There is no doubt that it is possible practically and theoretically. Not only we can connect two AC sources in series but also we can connect lots of AC sources with different characteristics (Frequency, Phase, Amplitude, ...) in series with together. This can get proved using "Fourier series" rules. However, the result may not be the one you expect. 

Hi Varun

generally if the AC source has no isolation, like isolator or if not connected via power combiner, one AC source will influence the performance of the second. since the source is not isolated a non linear effect like 3rd and 5ft order inter-modulation will occur between the sources.

hope this helps

Dan

When an AC circuit is given  source voltages consisting of a mixture of frequencies, the components in that circuit respond to each constituent frequency in a different way. Any reactive component such as a capacitor or an inductor will simultaneously present a unique amount of impedance to each and every frequency present in a circuit. You can utlise superposition theorem to calculate the effects separately considering one source at a time. You can find different amplitude & angle for each source. Beside this harmonic effects will effect the AC system performance.

In general it is possible. But at some moments it can happen that one of the sources habe the maximum positive voltage and the other the maximum negativ. Or otherwise. You have to consider this case and you will have to design your external load accordingly. 

Based on Fourier's theory, the instantaneous voltages and currents can be expressed respectively, by the sum of infinit number of components which differ by frequecies and magnitudes. The sum of voltages can be carried out in series and the one of currents in parallel. In complex plan, the root mean square quantities are the best representation of such waveforms. I agree with Mr R.Dilsh.

Two voltage sources with different frequencies cannot be connected directly because there will be phase shift from v(t) = VmSinwt. Therefore, the resultant won't bear same frequency.

Regarding connection of sources with different frequencies in series, I wonder how you can. Perhaps you may be referring to a situation as is reflected in a GATE 2016 problem as attached. This is a very interesting situation since one part of the circuit (the series leg) is a short to a frequency and the the other leg (the shunt part) is an open circuit to the other frquency. The net effect is to have the input voltage fully delvered at the output. This multifrequency source analysis is based on superposition which is applicable since the circuit is linear. It also shows that what you have in mind of tuning and extracting the required frequency is a possibility. 

Yes but , without distortion, constant frequency is needed for the efficient and economical working of power system equipments, with the signal distorted with injected another frequency signal, the distorted waveform, would stop the operation of equipments or would reduced the efficiency, though in communication signal are transmitted by frequency modulation, refer related papers or books,

Yes, you can connect two AC sources in series (provided you have them isolated). Please make sure their current ratings are same (or atleast close to each other like 1A and 1.2A certainly not 10A and 1A).

There is no doubt that it is possible practically and theoretically. Not only we can connect two AC sources in series but also we can connect lots of AC sources with different characteristics (Frequency, Phase, Amplitude, ...) in series with together. This can get proved using "Fourier series" rules. However, the result may not be the one you expect. 

I agree Majid Moradzadeh. If you do not have any syncronisation between the AC Generators you will get a radom wavesignal which can not be reproduced or predicted since you will always have slight tolerances in the frequency.

Dear Ronald Dilisch,

Thanks for your idea. I almost agree with you. However, it should be considered that, based on "Fourier series", when you get series two or more AC sources with different characteristics, the resulted signal is periodic and predictable. ''Synchronization"  must considered when you want to parallel two AC sources and it has its own discussions. In getting two signals series, the "Synchronization" is not a must but should be considered in order to get a desired signal.

Regards,

Majid Moradzadeh 

Dear Majid,

you are absolutely right. In theory. In the Fourierer series you have exact frequencies with a fixed proportion which can be transformed.  In praxis you will get a shift in the frequencies of the AC sources such as Inverters due to the drift of the components. This shift might be very small, but if it is very small it will lead to a chaotic behavior 

Kind regards

Roland

Dear Roland Dilsch,

Many thanks for your opinion. You are completely right and I think both of us are in the same way.

Regards,

Majid Moradzadeh

Jay Jagannath

No dear, it will raise high circulating current due to production of complex waveform.In this case you have to use two stage converter.

Hi,

Thanks to Majid and Roland, agree with their opinions. I think we need to consider the purpose the making series for 2 AC sources. If the purpose is to add energy(or power) from 2 sources, synchronization is necessary. If it is to analyse the resultant signal, synchronization is not necessary.

Regards

Than Tun 

Thanks to Majid and Roland, I agree with your opinions. Theoretically, it is possible to integrate multi-frequency sources. But practically, what are the constraints in multi-frequency source integration, as I am looking towards multi-frequency power transmission system design. 

Jay Jagannath Mr. Debashish Mishra,

I am using the two-stage structure for implementing ac sources.

Dear Varun,

In my opinion, 2 AC sources in series (without synchronization) will not be able to transfer energy to the load efficiently. Since 2 sources are in series, they have same current. Without synchronization, there will be 2 conditions at one particular time, as below.

(1) Voltage of source1 and voltage of source2 are in same polarity (both are positive or both are negative). In this condition, both sourcing will be giving energy, to the intended load.

(2) Voltage of source1 and voltage of source2 are in opposite polarity (one is positive and another is negative). In this condition, at least some part of energy from one source is going to another source instead of going to the intended load.

Therefore overall efficiency will be reduced. I think there is 50% chance for each condition over a considerable time period if 2 sources are independent. I will not be surprised, efficiency is reduced to 50% or below 50%, even without considering transmission loss. For example, max power of source1 is 1kW and max power of source2 is 1kW, max power of 2 sources in series will be 1kW instead of 2kW.

Without synchronization, will it be possible to avoid condition #2 ?

Dear Varun,

Sorry, I missed to address how to prevent burning IGBT last time. I assume you used IGBTs at the outputs of 2 sources. IBGT could burn by many different reasons, excessive current or voltage or temperature rise. Are they generating sinusoidal waveform? How did you connect 2 sources in series, for example using 2 isolation transformers, or using capacitors?

Convert it to DC for power addition and then convert back, that's the possible solution. In high power grids, power transfer is done by adjusting the frequencies so power transfers from high to low frequency