Recycling Power Through a Wye-Wye Transformer
- Thread starter JohnSmith1
- Start date
- Status
you mean these?
one of the few A+'s I ever received
usrd quite often in my profession
are you familier with the concept of power flow?
Power flow wil not be reversed in 480v bus but only be reduced consistent with power actually consumed by converter.
what does that mean?
sometimes the xfmr will see 480 and sometimes 208?
or the sum? Or the difference?
HOW does the converter consume power with no load?
it is the source and the load lol
For example consider a descending elevator with a regenerative inverter. What happens?
Where is the energy storage in the op's set-up?
where is the 'gravity' force?
where are 2 v's on a common winding/core?
Not the foggiest idea.
clarity
clarity
lol
I know you are joking.
for those interested
P = (Vs x Vr)/Z x sin(Vs ang - Vr ang)
P power transferred
Vs sending voltage and phase angle
Vr receiving "
Z between nodes/bus
power flows in direction of V drop
or
Is determined by phase angle
as can be seen if delta angle = 0, sin = 0, so no power flow/transfer
if delta ang = 90, sin = 1, max power transfer
clarity
lol
I know you are joking.
for those interested
P = (Vs x Vr)/Z x sin(Vs ang - Vr ang)
P power transferred
Vs sending voltage and phase angle
Vr receiving "
Z between nodes/bus
power flows in direction of V drop
or
Is determined by phase angle
as can be seen if delta angle = 0, sin = 0, so no power flow/transfer
if delta ang = 90, sin = 1, max power transfer
Last edited by a moderator:
JFletcher
Senior Member
- Location
- Williamsburg, VA
Insufficient design specs, parameters impossible to achieve, insufficient bank account, too little lead time on your end to allow mfg, etc. A whole slew of reasons.
Power returns to the source anyway. Returning it to the 480V side at 208V isnt possible, and building a 'return xfmr' to step the 208V back to 480V isnt necessary; the xfmr does that anyway.
It sounds to me like you need 208-480V test @ 100A but only want to pay for 5A worth of power, or are trying to use the xfmr as 95% of your load.
I dont know of any transformer that would allow you to feed it 480V on the high side and then connect a secondary 208V side back to a 208V primary tap. If you really want to try this on a small scale, simply buy a ballast for a street light that has numerous taps on the primary side (120,208,240,277,480) and two secondary taps (120,277) and try an experiment on a much smaller scale. My wag is that it wont work or will go poof in about half a second.
If you could recycle power in the manner you are describing, every xfmr in the country would be wired that way.
ggunn
PE (Electrical), NABCEP certified
- Location
- Austin, TX, USA
- Occupation
- Consulting Electrical Engineer - Photovoltaic Systems
I confess that I didn't read in detail what the OP is proposing, but what it seems to me from afar is that he is trying to beat the Laws of Thermodynamics. The best you can do is break even, and you cannot break even. There's no way around it.
my_mail_hub
Member
- Location
- Ohio
As I see it, the only way to do what your company is trying do would be to use banks of batteries or use capacitors, to store the excess power that passes through your equipment during the tests. The problem is you get a loss each time you step power down and up. Also if you are forced to convert if from ac to dc and back again. By the time you add in the cost of the necessary equipment, set up, maintenance, etcetera, it seems you'd probably be just as well off to go ahead and count your losses.
Sent from my LG-H811 using Tapatalk
Sent from my LG-H811 using Tapatalk
my_mail_hub
Member
- Location
- Ohio
The only way I see to do this would be to charge batteries or capacitors during your tests. You would have the complications of stepping up or down voltages or inverting the from ac to dc and back again, etcetera. By the time you add up the costs of equipment, set up, maintenance, and loss of converting power, it doesn't seem worth it.
The other idea I had was to try putting some sort of device on the secondary side of your coil that only let power flow one direction, back to the primary side. In electronics that is probably a cheap part. To prevent the flow of 480v in excess of 100a sounds expensive to me.
Sent from my LG-H811 using Tapatalk
The other idea I had was to try putting some sort of device on the secondary side of your coil that only let power flow one direction, back to the primary side. In electronics that is probably a cheap part. To prevent the flow of 480v in excess of 100a sounds expensive to me.
Sent from my LG-H811 using Tapatalk
JFletcher
Senior Member
- Location
- Williamsburg, VA
Even if that isnt his intention, wouldnt any harmonics get amplified or recycled ad infinitum running a distorted secondary waveform(s) back thru the xfmr primary side, providing 95-97% of the amperage?
