From: dshesnicky <dshesnicky@yahoo.com> Date: Sun Oct 6, 2002 9:41am Subject: Re: dumb questions - #1 and #2
> The dt006 board can be programmed with bascom-avr, codevision, or > imagecraft compilers. it means you build your hex file from > AVRStudio4, then download it with say the demo version of > bascom-avr I'll download a version of it and burn it onto a cdrom for the laptop on Monday. Funny that Atmel went to all the trouble of building the AVRStudio but it can't program a simple direct connection to the chip. Am I missing something here? > No, although I would have got something with a bit more amperage, > this will suit the dt006 I would think the amperage would only be an issue if adding alot of boards to the slots though. The 300 mA number was mentioned on your web site Don. Thanks for the replies guys. Don
From: dshesnicky <dshesnicky@yahoo.com> Date: Sun Oct 6, 2002 9:44am Subject: Re: dumb questions - gcc?
> I'll download a version of it and burn it onto a cdrom > for the laptop on Monday. What about avr gcc? Can that program the chip? Don
From: Don McKenzie <support2003@dontronics.com> Date: Sun Oct 6, 2002 9:46am Subject: Re: Re: dumb questions - #1 and #2
dshesnicky wrote: > > Funny that Atmel went to all the trouble of building the > AVRStudio but it can't program a simple direct connection > to the chip. Am I missing something here? trouble is every author uses a different pinouts. > > No, although I would have got something with a bit more amperage, > > this will suit the dt006 > > I would think the amperage would only be an issue if adding > alot of boards to the slots though. The 300 mA number was > mentioned on your web site Don. blush, did I? I dare say I did, just that most people want to add something more useable sooner or later when doing development. > Thanks for the replies guys. Don McKenzie E-mail: http://www.dontronics.com/e-mail.html Home Page: http://www.dontronics.com Add USB to your Product in 10 min. http://www.dontronics.com/giga.html The World's Largest Range of Atmel/AVR & PICmicro Hardware and Software Please place (or leave -->) "dt2002" in the subject or body of your message to guarantee your message getting through our spam filters.
From: Don McKenzie <support2003@dontronics.com> Date: Sun Oct 6, 2002 9:51am Subject: Re: Re: dumb questions - gcc?
dshesnicky wrote: > > > I'll download a version of it and burn it onto a cdrom > > for the laptop on Monday. > > What about avr gcc? Can that program the chip? > > Don I doubt it. Not for the dt006, which is also known as the sample programmer, as it was designed by Sample Electronics in Korea. The first web programmer I saw was from Jerry Meng in China, this uses 3 diodes to grab power from the printer port, and is basically a db-25 connector, three diodes and a socket, but I doubt that it will be found in standard software packages. The pony programmer sounds great, if only it didn't have that reset problem that Alex mentioned. Don McKenzie E-mail: http://www.dontronics.com/e-mail.html Home Page: http://www.dontronics.com Add USB to your Product in 10 min. http://www.dontronics.com/giga.html The World's Largest Range of Atmel/AVR & PICmicro Hardware and Software Please place (or leave -->) "dt2002" in the subject or body of your message to guarantee your message getting through our spam filters.
From: Ben Hitchcock <ben@wollongong.apana.org.au> Date: Sun Oct 6, 2002 10:40am Subject: Re: Re: dumb questions - gcc?
Hi, Just to throw my hat into the ring, I use a DT-003 with a couple of DT-104's. The DT-003 is connected to the parallel port of an old toshiba laptop. I'm sure you would be able to set up a similar system on the DT-006. I use avr-gcc (http://medo.fov.uni-mb.si/mapp/uTools/), and it compiles C code just fine. I then use SP12 (http://home.wanadoo.nl/electro1/avr/sp12.htm) to program the chips. The combination works really well, and allows you to program AVR's in a familiar language environment. Any questions, please feel free to ask. ben On Sunday, October 6, 2002, at 09:51 AM, Don McKenzie wrote: > > > dshesnicky wrote: >> >>> I'll download a version of it and burn it onto a cdrom >>> for the laptop on Monday. >> >> What about avr gcc? Can that program the chip? >> >> Don > > I doubt it. Not for the dt006, which is also known as the sample > programmer, as it was designed by Sample Electronics in Korea. > > The first web programmer I saw was from Jerry Meng in China, > this uses 3 diodes to grab power from the printer port, and is > basically > a db-25 connector, three diodes and a socket, but I doubt that it will > be found in standard software packages. > > The pony programmer sounds great, if only it didn't have that reset > problem that Alex mentioned. > > Don McKenzie E-mail: http://www.dontronics.com/e-mail.html > Home Page: http://www.dontronics.com > > Add USB to your Product in 10 min. http://www.dontronics.com/giga.html > The World's Largest Range of Atmel/AVR & PICmicro Hardware and Software > > Please place (or leave -->) "dt2002" in the subject or body of your > message to guarantee your message getting through our spam filters. > > ------------------------ Yahoo! Groups Sponsor > ---------------------~--> > Home Selling? Try Us! > http://us.click.yahoo.com/QrPZMC/iTmEAA/MVfIAA/CFFolB/TM > --------------------------------------------------------------------- > ~-> > > To Post a message, send it to: simmstick@e... > > To Unsubscribe, send a blank message to: > simmstick-unsubscribe@e... > > Your use of Yahoo! Groups is subject to > http://docs.yahoo.com/info/terms/ > >
From: Alex Shepherd <ashepherd@wave.co.nz> Date: Sun Oct 6, 2002 2:23pm Subject: Re: dumb questions - #1 and #2
> Yes I remember you talking about this in a previous message Alex. > > > Don, how about doing a small parallel port board like: > > http://www.mcselec.com/an_200.htm that has the tri-statable buffer > > chip on > > it and a 10 way connector that would allow a direct connection to the > > programming port on your simmsticks. > > > > Alex > > what would be the main advantage? Well if you choose the right pins then one of the programmer options in Atmel AVR Studio will support it directly and you would not have to resort to PonyProg etc. I guess you could request Atmel to support the DT006 directly... Alex
From: Alex Shepherd <ashepherd@wave.co.nz> Date: Sun Oct 6, 2002 9:26pm Subject: Re: dumb questions - #1 and #2
> The other option is to use the PonyProg2000 program from: > http://www.lancos.com/prog.html. I have used the 2.05a beta version which > works well except that it does not release the RESET line after programming > so you have to unplug the cable to run - which is a bit of a pain. I have > reported this to the author, but it has not been resolved yet. This is a bit > of a shame as the program works real well except for the problem with RESET > and can be scripted from the command line. I had a play with another > parallel port program that allowed me to manually set the port lines etc and > I found that once I toggled the RESET line the DT006 would run fine again so > maybe its worth just making a very small ap to do just this after pongprog > is used. Well I have just tried this by I downloading the TVicHW32 version 5.0 package from http://www.entechtaiwan.com/tvichw32.htm and hacked their LPT demo (LptTest.exe) to automatically open the LPT port, set pin 4 which is the reset line and then exit. I then made a batch file that executed a PonyProg script file (an e2s file) and then called the LptTest.exe utility. I am actually using the Jens File Editor as my IDE and have configured buttons to call the avr-gcc make script and just defined another button to call the prog.bat file. It works great!!! The only problem is the TVic software is not freeware so now that I know the concept works, I'll have a dig around for a free library that will let me build an ap to do the same thing. Alex
