Savel brain dump in English!

It is written:
“Intel, in partnership with several other companies which include the likes of IBM, Dell, Compaq, HP and Microsoft, recently introduced what will be the new standard for PC I/o in the years to come; PCI Express (formerly 3GIO or 3rd Generation I/O) is the name which it was given by PCI-SIG, the committee responsible for overseeing the PCI interface standard.

Approved as a standard on April 17 2002, PCI Express is an evolutionary upgrade to the existing PCI bus. It co-exists with the now ‘legacy’ PCI bus, but in terms of form, it is something completely new. ”

Now it is 2005. There is motherboards with PCI express x16, video cards with PCI express x16. And there is PCI express x1 slot in mainboards… But have you ever seen any PCI express device in the market? NO! So, why place that f**ked slots in motherboards? And why to remove simple PCI slots? My computer has one x16and 2 x1 PCI express slots. Thanks to God and my bright mind that I asked my computer with maximum simple PCI slots… Now all PCI express slots are empty and regular PCI is fully populated: SCSI controller, TV tuner, Firewire – IEE1394 controller, Fast LAN card. And now, I want to insert more cards to my computer and there is not place. Where are PCI express x1 devices?!?!

Maybe it is time to remove x1 slots, and rename PCIe to AGP EXPRESS!

From time to time I acquire “dead” LCD monitors. If you start monitor and see normal image and some horizontal or vertical line (black or color) this means that monitor is really dead. In very rare case you can press the LCD module and make the line disappear… But the line means that there is problem in very delicate part of LCD module. But if your monitor is not showing at all, or you see static pattern you have the chance. In fact about 80% of these monitors are repairable.

To begin repairing the LCD monitor you must have some soldering skills. Some spare LCD modules and some spare LCD controllers. These are used to test you new acquired LCD monitor modules. The LCD module can be broken (if have on with crack in the matrix), the only use of that part is to check if monitor is working. Also you must have spare modules for different sizes- 15″ and 17″ modules are not swappable. Also you need hacked back-light lamp generator to check that nice tube lamps.

If you see static pattern and LCD controller is OK, you must open the “non repairable” LCD module. You’ll find very few details. And secondary power supply- check them. Also check for missing parts. Also, with multimeter test the capacitors, resistors and etc. Look for open and short circuit. Also, good practice, is to examine the assembling. Sometime some tiny wire of flexible PCB jumps from its place.

The next image shows 17″ LCD module “Chungwa” CLAA170CE (or LM170E01 from LG Philips, LTM170??? from Samsung). In fact, don’t care who ever made LCD module. You can swap them! I’ve tested Philips modules with Samsung monitor controllers and vice versa. The LCD monitor controller ARE THE SAME. They differ only by software (firmware).

This monitor was not working at all. Few dull line and nothing. I disassembled it for spare parts. But before tearing it apart, I decided to play with it. At it began working. Maybe there was some bad contact in assembly or something. Another problem of this model- the back-light lamp connectors. And lamps were very worn out. So I soldered new lamps. (In other models the lamps are changeable, just remove 2 screws and pull out the lamps).

Now assembled the monitor without the case (I don’t have it!). That small PCB is the controller, and that strange white thing is power supply wrapped to isolating film. (230V lethal voltage!).

The next stage will be to find any suitable 17″ monitor case…

In my closet I found very old mobile phone. It is Motorola Associate 2000.

The power supply and antenna is lost. But the most interesting part, that the phone is still working. Of course I can’t make calls, but I can hear busy tone or after pressing few buttons I hear “This service is not available for you”. It is funny, as officially NMT-450 phone system is not working for about 10 years. As I know, the system is used in local Telecome for giving phone numbers where copper wires are too expensive- in distant villages or in places where all copper pairs are already used.

From time to time, I collect some interesting radio tubes, lamps and other strange devices. Here is small part of the collection:

From left to right. 8MHz crystal oscillator. Not very old- 1984. But look at these nice gold plated wires inside the bulb. Maybe it is some precision quartz crystal generator?

Next, the bulb with needle inside. It is so called spherical high intensity discharge lamp (HID lamp). Inside there is some mercury blob and high pressure xenon gas. (Osram XBO150 150W Xenon Short Arc Lamp). Such lamps are the brightest lamps ever made. This model is very small, only 150W. But this power is placed in 4mm gap. These lamps are used in theatre projectors. For example IMAX theatre uses 15000W water cooled lamp. These lamps must be powered with special equipment to stabilise lamp current.

And the last lamp is from other marginal side. This is optical multiplier. The purpose of such lamp is to detect very low light. These photomultipliers can detect single photons. This model is very small. Photon getting inside the tube smashed one electron from special coating. Then this electron smashes to special plate and strike more electron. When the electrons visits all the plates, the charge is detectable by external electronic components.

