Phase Shift ?

Status
Not open for further replies.
Besoeker said:
Transformers are not particularly good for harmonic attenuation, particularly lower order harmonics.
Many of the contracts we have had for VSD systems specify maximum THD on voltage at the point of common coupling generally in accordance with nationally accepted engineering recommendation.
There are a number of sites where we have had to include harmonic filters in our scope of supply for the overall installation to ensure contract compliance.
Usually we'd do this at the LV switchboard, generally 400V 3ph.
Sometimes, the supply is MV and each VSD has its own MV to LV transformer. In such cases, we put the filters on the MV side.
Click to expand...

The practice I was talking about is for example, if you have two drives fed from a common bus, you can insert a delta-delta transformer in front of one drive, and then insert a delta-wye in front of the other drive. The current at the common bus will look much like a normal sinusoidal waveform, because the 30 degree phase shift on one of the drive's current, help cancel the harmonic distortion of the other drive's current.
 
jim dungar said:
Make sure the UPS vendor is involved.

If the bypass transformer is fed from the same source as the UPS then, you will probably not have any problem.

If the transformer is fed from a different source then you need to be aware of a 'rare' mis-wiring issue often called 'rolled phases'. You are probably aware of the issue of phase rotation ABC vs ACB. Phase rolling is having a 'phasing' of CAB or BCA instead of ABC. A phase rotation meter will not detect rolled phases because they do have the proper rotation.

Always make sure that you take readings of all 9 different L-L voltage combinations prior to connecting two sources in parallel.
Click to expand...

I did speak to my Liebert guy and he said that all I needed was an isolation transformer that I could get from the local electrical distributor.
I found a GE Harmonic Mitigating Guard III transformer in the catalog with a 480 delta input and a 480/277wye output with a -30? phase shift.

This is a 100kVA UPS so I believe I would need a 112.5kVA transformer for the bypass.

We already have a two breaker bypass set up that can't bypass without dropping the load first.

He did warn me about the rotation and to check line to line between the UPS output and the bypass input.
 
tkb said:
He did warn me about the rotation and to check line to line between the UPS output and the bypass input.
Click to expand...

Without a 'scope', the only way to detect rolled phases is by checking voltage.
The nine correct L-L voltages are.
X1t -> X1ups = 0V
X2t -> X2ups = 0V
X3t -> X3ups = 0V
X1t -> X2ups = Vll
X1t -> X3ups = Vll
X2t -> X1ups = Vll
X2t -> X3ups = Vll
X3t -> X1ups = Vll
X3t -> X2ups = Vll
 
Last edited:
wirenut1980 said:
The practice I was talking about is for example, if you have two drives fed from a common bus, you can insert a delta-delta transformer in front of one drive, and then insert a delta-wye in front of the other drive. The current at the common bus will look much like a normal sinusoidal waveform, because the 30 degree phase shift on one of the drive's current, help cancel the harmonic distortion of the other drive's current.
Click to expand...
Sort of quasi 12 pulse.....
If you had two perfectly equal loads on the two drives, you would cancel 5th, 7th, 17th, 19th, 29th, 31st harmonics...odd multiples of 6 +/-1.
If the loads are not equal you will get residual 5th, 7th etc.
And actually, you wouldn't need the Dd, just the Dy. Or even just a non-isolating polygonal if you only want the shift. Been there. Done that.

But, even if perfectly balanced, the current isn't really like a normal sinusoidal waveform.

12P.jpg


This is based on level DC. In practice, it usually isn't and the waveform is actually worse.
 
ohmhead said:
WHY? and is this phase shift the the current & voltage shift in degrees on that phase winding of pri to sec . meaning matching phase or is this the difference between just voltage & current on a to c phase of both currents in each seperate winding on common tap ? can power factor effect this shift or loads ? were trying to understand what is phase shift and how or why its there basically ? take care
Click to expand...

There is no phase shift in voltage or current between each set of complementary windings. That is, for example, voltage and current of the primary winding(s) connected A to C will be [generally] in phase with the secondary winding(s) connected a to N The same goes for the other two sets of complementary windings. The only difference is in how the primary and secondary windings are connected to each other. Line-to-line voltage and current will be 30? out-of-phase, primary to secondary, because the secondary has to use two windings to supply each line-to-line load. Offhand, I can't think of any other way to explain it...

