more articles

content/content/articles/three-phase-inverter-ng.md

@@ -0,0 +1,24 @@
+---
+title: "Three Phase Inverter - Second Service"
+---
+
+I wrote in October about my first try to build a simple three phase inverter, see [here]({{< ref "three-phase-inverter.md" >}}). In the first try I used four MSP430 microcontroller, one for the PWM of each phase and one to coordinate the phase shift of the three phases.
+
+In this experiment I put everything on one STM32 microcontroller. Here I used the DMA feature to feed data into the PWM counter and I calculated the sine values at start-up time on the microcontroller. Additionally I put in the driver for a CAN interface, however, it is not yet supported in the firmware.
+
+![](/img_0140.jpg)
+
+From top to bottom you see the CAN driver, the STM32 board, opto coupler to separate logic and power part and then from right to left in the bottom half the low-side/high-side MOSFET drivers and the MOSFETs.
+
+![](/img_0144.jpg)
+
+The power supply consists of a traditional transformer and (top right) the rectifier and capacitors for the power part, together with the 12V regulator for the drivers and (top left) the regulators for 3.3V and 5V for the logic part.
+
+![](/img_0146.jpg)
+
+The motor is the same as in the earlier experiment - I don't have too much of them. And everything is put onto one board:
+
+![](/img_0143-e1482141676335.jpg)
+
+The code for this experiment is here: [https://gitea.hottis.de/wn/inverter2](https://gitea.hottis.de/wn/inverter2).
+

content/content/articles/three-phase-inverter.md

@@ -0,0 +1,43 @@
+---
+title: "Three Phase Inverter"
+---
+
+
+Already when I was still in school, about 30 years ago, I was curious to make an inverter using some MOSFETs. I actually was able to build a simple one phase inverter with rectangular signal shape (I used a NE555). Using this thing I drove a transformer to light a blub. However, all of these inverters I built passed by in fire.
+
+Now, I tried it again, not longer using MOSFETs but IGBTs with free-wheeling diode. Moreover, I used some microcontrollers and sine values to feed a PWM to get a sine-alike signal shape. And this time I was able with three phases to drive an asynchronous motor.
+
+![](/img_0053.jpg)
+
+The signal shaping is done with four MSP430 controllers, three as PWMs to drive the bridge and one to coordinate and control the three PWMs. The PWM controller is decoupled from the IGBT driver (IR2184) using optic couplers.
+
+![](/img_0054-e1476437702547.jpg)
+
+The bridge is a three phase IGBT module is a 6MB120F-060 I got for a few euros at ebay.
+
+![](img_0055-e1476437685461.jpg)
+
+To avoid high voltages in my setup I got a 24V async motor, also from ebay.
+
+![](/img_0056.jpg)
+
+The PWMs generate the signal from a sine table generated using Excel. Those I got this signal:
+
+![](/inverter0_2016-09-23-4.png)
+
+The main task of the coordinator is the start the PWMs with a phase shift of 120° (digital line 1, 2 and 3):
+
+![](/2016-10-13_1.png)
+
+Currently the PWMs start with random polarity. The interesting signals are the digital lines 4, 5 and 6.
+
+Sometimes the motor runs:
+
+![](/2016-10-13_works.png)
+
+But sometimes not:
+
+![](/2016-10-13_works_not.png)
+
+The firmware is available here [https://gitea.hottis.de/wn/inverter0](https://gitea.hottis.de/wn/inverter0) and [https://gitea.hottis.de/wn/inverter0ctrl](https://gitea.hottis.de/wn/inverter0ctrl).
+