Sine Waves, Instant Volt, and a headache...

Status
Not open for further replies.
D

Dweeber

Member
I have a theoretical problem and I need some good theory/experience. I have been informed that an automatic,solid-state transfer switch could parallel 2 independent 3-phase AC sources(208/120) out of phase(theta), without interruption and without damages. The loads are both 120 and 208. Believe it or not, I am having a hard time wrapping my mind around this concept.

The sine waves (180degrees out of synch) add up to exactly zero? Something doesn't seem right...thus the call for help.

I apologize if this is not the place for this request, and please feel free to erase it if it flies against the rules of this site.Thanks in advance for any assistance.
 
Not quite

Not quite

A solid state transfer switch does not parallel the sources but has a transfer time less than one cycle. So, if the loads are not rotating, you could switch from one source to another.

However, if there are motor loads, the motor voltage will be out of phase after the transfer and there will be a big surge of current as the magnetic field in the motor shifts from one phase angle to the next between the two sources.

When there are motors, having a dead time in an ATS is a good thing. It allows the motor EMF to decay and the controls to drop out before the motor is restarted.
 
If this is a closed transition switch, then it can "parallel" the two sources for a short time while transferring to avoid any power interruption on re-transfer to utility after an outage or for normal testing. It will monitor the two sources and close when they are within a few degrees of each other. It cannot connect two sources that are out of phase with each other.
Don
 
headache is getting worse....

headache is getting worse....

Beanland:

Allow me to clarify, I know what I want to say, but probably can't get it into enough detail for anyone to understand.

As far as the speed the switch can transfer, well, I am assuming instantly, I mean way less than a cycle.

The real-life scene is that there is a group of folks trying to convince me that a parallel situation, regardless of being synchronized, will not be a problem. I've gotten into a bit of heated debate over this....and need equations and laws of electricity on my side.

The really sad part is that there's really no way to experiment. The loads are vital to safety. Thus I need the background and expertise that this forum offers.

Respectfully,

D.
 
Oh boy, the headache is gaining ground...

Oh boy, the headache is gaining ground...

Don R:

Respectfully, and without a lot of explanation, and moving rapidly from theory to real-world, I agree the transfer is inhibited when it exceeds the transfer switch's acceptable limits. However, let's dial that setting out of the picture.

Take that very transfer switch. and remove it's prohibition, and then perform a parallel(ed) transfer, now what would be the result?

This the scenario that I am having trouble with. Even the slightest time that the parallel would occur, would result in some odd waveform, if they were out of phase.

I am of the understanding that matching two AC sine waves, out of phase, cause all sorts of real world problems. I tried two methods to solve, one was addition of 2 sine waves: e=E(sine(theta)),which lead me to nowhere if the sine waves were 180 degrees out; the second was to do it on graph paper, which still gave me a headache.

The driving force for my question arise from an obscure office, which is telling me that all is well, even out of phase. I can't buy into it without some understanding that 2 sources, out of synch, parallelled for some time would not cause some issue "downstream".

Thanks for the response.

D.
 
if you are trying to disprove a theory (there will be no damage), all you need to find is the exception. In this case, do the calculations for peak to peak additive waveforms. at the transition, or zero to peak. seems like those would be the two worst case scenarios ? if the equipment doesn't fry, sparks don't fly, and the plant blows under either of those scenarios, then the obscure office is correct ?
 
Well...I may have made a grave error...

Well...I may have made a grave error...

Nakulak,

Therein lies my problem. I cannot jeopardize the safety of others to prove myself right (or wrong for that matter). I do mean safety. There's an entire group of folks depending on a continuous supply (uninterruptible) 24/7.

The specifications are extremely tight. If I do my math correctly, we are talking of a few milliseconds or less, after that, consider the electrical system failed, and potentially, a lot of folks dead. It may seem melodramatic, but that's the hard truth.

I am trying to put this parallel-out-of-sync concept into the minds of folks that are in charge of the decisions, namely not me. I have to be able to prove myself on paper, and present this in a theoretical light, without being able to take things to a real world application. Such is the nature in which I work.

I have been told that it had worked once, but I am not privvy to the test parameters, and results, just yet.

I thank you for your response, I need all the help I can get. Sincerely,

D.
 
Take a simplified example:

1. Two voltage sources, 180 degrees out of phase. V1 = -V2
2. No load.
3. Source impedance of first source = Z1
4. Source impedance of second source = Z2

Put the four elements in a circuit, you will get a loop. Write the loop equation:

V1 - I?(Z1 + Z2) - V2 = 0
I = (V1 - V2)/(Z1 + Z2)
I = 2?V1/(Z1 + Z2)

Basically, if Z1 = Z2, the current will equal the fault current from one source.
 
Last edited:
You will not parallel two sources out of phase successfully without damage.
Are you sure you are talking about actual paralleling or very fast open transition like 4 milliseconds?
 
