Speed and rate of data transfer

[nitin_zilch]

suppose i want to transfer digital signals over 100 meter wire, the loic is implemented using two level, a +5 volts for 1 and 0 volts for logic 0. I want to transfer the data with speed 1000 Mbps which is 10^9 bits per second.

now i consider the worst case scenario that my data consist of series of 01010101... or 10101010... this mean the digital signal must change from +5 volt to 0 volts in 1/2(10^9) seconds. now given that the rate at which the signals are changed, will they not make to the end of the 100 meter wire, if they do, then how?

I do not think that repeaters are a better solution here, because if i consider that the signal travels 1 foot in 1 nanoseconds (as light would do), which is not possible, it would be like putting up thousands repeaters over 100 meters wire.

please help me, if i am wrong, and where?

 

[mfb]

There is no fundamental problem with multiple bits traveling in a wire at the same time - as long as they all move with (roughly) the same velocity, you can receive the bit stream at the other end. Dispersion will degrade the signal a bit, if the cable is too long.

 

[Mordred]

Couple of considerations as the standard for your goal on bit rates is pretty high.

What I mean by that is if you look at the ethrnet standards It'll become readily apparent some of the difficulties on obtaining the transfer speed you desire.
http://en.wikipedia.org/wiki/List_of_device_bit_rates.

Yes it is possible but you will need quality components. The link above may guide you to the type of components will work for you.

The other consideration is that 100m is typically the maximum distance for ethernet without repeaters. At least for Cat5, Cat5E with sheilded ends or Cat6 standard may be able to help gain your distance.

Location of wiring will also play a factor on digital signals you will want a routing of least EM interferance.

Good luck with your project what I posted above is just a few considerations. Others may have more to offer.

 

[nitin_zilch]

Couple of considerations as the standard for your goal on bit rates is pretty high.

What I mean by that is if you look at the ethrnet standards It'll become readily apparent some of the difficulties on obtaining the transfer speed you desire.
http://en.wikipedia.org/wiki/List_of_device_bit_rates.

Yes it is possible but you will need quality components. The link above may guide you to the type of components will work for you.

The other consideration is that 100m is typically the maximum distance for ethernet without repeaters. At least for Cat5, Cat5E with sheilded ends or Cat6 standard may be able to help gain your distance.

Location of wiring will also play a factor on digital signals you will want a routing of least EM interferance.

Good luck with your project what I posted above is just a few considerations. Others may have more to offer.

but i have shown above that the phase of the digital signal reverses before the charges moving in line could actually move even a foot

 

[nitin_zilch]

There is no fundamental problem with multiple bits traveling in a wire at the same time - as long as they all move with (roughly) the same velocity, you can receive the bit stream at the other end. Dispersion will degrade the signal a bit, if the cable is too long.

have you even read what i asked for

 

[Mordred]

He did read what you stated. Pulses can be sent down a line. In communiciations its not a steady state 1 or 0 throughout a line. Its multiple 1's and zero's or rather its differential voltages used to represent those 1's and zero's. Just like a sinusoidal wave form can travel down a line so can digital representatin signals.
Provided all your devices can send and receive data at 1 gigabit/sec then the line itself can handle it up to a maximum distance= to the line rating. Exceeding the line rating will slow the rate down through propogation delays. Hence why I posted the different standards to show what class of devices and wiring you should be looking at. Aslo keep in mind data can also be sent in various compression techniques. To increase data transfer one such is FSK

 

[skeptic2]

When you say you want to send digital signals over 100 meter wire, do you mean a transmission line of some sort with a defined impedance or literally over a wire with the return path through the ground? If you are talking about a transmission line I think you will find that 10101010... is not your worst case. Your worst case will likely be something like 11111111101111 or its inverse. This type of transmission is called NRZ for non-return to zero. The transmission line is essentially a low pass filter consisting of series inductance and shunt capacitance. If you send a long series of digits of the same value, the transmission line will tend to charge up to that value. Then when you try to send the opposite value, at high data rates, the transmission line may not be able to discharge enough to cross the threshold of the level detector at the receiver. This problem can be worsened by employing hysteresis in the detector.

The solution is to use RTZ (return to zero) coding. The best known of these is Manchester encoding. For every bit transmitted, there is both a 1 and a 0. This means that as the signal attenuates with distance, the average value stays about halfway between a 1 and a 0 and the level detector doesn't get fooled by a long string of 1s or 0s.

But there is a problem with Manchester encoding and that is that it requires double the bandwidth for the same data rate. That could easily erase any advantages from going to RTZ. However other RTZ coding schemes have been developed that do not require a larger bandwidth.