Looking at the math, he wants ~95-97% of the secondary amperage recycled to the primary side. Perhaps Im failing to understand the nature of the load that would require such a configuration. Motors started across the line could require something similar (huge inrush, little running power), but wouldnt they be self-regulating to the point that at t>0 they arent drawing 100A?
If one needed the ability to test, just for an instant, 100A on the secondary, wouldnt a much smaller xfmr in a more standard configuration work? Heck, a microwave xfmr could probably hold a 100A for a fraction of a second...
What the OP is describing sounds a lot like a short circuit. If the load needs 100A on the secondary, it will draw it. If it needs 5, it takes that. Running 5A thru the load and 95A back to the primary side.... I dont see how that is possible. Either the secondary will draw 100A, or you will have a short circuit from primary to secondary, tho at the same voltage, how would any current move at all? I still think it will do nothing useful or go poof, probably blow the primary windings due to circulating currents or harmonics.
If I'm completely off base Im willing to eat crow on all of this.
I hope the OP comes back to explain more in detail, because w/o it, this thread is pointless conjecture about a hundred variables/sets of info we lack.
You are a new member so welcome to you.
Cutting the losses...............
In the system I mentioned in post #13 you would would be throwing away or wasting the better part of 900kW.
Like the OP's case, it was a test set up which means a one time cost for the hardware and energy recovery multiple times. If you consider that, it isn't so hard to see the economic case for doing something rather than doing nothing.
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
I do hope we've not scared the OP away arguing the fine points of this design.
Ingenieur,
I think in post 48 you described a system that would function for the OP. What you show in post 14 _might_ work if the phase angles were correct, and what you show in post 35 I think has the 240V polarity swapped, but is the essence of what the OP was describing in single phase. IMHO connecting to primary taps is not necessary and makes things more complicated and risky, but is _possible_.
ggunn, my_mail_hub,
There is nothing mysterious or in violation of the laws of thermodynamics to 'recycle' power in a system that is being tested. If you have a device which is intended to _transfer_ power from input to output, only using up a small fraction to internal losses, then to test the device you have to do something with the output power. That output power could be dumped into a dummy load (eg a resistor bank) or it could be put to use.
For purposes of testing devices which transfer power, the best use is to use the output power as a _portion_ of your input power...but as has been noted this might not be trivial, and if you don't 'align' things properly then you 'recirculating dummy load' could easily become a 'bolted fault short circuit'.
To JohnSmith1,
I tried to answer in post 12, but on reading Besoeker and Ingenieur, I think I can expand.
In trying to 'regen' back to your 480V supply bus, you need the power output of your 'device under test' to be in proper phase and voltage alignment with that supply bus in order to get power to flow back to the supply bus. Get this wrong and the wrong amount of power will flow. Get this really wrong and you will be enjoying a fireworks show.
If the device you are testing is similar to a 'line interactive inverter' used in solar power production, then the device under test might itself provide all of the control necessary. You would provide rectified 480V to its input, connect its output to your transformer derived 208V bus, and the device would adjust output voltage and phase angle to supply appropriate power to the 208V bus. But we don't know the details of the device you want to test, and don't know how it will respond to the 208V bus.
As an exercise, I'd suggest designing a system to test a standard 480V delta to 208V wye transformer in place of your power converter.
I'll do the first step of the exercise: consider another standard 480V delta to 208V wye transformer, connected in parallel on both primary and secondary to the transformer that you want to test. As you can see, if the transformer is connected properly there will be _no_ current flow (other than magnetizing current on the primary side), and if the transformers are not paralleled correctly you will see very large current flows.
The next step in this exercise is to figure out what you would need to do in order to get controlled amounts of power to flow from 480V to 208V through one transformer, and from 208V to 480V on the other. Figure out this control and you are well on the way to figuring out the system you want to build.
To close I will repeat something: don't think of the direction of power flow as going from 'primary' to 'secondary'. For your application 'primary' will always be your power supply bus (the 480V), and power will happily flow from secondary to primary. You won't need 'primary taps' to regenerate power back to your bus.
