Schematic discrepencies?
RinksCustoms
Posts: 531
Edited-
In ordering parts from an outside scource to build a propstick/demo board hybrid for a·"new-gen" fuel injection project w/data logging, I noticed that the E3 spec'd "24lc256" is a 256Kbit E3. shouldn't the correct p/n be 24LC64?·Do the boys @ parallax have something up their sleeve for a built-in data logger or bigger internal ram for a gen2 PROpeller? I guess time will tell. In reading the manual (thank you guys @ parrallax for this beautiful chip and IDE by the way·- not saurcastic), the prop looks to·have 32K program memory, 8 cogs each w/~2K ram·, nice!
·I've also noticed that with·a theoretical·"virtual" page buffer (with a bit of code), one can utilize·multiple E3's on the same bus as the primary·E3. With some E3's, up to 8 chips and/or up to 1M per chip using one, or the primary,·I2C memory bus.
·
· The virtual page buffer i'm reffering to would store 32/64/128 bytes of data in a variable array (which for say·8 variables, would multiply your datalogging resolution x16) and when a counter reached a preset number of bytes (like 128), it would send the page to one of the external eeproms via I2C. And when it was time to recal data, you could use a loop to load 128byte pages sequentially, read 8 bytes at a time sequentially from each page to replay data. The E3 that parallax chose is made by microchip and has two modes for r/w data, a byte mode and a page mode. The 1024Kbit ver from Microchip·has a page write of 128bytes.
·Now please correct me if'n i'm wrong here-
·http://ww1.microchip.com/downloads/en/DeviceDoc/21941C.pdf·is the 1024bit version of the standard EE. It specs 4 devices per I2C bus giving·a total of·4.096 Mbits /· 2MBytes.
BTW, what would the·BIN addressing limits be on the prop to an external device? If the prop is 32bit, and uses LONG's, the theoretical addressing limit·should be FFFF FFFF, right? Or would the 9bits for "control/device address/R/W" have to be part of each long making the theoretical limit 7F FFFF? In either case, the 1024Kbit EE (or multiples for that matter) should be fully addressable by the propeller, right?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Definetly a E3 (Electronics Engineer Extrodinare!)
"I laugh in the face of imposible,... not because i know it all, ... but because I don't know well enough!"
Post Edited (RinksCustoms) : 8/3/2006 7:11:40 PM GMT
In ordering parts from an outside scource to build a propstick/demo board hybrid for a·"new-gen" fuel injection project w/data logging, I noticed that the E3 spec'd "24lc256" is a 256Kbit E3. shouldn't the correct p/n be 24LC64?·Do the boys @ parallax have something up their sleeve for a built-in data logger or bigger internal ram for a gen2 PROpeller? I guess time will tell. In reading the manual (thank you guys @ parrallax for this beautiful chip and IDE by the way·- not saurcastic), the prop looks to·have 32K program memory, 8 cogs each w/~2K ram·, nice!
·I've also noticed that with·a theoretical·"virtual" page buffer (with a bit of code), one can utilize·multiple E3's on the same bus as the primary·E3. With some E3's, up to 8 chips and/or up to 1M per chip using one, or the primary,·I2C memory bus.
·
· The virtual page buffer i'm reffering to would store 32/64/128 bytes of data in a variable array (which for say·8 variables, would multiply your datalogging resolution x16) and when a counter reached a preset number of bytes (like 128), it would send the page to one of the external eeproms via I2C. And when it was time to recal data, you could use a loop to load 128byte pages sequentially, read 8 bytes at a time sequentially from each page to replay data. The E3 that parallax chose is made by microchip and has two modes for r/w data, a byte mode and a page mode. The 1024Kbit ver from Microchip·has a page write of 128bytes.
·Now please correct me if'n i'm wrong here-
·http://ww1.microchip.com/downloads/en/DeviceDoc/21941C.pdf·is the 1024bit version of the standard EE. It specs 4 devices per I2C bus giving·a total of·4.096 Mbits /· 2MBytes.
BTW, what would the·BIN addressing limits be on the prop to an external device? If the prop is 32bit, and uses LONG's, the theoretical addressing limit·should be FFFF FFFF, right? Or would the 9bits for "control/device address/R/W" have to be part of each long making the theoretical limit 7F FFFF? In either case, the 1024Kbit EE (or multiples for that matter) should be fully addressable by the propeller, right?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Definetly a E3 (Electronics Engineer Extrodinare!)
"I laugh in the face of imposible,... not because i know it all, ... but because I don't know well enough!"
Post Edited (RinksCustoms) : 8/3/2006 7:11:40 PM GMT
Comments
-Phil
I've thought much about this, and plan to do a data aquisition for my brothers flat track car. I too, have been thinking about exactly how to do it. My conclusion has been that I2C is not necessarily the best way. I don't know how fast/much you want to log, but if it's slow, you could get away with the I2C.
Also, I would consider the SD card route. You can get the same speeds, but have TONS more space. Plus, various members on this forum are working out how to increase the number of files one can place on the cards (if you want broken datalogging, aka, turning the car off, different days, etc...). I can't recall now, but I think the current number of files is relatively low for loggin use, but could be adequate for a race weekend.
Also, you can have multiple outputs, we have 32 pins avail, and an FI system doesn't need that many pins, so you have some left over to have MANY I2C outputs. Or, should I say, multiple chains of 8 eeproms. Keep in mind that the ATMEL 1Mb eeproms can only have 2 per chain, for a total of 2Mb. If you use 8 512kb eeproms on one chain, you get 4Mb.
-Parsko