 

[jim hardy]

@ N'Zilch:

Maybe this is an easier question than we think.

You seem to be under the misconception that the sending end of the wire must be held "high" until the pulse reaches other end.

That's simply not so.

Imagine how a radar works - sends out a pulse and waits for the echo.
Same can be done with a wire, it's called Time Domain Reflectometry.

Send a narrow pulse down a wire and it'll get to the other end just fine.

I once stumbled across a computer memory that used a mechanical analog to that concept.
It was a coil of wire, perhaps fifty feet, with a loudspeaker voice coil at both ends.
The sending end applied a series of little twists to the wire, which propagated down it according to its stiffness and mass , per Newton's laws. When they got to receiving end they were detected and re-applied to sending end.
So this little memory could store hundreds of bits as a long serial "word".
Novel application of a mechanical delay line. What's most remarkable is, it actually worked.

old jim

 

[nitin_zilch]

@ N'Zilch:

Maybe this is an easier question than we think.

You seem to be under the misconception that the sending end of the wire must be held "high" until the pulse reaches other end.

That's simply not so.

Imagine how a radar works - sends out a pulse and waits for the echo.
Same can be done with a wire, it's called Time Domain Reflectometry.

Send a narrow pulse down a wire and it'll get to the other end just fine.old jim

how do you send a pulse down the wire, as much as i know, the signal is electrical, you would need a potential difference to so the same, and for that the potential must be maintained unless the signal has reached the other end.

In case of radar, you use the wave nature of radiation, therefore you can radiate and wait for the echo. however the same does not happen here.

and that is the reason that we need optical fiber to transmit data at higher rates.

tell me where i am wrong or answer to the basic question below

" how do you send a signal over a wire(co-axial or copper) whose frequency is large enough that the signal reverses it phase before the signal could reach its destination"

 

[davenn]

how do you send a pulse down the wire, as much as i know, the signal is electrical, you would need a potential difference to so the same, and for that the potential must be maintained unless the signal has reached the other end.

In case of radar, you use the wave nature of radiation, therefore you can radiate and wait for the echo. however the same does not happen here.

Having to ask that question, infers that you are probably in over your depth (wink)

Yes the same thing does happen, BOTH the pulse in the cable and the radar radio signal are electromagnetic radiation. They both travel at the speed of light minus appropriate velocity factors of the given mediums ... copper wire, air, or even optical fibre

So what Jim told you is correct
you still get that echo effect in a cable. In "another life" I worked for a telecommunications company. We would find cable faults using an instrument called a pulse echo locator and the principle of Time Domain Reflectometry that Jim mentioned, that way we could find out where down a long length of cable the fault was. The fault would provide an impedance bump in the cable and a % 'age of the pulse would be reflected back down the cable to the test instrument and it would give a digital readout of the distance to the fault

Dave

 

[nitin_zilch]

Having to ask that question, infers that you are probably in over your depth (wink)

Yes the same thing does happen, BOTH the pulse in the cable and the radar radio signal are electromagnetic radiation. They both travel at the speed of light minus appropriate velocity factors of the given mediums ... copper wire, air, or even optical fibre

Dave

i have an ADSL at my home, the wire runs through my telephone line, and what you say suggest that my line carries electromagnetic waves, not particle electron.

i know flowing electrons are also electromagnetic radiations, but i just want to confirm if the pulses we are talking of here is similar to transmitting current in the wire, under some potential difference, as we do in transmitting ac power to homes.

or we do it the other way?

 

[davenn]

electrons actually travel very slowly

do a search on these forums or in google and you will find quite a bit of
background on that subject
in AC power to the home, electrons do nothing more that cycle back and forward a few cm or so. Cant remember the actual distance but it isn't much
some one else on here will be sure to fill in the blanks

Dave

EDIT ... doing a search of the forums gave an indication that electron drift is ~ 1 mm / second
so in a 50Hz or 60Hz (depending on your country) AC mains supply, the electrons are oscillating back and forward just a tiny fraction of a mm

the pulse or other change in signal level is propagated down the wire, through the air or optical fibre cable as an electromagnetic wave, traveling at near the speed of light

 

[mfb]

how do you send a pulse down the wire, as much as i know, the signal is electrical, you would need a potential difference to so the same, and for that the potential must be maintained unless the signal has reached the other end.

You are trying to use an approximation usually done in electronics (speed of light is faster than anything else) in a system where this approximation is not valid any more.
You can send electromagnetic waves along a wire.

We would find cable faults using an instrument called a pulse echo locator and the principle of Time Domain Reflectometry that Jim mentioned, that way we could find out where down a long length of cable the fault was.

