Three phase high power low voltage high currents full-wave rectifier

[eneuro]

Hello,
Thanks to your ideas from this thread sometime ago Low power loss full-bridge rectifier after transformer 230VAC 50Hz with mosfets where there were some hints howto make synchronous rectification to avoid power losses on bridge diodes in the case of two phase AC currents, I'm working now on some idea to simplify my Hall efect based current sensor method and allow rectify AC low voltage currents in the range between 100W up to a few kW at 36V max and satrting from 12V.

I wonder, if maybe some ready ICs exists capable to rectify 3 phase and measure currents Hall efect sensors?

I'm considering also using existing (maybe new ICs) and use three two phase synchronous rectifiers to be able rectify 3 phase AC mains.
This shows such concept, where 3 two phase KBPC bridge rectifiers can be used this way, however of course this is not a best way to do it, but it only shows idea of maybe using a few two phase synchronous retifiers to rectify three phase?



High power rectifing to DC might be usefull during regenerative breaking of 3 phase BLDC motors, while the same mosfets which drives motor probably could be used to rectify back generated AC during electromagnetic brakeing, so trying to figureout "prior art" to see if some tricky ways to do it already exists? 

I've got some solution based on mentioned Hall effect current sensing and now trying to convert it to more compact smaller SMD design, so trying to figeout if maybe something interesting is on the market to rectify easy 3 phase high power which comes from BLDC motors (for safety resaons 36Vmax, so high currents might be present in the range of a few kW) , etc.... 

[T3sl4co1l]

LT makes a FWB controller for MOSFETs at line frequencies up to 600Hz.

Tim

[eneuro]

FWB controller for MOSFETs

Thanks for hints
Probably, you are talking about something like this LT4320 Ideal Diode Bridge Controller which I've found in this thread:
Re: Rectifier diodes. What data points generally determine efficiency?

Usually you need some safety margin just in case things go awkward in extreme cases, so we need, say, 10A and 40V rated parts. If you are using MOSFETs with a controller, IRF540N and IRF4905 are good options with at least 30A current rating and 55V voltage rating. If you need Schottky diodes, MBR1045 can be a good option.

Yep, adding Schottky diodes in parallel with Mosfet body diodes is good idea.
I've no idea howto use this single phase ideal bridge controler in three phase AC configuration, but there is something else_I'll have to figureout what could be possible fequencies needed to rectify, eg. in the case of a few thousand RPM alternator three phase output which is in my interest too 

I think, that maybe Hall efect sensing still can have advantage while I can make something like ideal half bridge controler, to drive high and low side mosfets based on direction on current present in the circuit and still use fast Schottky in parallel with mosfets body diodes, which will conduct current up to lets say 1A (eg. at 0.7Vf it will be only 1W loss or less, 2W per half-bridge, 6W per whole three phase three such drivers which is acceptable in high power application) and use opamp like this LM358D  Single Supply Dual Operational Amplifier to define this 0.5A-1A hysteresis in this opamp analog design, so when currents in the range of -0.5A..0.5A will flow througth such half-bridge then fast Schottky diodes will conduct, else above 0.5A( or 1A based on Hall current sensor resolution and design) opamp will turn on optoisolators in mosfets half-bridge drivers to bypass (short) given diode based on detected current direction 

This is similar to my oryginal concept here MPU was used for this, but now in autonomus analog version without need to use MPu to do this, however in the case of regenerative brakeing in BLDC motor it can be implemented in its motor driver itself, since the same mosfets and diodes can be used to rectify back power generated by slowing down BLDC motors.

Probably Hall effect current sensors will be fine there, and now considering different options available like this nice ACS714: Automotive Grade, Fully Integrated, Hall Effect-Based Linear Current Sensor IC with 2.1 kVRMS Voltage Isolation and a Low-Resistance Current Conductor , however 0.0012 Ohm resistance and we need to break power wire and make contacts on PCB for our power track 
Anyway, its footprint is really small and decetn current  of 66 to 185 mV/A output sensitivity  should easy allow create this to detect currents higher than 0.5A.

My SS495A based radiometric Hall effect sensor I've used so far had similar sensitivity and the advantage was that no need to break power wire since it could be winded over toroid core with this sensor inside, but of course thoss hand made sensors costs probably more in manual assembly, but it works and I can easy deetct positive and negative currents.

BTW: When looked at sample PCB with ACS714 like this not the cheapiest one ($11  ) : http://www.tme.eu/en/details/pololu-1187/sensor-modules/pololu/ I can't believe in this thing is capable for high currents-maybe those +/-30A, but 100A and more???



Nope, it ACS714 in this package looks hopeless for really high currents and I do not know, ~$2 SS495A with cut ferrite core still can be hand made for less than $3 and no need to solder high power wires connectors to PCB, while they go througth toroid core.
yep, it requires some work, but.... it looks much more powerfull for me.

Maybe, someone know other higher current rated Hall effect sensors, which allow put power wires inside this sensor (wtihout need to solder connectors) and still let detect currents as low as 1A without worry about Earth magnetic field (~0.5 Gauss 5*10^-5 T) interference ?

BTW:
Digikey pricing much better and for hundreds of such +/-30A ACS714's less than $3 per sensor, but in this package I do not know-I need this thing to operate at >100A and detect currents >=1A or better >=0.5A 
https://www.digikey.com/product-detail/en/ACS714ELCTR-30A-T/620-1260-1-ND/1955902

[eneuro]

This is waht I've  used so far in custom high power applications-as we can see no need to solder any high power connectors (needed in the case of ACS714), since high power wires with a few turns for better current sense can be put inside this thind and configured to pass AC in both directions, on only one AC/DC wire to provide current to half-bridge 



This one is used in 2.5kVA spot welder to control its primary current, so some dust visible, but you have to  those thick wires to brake this current sensor (didn't calculated its resistance but it is easy-the same as power wires used for windings) 

Probably, I will try add in smart way opamp (LM385D with a few resistors for mentioned histeresis and 5V voltage regulator close to it) somewhere on this 1inch2 (25mm x 25mm double layer PCB) and lets forget about those crappy looking ACS71x with thin package legs for high currents