Adjunct battery packs?

[danielsantos]

It's available on Ebay from salvaged cars. As I remember it costed me about $200.
And also I found original socket partNo (maybe it would be cheaper then buying whole cable from salvaged car)

Very awesome! I've recently discovered salvage parts on ebay; I got an entire "heating and cooling unit assembly" (the air handler in the cabin with blower, ducts, expander coil, etc.) for $65! Thanks for that tip!!!

I tried SPICE, but had difficulties trying to apply all MOSFET parameters from the datasheet.
In addition, the transformer parameters are unknown.
I think it's almost impossible to have the reliable simulation model.

There is MUCH to be said about doing a real-world test, but I'm a devout adherent to modeling. But without having the data on a part, I agree that it's almost impossible unless you test the part to discover it's characteristics. When I worked on an automotive transient protection circuit, it was handy because the standard transient protection faults (load dumps, etc.) are already coded into SPICE. Many electrical component mgfs supply SPICE models for their stuff.

As I know, in order to achieve more efficiency you need to lower a frequency of the DC-DC, and that reduces it's maximum power.
And building high-efficient 5kW DC-DC would be really tricky (also it will be large and expensive solution).

Too bad my oldest brother and I have been estranged since he turned into an arsehole. He's very good at switching PSUs. But there are many factors that go into switched PSU efficiency and while I'm not an expert, I know that there's:

1. RDS(on) - the drain-to-source resistance that looses power when the diode is on
2. the reverse recovery time and charge (trr and Qrr)
3. and the rise and recovery time (tr and tf).

I know there are MANY characteristics I may not quite appreciate just yet, but I intend to learn all of this in the coming months. Most of the power is lost when switching on and off because we go from very high to very low resistance and in that transition, we are converting electricity to heat. For example, the UJ4SC075005L8S has a rise and fall time of 39 and 13 ns, respectively.

I believe that using a lower switching frequency will result in a less stable voltage. I'm sure a capacitor can even that out but I'm not familiar with every consideration of this, beyond the fact that capacitors tend to have 90% efficiency so there are losses involved in that too.

I didn't got the idea.
In case you have auxiliary 400V battery, it's really problem if your battery is completely independent of the main battery.
Common solution is to have the same chemistry as in the main HV battery and switch them always together: they will have the same voltage, they will charge and discharge together, and in this case you will have very simple switching electronics.

I suppose that's the "bread and butter" of this project.

 

[FrancoVecchi]

I just saw this discussion. I also believe that a bidirectional DC/CDconverter is a good solution for an extender, the possibility of communicating with the machine's CANbus and regulating the withdrawal and charging of the extender's battery is very interesting. In terms of performance, I have created in past, a prototype of a non-isolated monodirectional converter of approximately 3KW with efficiency greater than 98% using SIC-MOSFETs. It wasn't easy, but now that I've discovered various problems with driving the MOSFETs, I think it's doable.