Scientific and Technical Subjects

  1. The Knowledge index
  2. Classical Mechanics
  3. How a delta sigma codec works
  4. The American National Software Failure
  5.  Connecting W98 & W3.11 with an Ethernet
  6.  The Effect of Electric Current on Human Flesh
  7. The Difficulty of Space Travel
  8. The Coming Monsters
  9. Lock-In Amplifiers

1. The Knowledge Index

It is time to end search engines and organize knowledge in a tree so that wrong result will not be an issue. Here is one small branch:

Physics
    Electrostatics
                    Sensors
                            Using the field effect transistor to detect electric fields

The electric field from a piece of plastic rubbed on hair is easily detected at a range of one foot. Objects like charged sweaters can be detected at a range of ten feet. The electric field due to the power line is apparent both inside and outside the house. On a clear day the electric field in the atmosphere can be readily detected.

Astrodynamics
    Orbital dynamics/orbital equations
                Programs for propagating orbits and satellite tracking.
                            See my orbit programs


Low Temperature
        Superconductors  See my file on Superconductors.

2. Classical mechanics

There is no knowledge more important than classical mechanics. You might as well just go learn Newton's laws. Folks used to say that if you knew Newton's laws and all the boundary and initial conditions then you would know all the past and future of the entire universe. This is still a good approximation. Learn why.

3. How a Delta Sigma CODEC Works.

The delta sigma codec has two parts. The front end is a delta modulator which is understandable to analog folks. The back end is a decimator. Its job is mathematical and it doesn't matter what you use to do it. I model the delta modulator as a comparator possibly followed by an integrator and then a D type flip-flop. The output of the flip-flop is the output of the circuit but it is fed back to a digital integrator of sorts. One way to model the digital integrator is as a up-down counter followed feeding a D/A. If the D flop Q output is considered to be a count up signal when it is a 1 and count down when it is a 0, then every clock causes the counter to either count up one or count down one. The output of the D/A goes to the other input to the comparator. I suppose when power is first turned on, the counter is sitting at 0 which makes the D/A put out the lowest voltage possible - the bottom rail. The signal ought to start off doing nothing which corresponds to a DC at half the power supply voltage, assuming there is only one supply. Then the comparator will put out a string of ones which pass through the flop and cause the counter to count up. The D/A output will ramp up and catch up with the input signal. All this time the D flop is putting out a bunch of ones. Finally the D/A passes up the input and the comparator switches to a zero which makes the counter count down one count. Thereafter the system hunts, jumping up and down one bit and putting out a string of alternating zeroes and ones. Now if the input tries to ramp up or down, as when an audio signal is present, then the system will follow it. As that happens the ones and zeroes will no longer be equal. A couple of extra ones means the voltage is rising and a few extra zeros means the voltage is falling. All is well as long as the input does not rise too fast. That is not likely though since the sample rate exceeds one megahertz. Also the counter and D/A is actually only a little capacitor and a big capacitor. The little capacitor takes a charge from the plus rail if it sees a one and a charge from the minus rail if it sees a zero. So this thing is putting out ones and zeros at a tremendous rate, little messages that say go up a hair or go down a hair. That is all that is necessary. To convert all the previous gibberish into nice binary count style sixteen bit words is the job of the decimator. The simplest decimator is just an adder. If you count 256 bits you get up to FF. So count the delta modulator output bit stream for 256 clock pulses. You will get a number somewhere between 0 and FF. This is the output of the Decimator. If the delta modulator is DC you will get 80 HEX. If the audio is a high frequency and large amplitude you will get a large number, like FE or 01. 0 and FF are the digital equivalent of banging against the rails, so adjust your gain accordingly. Now in the real world the things are much more complicated than this. Some delta modulators change their step size if they can't keep up. I don't know how they tell the decimator but it doesn't sound hard. Also the decimators have lots more bits than they need, so they do all kinds of averaging to use the extra bits before throwing them out. 

4. The American National Software Failure

There is a saying that if you were not there looking over the shoulder of the guy who wrote the software then you will never be able to operate it satisfactorily. This is testimony to the user unfriendly software and manuals.  Personal computers are impenetrable and harsh. Each new piece of hardware and software just made the mess worse.

Commands Missing from DOS

1. Tree Compare - This is like FC but compares all files in two directories and all the sub directories included.

2. Find Duplicates - There may be duplicate files scattered around on my hard drive. I don't know the names - just find them.

3. Tree size - Notice how Windows Explorer (was File Manager) does not reveal how much space is used in a folder and all included sub directories as a single number. Well how about "Myfolder contains 23 MB."