Any phase shifting due to a unity power factor on line-to-line loads is also shifted 30? primary to secondary. However, there will be a difference for non-unity-power-factor, line-to-line loads on the secondary because each line-to-line load is transformed through two secondary windings (which involves all three primary lines).
 
We see the two windings and we see that everyone says 30 degrees phase shift ?
Theory wise how or what makes it lag or shift by 30 degrees if its not the voltage to current on common pri to sec winding then what is shifted ?

Thinking about it as a electrician and not schooled as a engineer I always thought it was the voltage to current shift in the phase winding thats generating the secondary voltage upon transfer , and the crossing from primary to secondary induction part that made it lag due to magnetic field coil induction part and time in cycles to make the transfer to the secondary in a delta to wye transformer
theory wise id like to understand with out the math part ?



So we see you are saying its partly because the two windings and its now a 30 deg phase shift why ?
Why does the two windings make it shift in phase ?
Thanks for what you have explained i like to learn more on this subject its wondering why or how things work thats important to know so i ask, best to ya
 
jim dungar said:
Without a 'scope', the only way to detect rolled phases is by checking voltage.
The nine correct L-L voltages are.
X1t -> X1ups = 0V
X2t -> X2ups = 0V
X3t -> X3ups = 0V
X1t -> X2ups = Vll
X1t -> X3ups = Vll
X2t -> X1ups = Vll
X2t -> X3ups = Vll
X3t -> X1ups = Vll
X3t -> X2ups = Vll
Click to expand...

What is Vll for a voltage reading?
 
ohmhead said:
We see the two windings and we see that everyone says 30 degrees phase shift ?
...
Click to expand...
I have some errands to run today, so no time to go to any depth right now. I'll have to draw some diagrams to assist in the explaining. I'll have to get back to you later...
 
ohmhead said:
Why does the two windings make it shift in phase ?
Click to expand...
Because two primary line to line voltages are used to get one secondary line to line voltage, the secondary VLL won't be in phase with either of the primary line to line voltages.
 
ohmhead said:
Why does the two windings make it shift in phase ?
Click to expand...
Lemme see if I can add a few non-engineer's cents' worth, and if I'm incorrect, I have no doubt someone will correct me:

The voltage peaks across a transformer primary (or any load for that metter) that is powered line-to-line do not occur at the same time as the voltage peaks across a similar load that is powered line-to-neutral.

Similarly, when a primary is powered line-to-line (i.e., Delta), and the secondary is wired line-to-neutral (i.e., Y), the secondary voltage peaks are not in sync with the primary voltage peaks.

Th i8s is because the primary voltage across any line-to-line connection is the greatest when the two lines are of the same voltage above zero, but neither one is at peak during that moment.

One line's voltage is on the way up while the other line's voltage is on the way down, but neither line's voltage is at its peak at that moment. That's why line-to-line and line-to-neutral peaks aren't simultaneous.

There is no real delay between the primary and secondary windings of a single transformer. An individual transformer's voltage and current response can be considered instantaneous.

How did I do, EE's? :smile:
 
Phase &phasor sequence

Phase &phasor sequence

Well first thanks to the posts which help me understand .
I see the voltage cycle of time to voltage amplitude in angular points in degrees line to line delta is 120 degs. line to neutral wye is at 30 degs.

Its a function of time to amplitude in voltage ?
next i understand polarity in symmetry ac 1/2 cycle line to line delta its polarity is reversed upon transfer to secondary wye due to self induction of winding or coils .
meaning angular point in cycle to delta voltage applied to induced voltage in wye is opposite in polarity and can be posistive or negative at any point in time this limits peak voltage induced there not at the same level or peak voltage .

this effects the voltage delta to wye transfer meaning thats why we get 208 volts and not 240 volts upon transfer ?

so its all about time in cycles to polarity of that point in that cycle and when and where transfer of that voltage happens to be at that point delta to wye meaning just a voltage change ?

so what we are saying is it has nothing to do with coil inductance opposing the flow of current in that wye winding its all a function of phase & timing is that what you are saying ?
best to yas
 
ohmhead said:
Well first thanks to the posts which help me understand .
Click to expand...
Jut hope it does help... seems to be ;)

ohmhead said:
I see the voltage cycle of time to voltage amplitude in angular points in degrees line to line delta is 120 degs. line to neutral wye is at 30 degs.

Its a function of time to amplitude in voltage ?
Click to expand...
Not sure where you are getting 120 and 30, but...

