AC Sources in Parallel

[Peace Maker]

Can AC voltage sources (generators) be connected in parallel like DC
voltage sources?

One electrician told me it was impossible to connect two or more
generators in parallel for the purpose of powering appliances in a
house. He claimed the coils of both generators would be burnt in the
process.

Is that true?

I doubt his claim because DC sources can be connected in parallel.
And, since DC network analysis is similar to AC network analysis
(except in complex numbers arithmetic), I see no reason why AC sources
(generators) can't be connected in parallel.

I used Multisim Demo 2001 to simulate the connection of AC voltage
sources in parallel. The simulation failed to run because of an error,
which I haven't understood.

Kind regards
Segun

 

[Kevin Aylward]

Can AC voltage sources (generators) be connected in parallel like DC
voltage sources?

Pure DC sources cant be connected in parallel either. Its usually
equivalent to placing a S/C across the sources. Often DC sources have
isolating diodes to prevent one feeding the other. However, only one
source will actually provide the current.

This is the deal. If one source voltage is only *slightly* larger than
the other there will be I=delta_V/Rsource = very large current. That is,
one source will attempt to sink all of the current of the other source.

One electrician told me it was impossible to connect two or more
generators in parallel for the purpose of powering appliances in a
house. He claimed the coils of both generators would be burnt in the
process.

Is that true?

Essentially, yes.

If you use a control system, you may be able to overcome this issue.

I doubt his claim because DC sources can be connected in parallel.
And, since DC network analysis is similar to AC network analysis
(except in complex numbers arithmetic), I see no reason why AC sources
(generators) can't be connected in parallel.

That's because you haven't done I=V/R

However, you can usually connect transformer windings in parallel. Why
do you think this might be so?

I used Multisim Demo 2001 to simulate the connection of AC voltage
sources in parallel. The simulation failed to run because of an error,
which I haven't understood.

The error is because ideal voltage sources have zero internal
resistance. This means infinite amps. Spice will not let you connect
sources in parallel or inductors in parallel. Again, what current will
flow if you apply 1mv DC to an ideal inductor?

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

 

[John Popelish]

Can AC voltage sources (generators) be connected in parallel like DC
voltage sources?

One electrician told me it was impossible to connect two or more
generators in parallel for the purpose of powering appliances in a
house. He claimed the coils of both generators would be burnt in the
process.

Is that true?

It is certainly possible. however. the power grid is fed by many
generators fed in parallel, but they are precisely frequency and phase
matched before being paralleled. Once connected together, the
generators interact with each other to help distribute the load
between them. For instance, if one tries to go a bit faster (gets a
fraction of a cycle ahead of the others) than the others, that one
sees a big increase in load that slows it down and the others see a
decrease in load that lets them speed up a bit . If you just connect
two generators without having the instrumentation and controls to let
you make sure they are at the same frequency and phase, them you can
expect bad things to happen.

I doubt his claim because DC sources can be connected in parallel.

Not if they are not very well matched in voltage and have a little
resistance between them to limit the current that passes out of the
highest and into the lowest.

And, since DC network analysis is similar to AC network analysis
(except in complex numbers arithmetic), I see no reason why AC sources
(generators) can't be connected in parallel.

I used Multisim Demo 2001 to simulate the connection of AC voltage
sources in parallel. The simulation failed to run because of an error,
which I haven't understood.

Try putting a low resistance (say an ohm or two) between them. And
make the voltage of one slightly different (say a percent or so) than
the other and also the frequency (say, a percent or so).

 

[Peace Maker]

Pure DC sources cant be connected in parallel either. Its usually
equivalent to placing a S/C across the sources. Often DC sources have
isolating diodes to prevent one feeding the other. However, only one
source will actually provide the current.

This is the deal. If one source voltage is only *slightly* larger than
the other there will be I=delta_V/Rsource = very large current. That is,
one source will attempt to sink all of the current of the other source.


Essentially, yes.

If you use a control system, you may be able to overcome this issue.


That's because you haven't done I=V/R

However, you can usually connect transformer windings in parallel. Why
do you think this might be so?

LOL! Am I under some sought of examination by you? Your questions
(though insightful and constructive) give me that feeling. Is drilling
your potential customers part of your marketing strategy?