I used this to determine the length of a cable once.

 

[sophiecentaur]

There is a delay over any length of line, or Radio link (or even when you have a voice conversation with someone across the room). The delay for EM signals is at least 1ns for every foot. It does not matter at all, in most circumstances, except when there are multiple routes for signals and they may arrive 'out of step'. This needs to be taken into account.
The fact that the TV picture you're watching is a small fraction of a second later than when it was transmitted doesn't matter at all. The fact that the signals from a GPS Satellite are delayed on their journey is what makes the whole system work.

 

[Mordred]

In most communications. The process relies on differential oscillations. Essentially waves. Such as the case with your ADSL. They use a variety of communication protocols to compress data. The compression scheme is either an RTZ scheme or a NRTZ. If I recall The 802.11 is a form of RTZ but not positive on that.
It may be better if you can describe your project in more detail. The reason being the goals you set forth in the OP is not an easily obtainable rate of data transfer. Particularly if your not too familiar with communications. As the link above I posted on data speeds indicate.
Couple of valuable steps will help you out.
1) quality of wire is important as the medium your sending the signals down will as noted in everyones posts slow the signals down. Use a good quality low impedance cable.
Personally I recommend Cat 6
As we can see by that link above as being capable of those speeds.
2) the rate of switching of tranmitter and reciever as well as any repeaters.
3) Em noise. This usually isn't a problem as good cables are well shielded but its still a consideration.
4) you indicated 100 metres. This length is further than rs 232 signals. Though Rs232 can easily be converted to RS 485.
If your sending ethernet based then 100 metres is the limit for Cat 5. Although cat5e And cat6 can travel further they stll recommend the 100 metre limit. Without repeaters.
You will find that 5volt signals do not travel very far without losses. Hence for the distance I would recommend using 12 volt signals similar to RS 485.


Just some things to consider
As Jim stated it may be straight forward. However with enough considereations it should be do able

 

[nitin_zilch]

the answer to the question below will make my understanding better

" If i have a wire consisting of only neutrons, connected across the bulb, will the bulb light up when i apply a battery across it?"

 

[sophiecentaur]

How would that work? What would be holding the neutrons together, for a start?

 

[mfb]

There is no solid material with just neutrons. You can use isolators, however, to get a similar effect. Without movable charges, you don't get current flow.

 

[nitin_zilch]

How would that work? What would be holding the neutrons together, for a start?

There is no solid material with just neutrons. You can use isolators, however, to get a similar effect. Without movable charges, you don't get current flow.

but in case you have a charged battery and you connect it to a bulb through insulators, will the bulb glow? if yes, then why, if no, then why not?

 

[mfb]

No, it will not, there is no (significant) current flow.

 

[sophiecentaur]

This is turning into a nonsense scenario. Electric current is flow of charge. If there are no charges then how can there be a flow?
If you want to get more advanced then Maxwell's Equations involve the consideration of a 'displacement current' for a wave traveling through space but I don't think you are discussing at that level, are you?

 

[davenn]

This is turning into a nonsense scenario. Electric current is flow of charge. If there are no charges then how can there be a flow?
If you want to get more advanced then Maxwell's Equations involve the consideration of a 'displacement current' for a wave traveling through space but I don't think you are discussing at that level, are you?

agreed !

nitin_zilch you have been given a number of reliable responses in the first page of this thread. Stop going down a path that will destroy the prior quality of the thread

Dave

 

[nitin_zilch]

agreed !

nitin_zilch you have been given a number of reliable responses in the first page of this thread. Stop going down a path that will destroy the prior quality of the thread

Dave

i know i had i a misconception regarding the flow of current, which is highlighted in my question.

i want to know how electromagnetic wave is propogating and is used by the load, consider that the energy available to the load is not in from the kinetic energy of electrons?

 

[Mordred]

As you vary the voltage to the levels that are representing 1's and 0's. Whether that voltage is 5 volt 12 volt etc. Isnt really important.
Higher voltage is higher equates to higher current. Higher current. Signals themselves are varying levels of current. As one signal is sent then another the distance between the two signals stay the same. One does not catch up to the other. Much like sound waves travel through the air. Varying current flows the same way.

The rate of current flow depends on the conductivity of the material your using.
Every gate used in logic requires a certain amount of current to open the gate this is referred to in transistor logic as biasing.
The key is the amount of current per given time.

 

[sophiecentaur]

i know i had i a misconception regarding the flow of current, which is highlighted in my question.

i want to know how electromagnetic wave is propogating and is used by the load, consider that the energy available to the load is not in from the kinetic energy of electrons?