From: Don McKenzie <support2003@dontronics.com> Date: Mon Oct 7, 2002 6:26am Subject: Re: [dtdealer] A Proposal for a New DTMB-2 Bus (DonTronics MicroBus)
CC: Dealers SimmStick Group Carl Dietz <dietzer@a...> Mick Gulovsen wrote: > > Gday All, > > I like it... I assume that the PCI socket is .05" pin spacing... > > Mick I guess it is Mick, because of the length of the board. I suggested to Carl, that we post the message to the SimmStick user group for feedback, which he agreed to, however he doesn't monitor the group, so if anyone has feedback, can you also please CC it onto Carl. Cheers Don... > Don McKenzie wrote: > > >Hi Carl, thanks for the feedback. > >I am taking the liberty of sending this message to my dealer list for > >possible comment. > >Would you like me to circulate it to the SimmStick group also? > > > >You are one of the few customers that have taken the time to give good > >feedback on this subject. > >I'll read your comments over a couple of times to digest it all myself. > >:-) > > > >Cheers Don... > > > > > > > >>Hello Don, > >> > >>I'm sorry I could not write back to you sooner. I've been swamped at > >>work with major overtime for the last several months. I hardly have > >>time to sleep anymore, let alone keep up with my hobbies! I wanted to > >>thank you so much for your quick shipment of my order. I was amazed > >>that only took 6 days to get here in Texas! It takes longer than that > >>to get mail from my parents in the northern US! > >> > >>I am working on scale R/C model warships, and am trying to build my own > >>control systems. I have already decided that your SimmStick(TM) module > >>approach was the best way to go, as I could plug in only the modules I > >>need for a given ship. I plan to make modules for controlling servos, > >>DC motors, stepper motors as well as adding ultrasonic ranging, digital > >>compass, and even pressure sensors and inclinometers for the model sub I > >>want to build someday. It will probably take me a couple of years to > >>get all this built and working, since I have so little spare time. To > >>make things easier, I am focusing only on two CPUs, the ATmega128 and > >>the ATmega8, which have the horsepower and features to meet my needs. > >> However, the 64pin ATmega128 doesn't fit well with the existing > >>SimmStick(TM) module limitation of 30pins. I'm therefore very > >>interested in seeing an new, expanded DTMB bus. > >> > >>I read your thoughts on a new DTMB bus that you posted on your website. > >> If I may say so, I believe you addressed the concerns admirably, > >>especially the issue of the PCB thickness. As I plan to prototype my > >>own PCBs before paying to have nice ones made, the 0.05" thickness of > >>the SIMM board is a real problem for me. Don't get me wrong, the DT208 > >>is a great help with this (I've already ordered two of them), but it is > >>not an ideal long-term solution. > >>I know people have given you feedback saying not to change, and I know > >>my feedback is late in coming. For that I apologize. But I really > >>would like to see an expanded DTMB-2 bus solution. To that end, I have > >>come up with my own proposal which I'd like to ask you to read and > >>critique. If you decide its not worth it for Dontronics to pursue it, > >>that's OK. I'll still pursue some version of it for my own needs. I > >>just thought it would be nice if I could share my ATmega128 board > >>designs with others using SimmStick(TM) modules. The proposal is > >>attached to this email. I welcome any and all feedback, even if you > >>flat out think the proposal is stupid and never want to pursue it > >>yourself. > >> > >>Many thanks, > >> > >>Carl > >> > >>dietzer@a... > >>cdietz@f... > >> > >> ------------------------------------------------------------------------ > >>A Proposal for a New DTMB-2 Bus (DonTronics Micro Bus) > >> > >>The DTMB-2 bus was born out of desire to expand the original 30-pin > >>SimmStick(TM) bus, also known as the DTMB-1 bus, commonly used for PIC > >>and Atmel microcontrollers with 40 pins or less. There are several > >>reasons for updating/replacing the SimmStick(TM) bus: > >> > >> 1. The 30-pin SimmStick(TM) is too small for microcontrollers with > >> 40 pins or more, especially the new 64-pin devices like the > >> Atmel ATmega128. > >> > >> 2. Each year it is harder to find sources for 30-pin SIMM connectors. > >> > >> 3. The 30-pin SIMM sockets require a 0.05" thick PC board, which are > >> also hard to find sources for. > >> > >>I believe that the new DTMB-2 (DonTronics Micro Bus) should have the > >>following features: > >> > >> 1. Like the SimmStick(TM) bus, the new bus should support both a > >> card-edge connection and 0.100" header connection. The card-edge > >> connection must work with 0.062" (1/16") PC boards, eliminating the > >> problems caused by the 0.05" thickness of SIMM boards. > >> > >> 2. The new bus must be fully backward compatible with the SimmStick(TM) > >> bus. It must not only support the same signals on the bus, but be > >> able to physically interface with existing SimmStick(TM) modules and > >> motherboards. The purpose of the new DTMB-2 bus is to expand and > >> enhance the SimmStick(TM) bus with a standard that is easier to > >> support over time, not to eliminate the SimmStick(TM) bus outright. > >> > >> 3. It should support an optional external memory interface, compatible > >> both the ATmega series and the PIC17CXX series of parts. Such an > >> interface would require 19 pins, most of which could be used as > >> general purpose I/O if the user does not need an external memory > >> interface in his design. > >> > >> 4. It should add a new power pin called VCC, which is dedicated to 3.3V. > >> This will allow newer designs to take advantage of the lower voltage. > >> > >> 5. Should be expandable in the future without sacrificing SimmStick(TM) > >> compatibility, or compatibility with itself. > >> > >>I believe the following proposal for the new DTMB-2 bus satisfies all the > >>above requirements. I believe it maintains not only full compatibility > >>with the SimmStick(TM) bus, but allows boards designed for the new bus to > >>mate with existing SimmStick boards. > >> > >>I propose the following definition for the DTMB-2 bus: > >> > >> 1. The DTMB-2 bus would use the common 60-pin PCI card-edge connector, > >> for the following reasons: > >> > >> a) The PCI bus, commonly used in PCs, will be around for quite some > >> time. The older PC and PC-AT buses can still be found on many PCI > >> motherboards, nearly 15 years after the introduction of the PC bus. > >> I believe that the PCI connector will have similar longevity. > >> > >> b) The PCI card-edge connector supports 0.062" (1/16") PC boards. > >> These board thicknesses are available from all PCB suppliers. > >> This board thickness also allows the use of common 1/16" board > >> guides to provide vertical support for taller DTMB-2 modules. > >> > >> c) The 60-pin connector provides an additional 30 pins to the > >> existing SimmStick(TM) bus. This gives better support for CPUs > >> with 40 and even 64 pins. > >> > >> d) There also exists a 92-pin PCI connector, which is an extension > >> of the 60-pin connector. A board designed for the 60-pin PCI > >> connector will plug into the 92-pin connector. The 92-pin > >> connector can be used for expanding the 60-pin DTMB-2 bus design > >> in the future to a 92-pin design while maintaining full backward > >> compatibility with both 60-pin DTMB-2 and 30-pin SimmStick(TM) > >> buses. > >> > >> e) The 60-pin PCI connector is only 3.34" long, which is actually > >> shorter than the 3.8" 30-pin SIMM connector. Thus the 60-pin > >> PCI connector allows a DTMB-2 module to be 3.5" long (or less), > >> basically matching the footprint of a SimmStick(TM) module. > >> > >> 2. The DTMB-2 bus supports the use of dual-row headers, 2x30 (or even > >> 2x32 if that is more available). Careful assigning of the bus > >> pin-out will assign the original SimmStick(TM) bus to one 30-pin > >> row of the new 2x30 header, allowing the user to plug SimmStick(TM) > >> boards into the new DTMB-2 module. This would even allow the use > >> of a DT208 fat to skinny converter board to allow plugging a DTMB-2 > >> module into a 30-pin SIMM socket ! > >> > >> As an option, the DTMB-2 module may add a single row 30-pin > >> header on the top edge of the board, duplicating the new 30 pins > >> of the DTMB-2 bus and use either a single row or dual row header > >> on the bottom edge of the board for the SimmStick(TM) signals. > >> This would allow stacking of DTMB-2 modules for users who prefer > >> stackable modules to the motherboard/daughterboard approach. > >> > >> 3. The DTMB-2 bus adds support for an external memory interface, > >> consisting of 19 pins, which can be used by both the 8-bit external > >> memory interface of the Atmel AVR series and the 16-bit external > >> memory interface of the PIC17CXX series of parts. This requires > >> three control pins, called *RD/*OE, *WR/A0, and ALE/A7 (where the * > >> denotes an active-low signal) and 16 address/data pins, AD0-AD7 > >> and AD8/A8 - AD15/A15. Modules that use the external memory > >> interface will be called DTMB-2EM modules, where the EM stands > >> for External Memory. > >> > >> Although the DTMB-2 bus supports both the 8-bit Atmel AVR and > >> the 16-bit PIC 17CXX external memory interfaces, you can NOT > >> use both 8-bit and 16-bit DTMB-2EM modules simultaneously. > >> This is due to the 8-bit Atmel external memory interface using > >> A8-A15 as an output-only bus, which would cause drive conflicts > >> with AD8-AD15 on the 16-bit PIC17CXX series external memory > >> interface. > >> > >> If the user does not wish to use DTMB-2EM modules in his design, > >> then pins *WR/A0, ALE/A7, AD0-AD7 and AD8/A8 - AD15/A15 can be > >> used for general purpose I/O. The pin *RD/*OE, however, will be > >> reservered, with the requirement that it be tied high through a > >> pullup resistor and be jumpered so that it can be disconnected > >> from the CPU. Then if a DTMB-2EM module is plugged into a DTMB-2 > >> bus with modules that use the same pins as general-purpose I/Os, > >> the DTMB-2EM module can not drive out on pins AD0 - AD7 and > >> AD8/A8 - AD15/A15 and thus prevents drive conflicts from occuring. > >> > >> 4. The DTMB-2 bus adds a new 3.3V VCC pin. > >> > >> 5. The DTMB-2 bus adds two new serial I/O pins called SI2 and SO2 for > >> CPUs with two serial ports. > >> > >> 6. The DTMB-2 bus adds 8 new general purpose I/Os, called B0-B7. > >> > >> 7. The DTMB-2 bus interleaves the original 30 SimmStick(TM) bus > >> signals with the other 30 DTMB-2 signals. This supports using > >> one row of the dual row header as a regular SimmStick(TM) header. > >> It also makes it easier to support both PCI and SIMM connectors > >> on the same motherboard. > >> > >> 8. DTMB-2 PINOUT > >> > >>SimmStick(TM) DTMB-2(TM) > >>------------------------------- > >> 1 A1 1 A1 > >> 2 *RD/*OE > >> 2 A2 3 A2 > >> 4 *WR/A0 > >> 3 A3 5 A3 > >> 6 ALE/A7 > >> 4 PWR 7 PWR > >> 8 VCC (3.3V) > >> 5 CI/A4 9 CI/A4 > >> 10 SI2 > >> 6 CO/A5 11 CO/A5 > >> 12 SO2 > >> 7 VDD 13 VDD (5V) > >> 14 B0 > >> 8 *RES 15 *RES > >> 16 B1 > >> 9 GND 17 GND > >> 18 B2 > >>10 SCL 19 SCL > >> 20 B3 > >>11 SDA 21 SDA > >> 22 B4 > >>12 SI 23 SI > >> 24 B5 > >>13 SO 25 SO > >> 26 B6 > >>14 IO/A6 27 IO/A6 > >> 28 B7 > >>15 D0 29 D0 > >> 30 AD0 > >>16 D1 31 D1 > >> 32 AD1 > >>17 D2 33 D2 > >> 34 AD2 > >>18 D3 35 D3 > >> 36 AD3 > >>19 D4 37 D4 > >> 38 AD4 > >>20 D5/MOSI 39 D5/MOSI > >> 40 AD5 > >>21 D6/MISO 41 D6/MISO > >> 42 AD6 > >>22 D7/SCK 43 D7/SCK > >> 44 AD7 > >>23 D8 45 D8 > >> 46 AD8/A8 > >>24 D9 47 D9 > >> 48 AD9/A9 > >>25 D10 49 D10 > >> 50 AD10/A10 > >>26 D11 51 D11 > >> 52 AD11/A11 > >>27 D12 53 D12 > >> 54 AD12/A12 > >>28 D13 55 D13 > >> 56 AD13/A13 > >>29 D14 57 D14 > >> 58 AD14/A14 > >>30 D15 59 D15 > >> 60 AD15/A15 > >> > >> 9. DTMB-2 Connector Sources > >> > >> One source for the 60-pin PCI connector is Digi-Key, > >> part# 145154-4-ND. The connector is made by Amp/Tyco > >> Electronics, part # 145154-4. > >> > >> One source for the 92-pin PCI connector is Digi-Key, > >> part# 145165-4-ND. The connector is made by Amp/Tyco > >> Electronics, part # 145165-4. > >> > >> > > > > > > -- Don McKenzie E-mail: http://www.dontronics.com/e-mail.html Home Page: http://www.dontronics.com Add USB to your Product in 10 min. http://www.dontronics.com/giga.html The World's Largest Range of Atmel/AVR & PICmicro Hardware and Software Please place (or leave -->) "dt2002" in the subject or body of your message to guarantee your message getting through our spam filters.
From: Alex Shepherd <ashepherd@wave.co.nz> Date: Mon Oct 7, 2002 8:21am Subject: Using a DT006 to program another board with a 10-way ISP header
Hi, Has anyone tried to connecting a cable from the 10-way ISP header on a DT006 (with no CPU) to another board with a 10-way ISP header, using a straight through 10-way ribbon cable with IDC connectors, to program a CPU in the second board? Cheers Alex Shepherd
From: Don McKenzie <support2003@dontronics.com> Date: Mon Oct 7, 2002 8:58am Subject: Re: Re: dumb questions - gcc?
Ben Hitchcock wrote: > > Hi, > > Just to throw my hat into the ring, I use a DT-003 with a couple of > DT-104's. The DT-003 is connected to the parallel port of an old > toshiba laptop. I'm sure you would be able to set up a similar > system > on the DT-006. > > I use avr-gcc (http://medo.fov.uni-mb.si/mapp/uTools/), and it > compiles > C code just fine. > I then use SP12 (http://home.wanadoo.nl/electro1/avr/sp12.htm) to > program the chips. > > The combination works really well, and allows you to program AVR's in > a > familiar language environment. > > Any questions, please feel free to ask. > > ben Thanks Ben, just updated the dt104 page with that info. Don McKenzie E-mail: http://www.dontronics.com/e-mail.html Home Page: http://www.dontronics.com Add USB to your Product in 10 min. http://www.dontronics.com/giga.html The World's Largest Range of Atmel/AVR & PICmicro Hardware and Software Please place (or leave -->) "dt2002" in the subject or body of your message to guarantee your message getting through our spam filters.
From: Don McKenzie <support2003@dontronics.com> Date: Mon Oct 7, 2002 9:59am Subject: Re: Using a DT006 to program another board with a 10-way ISP header
Alex Shepherd wrote: > > Hi, > > Has anyone tried to connecting a cable from the 10-way ISP header on a > DT006 > (with no CPU) to another board with a 10-way ISP header, using a > straight > through 10-way ribbon cable with IDC connectors, to program a CPU in > the > second board? > > Cheers > > Alex Shepherd yes, have a look at: http://www.lawicel.com/simmstick/ispsimm.htm Don McKenzie E-mail: http://www.dontronics.com/e-mail.html Home Page: http://www.dontronics.com Add USB to your Product in 10 min. http://www.dontronics.com/giga.html The World's Largest Range of Atmel/AVR & PICmicro Hardware and Software Please place (or leave -->) "dt2002" in the subject or body of your message to guarantee your message getting through our spam filters.
