I am new to arduino signal processing and just trying to understand what Baud rate really means.
As I understand it, the baud rate speed is really the time that the processor takes when it is ON and when it is OFF. I attached an image as an example. For a slower baud rate, a 1 is represented by ON and voltage will stay ON and OFF for a long time as compared to a fast rate where the voltage will stay ON and OFF for a short time.
Am I understanding this correctly?
I am looking at transmitting a digital signal through the arduino over a long distance cable. The problem I am having right now is I believe the signal is being distorted because it is modulating too fast. I need the ON's and OFF's to last longer, to stretch out over time. As I understand it this would be as simple as slowing the baud rate down.
Could someone confirm this? Hope it makes sense, thanks.
Yes, Baud Rate is the rate at which the serial data is clocked. Notice that, minimally, there is only an Rx and a Tx line -- that is all that is needed to do Serial Transmission in the RS232 protocol. But, how does the receiver (the "Rx") know when it's receiving a "One" and when it's receiving a "Zero"? What if more than one "One" is transmitted at once? How does the Rx know how many "Ones" were intended, since all it receives is a continuous High voltage, for some period of time [actually, for a "One" the line goes "Low"]? That's where Baud Rate comes in. But setting the Transmitter and the Receiver at the same Clock rate, there is an agreement, between the two, at what rate the data is being transmitted. That way, that long low voltage period, gets properly chopped up into the correct number of "Ones". And, BTW: the "Start Bit" is what synchronizes the two clocks.
But, for a long line, the problem is likely not the pulse width, but the fact that the pulse gets distorted by things like reflections due to impedance mismatch, ringing due to dominant reactive components, hi-pass filtering due to parasitic capacitance, I2R losses, and etc/etc. Google "Transmission Line", and then step back, 'cuz it's a whale of a topic!!
Thanks guys for the quick responses. I know I am probably biting off more than I can chew, but It seems to me like this all possible, although very challenging. I am attempting to transmit image data through 4.5 KM of 15 Awg wire. What I have is an analog video camera deployed in a borehole , and the wire connecting the two is what is known as a wireline truck. To make things more complicating, 3 KM of the wire is wrapped up in a spool which will probably create all sorts of reflections and distortion, while the other 1.5 km is down the borehole. However before getting too depressed about the complicated nature of what I’m trying to do, I do know other companies are accomplishing this right now and achieving at least 1 image per second data rate, and I have seen that the images are very crisp and clear.
My plan is to use an Arduino (down the borehole in a sealed chamber) to slow down the pulse width. I will then amplify the digital signal up to 60 ish volts and send it up the wire to surface. My thinking is the wider and higher the signal is, the more clear it will be when it gets to surface. I plan to impedance Match to lower inflections. Not sure if this will all work but even if I can achieve one 30 kB image is very 5 minutes I am winning.
Simply amplifying the signal is what destroyed the first transatlantic telgraph cable.
The second was successful because the mathematics for "transmission lines" was worked out in the intervening years. Google it.
If you know the impedance of the cable, you can drive it properly. Your internet cable is probably about that distance to the CO and you get full-motion video on that cable, don't you? The cable spooled on the drum should make no difference other than length.
3 KM of the wire is wrapped up in a spool which will probably create all sorts of reflections and distortion,
No it won't.
My plan is to use an Arduino (down the borehole in a sealed chamber) to slow down the pulse width.
And how is that going to work?
If you are sending data out at a slower rate than it is coming in then the input buffer will fill up. When this exceeds the size of the buffer you will loose data. An Arduino Uno has only 2K of data memory so you can't get much of a buffer with that.
My thinking is the wider and higher the signal is, the more clear it will be when it gets to surface.
Your thinking is wrong. This is a digital signal so there is no degrading it either works or it does not.
Calculating the size of the buffer is simply a matter of knowing the input rate and output rate. Just like the overflowing bath problem you did in kindergarten. You know the one saying the taps input 2 gallons per minuet and the plug drains 0.5 gallons per minuet. The bath as a capacity of 80 gallons. How long is it before the bath overflows?
Grumpy_Mike:
You know the one saying the taps input 2 gallons per minuet and the plug drains 0.5 gallons per minuet. The bath as a capacity of 80 gallons. How long is it before the bath overflows?
Depends. If it's a nice warm bath, I might decide the impetus to luxuriate supersedes your fool experiment, and climb in, thus skewing the result.
dgrant123:
Thanks guys for the quick responses. I know I am probably biting off more than I can chew, but It seems to me like this all possible, although very challenging. I am attempting to transmit image data through 4.5 KM of 15 Awg wire. What I have is an analog video camera deployed in a borehole , and the wire connecting the two is what is known as a wireline truck. To make things more complicating, 3 KM of the wire is wrapped up in a spool which will probably create all sorts of reflections and distortion, while the other 1.5 km is down the borehole. However before getting too depressed about the complicated nature of what I’m trying to do, I do know other companies are accomplishing this right now and achieving at least 1 image per second data rate, and I have seen that the images are very crisp and clear.
My plan is to use an Arduino (down the borehole in a sealed chamber) to slow down the pulse width. I will then amplify the digital signal up to 60 ish volts and send it up the wire to surface. My thinking is the wider and higher the signal is, the more clear it will be when it gets to surface. I plan to impedance Match to lower inflections. Not sure if this will all work but even if I can achieve one 30 kB image is very 5 minutes I am winning.
Hmmm. Analog video camera means the signal is the old RS-170. So no digital involved and a low noise, wide band amplifier might just work.