I think the reasons why it's possible to connect, in parallel, the
secondary windings of two or more transformers are:

1. The AC sources connected to the primary windings are electrically
isolated from the secondary windings. So any dangerously high current
that may develop at the secondary windings is isolated from the AC
sources.

The additonal magnetic fluxes that such high current may develop at
the secondary sides of each transformer is counteracted by an equal
and opposite additonal magnetic flux developed in each primary coil.

The vectorial sum of the output voltages of the transformers must not
be zero for current to flow.

2. In actual fact, the combined inductive reactances offered by the
secondary coils prevent dangerously high current from circulating
along the secondary coils of the transformers.

The error is because ideal voltage sources have zero internal
resistance. This means infinite amps. Spice will not let you connect
sources in parallel or inductors in parallel. Again, what current will
flow if you apply 1mv DC to an ideal inductor?

Your second question...

In DC network analysis, an inductor is always seen as a short circuit.
Mathematically:
XL=2*pi*f
Since f = 0 (always for pure DC)
XL=0
I=1mv/0 = oo A

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

I would gladly love more light to be thrown on my answers to either
correct or improve them.

Mr. Aylward, I appreciate your posting of the insightful and
constructive questions. Please continue posting such at every
available opportunity.

Kind regards
Segun

 

[Don Kelly]

.. They must be in synchronism(same frequency) and it is a great benefit to
be able to control the "speed" which controls the real power, and also to
control the excitation or internal voltage to control reactive flow. It is
not a difficult problem provided you have the necessary controls. First of
all the incoming generator must be matched for voltage phase and frequency
with the on-line unit and then connected (at essentially 0 voltage across
the switch terminals)(this can easily be done with a set of lights ) and
then "speed" and voltage adjustments as needed. Having good metering helps.
If your generators do not have such control it is a problem but there are
devices available now which will do this for you- how well they work is
something that I don't know.

Please note that all AC grids have parallel generators- this is normal
operation.

 

[Don Kelly]

Pure DC sources cant be connected in parallel either. Its usually
equivalent to placing a S/C across the sources. Often DC sources have
isolating diodes to prevent one feeding the other. However, only one
source will actually provide the current.

--------
It's been done for over 100 years. For both AC and DC. How do you think a
multigenerator power system operates? The control needed is not complex and
preferrably consists of speed and excitation control. You are correct in
that internal impedance is important -in the AC case it does affect reactive
rather than power sharing. The basic idea is to match the voltage, frequency
and phase of the incoming machine to the on-line unit. The rest is a matter
of tweaking to get the desired distribution of load. A typical lab
experiment for students is to run a couple of AC machines in parallel with a
load and see what the effect of manual changes in load, prime mover speed
settings and excitation have on the performance.

 

[Kevin Aylward]

What part of "Pure" did you have trouble understanding?

What part of "If you use a control system, you may be able to overcome
this issue." did you have trouble understanding?

How do you
think a multigenerator power system operates?

Pretentious crap. Get real dude. You should learn to read the post, and
understand the point of the post.

The control needed is

not complex and preferrably consists of speed and excitation control.
You are correct in that internal impedance is important -in the AC
case it does affect reactive rather than power sharing. The basic
idea is to match the voltage, frequency and phase of the incoming
machine to the on-line u

{snip}

You completely missed the point. Its clear that the original poster is
not aware of even the most basic of electrical properties concerning
sources. That is, ideal current course in series don't work (infinite
volts generated), and ideal voltage sources in ¦¦ don't work (infinite
currents). This is the basic information required to know what the
essentials of the issues are. Sure, one can get around this issues with
a control system, but such systems were not relevant to the content of
post.

orginal text:

The question was whether or not one can simple put (of the store shelf)
generators directly in ¦¦. The answer to this question is most
definitely no.

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

 

[Bill Vajk]

Don Kelly wrote:

The question was whether or not one can simple put (of the store shelf)
generators directly in ¦¦. The answer to this question is most
definitely no.

Actually, that's not quite true.

To make such a connection one has to monitor the waveforms coming
off the generators (they will never be 100% synchronized while
running free) and when they match (waveforms line up) one can
parallel them.

The tandem set will, from that point onwards, self-synchronize.

 

[Peace Maker]

What part of "Pure" did you have trouble understanding?

What part of "If you use a control system, you may be able to overcome
this issue." did you have trouble understanding?


Pretentious crap. Get real dude. You should learn to read the post, and
understand the point of the post.