From: Fred Ledo <fredledo@atomic.net> Date: Mon Oct 7, 2002 9:31am Subject: Re: Re: [dtdealer] A Proposal for a New DTMB-2 Bus (DonTronics MicroBus)
Sounds real good !!! fred At 06:26 AM 10/7/02 +1000, you wrote: >CC: >Dealers >SimmStick Group >Carl Dietz <dietzer@a...> > >Mick Gulovsen wrote: > > > > Gday All, > > > > I like it... I assume that the PCI socket is .05" pin spacing... > > > > Mick > >I guess it is Mick, because of the length of the board. >I suggested to Carl, that we post the message to the SimmStick user >group for feedback, which he agreed to, however he doesn't monitor the >group, so if anyone has feedback, can you also please CC it onto Carl. > >Cheers Don... > > > > Don McKenzie wrote: > > > > >Hi Carl, thanks for the feedback. > > >I am taking the liberty of sending this message to my dealer list for > > >possible comment. > > >Would you like me to circulate it to the SimmStick group also? > > > > > >You are one of the few customers that have taken the time to give good > > >feedback on this subject. > > >I'll read your comments over a couple of times to digest it all myself. > > >:-) > > > > > >Cheers Don... > > > > > > > > > > > >>Hello Don, > > >> > > >>I'm sorry I could not write back to you sooner. I've been swamped at > > >>work with major overtime for the last several months. I hardly have > > >>time to sleep anymore, let alone keep up with my hobbies! I wanted to > > >>thank you so much for your quick shipment of my order. I was amazed > > >>that only took 6 days to get here in Texas! It takes longer than that > > >>to get mail from my parents in the northern US! > > >> > > >>I am working on scale R/C model warships, and am trying to build my own > > >>control systems. I have already decided that your SimmStick(TM) module > > >>approach was the best way to go, as I could plug in only the modules I > > >>need for a given ship. I plan to make modules for controlling servos, > > >>DC motors, stepper motors as well as adding ultrasonic ranging, digital > > >>compass, and even pressure sensors and inclinometers for the model sub I > > >>want to build someday. It will probably take me a couple of years to > > >>get all this built and working, since I have so little spare time. To > > >>make things easier, I am focusing only on two CPUs, the ATmega128 and > > >>the ATmega8, which have the horsepower and features to meet my needs. > > >> However, the 64pin ATmega128 doesn't fit well with the existing > > >>SimmStick(TM) module limitation of 30pins. I'm therefore very > > >>interested in seeing an new, expanded DTMB bus. > > >> > > >>I read your thoughts on a new DTMB bus that you posted on your website. > > >> If I may say so, I believe you addressed the concerns admirably, > > >>especially the issue of the PCB thickness. As I plan to prototype my > > >>own PCBs before paying to have nice ones made, the 0.05" thickness of > > >>the SIMM board is a real problem for me. Don't get me wrong, the DT208 > > >>is a great help with this (I've already ordered two of them), but it is > > >>not an ideal long-term solution. > > >>I know people have given you feedback saying not to change, and I know > > >>my feedback is late in coming. For that I apologize. But I really > > >>would like to see an expanded DTMB-2 bus solution. To that end, I have > > >>come up with my own proposal which I'd like to ask you to read and > > >>critique. If you decide its not worth it for Dontronics to pursue it, > > >>that's OK. I'll still pursue some version of it for my own needs. I > > >>just thought it would be nice if I could share my ATmega128 board > > >>designs with others using SimmStick(TM) modules. The proposal is > > >>attached to this email. I welcome any and all feedback, even if you > > >>flat out think the proposal is stupid and never want to pursue it > > >>yourself. > > >> > > >>Many thanks, > > >> > > >>Carl > > >> > > >>dietzer@a... > > >>cdietz@f... > > >> > > >> > ------------------------------------------------------------------------ > > >>A Proposal for a New DTMB-2 Bus (DonTronics Micro Bus) > > >> > > >>The DTMB-2 bus was born out of desire to expand the original 30-pin > > >>SimmStick(TM) bus, also known as the DTMB-1 bus, commonly used for PIC > > >>and Atmel microcontrollers with 40 pins or less. There are several > > >>reasons for updating/replacing the SimmStick(TM) bus: > > >> > > >> 1. The 30-pin SimmStick(TM) is too small for microcontrollers with > > >> 40 pins or more, especially the new 64-pin devices like the > > >> Atmel ATmega128. > > >> > > >> 2. Each year it is harder to find sources for 30-pin SIMM connectors. > > >> > > >> 3. The 30-pin SIMM sockets require a 0.05" thick PC board, which are > > >> also hard to find sources for. > > >> > > >>I believe that the new DTMB-2 (DonTronics Micro Bus) should have the > > >>following features: > > >> > > >> 1. Like the SimmStick(TM) bus, the new bus should support both a > > >> card-edge connection and 0.100" header connection. The card-edge > > >> connection must work with 0.062" (1/16") PC boards, > eliminating the > > >> problems caused by the 0.05" thickness of SIMM boards. > > >> > > >> 2. The new bus must be fully backward compatible with the > SimmStick(TM) > > >> bus. It must not only support the same signals on the bus, but be > > >> able to physically interface with existing SimmStick(TM) > modules and > > >> motherboards. The purpose of the new DTMB-2 bus is to expand and > > >> enhance the SimmStick(TM) bus with a standard that is easier to > > >> support over time, not to eliminate the SimmStick(TM) bus > outright. > > >> > > >> 3. It should support an optional external memory interface, > compatible > > >> both the ATmega series and the PIC17CXX series of parts. Such an > > >> interface would require 19 pins, most of which could be used as > > >> general purpose I/O if the user does not need an external memory > > >> interface in his design. > > >> > > >> 4. It should add a new power pin called VCC, which is dedicated > to 3.3V. > > >> This will allow newer designs to take advantage of the lower > voltage. > > >> > > >> 5. Should be expandable in the future without sacrificing > SimmStick(TM) > > >> compatibility, or compatibility with itself. > > >> > > >>I believe the following proposal for the new DTMB-2 bus satisfies all the > > >>above requirements. I believe it maintains not only full compatibility > > >>with the SimmStick(TM) bus, but allows boards designed for the new bus to > > >>mate with existing SimmStick boards. > > >> > > >>I propose the following definition for the DTMB-2 bus: > > >> > > >> 1. The DTMB-2 bus would use the common 60-pin PCI card-edge > connector, > > >> for the following reasons: > > >> > > >> a) The PCI bus, commonly used in PCs, will be around for quite > some > > >> time. The older PC and PC-AT buses can still be found on > many PCI > > >> motherboards, nearly 15 years after the introduction of the > PC bus. > > >> I believe that the PCI connector will have similar longevity. > > >> > > >> b) The PCI card-edge connector supports 0.062" (1/16") PC boards. > > >> These board thicknesses are available from all PCB suppliers. > > >> This board thickness also allows the use of common 1/16" board > > >> guides to provide vertical support for taller DTMB-2 modules. > > >> > > >> c) The 60-pin connector provides an additional 30 pins to the > > >> existing SimmStick(TM) bus. This gives better support for CPUs > > >> with 40 and even 64 pins. > > >> > > >> d) There also exists a 92-pin PCI connector, which is an extension > > >> of the 60-pin connector. A board designed for the 60-pin PCI > > >> connector will plug into the 92-pin connector. The 92-pin > > >> connector can be used for expanding the 60-pin DTMB-2 bus > design > > >> in the future to a 92-pin design while maintaining full > backward > > >> compatibility with both 60-pin DTMB-2 and 30-pin SimmStick(TM) > > >> buses. > > >> > > >> e) The 60-pin PCI connector is only 3.34" long, which is actually > > >> shorter than the 3.8" 30-pin SIMM connector. Thus the 60-pin > > >> PCI connector allows a DTMB-2 module to be 3.5" long (or less), > > >> basically matching the footprint of a SimmStick(TM) module. > > >> > > >> 2. The DTMB-2 bus supports the use of dual-row headers, 2x30 (or even > > >> 2x32 if that is more available). Careful assigning of the bus > > >> pin-out will assign the original SimmStick(TM) bus to one 30-pin > > >> row of the new 2x30 header, allowing the user to plug > SimmStick(TM) > > >> boards into the new DTMB-2 module. This would even allow the use > > >> of a DT208 fat to skinny converter board to allow plugging a > DTMB-2 > > >> module into a 30-pin SIMM socket ! > > >> > > >> As an option, the DTMB-2 module may add a single row 30-pin > > >> header on the top edge of the board, duplicating the new 30 pins > > >> of the DTMB-2 bus and use either a single row or dual row header > > >> on the bottom edge of the board for the SimmStick(TM) signals. > > >> This would allow stacking of DTMB-2 modules for users who prefer > > >> stackable modules to the motherboard/daughterboard approach. > > >> > > >> 3. The DTMB-2 bus adds support for an external memory interface, > > >> consisting of 19 pins, which can be used by both the 8-bit > external > > >> memory interface of the Atmel AVR series and the 16-bit external > > >> memory interface of the PIC17CXX series of parts. This requires > > >> three control pins, called *RD/*OE, *WR/A0, and ALE/A7 (where > the * > > >> denotes an active-low signal) and 16 address/data pins, AD0-AD7 > > >> and AD8/A8 - AD15/A15. Modules that use the external memory > > >> interface will be called DTMB-2EM modules, where the EM stands > > >> for External Memory. > > >> > > >> Although the DTMB-2 bus supports both the 8-bit Atmel AVR and > > >> the 16-bit PIC 17CXX external memory interfaces, you can NOT > > >> use both 8-bit and 16-bit DTMB-2EM modules simultaneously. > > >> This is due to the 8-bit Atmel external memory interface using > > >> A8-A15 as an output-only bus, which would cause drive conflicts > > >> with AD8-AD15 on the 16-bit PIC17CXX series external memory > > >> interface. > > >> > > >> If the user does not wish to use DTMB-2EM modules in his design, > > >> then pins *WR/A0, ALE/A7, AD0-AD7 and AD8/A8 - AD15/A15 can be > > >> used for general purpose I/O. The pin *RD/*OE, however, will be > > >> reservered, with the requirement that it be tied high through a > > >> pullup resistor and be jumpered so that it can be disconnected > > >> from the CPU. Then if a DTMB-2EM module is plugged into a DTMB-2 > > >> bus with modules that use the same pins as general-purpose I/Os, > > >> the DTMB-2EM module can not drive out on pins AD0 - AD7 and > > >> AD8/A8 - AD15/A15 and thus prevents drive conflicts from occuring. > > >> > > >> 4. The DTMB-2 bus adds a new 3.3V VCC pin. > > >> > > >> 5. The DTMB-2 bus adds two new serial I/O pins called SI2 and SO2 for > > >> CPUs with two serial ports. > > >> > > >> 6. The DTMB-2 bus adds 8 new general purpose I/Os, called B0-B7. > > >> > > >> 7. The DTMB-2 bus interleaves the original 30 SimmStick(TM) bus > > >> signals with the other 30 DTMB-2 signals. This supports using > > >> one row of the dual row header as a regular SimmStick(TM) header. > > >> It also makes it easier to support both PCI and SIMM connectors > > >> on the same motherboard. > > >> > > >> 8. DTMB-2 PINOUT > > >> > > >>SimmStick(TM) DTMB-2(TM) > > >>------------------------------- > > >> 1 A1 1 A1 > > >> 2 *RD/*OE > > >> 2 A2 3 A2 > > >> 4 *WR/A0 > > >> 3 A3 5 A3 > > >> 6 ALE/A7 > > >> 4 PWR 7 PWR > > >> 8 VCC (3.3V) > > >> 5 CI/A4 9 CI/A4 > > >> 10 SI2 > > >> 6 CO/A5 11 CO/A5 > > >> 12 SO2 > > >> 7 VDD 13 VDD (5V) > > >> 14 B0 > > >> 8 *RES 15 *RES > > >> 16 B1 > > >> 9 GND 17 GND > > >> 18 B2 > > >>10 SCL 19 SCL > > >> 20 B3 > > >>11 SDA 21 SDA > > >> 22 B4 > > >>12 SI 23 SI > > >> 24 B5 > > >>13 SO 25 SO > > >> 26 B6 > > >>14 IO/A6 27 IO/A6 > > >> 28 B7 > > >>15 D0 29 D0 > > >> 30 AD0 > > >>16 D1 31 D1 > > >> 32 AD1 > > >>17 D2 33 D2 > > >> 34 AD2 > > >>18 D3 35 D3 > > >> 36 AD3 > > >>19 D4 37 D4 > > >> 38 AD4 > > >>20 D5/MOSI 39 D5/MOSI > > >> 40 AD5 > > >>21 D6/MISO 41 D6/MISO > > >> 42 AD6 > > >>22 D7/SCK 43 D7/SCK > > >> 44 AD7 > > >>23 D8 45 D8 > > >> 46 AD8/A8 > > >>24 D9 47 D9 > > >> 48 AD9/A9 > > >>25 D10 49 D10 > > >> 50 AD10/A10 > > >>26 D11 51 D11 > > >> 52 AD11/A11 > > >>27 D12 53 D12 > > >> 54 AD12/A12 > > >>28 D13 55 D13 > > >> 56 AD13/A13 > > >>29 D14 57 D14 > > >> 58 AD14/A14 > > >>30 D15 59 D15 > > >> 60 AD15/A15 > > >> > > >> 9. DTMB-2 Connector Sources > > >> > > >> One source for the 60-pin PCI connector is Digi-Key, > > >> part# 145154-4-ND. The connector is made by Amp/Tyco > > >> Electronics, part # 145154-4. > > >> > > >> One source for the 92-pin PCI connector is Digi-Key, > > >> part# 145165-4-ND. The connector is made by Amp/Tyco > > >> Electronics, part # 145165-4. > > >> > > >> > > > > > > > > > > >-- >Don McKenzie E-mail: http://www.dontronics.com/e-mail.html > Home Page: http://www.dontronics.com > >Add USB to your Product in 10 min. http://www.dontronics.com/giga.html >The World's Largest Range of Atmel/AVR & PICmicro Hardware and Software > > Please place (or leave -->) "dt2002" in the subject or body of your > message to guarantee your message getting through our spam filters. > > >To Post a message, send it to: simmstick@e... > >To Unsubscribe, send a blank message to: simmstick-unsubscribe@e... > >Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
From: alxx@tig.com.au <alxx@ihug.com.au> Date: Mon Oct 7, 2002 0:44pm Subject: RE: Re: [dtdealer] A Proposal for a New DTMB-2 Bus (DonTronics MicroBus)
-----Original Message----- From: Fred Ledo [mailto:fredledo@a...] Sent: Monday, 7 October 2002 9:31 AM To: simmstick@yahoogroups.com Subject: Re: [simmstick] Re: [dtdealer] A Proposal for a New DTMB-2 Bus (DonTronics MicroBus) Sounds real good !!! fred At 06:26 AM 10/7/02 +1000, you wrote: >CC: >Dealers >SimmStick Group >Carl Dietz <dietzer@a...> > >Mick Gulovsen wrote: > > > > Gday All, > > > > I like it... I assume that the PCI socket is .05" pin spacing... > > > > Mick > >I guess it is Mick, because of the length of the board. >I suggested to Carl, that we post the message to the SimmStick user >group for feedback, which he agreed to, however he doesn't monitor the >group, so if anyone has feedback, can you also please CC it onto Carl. > >Cheers Don... > > > > Don McKenzie wrote: > > > > >Hi Carl, thanks for the feedback. > > >I am taking the liberty of sending this message to my dealer list for > > >possible comment. > > >Would you like me to circulate it to the SimmStick group also? > > > > > >You are one of the few customers that have taken the time to give good > > >feedback on this subject. > > >I'll read your comments over a couple of times to digest it all myself. > > >:-) > > > > > >Cheers Don... > > > > > > > > > > > >>Hello Don, > > >> > > >>I'm sorry I could not write back to you sooner. I've been swamped at > > >>work with major overtime for the last several months. I hardly have > > >>time to sleep anymore, let alone keep up with my hobbies! I wanted to > > >>thank you so much for your quick shipment of my order. I was amazed > > >>that only took 6 days to get here in Texas! It takes longer than that > > >>to get mail from my parents in the northern