Theatronics

eh, Sorry, like I said I slept through the (Heat Flows) part of the classes. (I lost it when we started talking about Resistance to Heat flow, I wondered if A high Heat flow through a High Heat flow resistance created something..? V = I * R so What = Heat * Insulation. All I could come up with is "Kool-aid Stands" then I passed out) Anyway, Yes they share the current. 4 devices on one side of the H-Drive (Lower Left lets say) carry 100A. Each will balance (Thanks to low Rds and Thermal compensation) a nearly equal share. (25A each) So here is my little dilemma: I could run 4 OZ copper and put nice thick traces on a PCB but I would prefer to mount these beasties off board. While I can use the TAB as one electrical connection and it seems capable of the current load (Again 25A) I am still concerned about the itty bitty legs of the TO-220 case and how to connect them. I hope that the package sinks enough heat away from the legs that they don't blow like fuses. I guess I'll just go with my current plan. Electronics isn’t beautiful! thanks -Mike

;D am I? ??? No. Just that the T&N theroms (When it comes to Resistor networks) are all about placement. In a straight line Series circuit there is no difference. But in a mesh where the currents will be divided amoung different branches, the LED might not even light because the potentials could result in unexpected polarity issues. Just forewarning that there are times when it is critical, but not here. Example:

You guys had me worried for a while with all that talk about Post Amp signal Mixing. Whew. Yea, the Way it happens is all thanks to the Zenith/GE pilot tone multiplex system. But if that fancy widget fails, you can still get a mono signal. The result is the combo of both left and right signals just like in the world of AM. And all that without shoring out your A-class amplifiers. But where is the fun in that. -Mike

Wow, now there is a doozer. Lets see. 1) Use a PIC that has the RS232 as part of the hardware. Some chips have it built in. The PIC16,17 and 18 families are good places to start. The Smaller PIC 12 chips can also be used but you will need to download the RS-232 source code from MicroChips web site to implement the interface. On the plus side, you get to pick the I/O pins. 2) get a MAX232 chip. (that’s the part number MAX232) Using a few capacitors and a single 5v supply it can create the +/- 12V signal needed for the RS232 standard. 3) the RS232 on your Freq Counter is most likely a very low order interface. SO you should be able to get by with just pins 2 and 3 plus a ground or signal return. On a DB9 that would be 2,3 and 5 while on a DB25 it's 2,3 and 7 Electronically that’s about as far as I can help without knowing more about what you are doing. One question though. Unless you are counting VERY high frequencies, Most PIC chips have a built in counter that can be triggered from a I/O pin. If you feed your signal to the chip all you need to do is disable the input, reset the counter, enable the input, wait 1 second, disable the input again, and then read the counter! Lots easier than the whole RS-232 interface stuff. The counters can count 16 bits on some chips and they run fairly quick. -Mike

from one software person to another. Series = Single step modal program Parallel = Multi-step / Time slice / Non-Modal Program ;D To really start you off on the right foot. Resistors are additive. If you need 10,000 ohms (10K) and you only have a box of 1000 ohm (1K) resistors. You could string 10 of them together and get a total of 10K ohms. IT wouldn't matter where the diode went. Ohm said: Current = Voltage divided by Resistance (I = E/R) He never said anything about order. Now Norton Thevenin, those two had a lot to say about resistors and order. Look them up. Good luck, and ask more questions. -Mike

This only applies to PNP / NPN Bipolar transistors. (3906 & 3904 examples) A handy way to remember how they work is to remember what they look like inside. a PNP is made of 3 regions, Positive, Negative and Positive again. NPN is the same but reversed. In a diode there are only 2 regions, P and N. Electron flow is from the N to the P. (Cathode to Anode) Basically a PNP is the same as two diodes "Pointing" at each other (Cathodes connected) and NPN has the Anodes connected "Pointing" Away from each other. Where the diodes meet is called the Base. The BASE to EMITTER flow is the same as the diode that makes it up. n-PN has the electron flow from the Emitter to the base (Base more positive than emitter) p-NP has the electron flow from the Base to the Emitter (Emitter more positive than Base) The current flowing between the collector and emitter, in simple DC terms, is the "Hfe" or "Beta" of the transistor times the current between the Base and Emitter. 0.01ma current at Base times a beta(Hfe) of 120 = 1.2ma current through Collector. (Remember the current at the Emitter is the BASE current PLUS the Collector Current in this case 1.2ma plus 0.01ma = 1.21ma) Anyone want to know what a Q point is? -Mike

