Paul and Sabrina's Cheap 3 Phase Inverter (AC Controller) with Field Oriented Control

[cts_casemod]

Crap thats big!

I need to put you two in the littlebox challenge!

https://www.littleboxchallenge.com

[P-hack]

Yah, high frequency is another brain bender, but that is where it has to go for smaller size. Still an 8" inductor for ~60kw to replace a lot of extra batteries seems to have its potential applications. Also the target has been a bit arbitrary, i.e. a specific motor might not need that much, another motor might need more. Paul, do you have something specific in mind motor wise? It is roughly toyota sized.

[cts_casemod]

Yes, I think we need to design something based on a real world scenario.

Technically Paul is designing this to allow 208V motors to be operated on 144V packs.

A 208V motor being operated at 300V is never going to draw 400Amps. Not even motor amps. Increasing the voltage will increase HP, but the amps will stay the same.

At 20HP the maximum battery current draw would be about 150Amps. A 18HP motor (50Hz) weights roughly 95KG. Anything above that is casted steel rather than alluminium, which is probably out of the scope for EV conversions.

have a look at joannes graphs on DIY Electric car. I believe his motor is 15HP running at nearly 500VDC.

I run a 10HP 208V motor at 409V and anything above 70A (battery side) is waste.

So perhaps an output of 200Amps at 600V would be more suitable for a high performance conversion, with most staying at 100Amps max.

The question now:

Is it really worth to maintain compatibility with a 144V system for those that are looking for a medium size performance (up to 60KW)? Probably yes!

Is it really worth to mantain compatibility with a 144V system for those looking at high performance (100KW or above)? Probably not!

That would allow a downsize from 400 (initially 600) to 200 input amps.

[MPaulHolmes]

Battery pack ------- inductor ---------- middle of 4th IGBT Half Bridge.

Top of 4th IGBT Half bridge is connected to the bus capacitor for the 3 phase inverter. Bottom of 4th IGBT is connected to ground.

Then you have the other 3 half bridges for the inverter. So, unless I'm misunderstanding something, the current through the inductor is the battery amps. So, for 144v, 60kw would require around 400amp through the inductor, unfortunately. But only 200amp for a 300v system (to boost to 600v 100amp).

So, you have 409vBattery * 70ampBattery = 29kW? What is your top speed?

[cts_casemod]

Originally Posted by MPaulHolmes

Battery pack ------- inductor ---------- middle of 4th IGBT Half Bridge.

Top of 4th IGBT Half bridge is connected to the bus capacitor for the 3 phase inverter. Bottom of 4th IGBT is connected to ground.

Then you have the other 3 half bridges for the inverter. So, unless I'm misunderstanding something, the current through the inductor is the battery amps. So, for 144v, 60kw would require around 400amp through the inductor, unfortunately. But only 200amp for a 300v system (to boost to 600v 100amp).

So, you have 409vBattery * 70ampBattery = 29kW? What is your top speed?

Probably the same as a stock polo with a 33KW 1Liter petrol engine. How much do people with a 500Amp DC motor pull at the highway?

Max top speed I reached was 45MPH miles on a test run in 3rd gear at 2500RPM, final current was 40Amps. Speed limit was 40MPH. I never really took the car to a real motorway, the closest is like 15 miles from here.

Converter wise:

The battery will provide a total of 400Amps @ 150V = 60KW
200Amps go directly to the output, 200Amps feed the boost converter.

The boost converter will provide 150VDC @ 200Amps = 30KW

The output will be 200Amps @ 150V from the battery + 200Amps @ 150V from the boost converter = 60KW.

Double the voltage, double the power equations above, to 120KW.


This differs from a transformer approach, which might be causing some confusion. If one were to use a transformer rather than an inductor, 100% of the power would have to go trough the transformer, therefore the output of the battery would be 400Amps @ 150V = 60KW and the output of the transformer (1:2 Ratio) would be 300V @ 200A = 60KW.

