56303 | 563xx | 5680 | 56805 | 5680x | 56F80 | 56F800DEMO | 56F805 | 56f807 | 56F830 | ADC | Bootloader | Codec | CodeWarrior | CW5 | CW6 | Debugger | DSP56303 | DSP56303EVM | DSP563xx | DSP5680 | DSP56800 | DSP56807 | DSP56858 | DSP56858EVM | DSP56F803 | DSP56F805 | DSP56F807 | DSP56F80x | DSP56F826 | DSP56F827 | DSP56F8xx | EVM | FFT | Flash_over_jtag | GPIO | Interrupt | Interrupts | JTAG | LCD | Linker | MCF5307 | Metrowerks | Modulus | MSCAN | PCMaster | PWM | Quad | Rif | RTOS | SDK | SPI

Question: The Motorola Application Note AN2049, Some Characteristics and Design Notes for Crystal Feedback Oscillators, discusses how to produce reliable clock circuits. It shows a different approach than that documented in 56800/56800E User's Manuals? Which is correct? Answer: This application note applies to the 68300 family of MCUs. It is not applicable to 56800/56800E DSPs. The clock circuits are different. Please follow the recommendations shown in the User's Manuals.

(You need to be a member of motoroladsp -- send a blank email to motoroladsp-subscribe@yahoogroups.com )

We have found that the recommended circuit in the DSP56F80x User's Manual (Figure 15-1) does not start up reliably at temperatures below 0 degrees C (32 degrees F), even on the latest Rev. D DSP56F807 chips (date code ZKAA0140). I verified that it was the oscillator itself and not the PLL or some other part of the chip by connecting a high-impedance, low-capacitance probe and looking at the oscillator output directly (ie before it goes through any other internal circuits). It was definitely the oscillator itself that was failing, not the PLL or some other part of the chip. We spent weeks of effort to try to get the on-chip oscillator to start up reliably at low temperatures, without success. We tried all possible combinations in the oscillator circuit, with 0, 1, or 2 capacitors of various values (10pF to 47pF) to ground (at EXTAL and/or XTAL), different feedback resistor values from 1M to 22M ohms, and with or without the series resistor (Rs in Motorola Application Note AN2049). None of these circuits started up reliably at low temperatures, so we reluctantly decided (as strongly recommended in Motorola Application Note AN2049) that the only solution was to use a separate oscillator circuit that was guaranteed to start up at low temperatures. With all due respect to the folks at Motorola, I have to say that in my opinion based on over 37 years experience in electronics, AN2049 does apply to any inverter-style oscillator circuit. It doesn't matter whether the oscillator is part of a DSP chip, a communication processor, or an 8-bit CPU, the design issues are exactly the same. The basic properties required to get a stable, reliable oscillator circuit are well-known, and unchanged from when they were developed over 80 years ago. It's just basic electronic theory which is taught in every high-school level electronics class. These properties are: a) the closed-loop gain of the circuit must be precisely 1.0, and b) the phase shift around the loop must be exactly zero degrees, or some multiple of 360 degrees (ie one full cycle) Even if Motorola fixes this problem so the crystal oscillator does work perfectly, it's still much better to use the external oscillator circuit. As is explained in Motorola Application Note AN2049, a process shrink by Motorola (which could happen at any time in the future) would likely result in the oscillator failing to start up reliably. We would then be faced with spending a lot more time and money to do this work all over again. For example, say that the DSP chips are currently made with 0.25 micron geometry. A year from now, Motorola changes to 0.16 micron geometry and all our products stop working and have to be redesigned. We do this, but then another year later Motorola goes to 0.12 micron geometry, and we have to repeat the process. And then ... You get the picture. Using the external oscillator circuit completely eliminates all of the above problems. We did find that the rise/fall time specifications of the DSP56F80x chips do not permit the use of any industry-standard oscillator circuits. None of the off-the-shelf commercially available oscillator circuits have rise/fall times that meet this specification (3 nanoseconds maximum) at frequencies below 10MHz. We resolved this problem by adding an inexpensive "speed-up" schmitt trigger (ON Semi NL17SZ14) to the design. The overall cost of this design is marginally higher than the simple crystal circuit, but it is absolutely guaranteed to work over all combinations of high/low temperature and power supply voltage. Plus, you are completely insulated from any process shrinks or other future changes to the DSP chips. For us, it's a no-brainer to use the external oscillator circuit in all of our new designs, whether they use a DSP chip, 32-bit MCU, or 8-bit microprocessor. Regards, Art Johnson Senior Systems Analyst PMC Prime Mover Controls Inc. 3600 Gilmore Way Burnaby, B.C., Canada V5G 4R8 Phone: 604 433-4644 FAX: 604 433-5570 Email: http://www.pmc-controls.com -----Original Message----- From: mwmann_at_motorola [mailto:] Sent: Thursday, September 05, 2002 4:52 PM To: Subject: [motoroladsp] Application Note AN2049 vs. 56800/56800E User's Manuals Question: The Motorola Application Note AN2049, Some Characteristics and Design Notes for Crystal Feedback Oscillators, discusses how to produce reliable clock circuits. It shows a different approach than that documented in 56800/56800E User's Manuals? Which is correct? Answer: This application note applies to the 68300 family of MCUs. It is not applicable to 56800/56800E DSPs. The clock circuits are different. Please follow the recommendations shown in the User's Manuals.   _____________________________________ /groups.php3