“Burr-Brown is known as an industry leader of high-precision, audio quality op amps. These ICs can be used in a variety of audio applications.” I am using OPA541 in metal TO-3 package. Continuous current output 5A (10A peek), internal power dissipation 125W … And it is using +-40V (80V) power supply. It gives lots of power for audio amplifier. And the quality of such power amp is very very good. I am using 4 chips, so I can bridge them or connect in parallel. The power supply for this monster must be powerful too. It is still under construction- now it is modified ATX. It gives me +-40V (80V total), but only 2A… Must redesign it… but current PSU is powerful enough to make my S-70 loudspeaker jump. And I love calm music. So all fans and radiators are useless.

The schematics are not complicated, even novice cam build it. I am using Texas Instruments (owner of Burr Brown) technical article No: SBOA082, here is part of schematics:

I divided schematics to 2 parts: one- preamp (is using computer as a source, pre amp is not useful) and power stage. Power stage is made from SMD parts to minimize the wire length and reduce possible layout problems. Here is photo of PCB:

See lots of ceramic capacitors on power lines and find black 0.1 ohm shunt power resistor. It is from OPA541 datasheet. It limits short circuit current. 0.1 ohm gives us 5A limit. This limits my output power, but I still have enough power for music. The shunt must dissipate about 2W of thermal power. Just place 000 resistor on it and enjoy full power and risk of short circuit.

I don’t what to write this day. Lots of hard work at yard. Made a nice paddock for my horse in neighbour’s yard .

So, I think I have the slowest Intel Pentium processor ever made. This nice CPU can have famous DIV bug and I thing it has lifetime Intel warranty for changing it to bug free. Nice solid ceramic chip, with lots of gold. Look at it:

It runs at amazing 60MHz speed. But what a design… I love old PGA chips. Collected small collection of them. From 386 to P4. The biggest one is Pentium Pro. Also I have some chips which is banned in USA! There is printing on it: not for resale in USA. It is old story, how Intel wanted to © his x86 technology…

I was tired of fans running and full speed and causing lots of noise. For my self-made audio amplifier I designed thermocontrolled fans. Why to move air if you amp is playing calm music? The total amp output power is about 400W, but when playing normal music while surfing the net it uses only a fraction of power. The power amplifier is using modified ATX power supply for +-40V. The fan of PSU is moving always, but the radiator is only warm. And the radiators of the amplifier are hot as corpse’s arse.
Here is the schematics and PCB layout:

The details used in device are very regular, found in old AT and ATX power supply. Only one comparator used from LM339. Notes for schematics: R3 is NTC thermoresistor. Use R5 trimmer to adjust the fan start temperature. R3 must be connected to head source- radiator. R1 is used for hysteresis. Can be removed if R3 is in good thermal contact with radiator. C1 and C2 are only to remove possible false operations from power supply spikes of RFI. If C1 is big enough, it can be used to start the fan for few turns on power up. To move the oil and to test if it is working. PCB to download.

It is linear FAN speed control, for more ecomonic way look at PWM fan speed control.

Look at this photo:

What you can see here? Some 168 pin DIMM? The first 3 looks like devices from same family… but they are very different. From left to right:
1. It is ROM! “Phaser 850 Ver 1.1 Roman Fonts. (c)2000 Xerox Corp. bla bla…”
2. It is 5V, buffered DIMM. Used is old Apple computer, also in some other exotic
computers.
3. It is PC100 3.3V DIMM. The one you use in PC world. (old pc…)
4. It is DSIMM used in Sun SPARCstation 10.

The first 3 differs only by position of the notches. The notches indicated voltage: 5V, 3.3V and reserved and type: unbuffered, buffered, reserved…

Having lots of second hand power mosfets? These interesting electronic components are very useful in hobby constructing. It replaces relays and simple transistors. But how to test them quickly?

All you need is simple tester. Almost every tester have buzzer for circuit continuity testing and in same configuration you can test diodes. The multimeter LCD show voltage dropout on diode. Now look at schematic diagram of N-channel mosfet.

You can see diode in drain-source path. Lets check it:

You can see 0.5V dropout on internal Schottky diode. In reverse there must be nothing if mosfet is open.

My multimeter shows “1″ if open circuit.
Now we must check the gate. There must be no connection in any other pin and in any other direction.


But here comes most interesting part. When we connect wires to gate, in one case we open device and in other we close. If power mosfet is good quality, the residual charge of testing leave the device open. So touch the gate and check the source- drain path.

It showing short circuit, as device is open ant the resistance is only fractions of ohms. Discharge the gate, and the device must close. P-channel devices work in same way, only polarity of gate is reversed.

I found LCD module (display) from Epson ECM-A0404. It looks like 20×2 module. Module has 16 pin connector… hmm… looks like standard LCD module.

I connected wires according usual pin-out. But, oops!, some sort of short circuit. At first glance, it looks like, that pin 1 and 2 are swapped. Reconnected wires and nothing… Took another module- it has small left-out of old PCB. It looks 2 and 3 connected to GND. And 1 maybe is Vcc (Power). Rewired again- nothing. Maybe I damaged module, switched module and again nothing.

For testing purposes connected standard LCD module (16×2) … and it works.

This is Win-Amp plug-in working on Windows XP and WinAmp5.08e. This means, that EPSON module is using some non standard pin-out. Short search in interned didn’t help. Only few people searching for pin-out of this LCD module.