Ah....I think the fog is lifting

Ah....I think the fog is lifting

Jghrist: Your response is what I was gunning for. Simple, direct, and easily understood. Thank you.

Ron: Your suggestion of a small "open" time between sources, albeit ridiculously fast, raises more questions for me. If there was this "open", and the two sine waves were not in sync, wouldn't the downstream components, have some inductive/capacitive time-dependent decay that would be somewhat unpredictable, let's say one source on the positive half of the wave, a ridiculousy small time gap, and reapplication of a second source on the negative part of the wave ( both 208 and 120)? I am not sure I'm even asking the right questions at this point. Thanks for the previous response, and any future response that can help me.

D.
 
There are solid state transfer switches called static transfer switches that do just that. Many of them can transfer in a 1/4 electrical cycle or less (<4ms).
They can transfer out of phase and the downstream equipment reacts depending on the type of load. Look at beanland's post.
 
I was involved in some expermential testing with a well know OEM and the Navy where we did just what you are discussiong, paralelling 2 sources out of phase. 180 degrees is out of the question, we blew up equipment seveal times between 10 and 15 degrees.

Now there are some autosynch devices on the market but i dont think that would fit your application. Please explain the syetem in more detail.
 
My Uncle Joe, who was a Seabee in the Big War, told me that they synched their generators with light bulbs between the two units. Light goes out, throw the switch!
 
This is the most asinine post I've ever seen on here - and there have been plenty.

Hopefully my translation is in error.

Here is what I think I heard:

1. The system in question is life critical. If the power fails, people die.

2. There is more than one "single point failure" mode.

3. The equipment that will be installed will not be fully tested under field conditions.

4. The person making the decisions does not understand the basic electrical physics necessary to judge the suitability of the equipment for the intended use.

5. The person making the decisions is is basing these decisions on unverified data from this web's participants.

If item 1 is true, please tell me that items 2 - 5 are not true.

... The really sad part is that there's really no way to experiment. The loads are vital to safety. Thus I need the background and expertise that this forum offers. ...
Click to expand...
I don't think that is what you need. I think you need an engineer that understands the system and the physics and has the integrity to say, "No, we are not installing a life-critical system with single point failure modes. We are not installing a life-critical system that can not be tested."

I'm not knocking the participants' advice, it's pretty good. I am appalled the FAA would even consider the installation of a life-critical system under these conditions.

Systems with single point failure modes fail - this is a fact. As I read the posts, it is is known this system failure may cause deaths - with sufficient certainty the OP is not willing to take the system out of service long enough to test it. Could someone please explain why this system, as outlined, would not be considered criminal.

carl

edited to change an inappropriate word.
put it back cause I couldn't make it read right
 
Last edited:
Did PDU static transfer testing some years ago, one of the test was an out of phase transfer. Now we were using resistive load banks, but with high speed recording equipment the engineer in charge performed the Static Transfer Switch operation, bump in the voltage sine wave for about one cycle. The resistive load (and fan motor and contactors which is minimal load) remained on line once again the current sine wave was distorted, but as I remember this was about a cycle.
 
Third try

Third try

Cuolter: I tried three times with a well writen response, but I time-out (?) before I can send it. Here the quick: The sentiment I read in your response is very similar to my reaction. Thanks, you have confirmed what I've been thinking.

Brian John: Thanks for your experience and results. I had an educated hunch that under particular circumstances, the result may be favorable.

D.
 
uh...wow..there's a lot of good here...

uh...wow..there's a lot of good here...

ZOG: Thanks for your experiences, I hope that I can do much better. I want to not blow up anything.

Rattus: The light bulbs (phase to phase) are a backup to the synchroscope, which is a manual backup to the automatic-synch equipment, so hopefully I won't have to verify my abilities to guess when the voltage is in phase. That simplistic technology is a proven, your uncle is/was right.

Again, to all that have made an effort to enlighten me, in real-word and theoretical electrical applications, I thank you.

D.
 
Please allow me to replace the word "asinine" in post 14 with "fearsome" "Asinine" was the wrong word.

carl
 
sure....

sure....

Coulter: About the "assinine", well, I assumed (you probably know where that is going), that when the preposterous nature of my query was called into question,that I was not the "assinine", rather the subject at hand. I appreciate the correction, regardless (even if I'm wrong). Thanks,

D.
 
You just can't simply parallel two sources that are out of phase.

But you can do almost anything with "solid state". It would even be possible for a transfer switch to rectify two AC inputs to DC, and then convert them back to AC in phase. But then you basically have a UPS minus the battery.

I would ask to see some info on this "solid state switch". Manufacturer and Model number, etc. That should either enlighten you or give you some ammo to say heck no.

Steve
 
Status
Not open for further replies.
Top