(You need to be a member of motoroladsp -- send a blank email to motoroladsp-subscribe@yahoogroups.com )

Hi All, Two things: 1. Can anyone recommend a good way (read low component count) to allow the DSP56F805/7 analog input to handle a +/-10V analogue input (preferably differential). 2. Has anyone tried the IsoPod from NewMicros (www.isopod.net) which is basically a 1"x3" DSP56F805 board with RS232, RS485, CAN buffers and power supplies with most signals (No address or data busses) present on headers. It is programmed using a new form of Multitasking Forth (IsoMax). Regards, Peter Elliot

(You need to be a member of motoroladsp -- send a blank email to motoroladsp-subscribe@yahoogroups.com )

view schematic attach files 10Vin.pdf regards, alex --- In motoroladsp@y..., email@p... wrote: > Hi All, > > Two things: > > 1. Can anyone recommend a good way (read low component count) to allow > the DSP56F805/7 analog input to handle a +/-10V analogue input > (preferably differential). > > Regards, > > Peter Elliot --------------------------------- Type: application/pdf

(You need to be a member of motoroladsp -- send a blank email to motoroladsp-subscribe@yahoogroups.com )

Hi Alex, af> view schematic attach files 10Vin.pdf Thanks for the schematic. It is similar (but more complete) to what I was thinking. I didn't realize you could run it off a single ended 3.3V - most of my circuits used all sorts of weird voltages ;-) I'll give your circuit a whirl when I get my IsoPod for testing ($99 56F805 card with built in 'Forth'! - time to brush up on my Forth and state programming programming techniques). Thanks again, Regards, Peter

(You need to be a member of motoroladsp -- send a blank email to motoroladsp-subscribe@yahoogroups.com )

Can you provide a rule (ratio of R5 & R6 value to R1 & R2 value?) for modifying this circuit for other input ranges? What if I want an input range of +/- 500mV? Thanks! --- In motoroladsp@y..., alex fusti <alfusti@y...> wrote: > view schematic attach files 10Vin.pdf > > regards, > > alex > --- In motoroladsp@y..., email@p... wrote: > > Hi All, > > > > Two things: > > > > 1. Can anyone recommend a good way (read low component count) to allow > > the DSP56F805/7 analog input to handle a +/-10V analogue input > > (preferably differential). > > > > Regards, > > > > Peter Elliot > > ---------------------------------

(You need to be a member of motoroladsp -- send a blank email to motoroladsp-subscribe@yahoogroups.com )

Hi All, Modifying schematic 10Vin.pdf R1 = R2 R5 = R6 dVout = +/- 1.5V Vout = 1.65 + dVout dVout = dVin * R5 / R2 Gain = R5 / R2 = dVout / dVin Examples: Vin = +/- 500mV dVin = 500mV dVout = 1.5V Gain = 1.5V / 500mV = 3 = R5 / R2 R2 == 10K R5 = R2 * 3 = 30K Vin = +/- 10V dVin = 10V dVout = 1.5V Gain = 1.5V / 10V = 0.15 = R5 / R2 R2 == 10K R5 = R2 * 0.15 = 1.5K Regards, Alex --- In motoroladsp@y..., "rlmcintosh" <yahoo@m...> wrote: > Can you provide a rule (ratio of R5 & R6 value to R1 & R2 value?) for > modifying this circuit for other input ranges? What if I want an > input range of +/- 500mV? > Thanks! > > --- In motoroladsp@y..., alex fusti <alfusti@y...> wrote: > > view schematic attach files 10Vin.pdf > > > > regards, > > > > alex > > --- In motoroladsp@y..., email@p... wrote: > > > Hi All, > > > > > > Two things: > > > > > > 1. Can anyone recommend a good way (read low component count) to > allow > > > the DSP56F805/7 analog input to handle a +/-10V analogue input > > > (preferably differential). > > > > > > Regards, > > > > > > Peter Elliot > > > > > > > > --------------------------------- > >

(You need to be a member of motoroladsp -- send a blank email to motoroladsp-subscribe@yahoogroups.com )