Thermodynamics. Now there is a fun subject. ::) Well I wish it was. Having to build my first SUPER circuit I am having trouble handling the high current and temps created and I was wondering if anyone had a good solution for this type of challenge. The design is a H-Bridge that is rated at 100A / 100V DC it is built using 16 N channel MosFet parts in TO-220 cases. I had thought about 'clamping' 4 of them at time between 2 aluminum plates and using the metal tab as one of the connections for the power (Battery or Motor) Then I could bolt the 8Ga Wire to the metal plates. While the trigger lead is just a low current signal it can be soldered to a wire with shrink tubing over it and run back to the board. The trouble comes with the 3rd and last lead. How do I connect 8ga (AWG) wire to a little to-220 case lead. In effect I am trying to solder a 8ga wire to a 22ga one. One method I had seen involved bolting the remaining lead to one of the plates and then insulating the two plates. This would result in a 'Chip Sandwich' The top and bottom plates would be the POWER buss and the Trigger gate connections would be soldered directly to the TO-220 case pins. Any better ideas out there? Thanks, -Mike

All correct: To expand.. In a 4 channel system, both the Tx and RX units are tuned to the same freq. AM/FM doesn't matter Just assume there is a wireless link between the two for now. Both units also expect to work with a fixed number of channels, in this case 4. The pulse train would run 4 pulses with a long pause between them to signal the end of the sequence. Each pulse is between 1 and 2 milliseconds long. Most RC systems accept that 1.5ms means Centered while 1 and 2 represent the far left and far right positions of a servo. The breaks between each pulse and the long break between the sequences can vary depending on the manufacturer. Some use an edge triggered sync system and the break between the pulses is nothing more than a pulse. A simple sequencer takes the pulses and 'routes' them to different outputs on the servo driver. In fact if you watched the signals coming into the Rx unit and the signals going to out the servos, they would line up. While Pulse 1 is being routed to Servo 1, The outputs on 2-4 are idle. Hope that helps some also... -Mike The breaks between each pulse and the long break between the sequences can vary depending on the manufacturer. Some use an edge triggered sync system and the break beween the pulses is nothing more than a pulse. A simple sequencer takes the pulses and 'routes' them to different outputs on the servo driver. In fact if you watched the signals coming into the Rx unit and the signals going to out the servos, they would line up. While Pulse 1 is being routed to Servo 1, The outputs on 2-4 are idle. Hope that helps some also... -Mike

The old Standard was to place the digital power pins in the corners of the chips. in the 70s this made board layout a bit easier. Power buses ran between the chips in rows and they were much larger traces than the digital signals that could sneak in and out between the pins. Analogue chips didn't live in a world of straight rows and data busses. The pin placement was driven by design needs. Some parts were designed as direct replacements of older parts. Other parts had pin layouts based on expected use. For instance, if possible, two related pins might be placed near each other to simplify PCB layout. There are a few industry standards but they are not rules, just traditions. Just like most people like to mount through hole parts all on one side of the PCB and call it the top. If a part doesn't fit one way, why not try mounting it from the bottom? Presto, a reverse pinout! -Mike

There are a few other clever parts to some driver chips. A few models supply a built in PWM generator. I like the HIP4080A for the projects I have worked on. It has a feed back loop analogue mode that generates the PWM onboard based on a voltage input target. Although I have found that mode a bit tricky to use. I prefer to use it in a digital mode where I can feed it a direction and PWM signal and the chip handles all the rest. The most valuable part of the chip is a charge pump that creates the high voltage needed to trigger the MosFets (N-Channel) at the top of the bridge as they need to see slightly more than the source voltage. The charge pumps on the chip create a little more than the input voltage Here is a link to a typical datasheet for this chip : http://www.intersil.com/data/an/an9404.pdf

When you say the port is on a PCB and connected to a 'Chip' Is this perhaps a UART or a Driver chip? For smaller projects I like the MAX232. I have had a lot of luck with an inexpensive logic probe. It shows 2 LEDs (Red=High, Green=Low) It is easy to follow a serial signal through a circuit just with that tool. I the device connects to a terminal of any kind, is it possiable to remove the driver chip/Uart from a socket and jumper the TX and RX signals to create a Loopback test. This will confirm that the problem is NOT in the circuitry leading up to the IC. Send more details, This should be easy to hunt down. ;) -Mike