Now that's a real misconception which you need to address. The KE of electrons is an infinitessimal contribution to any energy transferred by 'electricity'. If you have a problem with that then consider a bicycle chain. Is it the KE of the links of the chain that transfers the power from your legs to the wheels? If it were, then the chain would act like a flywheel and the bike would carry on for ages if you removed your feet from the pedals.
The conduction electrons constitute a tiny fraction of the mass of the wires (1/100,000 say) and the drift speed is around 1mm/s. Work out the KE involved.
I think that you are looking for a very convenient and familiar model to describe electromagnetic effects; there isn't one. Read about it - and then read some more, if you want to get this sorted out in your head. Don't just ask more and more wild questions about it. It isn't an efficient way to learn.

 

[carlgrace]

Back to the original question, it will be quite impossible to send full CMOS-level (5V) signals down 100 meters of any reasonable wire at 1 Gbps. You need a SERDES (serializer-deserializer) to make this happen. Here is a good place to start if you're actually interested in digital communications:

http://www.xilinx.com/publications/archives/books/serialio.pdf

I suspect this was just a "theoretical" question. If so, you need to think about the concept of latency. That is, the information can still get from point a to point b but it doesn't have to be instantaneous. Thinking about electrons in a wire is usually not the most useful way to think about a digital communications link. Thinking about EM waves is usually far more fruitful.

As an example of latency, we can send signals to the rover on Mars but there is a 12 minute (or so) latency (depending on orbit). But the information still gets through. How is that possible?

 

[sophiecentaur]

Back to the original question, it will be quite impossible to send full CMOS-level (5V) signals down 100 meters of any reasonable wire at 1 Gbps. You need a SERDES (serializer-deserializer) to make this happen. Here is a good place to start if you're actually interested in digital communications:

http://www.xilinx.com/publications/archives/books/serialio.pdf

I suspect this was just a "theoretical" question. If so, you need to think about the concept of latency. That is, the information can still get from point a to point b but it doesn't have to be instantaneous. Thinking about electrons in a wire is usually not the most useful way to think about a digital communications link. Thinking about EM waves is usually far more fruitful.

As an example of latency, we can send signals to the rover on Mars but there is a 12 minute (or so) latency (depending on orbit). But the information still gets through. How is that possible?

In fact, the black box approach is usually the most successful in engineering. You do the link budget which tells you your likely carrier to noise ratio. This will tell you the likely error statistics for your choice of coding and modulation and you then decide whether it's a pass/fail.

Life is too short to think in terms of studying every link in every chain.

 

[carlgrace]

In fact, the black box approach is usually the most successful in engineering. You do the link budget which tells you your likely carrier to noise ratio. This will tell you the likely error statistics for your choice of coding and modulation and you then decide whether it's a pass/fail.

Life is too short to think in terms of studying every link in every chain.

How true that is. Start with an ideal model and only break open the model as each point fails. Always good advice.

I would say that in a serial link you would be well advised to think more in terms of eye opening rather than C/N. In a serial link it is usually dispersion that kills you, not noise.

Still, if the goal is "understanding" in some sense, then EM waves are a better paradigm than discrete electrons.

 

[sophiecentaur]

Electrons are probably the worst possible introduction to 'electricity'. Never touched them in my early electrical learning. Nowadays they are used by people who have very little understanding of Science to confuse kids with their teaching. And it's not even their fault.

 

[the_emi_guy]

On the topic of latency of data communications links here are some stats you may find interesting:

At 100 gigabit Ethernet, the spacing between bits on the fiber optic cable is a mere 2mm.
A long reach (40Km) 100GBE link will contain 20 million bits in flight.
Corning Glass has shipped over 30 million Km of long haul (LEAF) fiber.

Thus, at any frozen moment in time, this corresponds to 15 terabits of data in flight within these cables.

 

[sophiecentaur]

On the topic of latency of data communications links here are some stats you may find interesting:

At 100 gigabit Ethernet, the spacing between bits on the fiber optic cable is a mere 2mm.
A long reach (40Km) 100GBE link will contain 20 million bits in flight.
Corning Glass has shipped over 30 million Km of long haul (LEAF) fiber.

Thus, at any frozen moment in time, this corresponds to 15 terabits of data in flight within these cables.

There is a directly similar acoustic situation. When you shout at someone in the distance, your whole sentence may be hanging in the air, on its way to them after you close your mouth finally.
Latency is only a problem when synchronism is important. This can be on a circuit board where line lengths on a Bus need to be equalised to within a mm or two.