Sure, why not? It all depends what the loads do, right?

Let's say you transmit simultaneously signals (which could be nothing more than power) at 50 Hz and 1 KHz, on the same transmission line. Then you have loads that filter out the high frequencies or the low frequencies. In effect, what you have done is multiplex two different power transmission systems on the same cables.

Many power supplies go to great lengths to filter out high frequency components from the grid, and to avoid sending high frequency noise back to the grid. So already, power supplies are aware of frequency bands. Just make them more so.

Schemes like power over Ethernet (POE) assume that power is DC only, and anything at higher frequencies is "signal." But there's nothing to preclude creating more than just one "power" frequency band, just as you can do when frequency-dividing signal bands. In short, you transmit power on the cable at different frequencies (much as you can do with signals), and you put a band pass filter at the front end of each power supply. (I'm clearly ignoring practical problems which would arise at high power levels, with the filter components!)

Sure, why not? It all depends what the loads do, right?

Let's say you transmit simultaneously signals (which could be nothing more than power) at 50 Hz and 1 KHz, on the same transmission line. Then you have loads that filter out the high frequencies or the low frequencies. In effect, what you have done is multiplex two different power transmission systems on the same cables.

Many power supplies go to great lengths to filter out high frequency components from the grid, and to avoid sending high frequency noise back to the grid. So already, power supplies are aware of frequency bands. Just make them more so.

Schemes like power over Ethernet (POE) assume that power is DC only, and anything at higher frequencies is "signal." But there's nothing to preclude creating more than just one "power" frequency band, just as you can do when frequency-dividing signal bands. In short, you transmit power on the cable at different frequencies (much as you can do with signals), and you put a band pass filter at the front end of each power supply. (I'm clearly ignoring practical problems which would arise at high power levels, with the filter components!)

Yes this is what in the case of power line communication (PLC).

Of course yes, in olden days , wave traps were used to filter out some communication signals (approx. 300Hz) from transmission lines, for the purpose of communication between substations . Now they use cell phones. So, we had been using the same transmission line for ac power transfer as well as for communication , at two different frequencies. Therefore , it must be supported for 50 and 60 Hz as well , provided you have a proper filters .

Wireless power transmission is a big researches area, Its already proven it can be done (although in small scale), The theory behind is mostly induction so its used in small distance,

For the same theory to work in power transmission in a wire, I ques you must need a special type of cable which can react to different signals with different frequency , 

The question seems to involve the transmission of power and not signals on top of power which has been done for decades. The issue then becomes load. The same conductor can only handle a maximum load no matter what the frequency is used to transmit the power. When the combined power becomes too high, the wire will fail no matter what the frequency used.

The RMS load limits (current and voltage) of the transmission

medium is the first consideration; the high-efficiency separation

at the load is a quite different problem; and the combining of

the generator powers is a challenge of another sort again.

Obviously, the closer the frequencies, the more challenging

is their separation, especially if (and assuming) the cost has

to be trivial. 50Hz and 60 Hertz are probably far too close in

meeting that objective.

Barrie

It possible through transformer which has no.of inputs (no.of inputs depends upon no.of power frequencies has to be sent) and single output(it gives resultant waveform of flux which is less than that of peak flux carried by transformer).But consider the skin effect, high frequency signals try to escape from the conductor. And separation of power frequencies is difficult task which involves passive components and those passive components are not to be effects any other parameters of power system.And while designing frequency mixing transformer consider the core losses of transformer.

Yes . It is possible using FDM

Thanks to all the experts for positive answer. Now please explain us the procedure for the power transmission by two methods I can think as of now:

(1) Mixing of power supplies at different frequencies and voltages at one end of transmission lines and separating them at the other end; and

(2) Inject high voltage single frequency A/C power supply at one end of transmission line and  step down it at the other end, thereafter convert it to DC power supply which can be used to generate A/C power supplies at different frequencies using voltage/current oscillators.

The output A/C voltage should be 110V, 220V, 615V, 6.6KV or higher.

As Alberti Manfredi rightly suggested that 50 Hz and 1 kHz would be no problem. However, in your original question you mention 50Hz and 60Hz. To put both these simultaneously on one line may be somewhat problematic as the receiving equipment may not be able to discriminate between the two. Filters are not likely to help as a relatively high a Q would be required making (1) difficult. The conventional way (2) would be better and for this there already exists equipment and known technology.

As i mentioned in the previous answer  take a transformer having two primary windings and one secondary winding.50 Hz,60Hz  signals connected to primary windings separately (or take frequencies which make resultant wave nearer to square wave like fundamental + 3rd harmonic + 5th harmonics etc) o/p  voltage wave combines both waveforms and gives resultant wave.

          Generally predefined(assumptions at the time of design) conditions are not existing at load end it may varies time to time,so it is better to prefer 2nd method which was mentioned in your question

For widely separated frequencies at similar voltages, it wouldn't be too hard but mixing 110V with 6.6kV will give you major problems. Just attaching both to two primaries on a transformer will result in the 6.6kV being fed back into the 110V output of the other generator which will probably blow it up.