4. A real file finder that finds "all files of type txt containing the words 'software history' but not the words 'for sale' made between the dates 6-10-98 23:50 and 6-11-98 01:15."

Other Problems with Microsoft Software

The Windows Explorer like the original File Manager copied the DOS DIR command and none of them list the size of directories. Why the business community allows this failure is beyond me.

The new Microsoft Internet Explorer has a unique address book format called FAVORITES. The new FAVORITES is not compatible with the old FAVORITES and of course it cannot use Netscape BOOKMARKS files. On the other hand the new Netscape BOOKMARKS can read the old BOOKMARKS and the old FAVORITES and possibly other files as well using an automatic interpreter called IMPORT. Explorer does not have IMPORT.

Outline for writing code correctly

Good programs are mostly comments. Comments should explain why a program is being written, who it is for, what equipment it runs in, what each line does, what the interfaces are, what the interfacing equipment is, maybe even how it works, and finally who wrote it and the date of each change. Remember one time I wrote detailed instructions which were sufficient for a novice and perfect stranger. A year later I found out who that stranger was. It was me, and every detail was criticized for lacking justification.

Outline for writing the manual correctly

A software manual must define the operating environment for the software. It must describe how the equipment is prepared to run the software. It must describe how to start the software and how to stop it once started.

Ninety percent of all programming effort is identifying the interfaces and the requirements. Learning and writing code is the easy part.


4. Coax Ethernet for  Windows 98  to Windows 3.11

Coax was chosen because it is both cheap and well understood. Don't forget the 50 ohm anti reflection terminations at each end.

 a. Network setup for the Celeron 333 using Windows 98

Go to Start, Settings, Control Panel, Network.
Click on Client for Microsoft Networks. Press Add, Service.
Select File and printer sharing for Microsoft networks.
Allow it to read the CD and restart.

Go to Start, Settings, Control Panel, Network.
Under the Configuration tab highlight Client for Microsoft Networks and press Properties.
In the properties window "Logon validation" avoid the logon square. It is for networks with servers. Put in the work group name. I used workgroup. In the box below "Network Logon Options" choose "Logon and Restore Network Connections."

Go to Start, Settings, Control Panel, Network.
Under the Configuration tab highlight
File and printer sharing for Microsoft networks. Press
Properties and Verify that Browse Master is Automatic.
Close out the Control Panel.

Open the desktop Network Neighborhood and verify that the
other computers in the network are there.

In Windows Explorer highlight the C drive and pull down
Files, Sharing. Do not use the drive letter as the name or
the other computer operator may get confused about whose
drive is whose. Access type is Full for me and Passwords
is left blank.

b. Setting up the '486 Using Windows 3.1.1

Only famous name like 3-Com cards can install and they must use the
windows disk 7 and 8 drivers. Hardware supplied drivers and 3-Com's
internet downloadable drivers do not work.
The I/O address was 300 and ROM memory DC00 and interrupt 5 for both
486 3-Com cards.

Method for connecting 3-Com Ethernet cards.
These cards are connected by coax cable and require 50 ohm
termination. There is no hub, server or anything between the machines.
Configure and test the cards using 3c503.exe. This 3-Com program does
complete diagnostics for the card and network.
Configure I/O address 300, ROM address DC00, Interrupt 5 using the
diagnostic tests. 1. Internal, 2. Adapter, 3. Echo.  For Echo,
configure the card on the other computer using Echo server setup.

Under Windows Networking choose Install Microsoft Windows Network.
Turn on the sharing. Under adapters choose 3-Com Etherlink II or
IITP 8 or 16 bit NDIS2 compatible.
For protocol choose IPX/SPX transport with NETBIOS
and Microsoft NetBEUI.
Set Base I/O port 0x300, ROM address DC00, Interrupt 5 and
location onboard. User name will become the login user, and
work group was workgroup, Computer name can be anything since it is
not used.

Add:
To Autoexec.bat:   C:\windows\net start
To Config.sys: DEVICE=ETH503.sys /I:5

When restarting it says Windows has modified your Config.sys
system.ini, and protocol.ini, and put the old ones as .000.

In Main\Control Panel\Network choose a computer name and a
work group name, if this was not done above.
The computer name must be different for each computer while the
work group name must be the same.
In File Manager to share hard drives under Disk choose Share as.
Pick a proper name which is not a drive letter so you won't get
mixed up with the hard drive on the other computer.
Then go to the other computer and choose Connect network drive
in the same menu.
Because I had used interlnk and intersvr to serial link the other
machine and had already two floppies and two hard drives, I got up
to drive h!
The other machine had a parallel zip and got up to g.

