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

[P-hack]

Just a reminder, we had arbitrarily set the upper limit @ ~36kw for a 144v battery, so that limit has been my focus primarily (510v@70a to beat on some 18kw 200-240v motors with at 2x rated speed), which I assume you would gear down for 2x torque if you like.

I do like the boost topology best (short of rewind), it beats on the source the least and allows a direct path to the motor when all those volts aren't needed.

Can you expand on the center tap=3 igbt idea?

[cts_casemod]

Right, lets stick with that then.

Originally Posted by P-hack

I do like the boost topology best (short of rewind), it beats on the source the least and allows a direct path to the motor when all those volts aren't needed.

Buck works just as well. The inverter is buck by default.
With a mirrored supply, the ground is floating when the converter is operating. When not, a reverse diode across the capacitor completes the path. The output will be Vbat - 0.7, just like in a boost converter.

In case of a failure on the power switch, the input gets shorted to the output trough the inductor rather than shorting the supply.


Ignore the 3 IGBT's idea. Just me thinking out load.

[P-hack]

Originally Posted by cts_casemod

In case of a failure on the power switch, the input gets shorted to the output trough the inductor rather than shorting the supply.

This is what I know as an inverting buck boost (not a buck):

So what happens to the output when the switch gets shorted? zero volts you say?

[cts_casemod]

This diagram will help visualize what I have in mind.



The addition of the extra IGBT has two functions. If the converter ever goes offline for some reason, the current will flow trough the IGBT diode into the inductor and closes the loop to GND. The same happens if that particular switch shorts (shorting the primary would make smoke as mentioned).

The orientation means a standard half bridge could be used, so it would simplify the design, with the center point attached to GND via an inductor and the negative attached to the DC-Link as the motor IGBT's. A 4 IGBT approach as per Paul's idea, but requiring no other extra components, other than the inductor.

[P-hack]

please don't call that a buck converter more like an upside down boost converter with the smoothing cap in the wrong place(b). Many of the topologies are electrical anagrams of sorts, but the terms have specific meanings. ref: Regulator Topologies for Battery-Powered Systems - Tutorial - Maxim



also these and their right side up counterparts can all use a half bridge. The question is can they still with zvs/or zcs or whatever (though zvs needent be a requirement, only a lot of volts at 36kw is).

[cts_casemod]

Thats a buck boost converter as described on the document I posted on #1599. The second diode is replaced by an active switch.

The output is a negative voltage with amplitude from 0 to -Viin, which can be used to sum with with the input.

Are you saying Texas instruments is using the wrong terminologies!?

[P-hack]

I think what you are trying to say is you can use a buck CONTROLLER in an inverting buck boost, but then you drew an upsidedown (inverted) boost as referenced in the maxim document. How you arrange and connect the components matters. You didn't connect the load to v-out and ground, you connected it to v-in and v-out(and now the capacitor is out of place). I thought you were talking about a buck converter, then an inverting buck/boost converter, when really a low-side boost converter is what you meant (with the inductor and diode on the bottom and the diode reversed). Instead of boosting the top rail, it boosts the bottom rail more negative.

note: either boost or inverted boost can buck in the other direction if you use a half bridge. There is no forward buck, different control scheme. review (b) in the previous picture.

[cts_casemod]

I disagree. The output of my circuit is a negative voltage. The capacitor is not reversed. It is polarized to receive a negative voltage.

The output is -Vin and GND = -Vbat.

If the output is attached to VCC rather than GND, the output is +Vin and -Vin = 2*Vin



One could also have an output of 2xVin using this circuit.



Both are buck-boost converters. They are not a pure buck, they are not a boost either and most certainly the components are not in the wrong place!

The maxim regulator you posted takes a negative voltage and converts it to a positive, hence the component values are the other way round.

[P-hack]

what maxim called an "upside down" or inverted boost, ti (probablydepending on author) calls a negative boost:
http://www.ti.com/lit/an/slyt516/slyt516.pdf

"1. The converter should have external compensation to
accommodate the different control algorithm associated
with the boost converter, which will be discussed later."


cout isn't doing anything very useful here, needs to be configured as a normal negative boost converter:

[P-hack]

one more try at explaining it attached (ground symbol doesn't matter):