4 ms·

Basics: Introduction to Zener Diodes - Evil Mad Scientist Laboratories

13 points by th0ma5 15y ago

losethos 15y ago
So, I make an EEPROM writer device for toner cartrige chips. Some chips don't respond at 5V but do at 4.8V or 4.5V... it's kinda weird. If you don't know what's going on, it's usually best to figure-it-out and not do African engineering. It's weird they needed different voltage. I don't think 3.3V worked. They were ASIC special EEPROMS with "copyright Xerox" write-protected burned into them. Anyway, so I learned about power audio circuits using diodes voltage drops to fix cross-over issues. I get the notion I can place several diodes with taps to get jumper-selectable cheap voltage settings with only one voltage regulator at the top. I email the surface-mount diode company and ask minimum voltage drop because it's not listed, just max. I get clueless rep who tells me "oh sure sure". I cannot test with real devices -- just have to build our final designs. We get the boards and they do not have much of a voltage drop. It wasn't just from light loading. It mostly worked because the strange chips were rare. What a miserable product, but it worked just fine. As I recall, I tested with a 1N4006 or whatever thru-hole axial diode and got .5V. I think I put a loading resistor on the supply that dwarfed the power drawn by the EEPROM stuff. I don't remember exactly, but I know I tested. I can name parts to prove my EE knowledge. 2N2222 NPN transistors LM741 op-amp TI555 one-shots 74LS244 octal buffers -- good for address lines 74LS245 octal bidirectional buffers -- good for data lines TIP120 darlington switching transistors good for stepper motor drivers. Happy? I know lots more. God says... C:\LoseThos\www.losethos.com\text\QUIX.TXT nd, in which are recorded the famous deeds of King Arthur, whom we in our popular Castilian invariably call King Artus, with regard to whom it is an ancient tradition, and commonly received all over that kingdom of Great Britain, that this king did not die, but was changed by magic art into a raven, and that in process of time he is to return to reign and recover his kingdom and sceptre; for which reason it cannot be proved that from that time to this any Englishman ever killed a raven? Well, then, ---- Those parts are out-dated. I don't think I'd like modern engineering without wire-wrap to test. FPGAs are cool. You know what, FPGAs seem really cool, but are probably only good for DSP image processing and stuff. If you make a CPU it's probably worthless. The notion of parallel computing engines to go along with CPU like they use GPUs... that sure sounds fun, but is probably worthless. DSP filters sounded cool to me at first, but are quickly discovered to be dull in practice.

13 points by th0ma5 15y ago

So, I make an EEPROM writer device for toner cartrige chips. Some chips don't respond at 5V but do at 4.8V or 4.5V... it's kinda weird. If you don't know what's going on, it's usually best to figure-it-out and not do African engineering. It's weird they needed different voltage. I don't think 3.3V worked. They were ASIC special EEPROMS with "copyright Xerox" write-protected burned into them. Anyway, so I learned about power audio circuits using diodes voltage drops to fix cross-over issues. I get the notion I can place several diodes with taps to get jumper-selectable cheap voltage settings with only one voltage regulator at the top. I email the surface-mount diode company and ask minimum voltage drop because it's not listed, just max. I get clueless rep who tells me "oh sure sure". I cannot test with real devices -- just have to build our final designs. We get the boards and they do not have much of a voltage drop. It wasn't just from light loading. It mostly worked because the strange chips were rare. What a miserable product, but it worked just fine. As I recall, I tested with a 1N4006 or whatever thru-hole axial diode and got .5V. I think I put a loading resistor on the supply that dwarfed the power drawn by the EEPROM stuff. I don't remember exactly, but I know I tested. I can name parts to prove my EE knowledge. 2N2222 NPN transistors LM741 op-amp TI555 one-shots 74LS244 octal buffers -- good for address lines 74LS245 octal bidirectional buffers -- good for data lines TIP120 darlington switching transistors good for stepper motor drivers. Happy? I know lots more. God says... C:\LoseThos\www.losethos.com\text\QUIX.TXT nd, in which are recorded the famous deeds of King Arthur, whom we in our popular Castilian invariably call King Artus, with regard to whom it is an ancient tradition, and commonly received all over that kingdom of Great Britain, that this king did not die, but was changed by magic art into a raven, and that in process of time he is to return to reign and recover his kingdom and sceptre; for which reason it cannot be proved that from that time to this any Englishman ever killed a raven? Well, then, ---- Those parts are out-dated. I don't think I'd like modern engineering without wire-wrap to test. FPGAs are cool. You know what, FPGAs seem really cool, but are probably only good for DSP image processing and stuff. If you make a CPU it's probably worthless. The notion of parallel computing engines to go along with CPU like they use GPUs... that sure sounds fun, but is probably worthless. DSP filters sounded cool to me at first, but are quickly discovered to be dull in practice.