Line voltage peaks of the primary source occur A@0?, B@-120?, and C@-240?. Keep in mind this is the line voltage and not the line-to-line voltage. The phasors indicate the degree at which the positive peaks occur and reoccur every cycle using A@0? as the reference.

When you have line-to-line loads on the primary supply (i.e. the primary windings), the positive line-to-line peaks occur when the line voltage waveforms are at their greatest distance apart... which happens to be 30? after the leading waveform's positive peak and 30? before the trailing waveform's negative peak. (Mr. Fine's explanation is in error in this regard.) Note the phasor for the C-A winding is at -30? with respect to A@0?.

ohmhead said:
next i understand polarity in symmetry ac 1/2 cycle line to line delta its polarity is reversed upon transfer to secondary wye due to self induction of winding or coils .
meaning angular point in cycle to delta voltage applied to induced voltage in wye is opposite in polarity and can be posistive or negative at any point in time this limits peak voltage induced there not at the same level or peak voltage .
Click to expand...
If I understand you correctly, you are correct about the polarity due to induction. The voltage (and current, I should add) peaks occur at the same time with opposing polarity in primary and secondary matched windings. The magnitude of voltage and current peaks are transfomed by the ratio of turns in each coil as nearly equal energy.

ohmhead said:
this effects the voltage delta to wye transfer meaning thats why we get 208 volts and not 240 volts upon transfer ?
Click to expand...
Well, I don't quite see it that way. I see it more as winding voltages. For instance 480V on primary winding and 120V on the secondary winding. 208V is just a function of being a wye secondary. For the sake of the discussion, if you reconnected the secondary in a delta configuration, the output would be 120V 3? 3W (and, btw, no phase shift... unless of course you reconnected in reverse polarity, then it would have a 180? phase shift).

ohmhead said:
so its all about time in cycles to polarity of that point in that cycle and when and where transfer of that voltage happens to be at that point delta to wye meaning just a voltage change ?
Click to expand...
Now you're back on track ;)

ohmhead said:
so what we are saying is it has nothing to do with coil inductance opposing the flow of current in that wye winding its all a function of phase & timing is that what you are saying ?
Click to expand...
You words and mine are different, so it's kind of hard to say that is what I'm saying... but I think you understand for the better part what I am saying :rolleyes::grin:
 
Phase shift

Phase shift

Well SMART $ ,we agree about the winding & turns ratio on wye to give 208 volts dont know why i even thought that ? But to me learning is about mistakes you dont forget them and we appreciate the help and the time to answer my many questions we now have a good solid understanding of phase shift in a new way i will read more now into this but its clear .
SMART $ & LARRYFINE good posts best to both of yas
 
Smart $ said:
When you have line-to-line loads on the primary supply (i.e. the primary windings), the positive line-to-line peaks occur when the line voltage waveforms are at their greatest distance apart... which happens to be 30? after the leading waveform's positive peak and 30? before the trailing waveform's negative peak. (Mr. Fine's explanation is in error in this regard.) Note the phasor for the C-A winding is at -30? with respect to A@0?.
Click to expand...
Is my error because I did not mention (but implied) that the two line voltages would be of opposite polarity?

Am I still correct that the two lines are of equal voltage (and opposite polarity) at that moment of line-to-line peak?
 
LarryFine said:
Is my error because I did not mention (but implied) that the two line voltages would be of opposite polarity?

Am I still correct that the two lines are of equal voltage (and opposite polarity) at that moment of line-to-line peak?
Click to expand...

I didn't catch the implication, but if you say it's there...;)

Yes, they are equally opposing voltages at that moment... with respect to their zero crossing level as the reference.
 
Smart $ said:
I didn't catch the implication, but if you say it's there...;)
Click to expand...
Well, I was thinking it. Can't read my mind? :D
LarryFine said:
. . . the primary voltage across any line-to-line connection is the greatest when the two lines are of the same voltage above zero, but neither one is at peak during that moment.
Click to expand...
I should have said "the same voltage away from zero."

One line's voltage is on the way up while the other line's voltage is on the way down, but neither line's voltage is at its peak at that moment. That's why line-to-line and line-to-neutral peaks aren't simultaneous.
Click to expand...
Maximum voltage between two conductors would require them to be of opposing polarity, so there's the implication. :roll:


That's my story, and I'm stickin' with it! :cool:
 
Status
Not open for further replies.
Top