1) What is the limit for the lenth of a single wire? Right now the maxim length is around one feet(!) and I don't like it (these long wires do however emerge for "slow-rising" PROM data outputs. The fast CPLD output address wires are a bit, but not much, shorter). But the design looks so good now. From above, that is.
There is no theoretical limit for wire length, but signal integrity issues might arise. Here are some potential problems to keep in mind:
- Without termination resistors, signals can reflect and bounce multiple trips back and forth on each wire. The longer the wire, the longer the 'settling time' for reflections to diminish. The net effect is increased propagation delay, which means reduced timing margins. With a 12-inch (30 cm) wire, the round trip delay for a reflection is roughly 3.6nS. If you slow down the rise and fall time of the CPLD outputs, this will help hide the reflections in the (slow) rise/fall of the signal.
- Any signals which are edge-sensitive (clocks, write pulses, strobes, etc.) cannot tolerate excessive reflections or ringing. Signal edges must be clean, and adequate noise margins must be maintained in the face of undershoot. If you have signal integrity problems with edge-sensitive signals, there are three simple counter-measures.
1. Series termination -- a 33-ohm resistor near the source pin will reduce rise/fall time and attenuate reflections. Do not use series termination with signals which have multiple load pins, this is only for point-to-point single-source single-load signals.
2. Reduce signal drive current or "strength" -- this is essentially the same as series termination, it increases the output impedance of the signal driver. It also reduces the rise/fall time, which helps moderate signal reflections.
3. With clock signals (or other edge-sensitive signals) with multiple loads, you have one available option: parallel termination. In your application, I would suggest 100 ohms (to ground) at the very end of the wire furthest from the driver as a first-trial value. Some experimentation may be required to optimise the parallel termination resistor value.
With any signals with multiple load connections, make sure you wire these pins in a daisy chain rather than radially. The signal source must be at one end of the daisy chain. Stubs are allowed, as long as the stub is short (2 inches or less is my guess, in your application). Keep stubs as short as possible.
2) Is there any problem in switching two voltages (3V+5V) on "simultanously" using a 2P switch (I have however a quite large delay in the "weird" reset glitch)? I am thinking of putting the supply outside the "Mother Board". This makes it neat, but is it wise?
With a mechanical switch, there is little practical difference between a "delayed" contact make/break and simply turning only one of the supplies on at a time. This may, or may not, be a problem. Some devices have a tolerance for inputs driven to an over-voltage level referenced to GND, regardless of supply voltage to the input device. On the other hand, some devices express input over-voltage tolerance relative to the input device's positive supply voltage. You will need to check the device datasheets.
You may have a problem (it's usually called 'latch-up'), but you are probably OK.
Have you considered powering the 3V rail from a regulator which is in turn powered by the 5V rail? This will give you simultaneous power-up (and power-down) of both supply rails while requiring only a single-pole power switch (for the 5V rail).
I haven't given you much in the way of definite answers. Once in a great long while, I am (barely) overcome with just enough humility to confess that I'm not absolutely confident that I can provide definitive (and correct) answers.
Good to hear of your progress, Roger. Good luck to you!
-- Bob Elkind
SIGNATURE:
README for newbies is here: http://forums.xilinx.com/t5/New-Users-Forum/README-first-Help-for-new-users/td-p/219369
Summary:
1. Read the manual or user guide. Have you read the manual? Can you find the manual?
2. Search the forums (and search the web) for similar topics.
3. Do not post the same question on multiple forums.
4. Do not post a new topic or question on someone else's thread, start a new thread!
5. Students: Copying code is not the same as learning to design.
6 "It does not work" is not a question which can be answered. Provide useful details (with webpage, datasheet links, please).
7. You are not charged extra fees for comments in your code.
8. I am not paid for forum posts. If I write a good post, then I have been good for nothing.
