EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: tsaG on March 08, 2018, 05:47:10 pm
-
Hello,
I do have an artificial horizon that would need 36V of rotating voltage at 0,4A and 500Hz. The problem is my power supply (battery) only delivers 12V.
I would like to design and build a power supply that is small and light enough to put it behind a cokpit panel (or in the motor compartment).
These are the steps (I think) I have to take:
1. Boosting the voltage. 36V AC means 50.9V. I would like to remove the negative voltage so I have to multiply the voltage by two -> 101.8V
2. Creating a phase of AC (0 - 101.8V). I would do this using a half H-Bridge per Phase which is controller by a PWM Signal from a microcontroller
3. Synchronize all phases with a phase shift of 120° -> done by the microcontroller.
4. (Maybe) add a filter to the output
There is even more to consider like dead time between the phases and so on (I think).
This is the circuit for the switching I planned to use
(https://www.dropbox.com/s/enn59odsnujag2k/switch_circuit.PNG?dl=1)
Here is a Link to the "Datasheet" of the horizon. Sorry, it is in german but it also doesnt give more information than I mentioned above :)
http://www.germanluftwaffe.com/archiv/Dokumente/ABC/w/Wendehorizont%20Fl.22415-1.pdf (http://www.germanluftwaffe.com/archiv/Dokumente/ABC/w/Wendehorizont%20Fl.22415-1.pdf)
Im quite new to the AC world and these three phase motors. Isnt there an easier way to do this? I found the DRV8313 three phase motor driver chip. It says it is for BLDC motors but the internal block diagram is similar to my schematic I posted above. The main difference, correct me if Im wrong, between Asynchronous motors and BLDC motors (apart from the AC) is that the BLDC drivers need some "Feedback" from the 3rd phase that is not active. Or would it possible to use this driver? It is only rated to 60V but maybe it will run sufficently at that Voltage :-//
Thank you!
Patrick
-
the DRV8313 should work it is basically the same as your circuit + all the stuff you would need to add
three half bridges, each with an enable and an input to turn on top or bottom fet
-
FYI, the German “Drehstrom” does not translate to “rotating current” in English. (“Rotary current” is a kind of ocean current.) It’s called three-phase power. (Or, hypothetically, whatever number of phases you use, if more than 3.) This terminology issue may affect how many people look at your thread. ;)
-
This is the circuit for the switching I planned to use
if you connect Q1,2,3 like that, you will instantaneously fry the mosfet's body diodes, probably the mosfets themselves. they are all upside down, even if you reconnect them correctly, you'll never turn them on because Q7,8,10 tied to the Vcc. they should be to GND from pfets gates..
-
36V AC is around 51 Vpp; with a H-bridge topology like this, you're fine with 55ish volts; some headroom is always nice.
I'd just use a boost converter; the actual load is only 15 watts or so. Try TI's power supply designer thingy.
And I'd normally suggest a monolithic driver like the DRV8313 that langwadt mentioned, but I dislike operating parts that close to their max voltage (say 55V on a 65V abs max part). Consider using a proper MOSFET driver (here (https://www.digikey.com/products/en/integrated-circuits-ics/pmic-gate-drivers/730?FV=1c0011%2C1c0002%2C1c0003%2C1c0006%2Cmu108V|1025%2Cmu1200V|1025%2Cmu200V|1025%2Cmu600V|1025%2Cmu620V|1025%2Cmu95V|1025%2C1b880007%2C1e280002%2C1f140000%2Cffe002da%2C45000e0%2C45001a1%2C450109c%2C450025c&quantity=&ColumnSort=1000011&page=1&stock=1&pageSize=50) are some suggestions) and discrete transistors. I recommend looking for parts that are fine with higher switching speeds; some are only rated to 20 kHz or so, which can be a bit low on a 500 Hz output.