Power Supply for capacitor bank

[Silent Jack]

Its supply is from either a 12V or 14.8V battery, I'm leaning towards the 14.8V.  The capacitor bank that the PS is supposed to charge is 2x 400V 470uF capacitors in parallel for 400V 940uF.  Am I misunderstanding, I thought the PS MOSFETs/IGBTs would need to match the capacitor voltage, do they instead need to match the battery?  Or rather are there different segments?  They are for switching, correct?

 

[Hero999]

It depends on the circuit, look at the maximum voltage across the MOSFET.

 

[Silent Jack]

I have a circuit diagram that I've been working off of to modify to meet my requirements.  Would that be helpful to post as far as discerning things?

 

[Hero999]

Look at the voltage across the MOSFET when it's in its off condition and it will give you an indication of the maximum voltage rating.

For example in a typical boost converter the maximum voltage across the switch (MOSFET, BJT, etc.) is no more than the input voltage.

http://en.wikipedia.org/wiki/Boost_converter

 

[Silent Jack]

The attached picture is of the diagram I am lableling pertaining to the modifications I am making so far.  It is incomplete.  I am still translating and adjusting.  I have to give credit to the design's author rwilsford, who shared it.  I have been cross referencing with other open and basic designs, along with datasheets to adapt it.  Sadly, there still is a good bit I feel I don't understand as well as I should like.

View attachment 41091

 

[Hero999]

It looks pretty straightforward and uses the same topology as the circuit from Wikipedia,  I posted above.

The 555 acts as an oscillator which runs until the required voltage is reached and is shut-off by the comparator. The shut down occurs when pin 4 is taken to 0V, which happens when the voltage on Rdiv2 is greater than the voltage on VR1.

I'd probably consider removing the 555 and using the spare comparator on the LM393 as the oscillator.

Why use JPG for schematics? It's designed for photographs and is lossy so goes fuzzy after a few saves, especially if the quality is set low. You should use PNG or GIF for drawings. I've converted your schematic to a PNG and removed the fuzz.

View attachment 41092

 

[Silent Jack]

I appreciate you reformatting the image to a more suitable file format so that it will be of more help to others reading our ongoing discussion.  How would you modify the circuit to use that portion of the LM393 as you propose?  I was unaware this was possible.  Also, since the bank is now a fraction of the size it was originally designed at,  I was considering adding a second identical bank and charging it alongside the other so as to reduce service downtime.  What is the simplest way to accomplish this?

 

[Hero999]

The LM393 can be configured as an oscillator so a 555 is not needed, here's an example.

Two capacitor banks can easilly be connected in parallel.

View attachment 41100

 

[Silent Jack]

Perhaps you could explain a little of how this works?  Also, what do all those open ends connect to in your diagram?

 

[Hero999]

What open ends? There's only one which is the output and that's because I forgot to label it.

U1 is an oscillator and performs exactly the same function as the 555 in your circuit, look up Schmitt trigger comparator oscillator on Google for more information.

U2 is a comparator which shuts off M1 when the output voltage is high enough, similar to the comparator in your circuit which stops the 555 oscillating.

V2 represents a voltage reference which can be an LM78L05, TL431 or just a potential divider from the 12V power supply.

 

[Silent Jack]

Thank you for the additional explaination, it is helping.  The part I asked about being "open" are the highlighted areas of this image.  I don't recognize this notation or symbol.  I've made certain to keep it in the more legible PNG format that you recommended.  Overall, I am endevouring to improve my overall understanding of symbols, circuit diagrams, and the flow and function of the circuit and components.  I appreciate your continued efforts in aiding my ellucidation.

View attachment 41103

 

[Hero999]

I tell you what: guess?

Even if you don't understand the symbol, it shouldn't be hard to figure it out by looking at similar schematics.

 

[Silent Jack]

Common ground, seems a sensible solution.

 

[Silent Jack]

Been reading through data sheets and IC layouts/circuit diagrams.  Would pins 1-4 be one half and pins 5-8 be the available portion that you mentioned substituting for the 555?

Went back and added the diagram I was referencing.  Thought it might help.

View attachment 41105

 

[Hero999]

Please think more before asking questions.

 

[Silent Jack]

Well, the diagram you posted has two LM393's much like the one I am working from.  Mine labels the pins used on each and uses differing symbols, suggesting different purpose.  U1, in my diagram has pins 8 & 4 connected; these are designated as Vcc & gnd respectively.  While U1.1 specifies that pins 1, 2 & 3 are connected; designated Output1, IN1(-) & IN1(+).  I note, based on the diagram that pins 1, 2 & 3 seem to correspond to pins 7, 6 & 5.  If these two pin sets are capable of the same things, and if as it appears there is no overlap in pin usuage, why are two used?  Would it not be possible to use a single chip to perform the needed functions?  Also, which portion of the IC supplants the use of the 555 you mentioned?

 

[Hero999]

Of course a single chip can be used, there are only two on my diagram because LTSpice uses separate symbols with their own power supply pins.

 

[Silent Jack]

Unless it were a topology issue.  I'll work on redrafting it and see how it works out.

 

[Silent Jack]

Another thought.  How much affect will the inductor have on nearby ICs and semiconductors?  Should I do anything to isolate or shield it?  And if the affects are significant enough on certain components, what is the most effective way to rearrange them?

 

[Hero999]

Please provide more information.