More info
- From: "rogerdff" <rogerdff@xxxxxxxxxxx>
- Date: Thu, 14 Aug 2008 09:49:54 -0500
First of all, thanks for the replies so far.
Actually, we need to develop a final product with all the embedded
hardware designed in a case: PCB for the controller and for the
input/output modules.
The NI solution is very interesting and fast, but as far as I could
understand I will have to start from "zero" to design the final hardware.
For this reason, it is important to start working with a development kit
that offers most of the needed devices.
The Labview is being considered to develop a prototype. A DAC board is
being considered to generate the analog signals (simulating a conditioned
signal) used for the ADC inputs. Then, the digitalized signal is going to
be fed (using SPI, Parallel, etc) to the controller.
For the controller, I have seen so far two options:
1. From Atmel, the DIOPSIS 940HF
2. From TI, the TMS320F3812
cause both combine DSP and uC.
For the ADC, I am looking for an evaluation board that offers simultaneous
sampling for many channels (preferably as many as possible, since we need
at least 40 analog channels). So far, I found some options from TI:
- ADS8365, but uses parallel output
- ADS1178, that uses SPI
Since a lot of data is to be stored, it is also important to have at least
16-32Mb of SRAM, and a higher mass storage media such as SD cards of
2-4GB.
Digital input varying from 24 to 250Vdc is also needed: at least 2 times
the number of analog channels, which means at least 80 DI. Again, for the
prototype, a DO board with Labview is being considered.
Digital output for alarm purposes is also considered.
Communications features (Ethernet, USB) is basic requirement, as it is
also a keypad, LCD, etc.
When the prototype with the "final" firmware is designed, then the final
PCB input modules together with the controller PCB will be developed. The
test will be carried out using a digital box capable to generate the real
analog values coming from the field: currents ranging from 1 to 20pu (1 pu
equals to 1 or 5A) and voltages ranging from 1 to 2pu (1 pu equals to 66 or
115V).
But for prototyping, the ADC board and the controller are the critical
hardware components that I need to define until next week.
Well, that is the "architecture" we have envisaged so far. Any ideas are
welcomed.
Hope I was clear enough. Any other detail, please, just ask me.
Thanks again for all.
.
- Prev by Date: Re: Please help with Circular Convolution!
- Next by Date: Re: Lightning and Fourier transform of an impulse
- Previous by thread: Re: NEWBIE QUESTION: "DSP For Scientists and Engineers." No Maths?
- Next by thread: circular array beamforming
- Index(es):
Relevant Pages
|