US! > > >> > > >>I am working on scale R/C model warships, and am trying to build my own > > >>control systems. I have already decided that your SimmStick(TM) module > > >>approach was the best way to go, as I could plug in only the modules I > > >>need for a given ship. I plan to make modules for controlling servos, > > >>DC motors, stepper motors as well as adding ultrasonic ranging, digital > > >>compass, and even pressure sensors and inclinometers for the model sub I > > >>want to build someday. It will probably take me a couple of years to > > >>get all this built and working, since I have so little spare time. To > > >>make things easier, I am focusing only on two CPUs, the ATmega128 and > > >>the ATmega8, which have the horsepower and features to meet my needs. > > >> However, the 64pin ATmega128 doesn't fit well with the existing > > >>SimmStick(TM) module limitation of 30pins. I'm therefore very > > >>interested in seeing an new, expanded DTMB bus. > > >> > > >>I read your thoughts on a new DTMB bus that you posted on your website. > > >> If I may say so, I believe you addressed the concerns admirably, > > >>especially the issue of the PCB thickness. As I plan to prototype my > > >>own PCBs before paying to have nice ones made, the 0.05" thickness of > > >>the SIMM board is a real problem for me. Don't get me wrong, the DT208 > > >>is a great help with this (I've already ordered two of them), but it is > > >>not an ideal long-term solution. > > >>I know people have given you feedback saying not to change, and I know > > >>my feedback is late in coming. For that I apologize. But I really > > >>would like to see an expanded DTMB-2 bus solution. To that end, I have > > >>come up with my own proposal which I'd like to ask you to read and > > >>critique. If you decide its not worth it for Dontronics to pursue it, > > >>that's OK. I'll still pursue some version of it for my own needs. I > > >>just thought it would be nice if I could share my ATmega128 board > > >>designs with others using SimmStick(TM) modules. The proposal is > > >>attached to this email. I welcome any and all feedback, even if you > > >>flat out think the proposal is stupid and never want to pursue it > > >>yourself. > > >> > > >>Many thanks, > > >> > > >>Carl > > >> > > >>dietzer@a... > > >>cdietz@f... > > >> > > >> > ------------------------------------------------------------------------ > > >>A Proposal for a New DTMB-2 Bus (DonTronics Micro Bus) > > >> > > >>The DTMB-2 bus was born out of desire to expand the original 30-pin > > >>SimmStick(TM) bus, also known as the DTMB-1 bus, commonly used for PIC > > >>and Atmel microcontrollers with 40 pins or less. There are several > > >>reasons for updating/replacing the SimmStick(TM) bus: > > >> > > >> 1. The 30-pin SimmStick(TM) is too small for microcontrollers with > > >> 40 pins or more, especially the new 64-pin devices like the > > >> Atmel ATmega128. > > >> > > >> 2. Each year it is harder to find sources for 30-pin SIMM connectors. > > >> > > >> 3. The 30-pin SIMM sockets require a 0.05" thick PC board, which are > > >> also hard to find sources for. > > >> > > >>I believe that the new DTMB-2 (DonTronics Micro Bus) should have the > > >>following features: > > >> > > >> 1. Like the SimmStick(TM) bus, the new bus should support both a > > >> card-edge connection and 0.100" header connection. The card-edge > > >> connection must work with 0.062" (1/16") PC boards, > eliminating the <snipped a bit> Thats a great idea. Need to have vcc5 (5 volts) and vcc3.3(3.3 volts) and maybe also vcc1.8 ? also additional grounds minimum of 2 , maybe have more ? maybe pwr gnd signal gnd pci should be around for a while in one form or another. What connector is Intel going to use for pci x or whatever they are calling the next iteration? (PCI-X is 64 bit, so 90 or 120 pin connector ?) Another option is ide connectors. What about the pc/104 connectors ? An existing standard. Going to be around for a while. Designed for industrial uses. pc/104 uses 40 or 60 pin connectors http://www.pc104.org/technology/reg_info.html pc/104 plus uses 120 pin connectors http://www.pc104.org/technology/plus_info.html scroll down for comiparison of pci form factors. Another thing to look at is having builtin or pins/connections reserved for usb making it easier for when parallel and serial ports finally get removed from pc's. Starting to get increasing numbers of new pc's and laptops with no parallel or serial ports having usb or easily added option would eliminate the need for multiple adaptors. Few links of interest maybe http://www.meritec.com/Pages/products/pci/ (scrolldown for specs) http://www.pcisig.com/home http://www.pcisig.com/specifications http://sunsolve.sun.com/handbook_pub/General/PCI_Info.html http://www.nullmodem.com/PCI.htm http://www.nullmodem.com/USB.htm a few nice usb links http://www.howstuffworks.com/pci2.htm http://www.douglas.com/hardware/pcbs/pc104/ (those flash disc adaptors look good) http://www.pc104.org/technology/pc104_tech.html Alex
From: Don McKenzie <support2003@dontronics.com> Date: Mon Oct 7, 2002 4:46pm Subject: Re: A Proposal for a New DTMB-2 Bus (DonTronics MicroBus)
Carl Dietz wrote: > > Don McKenzie wrote: > > > CC: > > Dealers > > SimmStick Group > > Carl Dietz <dietzer@a...> > > > > Mick Gulovsen wrote: > > > > > >> Gday All, > >> > >> I like it... I assume that the PCI socket is .05" pin spacing... > >> > >> Mick > >> > >> > > I guess it is Mick, because of the length of the board. > > I suggested to Carl, that we post the message to the SimmStick user > > group for feedback, which he agreed to, however he doesn't monitor > > the > > group, so if anyone has feedback, can you also please CC it onto > > Carl. > > > > Cheers Don... > > > > > Don, > > Yes, it is 0.05" spacing. > > I'm attaching two PDF's I downloaded from Digi-Key. One is the AMP > drawing of the 60pin connector (shows both the connector and the board > footprint), the other is their general specification for the entire > line of 0.05" spacing connectors. It has more data than the drawing > provides. > > The only drawback I see to the PCI connector is that the board edges > must be beveled at least 20 degrees, as the drawing shows. However, > beveling more than 20 degress would hurt nothing, as long as you > didn't file away the copper pads. This can be done with a file, a > grinder, a dremel with a grinding head, etc. Anybody can do it with > little effort. > > FYI, if I remember right Digi-Key sells the 60pin PCI connector I > mentioned for US$2.34 in quantities of one. > > Carl > > P.S. I only replied to you on this. If you want me to just do a > 'Reply All' in the future, just let me know. I just didn't feel right > sending email to your dealer list without your prior permission. > > Name: 208856.pdf > 208856.pdf Type: Acrobat (application/pdf) > Encoding: base64 > > Name: 158778.pdf > 158778.pdf Type: Acrobat (application/pdf) > Encoding: base64 Hi Carl, I have added the info and PDF's you have given me to http://www.dontronics.com/dtmb.html you won't be able to reply to the dealer list without being a member, but I'll CC this to the SimmStick group. I think you can post without being a member to the SimmStick group. I know I went through various settings when I was getting spam and viruses on the SimmStick group. Don McKenzie E-mail: http://www.dontronics.com/e-mail.html Home Page: http://www.dontronics.com Add USB to your Product in 10 min. http://www.dontronics.com/giga.html The World's Largest Range of Atmel/AVR & PICmicro Hardware and Software Please place (or leave -->) "dt2002" in the subject or body of your message to guarantee your message getting through our spam filters.
