Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
Nominally equal panels may exhibit a lot of variation in output
if any shading or temperature difference. Without at least some
blocking diode apiece, you could see current circulate between
panels rather than all going to the inverter / charge controller
(or internal junction currents)...
Using a discrete FET for a ~1A output and near 1:1
ratio seems excessive. Easy enough to integrate
at a ~ 2A pulse / avg current level.
I imagine that getting across the isolation boundary
forward and back, is the greater challenge. But might
consider something like just hard chopping the VIN...
Maybe since you do not know, consider modules which combine the two. 100kHz does not suggest WBG devices to me and 200V is pretty low end for SiC. Gate loop design/layout is a solved problem in a module.
You think the industry is monolithic? Somehow moving in unison, or should? Toward a common goal and solution?
Grappling at every level from factory to design shop, over nickels and dimes is more like it. ;
Might suggest a DPDT relay and using diode logic on the coil.
Bidirectionality in the current path comes for free then.
Coil current vs "electronic commutation" circuit current
and complexity is your trade.
Energy has to go somewhere.
Your output section lacks the "flywheel" inductor that would force secondary current to circulate, so energy will escape the core through the primary instead of bleeding the core between power "slugs".
Also the output looks like no DC path to ground. Primary neither.
Question what a long chain organic molecule would be in
die attach paste for. I prefer eutectic DA, will use Ag-glass
if I don't care about ohmic connection to the "handle".
Seems more like a virtue-signaling marketing thing than
any assembly-process value. They're selling "free" and it's...
You need two things.
A value proposition and a customer.
There are high fixed costs once you start getting real.
Probably > $1M to get from fog to qualified chip that
can go onto a board and work. Do not neglect the
design-in collateral, which can be a huge added cost
when you're pushing...
For abuse testing I think a boat anchor iron 50Hz
transformer is not a bad idea. Something that you
can "screwdriver short" and not have to stuff the
smoke back in.
You might find a way to use one of those with the
mods / approaches I suggested, to fix the line
variation interest.
As...
Sheet rho and tempco are the big deals. Tolerances maybe, if they go to pin electricals.
I prefer isolated thin film resistors,(here, your poly types) except in ESD / high current roles where pulsed thermal input might "program" a low volume, poor thermal conduction thin film resistor (aka fuse).
It's always good to knock off the easy ones first, and see what else then gets "found". Like those missing terminals and the single connection net. These are probably real and fixing them may make other things fall into place.
Lather, rinse and repeat.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