For printing Print Manager has to be running and open.
Maybe the sharing has to be reselected while Print Manager
is open. A list of printers shows up on the machine with
the printer.


5. The Effect of Electric Current on Human Flesh

a. Disclaimer

Don't blame me if you fry yourself. I did these experiments already and they never need to be done again. Just take my word for it. Most folks even electronic engineers have a weak understanding of what it takes to get shocked. So here it is.

b. Getting the current

Most folks believe that high voltage is the thing to be afraid of. Other folks believe that high current is the thing to fear. I got the inside story when I was a kid. For those who believe in causality (cause and effect), I used to believe that voltage causes current. That idea comes natural considering the hydraulic analogy. Doesn't pressure cause flow? Well if voltage is like pressure and current is like flow, then I am right. Well some scientists just say voltage and current are "associated" with each other. Any way I claim that you need voltage ACROSS YOU to get current THROUGH YOU. It doesn't do a bit of good (assuming you are trying to get fried) to grab a million volt (DC) wire, if you are otherwise insulated. You will just sit there and FLOAT at the wire's potential.

There is a similar story for current. All the current in the world will not hurt you if it is going through someone else. Touching a big old bus bar that is carrying a lot of current at low voltage will not hurt you.

c. Feeling the current

To feel it you have to make the current flow through you. That usually means two connections. Before you even try to get connected, lets talk about impedance. Devices that can produce high voltage but only tiny currents are called high impedance. Static electricity is like that. You can get up out of a chair and get charged up to ten thousand volts. Walk over to a grounded pipe and touch it. If you can see the spark it is something like five thousand volts. If you feel it, that is more like ten thousand volts. If it hurts that is more like thirty thousand volts. Notice you didn't feel the ten thousand volts on you while you carried it across the room. I have a 750 volt power supply that has such a high impedance that I can touch across it and not even feel a thing.
Now if the high voltage source is low impedance look out. I heard about a squirrel that got connected across the terminals of a 13kv transformer. There was a loud bang and a white flash. Little black pieces of the squirrel came raining down.

Then there are low impedance devices. These devices can produce a large current but have only very small voltages. A battery powered flashlight is almost a low impedance device. Try holding a metal flashlight while it is on. You will not feel anything even though there is a current through the metal that you are touching. There is even a voltage across it end to end. You can touch both ends and not feel anything, but there is a tiny current through you.

If you find a low voltage high impedance source of electricity, that is a pretty weak signal. You will probably need amplifiers to detect it.

I had a debate about low voltage high current sources one time at the University of New Orleans. Harry told me that a device that can give a high current even at a low voltage would be dangerous. So I brought in a bank of a dozen capacitors totaling 20,000 microfarads, connected by two big old bus bars a half inch apart and not insulated. I asked him if my toy was dangerous. He said it should definitely shock me. So we went back into a physics lab and connected up a DC power supply and a voltmeter. It had a current meter so we could watch the capacitors charge.

I turned it on, brought the capacitors up to 12v, and asked "Will I feel that?" He said yes, so I touched both bus bars. I felt nothing. My hands were dry.

I brought the voltage up to 20v. "Will that shock me?" "Yes" I touched again and felt nothing.

I increased the voltage to 30v. "Will that shock me?" "Yes" I touched again and felt nothing.

I increased the voltage to 50v. "Will that shock me?" "Yes, That should be a horrible shock. Look out!" I put my hand flat on the bus bars and felt nothing. "Are you satisfied now that I am right?" He was still not certain.

I decreased the voltage of the power supply by turning the knob. Suddenly the voltmeter went to zero. The capacitors had discharged through a sneak circuit in the power supply. I played with the knob. The power supply was dead. A transistor had been blown away-shorted. I was in trouble. We got a schematic and figured out how this Disaster had happened. I offered to fix it but the old guy who ran the shop just wanted us to go away. I have much bigger capacitors now, totaling a half a Farad, and I am scared of shorting them. I keep them charged anyway. It is good for electrolytic capacitors

One time I made measurements to find out what current feels like. I connected a milliammeter in series with one hand and used the other to adjust the voltage of a DC power supply.

I found out that I could not feel currents below 0.1 ma. This fact is used by the familiar touch lamps that turn on and off when you touch them. The current is just too small to feel.

At about one ma I started to feel something. It didn't really hurt, but it was definitely detectable.