From: Fred Ledo <fredledo@atomic.net> Date: Tue Oct 8, 2002 1:33am Subject: Re: Re: [dtdealer] A Proposal for a New DTMB-2 Bus (DonTronics MicroBus)
60 pin /2 sides = 30 pin per side 30 x .1 = 3 inches ? At 06:26 AM 10/7/02 +1000, you wrote: >CC: >Dealers >SimmStick Group >Carl Dietz <dietzer@a...> > >Mick Gulovsen wrote: > > > > Gday All, > > > > I like it... I assume that the PCI socket is .05" pin spacing... > > > > Mick > >I guess it is Mick, because of the length of the board. >I suggested to Carl, that we post the message to the SimmStick user >group for feedback, which he agreed to, however he doesn't monitor the >group, so if anyone has feedback, can you also please CC it onto Carl. > >Cheers Don... > > > > Don McKenzie wrote: > > > > >Hi Carl, thanks for the feedback. > > >I am taking the liberty of sending this message to my dealer list for > > >possible comment. > > >Would you like me to circulate it to the SimmStick group also? > > > > > >You are one of the few customers that have taken the time to give good > > >feedback on this subject. > > >I'll read your comments over a couple of times to digest it all myself. > > >:-) > > > > > >Cheers Don... > > > > > > > > > > > >>Hello Don, > > >> > > >>I'm sorry I could not write back to you sooner. I've been swamped at > > >>work with major overtime for the last several months. I hardly have > > >>time to sleep anymore, let alone keep up with my hobbies! I wanted to > > >>thank you so much for your quick shipment of my order. I was amazed > > >>that only took 6 days to get here in Texas! It takes longer than that > > >>to get mail from my parents in the northern US! > > >> > > >>I am working on scale R/C model warships, and am trying to build my own > > >>control systems. I have already decided that your SimmStick(TM) module > > >>approach was the best way to go, as I could plug in only the modules I > > >>need for a given ship. I plan to make modules for controlling servos, > > >>DC motors, stepper motors as well as adding ultrasonic ranging, digital > > >>compass, and even pressure sensors and inclinometers for the model sub I > > >>want to build someday. It will probably take me a couple of years to > > >>get all this built and working, since I have so little spare time. To > > >>make things easier, I am focusing only on two CPUs, the ATmega128 and > > >>the ATmega8, which have the horsepower and features to meet my needs. > > >> However, the 64pin ATmega128 doesn't fit well with the existing > > >>SimmStick(TM) module limitation of 30pins. I'm therefore very > > >>interested in seeing an new, expanded DTMB bus. > > >> > > >>I read your thoughts on a new DTMB bus that you posted on your website. > > >> If I may say so, I believe you addressed the concerns admirably, > > >>especially the issue of the PCB thickness. As I plan to prototype my > > >>own PCBs before paying to have nice ones made, the 0.05" thickness of > > >>the SIMM board is a real problem for me. Don't get me wrong, the DT208 > > >>is a great help with this (I've already ordered two of them), but it is > > >>not an ideal long-term solution. > > >>I know people have given you feedback saying not to change, and I know > > >>my feedback is late in coming. For that I apologize. But I really > > >>would like to see an expanded DTMB-2 bus solution. To that end, I have > > >>come up with my own proposal which I'd like to ask you to read and > > >>critique. If you decide its not worth it for Dontronics to pursue it, > > >>that's OK. I'll still pursue some version of it for my own needs. I > > >>just thought it would be nice if I could share my ATmega128 board > > >>designs with others using SimmStick(TM) modules. The proposal is > > >>attached to this email. I welcome any and all feedback, even if you > > >>flat out think the proposal is stupid and never want to pursue it > > >>yourself. > > >> > > >>Many thanks, > > >> > > >>Carl > > >> > > >>dietzer@a... > > >>cdietz@f... > > >> > > >> > ------------------------------------------------------------------------ > > >>A Proposal for a New DTMB-2 Bus (DonTronics Micro Bus) > > >> > > >>The DTMB-2 bus was born out of desire to expand the original 30-pin > > >>SimmStick(TM) bus, also known as the DTMB-1 bus, commonly used for PIC > > >>and Atmel microcontrollers with 40 pins or less. There are several > > >>reasons for updating/replacing the SimmStick(TM) bus: > > >> > > >> 1. The 30-pin SimmStick(TM) is too small for microcontrollers with > > >> 40 pins or more, especially the new 64-pin devices like the > > >> Atmel ATmega128. > > >> > > >> 2. Each year it is harder to find sources for 30-pin SIMM connectors. > > >> > > >> 3. The 30-pin SIMM sockets require a 0.05" thick PC board, which are > > >> also hard to find sources for. > > >> > > >>I believe that the new DTMB-2 (DonTronics Micro Bus) should have the > > >>following features: > > >> > > >> 1. Like the SimmStick(TM) bus, the new bus should support both a > > >> card-edge connection and 0.100" header connection. The card-edge > > >> connection must work with 0.062" (1/16") PC boards, > eliminating the > > >> problems caused by the 0.05" thickness of SIMM boards. > > >> > > >> 2. The new bus must be fully backward compatible with the > SimmStick(TM) > > >> bus. It must not only support the same signals on the bus, but be > > >> able to physically interface with existing SimmStick(TM) > modules and > > >> motherboards. The purpose of the new DTMB-2 bus is to expand and > > >> enhance the SimmStick(TM) bus with a standard that is easier to > > >> support over time, not to eliminate the SimmStick(TM) bus > outright. > > >> > > >> 3. It should support an optional external memory interface, > compatible > > >> both the ATmega series and the PIC17CXX series of parts. Such an > > >> interface would require 19 pins, most of which could be used as > > >> general purpose I/O if the user does not need an external memory > > >> interface in his design. > > >> > > >> 4. It should add a new power pin called VCC, which is dedicated > to 3.3V. > > >> This will allow newer designs to take advantage of the lower > voltage. > > >> > > >> 5. Should be expandable in the future without sacrificing > SimmStick(TM) > > >> compatibility, or compatibility with itself. > > >> > > >>I believe the following proposal for the new DTMB-2 bus satisfies all the > > >>above requirements. I believe it maintains not only full compatibility > > >>with the SimmStick(TM) bus, but allows boards designed for the new bus to > > >>mate with existing SimmStick boards. > > >> > > >>I propose the following definition for the DTMB-2 bus: > > >> > > >> 1. The DTMB-2 bus would use the common 60-pin PCI card-edge > connector, > > >> for the following reasons: > > >> > > >> a) The PCI bus, commonly used in PCs, will be around for quite > some > > >> time. The older PC and PC-AT buses can still be found on > many PCI > > >> motherboards, nearly 15 years after the introduction of the > PC bus. > > >> I believe that the PCI connector will have similar longevity. > > >> > > >> b) The PCI card-edge connector supports 0.062" (1/16") PC boards. > > >> These board thicknesses are available from all PCB suppliers. > > >> This board thickness also allows the use of common 1/16" board > > >> guides to provide vertical support for taller DTMB-2 modules. > > >> > > >> c) The 60-pin connector provides an additional 30 pins to the > > >> existing