As has been mentioned, you may also need to avoid say 60Hz signals appearing on the input of a load that expects 50Hz. While many devices will happily work from either, providing both will produce a 10Hz beat. That will seem like amplitude modulation of a 55Hz supply to the load and it would then need reservoir capacitors 6 times larger than the load designer expected. The Q required for filters would be achievable using small electronic devices (e.g. op-amps of switched capacitor filters) but to handle the power you would probably expect to use a passive solution and that is not achievable any a reasonable cost.

A more practical solution may be to distribute a single frequency and change the output to what the loads expect using inverters.

Power supply frequency can cover 50/60Hz (grid), 50/3Hz (railway) and 400Hz (some special application, ship, aircraft, etc). If you go through the development history of the power grid, the mix of frequency for power transmission is very hard to be applicable. The cost for power quality control, resonance and Var compensation, is complex and huge. If you think about the switching and breaking of high voltage circuit breaker, you can understand the seperate type test needed for 50Hz/60Hz. It can't be simply treated as signal processing, it's power, it will increase the complexity of power grid and increase the infrastructure cost.

.

I agree. The original question/suggestion and its follow-up

suggest a lack of understanding about the huge domain of 

power engineering, leading to the general answer of "The

transmission of alternate electric currents having different

frequencies possible in a single conducting wire" is only

possible in theory, and even then, not a simple matter. I

suggest there are other more practical solutions, such as

using the "single conducting wire" to transmit DC power

and then converting this into the various required outputs

using suitable converters at the load end of this "wire" -

which of course can never be a "single" conductor in the

first place, but must be a pair.

.

Barrie

If some of the appliances need to run on 50Hz while others requires 60Hz, it would probably be more economical overall to supply a single frequency and then generate the other at the appliance end. Motor-generator convertors do this very efficiently and the generated frequency could be phase locked to the distribution frequency if the concern here is timing stability. Only a single convertor is needed per site if adequately rated.

Thank you Dr. Barrie Gilbert for minute observation.

Agreed with you that a pair of wire is needed for conducting electricity. There is not a lack of understanding. The three phases A/C current can be transmitted to three different areas  run as single phase to run single phase appliances in an area. This single phase current transmission is done just laying down a single conducting wire and the second one is earthing.

Your suggestion for DC power transmission over long distance is generally not acceptable by electric power engineering as this will lead to significant electric power loss in transmission. This is the reason invention of electric transformer became boon to us.

My question is straight forward: Whether it is practically possible to transmit electric power through single conducting wire (assuming earthing as second wire)  having combination of  a number of currents having different frequencies (also different phase angles)  generated by multiple/same electric generators at different voltages. This will eliminate a significant number of electric transmission lines and towers giving benefits in reduction in cost and risk I agree with most of the experts about the limitation on power load carrying capacity of the transmission wire.

Sir one question came in my mind just now. Suppose, instead of earthing, we use the neutral wire connection coming out of the star-connected winding of a generator to make pair of wire for electric power transmission. Whether this will save electricity? which is better? use of earthing or neutral wire coming out of the generator.

With best regards,

-Sohail

Hi Sohail,

If you talk about AC/DC, 3-phase AC transmission is good selection for certain long distance power transmission, and neutral line is almost zero volt if 3-pahse is balanced. For transmisson, the line is 3 phase line + 1 ground line (normally line you can see 3+2 lines, it's only related to protection), no neutral line there. Here, the earth gound line is protection, and is very important for transmission grid. It's used to protect and shield the power line from lightning direct strike. If you want to use ground line, ok, you just put it facing to lightning.  

For power transmission, actually, there is one recommended distance/power load for each voltage level. 500kV AC is normally used for distance less than 800km. According the new development of HVDC classic / HVDC light, the ohmic loss + coverter loss is comparable to AC, e.g 1000km. If for 2000km, 800-1000kV HVDC is the better choice. While, we can also to use HVAC, 1100kV is ready in China.

If you wish to transmit 3 phase over a long distance then the usual way is to use delta (3 lines + earth) instead of star (4 lines plus earth). As most of the previous answers mention: several frequencies/voltages may be theoretically possible but in reality impractical.

The principles of long distance transmission are well documented in many text books.  "The three phases A/C current can be transmitted to three different areas run as single phase to run single phase appliances in an area. This single phase current transmission is done just laying down a single conducting wire and the second one is earthing." That approach will produce very large losses in the ground return since it will not be possible to balance the phases between different geographical areas. I believe that is why Barrie Gilbert wrote ".. which of course can never be a "single" conductor in the first place, but must be a pair.".

The original question appears to be specifically about whether power at different frequencies can be mixed on one line. The simple answer is that it would be very difficult because (as has been said) separating the desired frequency from the other(s) at the load end would need a power filter with unrealistic Q factor for devices that need one specific frequency.

If the load is not affected by the frequency, there seems to be no benefit from supplying multiple frequencies. The distribution losses depend on the r.m.s. current and that would be approximately the sum of those for the various frequencies (average over a full cycle of the lowest beat frequency).

