to increase volts you must transform it with a transformer - an AC device. Although the volts would increase your current has to decrease - you cannot get more out than you put in.

Rock on with PSPICE!

Thanks Winfield..it was helpful..But i am in still confusion when i am doing through simulation..as u showed in a green page..Circuit defining for energy harvest, my source is a 0v to 0.8V pulse source with milliamps..can u draw the actual circuit how it looks..is it the same joule thief circuit..and can you tell me what is the .model statement for the depletion mosfet for this application

Thanks

Whoa, you would like everything handed to you on a platter! I like SPICE, but it has its limits. This is especially true when it comes to models. With respect to the little circuit I showed you, I would feel better if you purchased a few parts and measured the relevant parameters, g_m and V_GS vs I_D, for example, rather than asking about SPICE models.

But since you asked, most SPICE MOSFET models are a disaster at low currents, where linear operation happens, because it's the subthreshold region. Bottom line, you have to create your own special models.

You may have better luck with JFET models. One important thing that's usually missing from both model types: noise.

Without noise you can't hope to SPICE-model the circuit, because it won't start oscillating. What's more, without all the parameters just right, it won't bear much resemblance to the real circuit. You'll probably be better off with analytical approaches to analyzing the circuit.

Hi, I know this post was some time ago but now i'm trying to implement same concept.

But I dont want to depend on a IC like LTC3108 to do all work, because only low current possible.

I also saw a new IC BQ25570 which is able to do a very efficient harvest from solar panel and charge battery and also provides balanced output to battery and load.

The schematic u draw with transformer is very good boost converter, I'm planing to use that concept because simplicity and low cost. But maybe less efficient than new concepts because transformer coupling losses. The BQ25570 has independent buck boost controllers one coil to store input energy and other coil to store energy at output.

The boost topoly using transformer and resonance based on Lsecundary C is simple, stable and low EMI emissions( I think so, because buck boost controllers should have some kind of filter to eliminate peaks on mosfet transitions...)

Anyway, the transformer can be designed directly on pcb one coil inside other (outide primary) Just need to calculate impedance from pcb coil spiral primary and secunday it can be done easy here http://webench.ti.com/wb5/LDC/?DCMP=ldc&HQS=sva-psp-ssp-ldc-awire-20150819-lp-webenchcoil-wwe#/spirals

This way allows to do low cost tranformer for own propouses. Since coils are planar and one in center of other coupling is greater and also can use ferrite disk to force field at pcb level.

Any reports about this kind of planar tranformers are welcome

Back to tranformer boost topology, which considerations should I have to have max energy transfer? The primary impedance plus mosfet rds on should be equal to voltage source internal resistence to archive max power tranfer?

Hope this topic keeps views nowdays

Regards