[MPaulHolmes]

I think freeway battery amps at 144v are around 60-70amp for a DC controller, for constant cruising. I'm not sure about that, but I've heard that thrown around.

Oh I hope you are right. I just don't understand it though. I mean, if I have an inductor in series with the the battery bank B+, current out of the battery is the same as the current through the inductor, right?

[P-hack]

that is causing me some grief too, every boost simulation pegs the average inductor current as being equal to battery current (I checked last time you mentioned it), are we not counting dc bias current for some reason? We do that when determining effective inductance (and conductor area).

[cts_casemod]

Originally Posted by MPaulHolmes

I think freeway battery amps at 144v are around 60-70amp for a DC controller, for constant cruising. I'm not sure about that, but I've heard that thrown around.

Cool, how much will they peak? I heard 15KW is enough to 60MPH cruizing in a straight line, but I am curious how much acceleration they can have.

I can only tackle very step hills at 24-28MPH, with a continuous current of about 45Amps. More would overheat the motor, unless it's just required for acceleration during a few seconds, for example, after a stop. This is the good thing about voltage - reduce one gear and use the extra HP at higher voltage and less current!

Originally Posted by MPaulHolmes

Oh I hope you are right. I just don't understand it though. I mean, if I have an inductor in series with the the battery bank B+, current out of the battery is the same as the current through the inductor, right?

When the freewheeling diode is active, current is the same as the output. 200Amps. The inductor current sums with the battery voltage to provide the output.

When the switch is active the diode is reverse biased and the current flows trough the inductor instead, with the load being supplied by the capacitors.

So you have a constant output from the battery. 50% of the time it feeds the load in series with the inductor, while the remaining 50% is charging the inductor by building its magnetic field.

Notice on post 1525 the output current is just slightly less than the peak inductor current, not twice. This has to do with the dead time between charge and discharge.

If we want to be really scientific, yes for a 50% duty cycle and an output of 200Amps one has to transfer 400A during half the time. But the average continuous current is still 200A. In fact 200Amps AC.



Bottom end: This works similarly to a flying capacitor setup.
The inductor is providing half of the output and therefore only half the voltage, at the same amperage as the output. The output voltage comes from the sum of the voltages from both the battery and inductor, but since the battery is feeding both the load and the inductor its effective output is twice the output amperage. Energy conservation law

[freebeard]

Oh I hope you are right. I just don't understand it though.

I think everyone is filtering what is really going on through a Dominate Paradigm. The guy who figured this all out was Steinmetz. It's described in the first hour of Eric Dollard's History and Theory of electricity.

Meanwhile, I keep bumbling along—I see the current episode of EVTV addresses cases of transmission failure in conversions (and Teslas?), and suggests that they're built to absorb power that 'spools up' over ~500-800ms, while the electric controller can force full power in ~5-10ms, leading to shock failures. It's at 1:25:30 as part of a typically long-winded discussion of sizing motors and multi-gearing vs single speed.

[cts_casemod]

Originally Posted by freebeard

I think everyone is filtering what is really going on through a Dominate Paradigm. The guy who figured this all out was Steinmetz. It's described in the first hour of Eric Dollard's History and Theory of electricity.

Meanwhile, I keep bumbling along—I see the current episode of EVTV addresses cases of transmission failure in conversions (and Teslas?), and suggests that they're built to absorb power that 'spools up' over ~500-800ms, while the electric controller can force full power in ~5-10ms, leading to shock failures. It's at 1:25:30 as part of a typically long-winded discussion of sizing motors and multi-gearing vs single speed.

Thats because most controllers are VSI (voltage source inverters). For an induction motor any small percentage of error on the slip angle can cause huge amounts of torque for a short duration. I had some regeneration control issues due to that as well. An error in the control loop or the field building up very fast during regeneration could cause short bursts of very high torque.

The solution is to use a VSI, but add a cycle by cycle current limit, which is generally implemented in hardware. (software can generally do average current only). I have that on my TODO list for my inverter.