At five milliamperes the current hurt, and at ten ma it hurt so bad I couldn't stand it any more. I figure 100 ma would do real damage. I heard that shock treatment uses hundreds of ma and that scares me.

Now to change to voltage. I never measured the current but for DC, with dry fingers, I start to feel it somewhere between 70 and 90v. At 100v it is getting bad, and 105 is all I can stand. The strangest feeling though is that only one finger hurt when the current was steady. Then when I pulled my hand away and free, the other finger suddenly hurt instead for a few seconds. DC is strange. For AC lower voltages can be felt.

I once saw a government chart of pain or damage vs frequency. It was remarked by a buddy that 60 Hz (the standard AC power frequency) was the most dangerous frequency that could be found.

In my time I have had my share of accidents. Once a tube circuit power transformer shorted out while I was connected between the earth ground and chassis ground. I later measured 350 VAC. The current was from one finger on one hand to another finger on the other hand. It was so bad that I could not move either arm to get away. On my left hand the metal probe was actually making a small hole in my skin about 2 mm diameter. I saw the hole growing and for a second could do nothing. Finally, I stood up and pulled away breaking the circuit. I have since noticed that the outermost millimeter of skin accounts for the majority of the resistance measured by a voltmeter. Once an injury has broken the skin the current may increase markedly.

Another time I had a 2 microfarad capacitor charged to 1000V just laying around. I was looking for something and my chin touched a wire sticking up. The other contact was a hand. The effect was instant. I saw a flash of light like the A bomb went off. I felt like someone hit me with a bat on the back of the neck. It was all over before I could even think of moving.

Some years later I was doing an RF experiment with 10 watts at 146 MHZ from a transmitter. The capacitor was held between two fingers with the power applied and it was arcing to one finger. I could smell my flesh burning for some time but deliberately would not let go in order to finish the experiment, which took 15 seconds. Afterward there was no visible mark and I concluded that the injury was much less significant than a mosquito bite. If you want to smell your own flesh burning just rub your hands together until they get warm and quickly sniff the hot surface. Another way is to put your hand directly over a camera flash strobe that is battery powered. (Not the kind with the burning bulbs. That will "fry" you.) Immediately after the flash, smell your hand.

Many years later I was a little safer, working with only one hand. My hand got connected across a similar capacitor. It was a light touch and a tiny piece of flesh exploded into a ball of yellow fire. The ball was about an inch in diameter. Afterward there was a visible mark on my skin about 0.5 mm diameter. So the color of exploding human flesh is orange from the sodium in it.

5. The Difficulty of Space Travel

Most folks don't realize how tough a place space is to live in. That includes the Moon and planets. The South Pole is a piece of cake in comparison. In space you have to carry your own air, water, and food. We still do not know how to recycle these things enough to make a difference. Space is full of radiation. If you stay there long enough you will be well cooked. Shielding is too expensive. At least at the South Pole you can go outside and breathe the air and pull off your gloves and feel your house. Try that in space. Then there is the propulsion. If your engine fails you won't even come back in pieces. I would rather be in a plane whose engine failed, than a rocket that lost 10% of its thrust. If you want to shoot for the stars you are really crazy. Any conceivable propulsion system will take millions of years to get there, and by then even the bugs that ate you after you died will be long dead. The electronics will be ruined by the radiation, and the fuel tanks shot full of holes by grains of sand that you happened by. But the worst thing is that if you are moving with any reasonable speed relative to your local space junk, and it might not be your fault, any small object that hits you, say a BB, will hit with the force of a stick of dynamite. A direct hit by a bowling ball sized object will act like a nuke. So forget it. Little green men will have similar problems so they will never come here.

6. The Coming Monsters

Folks in the United States think America was chosen by god to be eternal. Better study your history again. Not even an animal can claim that destiny. The hard school of evolution is not going to go away just for your funny ideas. Change is permanent. And to make things even worse there are folks like me who are fooling with genes, cloning all sorts of evil things. Someday we are going to start modifying human genes and people will never be the same again. If you want to face it squarely just remember the scientific rule from quantum mechanics- "If it can happen it will happen." Humanity cannot withstand this forever. The result of drastic meddling, can only be called MONSTERS. They will be here before you know it and they will inherit the Earth.

7. Lock-In Amplifiers

The lock in amplifier, sometimes called a boxcar averager, is an amazing device. It uses a phase sensitive detector to pull a constant frequency AC signal out of noise. Because the bandwidth can be made very small, it can detect very tiny signals. The downside of this of course that you have to wait a long time to get the result - usually a yes /no decision about the presence of the signal.
.