I would like to increase the current capacity of a DC - DC Converter.
The current capacity of the circuit I have is 5 amps. I would like to increase this to 100A. Here is the circuit View attachment ArduinoSolar.pdf
Apart from changing R1 to keep in line with the new current flow, or maybe changing the whole current sensor circuitry to a simple hall effect sensor.
And changing the FETs to handle the new current flow.
Does the inductor need to be a different value to continue to operate as normal with the new current flow? I know the wire size needs to be increased and that I will need a larger toroid to accomodate.
Your inductor may have the same value, but certainly would
need a lot more core and probably a lot more copper cross-
section. But it's also likely that 100A-capable power devices
will want to be switched slower (or, maybe you parallel a lot
of smaller ones with their gate drive split). That would drive
the inductor value up, if fSW goes down. Every core has a
volt-seconds limit, and your seconds/cycle are going up.
Start with finding a core that can take 2X your designed current
without saturating.
You could also look at making a parallel "power farm" based on
the existing design, replicated 20 times with the fanin load
appropriately reduced, and a single control head. This has
economies of scale, perhaps, and some measure of redundancy
/ ability to replace just one failed section (if you can find the
bad one, and it hasn't taken out its neighbors domino-style).
You could also look at making a parallel "power farm" based on
the existing design, replicated 20 times with the fanin load
appropriately reduced, and a single control head. This has
economies of scale, perhaps, and some measure of redundancy
/ ability to replace just one failed section (if you can find the
bad one, and it hasn't taken out its neighbors domino-style).
I was thinking this last night, just before you posted. Could you please clarify fanin load?
I am thinking that 10amp per parallel circuit. Just get mosfets that will handle 10 amp and toroids for 10 amp. Then I only need 10 in parallel.
If I am correct then I should breakout at the junction of C2 and R3. The parts I need to replicate and parallel would be as follows,
R3
R10
Q1
Q2
Q3
D1
D2
C7
C3
L1
Then create an identical circuit 10 times and connect all the L1/C3 junctions to battery positive.
Would I need a small resistance, say .01ohm, between each output and battery positive to balance the load in each leg? Like in a linear PSU with multiple transistors?
I think I may need some driver transistors for the PWM outputs, as the IR2104 may not have the capacity to run all the mosfets, is this correct?