As the title reads, I'm trying to understand the best (Practical) way to synchronize a diesel generator with a pure sine wave inverter, in order to implement an automatic synchronizer based on microcontrollers.

I've searched alot for references, books and articles related to this topic and I came up (theoretically) with these conclusions :

To Synchronize any two sources of single phase AC power, these conditions must be achieved:

  1. Phase angle between the two wave forms must match.
  2. Voltage on both terminals of the sources must match.
  3. Frequency between the two sources must match.

Now, frequency synchronization seems to be achieved since both the inverter and the generator is designed to output 50 Hz pure sine wave, Phase angle condition can be achieved by studying the initial state of the inverter wave then start the inverter based on this method in order to make the phase difference between the two sources zero, for voltage It seems to be more complicated to match the voltage of both sources (since both sources are not designed by me /and seems to me that I can't control their voltage output/ neither the generator nor the pure sine wave inverter, I bought them both) so here lies the problem , how can this be done ?

Also I want to know (please) what's the effect of slightly different values in each condition of synchronization, and is my approach to this goal totally wrong ?

This is my first time doing something like this, but not the first time I deal with 220 volts AC, so any advice would be much appreciated.

  • \$\begingroup\$ To the extent that this is possible, it is off topic. My feeling is that it is actually not possible. If you need this type of phase lock, you should use an inverter rather than a generator. The problem is that generators require some time to react to changes in the load. During that reaction time, something will be off, either the phase/frequency or the amplitude. \$\endgroup\$ – mkeith Apr 7 '16 at 4:06
  • \$\begingroup\$ You're right, the response of generator is not instantaneous, but still there are Synchronizers on the market that do this job :-( @mkeith \$\endgroup\$ – Bilal Sulaiman Apr 7 '16 at 4:10
  • \$\begingroup\$ I have read that honda inverter generators will automatically sync with any AC source. You actually connect the AC source to the inverter/generator before you start it. It will detect and syncrhonize automatically, then stay in sync. I am sure there is some range of frequencies where this will work. If the frequency is too far off it will probably disconnect. \$\endgroup\$ – mkeith Apr 7 '16 at 4:15
  • 1
    \$\begingroup\$ That's exactly what I am trying to implement one way or another, detect and synchronize automatically @mkeith \$\endgroup\$ – Bilal Sulaiman Apr 7 '16 at 4:17
  • \$\begingroup\$ Electromechanical generators, once synced, will be kept in sync - if it starts to fall out of sync, current will flow "backwards" and the grid will push the generator back into sync. Much like being on a common drive shaft. Inverters, not so much: current can't flow backwards through switched silicon. The traditional way to sync was a light bulb across each phase. Here youtube.com/watch?v=RGPCIypib5Q and here youtu.be/pdKAM2Xrtjc?t=5m6s \$\endgroup\$ – Harper - Reinstate Monica Apr 8 '16 at 7:36

First, you have to control one of the sources. I don't think there is anything to do about power network, but you must be able to control the diesel. There should be some throttle something.

Actually, it could be a very nice control project. You should measure the phase and by changing throttle keep the phase error zero. Like a diesel PLL.

But a more practical way is to convert the diesel output to DC and then with an invertor just create any sine wave you want.

  • \$\begingroup\$ Thanks alot. It never occured in my mind to convert generator's AC to DC then invert it back again. This method actually might be of a practical use for synchronization, but thinking about the pricing (no KW is taken into consideration) an AC-DC converter for a generator, then building up a pure sine wave for that AC-DC converter will probably force me to buy an inverter that has the sync feature built-in. \$\endgroup\$ – Bilal Sulaiman Apr 10 '16 at 5:46

Buy an inverter which has the capability to synchronize to external ac sources built-in. I don't think there is a simple way to add this sort of functionality to an existing inverter.

  • \$\begingroup\$ Honda inverter generators are designed to be paralleled. Buying a honda might solve the problem, but they don't have a diesel model as far as I know. \$\endgroup\$ – mkeith Apr 9 '16 at 18:05
  • \$\begingroup\$ I have not tried this myself. However, a grid tie inverter is designed to do exactly what you want: synchronize to an external ac source and provide power. The whole point is to be able to push power out to the electrical grid without disrupting it. There may be other inverters that are designed to do this as well which will also work when there isn't an ac source to synchronize to. Anyway, my point is that you need to find an inverter which has this capability built in as I don't think there is a way to augment an existing inverter. \$\endgroup\$ – alex.forencich Apr 9 '16 at 18:08
  • \$\begingroup\$ The grid tie inverter assumes that the grid has extremely low impedance. It phase locks and then injects sinusoidal current into that very low impedance. If this causes a large voltage rise, that is a fault condition. It will not back off and regulate. It will disconnect. And, as I said, it attempts to constantly confirm that the grid voltage is present and healthy. The conventional thinking is that if the inverter continues to supply power after the grid goes down, it could accidentally electrocute a utility worker. So some effort has been put into detecting this condition. \$\endgroup\$ – mkeith Apr 9 '16 at 18:14
  • \$\begingroup\$ Fair enough. So don't buy a grid tie inverter, buy one that has the synchronization capability baked in. \$\endgroup\$ – alex.forencich Apr 9 '16 at 18:16
  • \$\begingroup\$ I am going to delete my first comment, because it has become misleading now. I think the rest may still have value so I will leave it. \$\endgroup\$ – mkeith Apr 9 '16 at 20:02

This is a good thing to work upon. Actually, you need to have full control on one of the sources. I have many ideas for which you can work on it. But I want to mention something very important before proceeding. First of all, the inverter must be giving purely sinusoidal output otherwise there will be loss of power continuously during the parallel operations and heating may occur. Now: I suggest you to go with controlling the Diesel generator. As you have mentioned from your literature review, the parallel operation can be achieved by satisfying above conditions. I have ideas that can be used for implementation of this project. The frequency of the diesel generator can be controlled by the governor control which is a valve or choke like thing present on almost every generator. You should dismount it and place a servo motor mechanism on it so that you should be able to electrically control the throttle of the governor ( basically you will be changing fuel intake in order to increase the speed of the generator). Secondly, You should use an Auto Transformer whose tap can be changed by a similar servo motor. This auto transformer will be used to balance voltages. Then you can develop a light bulb test or dark bulb test like mechanism which is available across the internet. Using all this, you can have a circuit breaker between the terminals of generator and inverter. So your system will be continuously monitoring and changing parameters of voltage and frequency until they are balanced. As soon as there is a match, the breaker is closed. And You have successfully synchronized your generator with the inverter. Note: Always keep protections in your mind for example generator reverse power flow protection and over current protection etc. This will reduce the chances of failure of the system and ensure safety.


The conventional way to do this (three phase sets) is to set the no load voltages to be the same (to within a volt or so), set the full load droop to be the same (at whatever 100% load is for each machine, this controls the reactive power sharing), and set the frequency to be as close as you can, then when the two machines drift into sync, close the switches...

The governor response then controls how much real power each machine contributes, and both machines run at the same frequency due to feedback from the torque produced by the rotor angle.

I would note that most inverters are actually current sources, you need one that has a voltage control loop as well to allow you to control the reactive power production and hence bus voltage.

Synchronisation of a small machine with the grid (which is effectively an infinite power sink) is actually much easier then synchronisation between a small number of machines of reasonably equal size, there is MUCH more inertia in the grid, so if tends to resist frequency changes.

I am not at all sure that synchronising a single phase machine is really worth the trouble, they tend to be small going on tiny, and the automatic phase locking once in sync will be a far more chancy proposition in the absence of a rotating three phase field.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.