-Jon
Ingenieur,
I think in post 48 you described a system that would function for the OP. What you show in post 14 _might_ work if the phase angles were correct, and what you show in post 35 I think has the 240V polarity swapped, but is the essence of what the OP was describing in single phase. IMHO connecting to primary taps is not necessary and makes things more complicated and risky, but is _possible_.
ggunn, my_mail_hub,
There is nothing mysterious or in violation of the laws of thermodynamics to 'recycle' power in a system that is being tested. If you have a device which is intended to _transfer_ power from input to output, only using up a small fraction to internal losses, then to test the device you have to do something with the output power. That output power could be dumped into a dummy load (eg a resistor bank) or it could be put to use.
For purposes of testing devices which transfer power, the best use is to use the output power as a _portion_ of your input power...but as has been noted this might not be trivial, and if you don't 'align' things properly then you 'recirculating dummy load' could easily become a 'bolted fault short circuit'.
To JohnSmith1,
I tried to answer in post 12, but on reading Besoeker and Ingenieur, I think I can expand.
In trying to 'regen' back to your 480V supply bus, you need the power output of your 'device under test' to be in proper phase and voltage alignment with that supply bus in order to get power to flow back to the supply bus. Get this wrong and the wrong amount of power will flow. Get this really wrong and you will be enjoying a fireworks show.
If the device you are testing is similar to a 'line interactive inverter' used in solar power production, then the device under test might itself provide all of the control necessary. You would provide rectified 480V to its input, connect its output to your transformer derived 208V bus, and the device would adjust output voltage and phase angle to supply appropriate power to the 208V bus. But we don't know the details of the device you want to test, and don't know how it will respond to the 208V bus.
As an exercise, I'd suggest designing a system to test a standard 480V delta to 208V wye transformer in place of your power converter.
I'll do the first step of the exercise: consider another standard 480V delta to 208V wye transformer, connected in parallel on both primary and secondary to the transformer that you want to test. As you can see, if the transformer is connected properly there will be _no_ current flow (other than magnetizing current on the primary side), and if the transformers are not paralleled correctly you will see very large current flows.
The next step in this exercise is to figure out what you would need to do in order to get controlled amounts of power to flow from 480V to 208V through one transformer, and from 208V to 480V on the other. Figure out this control and you are well on the way to figuring out the system you want to build.
To close I will repeat something: don't think of the direction of power flow as going from 'primary' to 'secondary'. For your application 'primary' will always be your power supply bus (the 480V), and power will happily flow from secondary to primary. You won't need 'primary taps' to regenerate power back to your bus.
-Jon
As I and others have said
do not mix voltages on the primary, risk or short or device damage, or both
possibly a 208-480 inverter to 480 bus xfmr may do it
Or you need a load, preferably storage (battery, caps, etc)
invert that back into the 480 bus
The concept of loading the device to 100% and putting the output back onto the supply grid is common and sound
imho your method is not
have you done a cost analysis: wasting the power vs recover?
using it for heating (or cooling) during test cycle, therefore reducing consumption by the normally used hvac equipment?
I suggest you consult with an ee or a firm that mfgs ss power test equipment
do not mix voltages on the primary, risk or short or device damage, or both
possibly a 208-480 inverter to 480 bus xfmr may do it
Or you need a load, preferably storage (battery, caps, etc)
invert that back into the 480 bus
The concept of loading the device to 100% and putting the output back onto the supply grid is common and sound
imho your method is not
have you done a cost analysis: wasting the power vs recover?
using it for heating (or cooling) during test cycle, therefore reducing consumption by the normally used hvac equipment?
I suggest you consult with an ee or a firm that mfgs ss power test equipment
Last edited:
Phil Corso
Senior Member
- Location
- Boca Raton, Fl, USA
Phil Corso
Senior Member
- Location
- Boca Raton, Fl, USA
Gentlepeople...
I believe most of of you have overlooked the three basic principles (or tenets) of power transmission. Here they are:
1) Power is transmitted if the two "Bus-Angles" differ.
2) VAr is transmitted if the two "Bus-Voltage" magnitudes differ.
3) Power and VAr are transmitted simultaneously, if both Magnitude and Angle, differ.
Regards, Phil Corso
I believe most of of you have overlooked the three basic principles (or tenets) of power transmission. Here they are:
1) Power is transmitted if the two "Bus-Angles" differ.
2) VAr is transmitted if the two "Bus-Voltage" magnitudes differ.
3) Power and VAr are transmitted simultaneously, if both Magnitude and Angle, differ.
Regards, Phil Corso
- Status