The control needed is
{snip}

You completely missed the point. Its clear that the original poster is
not aware of even the most basic of electrical properties concerning
sources. That is, ideal current course in series don't work (infinite
volts generated), and ideal voltage sources in ¦¦ don't work (infinite
currents). This is the basic information required to know what the
essentials of the issues are. Sure, one can get around this issues with
a control system, but such systems were not relevant to the content of
post.

I am aware of basic electronics, which you implied I wasn't. I posted
the original message out of curiosity in knowing how several power
plants (hydro, thermal) in my country are fed to the national control
center (NCC) and then re-distributed to other parts of the country.
That's why I emphasized the use of "generators". When the electrician
told me that it was impossible to connect two similar generators we
had to power appliances in our house, I then wondered: "So, how come
the national power grid works?" This probably got me confused in
erroneously mentioning that DC sources can be connected in parallel.

Your posts and others have informed me more on the processes involved.

Anyway, thanks for making certain things more obvious to me concerning
DC sources and their internal resistance.

I was expecting you to comment on the answers I returned to your
questions.

Kind Regards
Segun

 

[Kevin Aylward]

I am aware of basic electronics, which you implied I wasn't.

Well, with all due respect... My point here is that basic electronics
surely includes V=IR. You can't get much more simpler. An application of
this formula immediately shows what the issue is. A delta V between
generators of only 1V, AC or DC, theoretically, could result in huge
currents from one generator feeding the other because the internal
resistance is so low.

I posted
the original message out of curiosity in knowing how several power
plants (hydro, thermal) in my country are fed to the national control
center (NCC) and then re-distributed to other parts of the country.
That's why I emphasized the use of "generators". When the electrician
told me that it was impossible to connect two similar generators we
had to power appliances in our house,

If you get say, two petrol generators from the hardware store, and
connect them in ¦¦, its pretty much guaranteed to blow a fuse, so the
electrician is correct. Voltage, frequency and phase will not be
matched.

The national grid is a different issue. Steps are taken such that the
generators can be connected together.

I then wondered: "So, how come
the national power grid works?" This probably got me confused in
erroneously mentioning that DC sources can be connected in parallel.

Your posts and others have informed me more on the processes involved.

Anyway, thanks for making certain things more obvious to me concerning
DC sources and their internal resistance.

I was expecting you to comment on the answers I returned to your
questions.

I have been a bit busy. Still am..

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

 

[Kevin Aylward]

Actually, that's not quite true.

Yes it is, assuming that the generators have been *specifically*
designed to be || capable.

To make such a connection one has to monitor the waveforms coming
off the generators (they will never be 100% synchronized while
running free) and when they match (waveforms line up) one can
parallel them.

The tandem set will, from that point onwards, self-synchronize.

You are moving the goalposts. What part of "directly connected, off the
store shelf generaters" are you having trouble with? Direct connection
means || the power plug sockets only.

Sure, we have *already* pointed out that *if* you synchronise voltage,
frequency and phase, than you can || such generators. Synchronising
voltage requires a control circuit to makethem exactly equal.

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

 

[Don Kelly]

--
Don Kelly
[email protected]
remove the urine to answer

Yes it is, assuming that the generators have been *specifically*
designed to be || capable.


You are moving the goalposts. What part of "directly connected, off the
store shelf generaters" are you having trouble with? Direct connection
means || the power plug sockets only.

Sure, we have *already* pointed out that *if* you synchronise voltage,
frequency and phase, than you can || such generators. Synchronising
voltage requires a control circuit to makethem exactly equal.

This control circuit could be nothing more complex than a voltmeter and or a
light across the open circuit- exactly equal is desired but not absolutely
necessary. Look for and adjust for slow voltage swings and refine until
voltage is nearly 0 and holding. Such a setup was in use for many years for
manual synchronising of alternators-( lights alone, for a 3 phase machine
can give close enough indication of relative speed phase and voltage). A
difference in voltage will cause a reactive flow between the generators and
as long as it is not excessive- no problem. However it would be *very*
desirable to have voltage control and speed control of at least one of the
machines in order to balance real and reactive loads.
However, for the small home generators, it really is simpler to buy an
electronic device designed for this. As far as I can tell it is basically a
rectifier/inverter setup dessigned for ease of use.

Don Kelly