Alternator modifications

[ConnClark]

I am still working on it. I think I have something that I can build and test. The controller will will consist of 6 op amps, 30 resistors, 8 capacitors, and 18 diodes (not counting the 6 diodes and MOSFETS in the bridge circuit).

[ConnClark]

Got the last issue resolved. The circuit should cost about $30 to build in small quantities. Here is the schematic I used to simulate it with.

[Josh8loop]

Fellas,

This is really intersting stuff! I have a small wind turbine at my house that is based on a PMA(Permanent Magnet Alternator). It is more efficient than it's rotor excited counterparts such as car alternators that use up to 30-50 watts to get going. The only issue I have had with mine is that the voltage varies with rotor RPM which is to be expected of course. I am using very low voltage drop Schottkey diodes to rectify the 3 phase output to DC. Since the DC voltage varied so much, I needed a way to tame it and make it usefull. I ended up finding some DC-DC converters on Ebay that have the unique ability to take anywhere from 18-75 Vdc and put out a constant user adjustable voltage output. Overall the system works quite well. I wonder if rectifying the car alternator output with some low voltage drop diodes(like mentioned above) and coupled with using DC-DC converters could increase the alternators efficiency a bit?

[ConnClark]

Originally Posted by Josh8loop

Fellas,

This is really intersting stuff! I have a small wind turbine at my house that is based on a PMA(Permanent Magnet Alternator). It is more efficient than it's rotor excited counterparts such as car alternators that use up to 30-50 watts to get going. The only issue I have had with mine is that the voltage varies with rotor RPM which is to be expected of course. I am using very low voltage drop Schottkey diodes to rectify the 3 phase output to DC.

Just as a side note I am seeing about 1/2 of a Volt increase over Low Forward Voltage Drop (LFVD) Schottky diodes alone. As power is equal to Volts squared over resistance you might find this circuit very attractive. Right now in my circuit for a vehicle I'm finding you don't want to use MOSFETS with more than 50 nano Culombs of Gate charge. In a windmill application with lower RPM you could probably get away with more.

Since the DC voltage varied so much, I needed a way to tame it and make it usefull. I ended up finding some DC-DC converters on Ebay that have the unique ability to take anywhere from 18-75 Vdc and put out a constant user adjustable voltage output. Overall the system works quite well. I wonder if rectifying the car alternator output with some low voltage drop diodes(like mentioned above) and coupled with using DC-DC converters could reduce the alternators efficiency a bit?

I think you mean improve its efficiency The only problem I see with your plan is that by changing a standard alternator to a permanent magnet rotor is it will always try to generate as much power as it can. The DC to DC converter will have to waste some of it to throttle it down.

Actually there are permanent magnet alternators. They can get the efficiency into the mid to high 80% range. They are of course much more expensive and complicated.

[Josh8loop]

Connclark,


Yes, indeed I did mean increase Thanks for the catch! Sure be nice to find a way to get rid of the blasted field coil energy consumption! How much more efficient is your active rectification?




















..

[ConnClark]

Josh8loop,

I don't have an efficiency figure for Schottky diodes alone but over standard silicon barrier diodes its about a 20% improvement, and that is not counting the reduction in field coil current you will get

[Josh8loop]

Originally Posted by ConnClark

Josh8loop,

I don't have an efficiency figure for Schottky diodes alone but over standard silicon barrier diodes its about a 20% improvement, and that is not counting the reduction in field coil current you will get

That's really pretty good! Certainly good point about the reduction in field current, and perhaps also reducing eddy current losses in the aluminum framework(albeit perhaps small). Another thing is that as you reduce the rotor current, there will be less heat generated there which means that we may be able to encapsulate it for less windage loss.

I thought you were going to active rectification using low on resistance fets and driving the gate with some type of circuitry?












..

[ConnClark]

Originally Posted by Josh8loop

I thought you were going to active rectification using low on resistance fets and driving the gate with some type of circuitry?
..

I am. I had to have some diodes though. FETs won't work very well on their own. They can act like a diode for start up but they have a limit on the amount of current you can pass through them. I still need some diodes in parallel to the FETs. Also the diodes will serve as a back up if the FETs get fried.

[Josh8loop]

Just learned about the "Smart Alternator" Kia and other companies use. As you probably know it's quite simple in principle, and would be easy to implement with a couple sensors and a simple PLC setup.

[t vago]

Originally Posted by Josh8loop

Just learned about the "Smart Alternator" Kia and other companies use. As you probably know it's quite simple in principle, and would be easy to implement with a couple sensors and a simple PLC setup.

Kia is still using a standard alternator. The only thing they're doing differently, WRT other late model cars, is that they're automatically varying alternator output based on whether the car is accelerating or decelerating.

ConnClark is trying to increase the efficiency of the alternator itself.