My question is what are you trying to achieve? If it is to reduce wiring in a situation where you must distribute power at multiple frequencies because of your generation plant, you will pay a penalty in increased distribution losses and very expensive filtering (which may not even be possible). If you were hoping to reduce losses, the currents sum so you will not get a benefit. If you just need multiple frequencies at the appliances, recreating the frequencies required where they are needed would be more efficient and very much cheaper.

SOHAIL AHMAD KHAN:

.

Has the focus of your question has changed somewhat,

to: "can existing  power lines be used do double duty?"

.

Before any other consideration, one would need to delve

deeply into the design of power transmission systems, to

estimate how much more current could be handled. Also,

one would need to revisit the design of the transformers

already in the transmission path, and probably much else.

. 

With reference to 'losses in transmission', this matter must

take into consideration all the other additive losses both at

the power-combining stage, and in the separation stages.

.

Barrie

@Vasile Surducan, FDM stands for Frequency Division Multiplexing.

.

Vasile:

You write:

.

"Multiple frequencies can coexist on the same power grid line,

and even can be transferred more or less via power transformers.

Using the same phase is simplifying everything."

.

This is an interesting assertion. It seems to be saying that the

idea has already be proven - somewhere. I am surprised that

a significant range of frequencies can be optimally managed

by high-power transformers, particularly if the f-ratio is large.

.

Sub-cycle phase locking can help only a little in lowering the

crest  factor, but not enough in practice to be useful. This is

readily calculated, of course, but the attached file shows the

case of two equal 1kV phase-locked sine waves having an

f-ratio of 5. The crest-factor will never exceed 2 in this case.

.

Barrie (Hope to see you later in the year)

The reality is that a power transmission Line do transmit various frequency, remember the presence of harmonics in power system. Now a high frequency signal (f >> 50 0r 60Hz) of very  magnitude can be transmitted simultaneously through a power transmission line and this is what exactly done in communication system through power transmission line. A high frequency signal of very low amplitude is injected in the   High  Voltage Power frequency  transmission line and this high frequency signal can be extracted through capacitive coupling at any point as required.

While it is theoretically feasible to combine 50Hz and 60Hz currents in a single network, you would need very expensive filters at either end to first combine and then separate the currents. Furthermore peak network capacity would be limited by the respective filter capacities

A more effective solution would be to use a 50HZ-60HZ (or vice versa) converter where needed and matched to the expected load. This would keep the network simple, optimised, and costs will be significantly less.

Yes, for this a good model needs a modulator that would modulate so that frequency gap between the signals become large and then multiplex them in an orthogonal manner so that they will not interface.   At receiver you have to apply the inverse process by demultiplexing in similar manner and then demodulation will be applied.

Sohail needs to find a quiet place and do some sums.

He's not getting much practical help from the majority

of respondents. But can someone "beyond the gate"

recommend a few good books on power engineering?

Surely it is possible. Remember high-frequency power line communication channels. Standard frequency (50 or 60 Hz) currents and high-frequency (50 kHz and more) signals are simultaneously transmitting through the phase wires of power transmission lines. 

Tahir:

That's not the appropriate analogy. You surely appreciate

that there's a world of difference between the transmission

of data on power lines (a standard practice), and the use 

of such lines to manage two different power frequencies.

Barrie

Isn't this what happens in a radio antenna?  Multiple RF frequencies hit the antenna with varying electric fields that create AC currents with different frequencies which the radio tuner, through resonance, sends to the radio's amplifier depending on which radio station the user wants to hear?

If you want to connect two power transmitters working on different frequencies to a single antenna, you need a combiner unit to prevent the output of the first amplifier entering the output stage of the second and blowing it up, and vice versa. The combiner has to discard a significant fraction of the power from each amplifier to do that.

We are talking about transmission of electric power and not the electric signal. Theses two are different. Research on wireless transmission of high electric power at high voltage is in nascent stage and not commercialized yet.

I think the main constraint will be the I2R losses in the cable. Assuming you dedicate half the capacity to each frequency, and can overcome technical problems, you will have a big increase in capital costs in generators, transformers. You will also probably still need a filter at each terminal, further increasing costs.

Most equipment can operate at either frequency with a slight performance loss. But if dual frequency power availability is critical, it is likely to be cheaper to simply run a second cable.

Sohail is correct, the transmission of power is very different from adding low-power data signals at much higher frequencies.

Losses from I2R heating of the cables will be no worse than when transmitting a single frequency with the same RMS current.

As I said on the 3rd June, the problems are combining the two power sources at the generator end and separating them at the load end. Using LC filters to block one while passing the other needs vary large components to operate at such low frequencies, and the Q will not be good enough to effectively separate 50Hz from 60Hz for any simple implementation.

Most loads are not sensitive to a difference of just 10Hz anyway so there doesn't seem to be any point in doing this.

JdF: But if dual frequency power availability is critical, it is likely to be cheaper to simply run a second cable.

Cheaper still to run a rotary convertor local to the load.

One issue that has not been noted in this

thread is that the peak value of the voltage

will depend on the phase synchronization

between the two or more waveforms of the

power voltages. This is never an issue in

cases where there is only one voltage.

Barrie