SimmStick(TM) bus. This gives better support for CPUs > > >> with 40 and even 64 pins. > > >> > > >> d) There also exists a 92-pin PCI connector, which is an extension > > >> of the 60-pin connector. A board designed for the 60-pin PCI > > >> connector will plug into the 92-pin connector. The 92-pin > > >> connector can be used for expanding the 60-pin DTMB-2 bus > design > > >> in the future to a 92-pin design while maintaining full > backward > > >> compatibility with both 60-pin DTMB-2 and 30-pin SimmStick(TM) > > >> buses. > > >> > > >> e) The 60-pin PCI connector is only 3.34" long, which is actually > > >> shorter than the 3.8" 30-pin SIMM connector. Thus the 60-pin > > >> PCI connector allows a DTMB-2 module to be 3.5" long (or less), > > >> basically matching the footprint of a SimmStick(TM) module. > > >> > > >> 2. The DTMB-2 bus supports the use of dual-row headers, 2x30 (or even > > >> 2x32 if that is more available). Careful assigning of the bus > > >> pin-out will assign the original SimmStick(TM) bus to one 30-pin > > >> row of the new 2x30 header, allowing the user to plug > SimmStick(TM) > > >> boards into the new DTMB-2 module. This would even allow the use > > >> of a DT208 fat to skinny converter board to allow plugging a > DTMB-2 > > >> module into a 30-pin SIMM socket ! > > >> > > >> As an option, the DTMB-2 module may add a single row 30-pin > > >> header on the top edge of the board, duplicating the new 30 pins > > >> of the DTMB-2 bus and use either a single row or dual row header > > >> on the bottom edge of the board for the SimmStick(TM) signals. > > >> This would allow stacking of DTMB-2 modules for users who prefer > > >> stackable modules to the motherboard/daughterboard approach. > > >> > > >> 3. The DTMB-2 bus adds support for an external memory interface, > > >> consisting of 19 pins, which can be used by both the 8-bit > external > > >> memory interface of the Atmel AVR series and the 16-bit external > > >> memory interface of the PIC17CXX series of parts. This requires > > >> three control pins, called *RD/*OE, *WR/A0, and ALE/A7 (where > the * > > >> denotes an active-low signal) and 16 address/data pins, AD0-AD7 > > >> and AD8/A8 - AD15/A15. Modules that use the external memory > > >> interface will be called DTMB-2EM modules, where the EM stands > > >> for External Memory. > > >> > > >> Although the DTMB-2 bus supports both the 8-bit Atmel AVR and > > >> the 16-bit PIC 17CXX external memory interfaces, you can NOT > > >> use both 8-bit and 16-bit DTMB-2EM modules simultaneously. > > >> This is due to the 8-bit Atmel external memory interface using > > >> A8-A15 as an output-only bus, which would cause drive conflicts > > >> with AD8-AD15 on the 16-bit PIC17CXX series external memory > > >> interface. > > >> > > >> If the user does not wish to use DTMB-2EM modules in his design, > > >> then pins *WR/A0, ALE/A7, AD0-AD7 and AD8/A8 - AD15/A15 can be > > >> used for general purpose I/O. The pin *RD/*OE, however, will be > > >> reservered, with the requirement that it be tied high through a > > >> pullup resistor and be jumpered so that it can be disconnected > > >> from the CPU. Then if a DTMB-2EM module is plugged into a DTMB-2 > > >> bus with modules that use the same pins as general-purpose I/Os, > > >> the DTMB-2EM module can not drive out on pins AD0 - AD7 and > > >> AD8/A8 - AD15/A15 and thus prevents drive conflicts from occuring. > > >> > > >> 4. The DTMB-2 bus adds a new 3.3V VCC pin. > > >> > > >> 5. The DTMB-2 bus adds two new serial I/O pins called SI2 and SO2 for > > >> CPUs with two serial ports. > > >> > > >> 6. The DTMB-2 bus adds 8 new general purpose I/Os, called B0-B7. > > >> > > >> 7. The DTMB-2 bus interleaves the original 30 SimmStick(TM) bus > > >> signals with the other 30 DTMB-2 signals. This supports using > > >> one row of the dual row header as a regular SimmStick(TM) header. > > >> It also makes it easier to support both PCI and SIMM connectors > > >> on the same motherboard. > > >> > > >> 8. DTMB-2 PINOUT > > >> > > >>SimmStick(TM) DTMB-2(TM) > > >>------------------------------- > > >> 1 A1 1 A1 > > >> 2 *RD/*OE > > >> 2 A2 3 A2 > > >> 4 *WR/A0 > > >> 3 A3 5 A3 > > >> 6 ALE/A7 > > >> 4 PWR 7 PWR > > >> 8 VCC (3.3V) > > >> 5 CI/A4 9 CI/A4 > > >> 10 SI2 > > >> 6 CO/A5 11 CO/A5 > > >> 12 SO2 > > >> 7 VDD 13 VDD (5V) > > >> 14 B0 > > >> 8 *RES 15 *RES > > >> 16 B1 > > >> 9 GND 17 GND > > >> 18 B2 > > >>10 SCL 19 SCL > > >> 20 B3 > > >>11 SDA 21 SDA > > >> 22 B4 > > >>12 SI 23 SI > > >> 24 B5 > > >>13 SO 25 SO > > >> 26 B6 > > >>14 IO/A6 27 IO/A6 > > >> 28 B7 > > >>15 D0 29 D0 > > >> 30 AD0 > > >>16 D1 31 D1 > > >> 32 AD1 > > >>17 D2 33 D2 > > >> 34 AD2 > > >>18 D3 35 D3 > > >> 36 AD3 > > >>19 D4 37 D4 > > >> 38 AD4 > > >>20 D5/MOSI 39 D5/MOSI > > >> 40 AD5 > > >>21 D6/MISO 41 D6/MISO > > >> 42 AD6 > > >>22 D7/SCK 43 D7/SCK > > >> 44 AD7 > > >>23 D8 45 D8 > > >> 46 AD8/A8 > > >>24 D9 47 D9 > > >> 48 AD9/A9 > > >>25 D10 49 D10 > > >> 50 AD10/A10 > > >>26 D11 51 D11 > > >> 52 AD11/A11 > > >>27 D12 53 D12 > > >> 54 AD12/A12 > > >>28 D13 55 D13 > > >> 56 AD13/A13 > > >>29 D14 57 D14 > > >> 58 AD14/A14 > > >>30 D15 59 D15 > > >> 60 AD15/A15 > > >> > > >> 9. DTMB-2 Connector Sources > > >> > > >> One source for the 60-pin PCI connector is Digi-Key, > > >> part# 145154-4-ND. The connector is made by Amp/Tyco > > >> Electronics, part # 145154-4. > > >> > > >> One source for the 92-pin PCI connector is Digi-Key, > > >> part# 145165-4-ND. The connector is made by Amp/Tyco > > >> Electronics, part # 145165-4. > > >> > > >> > > > > > > > > > > >-- >Don McKenzie E-mail: http://www.dontronics.com/e-mail.html > Home Page: http://www.dontronics.com > >Add USB to your Product in 10 min. http://www.dontronics.com/giga.html >The World's Largest Range of Atmel/AVR & PICmicro Hardware and Software > > Please place (or leave -->) "dt2002" in the subject or body of your > message to guarantee your message getting through our spam filters. > > >To Post a message, send it to: simmstick@e... > >To Unsubscribe, send a blank message to: simmstick-unsubscribe@e... > >Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
From: dshesnicky <dshesnicky@yahoo.com> Date: Tue Oct 8, 2002 2:23am Subject: Re: dumb questions - overvoltage on the 7805
> No, although I would have got something with a bit more > amperage, this will suit the dt006 FYI... I found this reference for over voltage levels on the New Zealand Reflections site: http://www.reflections.co.nz/micro/Boards/DT001/DT001partslists.htm "Don't go too high with the input voltage as you may go over the 7805 specs and heat up the device to destruction. I would suggest that 18 Volts is as far as you should go." Don
From: dshesnicky <dshesnicky@yahoo.com> Date: Tue Oct 8, 2002 2:38am Subject: Re: dumb questions - gcc?
> I then use SP12 (http://home.wanadoo.nl/electro1/avr/sp12.htm) to > program the chips. Ben, Can the SP12 be used to program with any compiler's output ie. you just need the hex or bin file? Does gcc have an assembler as well? I think it has to if I remember correctly. I've downloaded it, sp12,and bascom-avr for tests tonight. Don
From: Don McKenzie <support2003@dontronics.com> Date: Tue Oct 8, 2002 8:51am Subject: [Fwd: Re: [Fwd: RE: Re: [dtdealer] A Proposal for a New DTMB-2Bus (DonTronics MicroBus)]]
forwarded to the SimmStick Group. Don McKenzie wrote: > <snipped a bit> > > Thats a great idea. > > Need to have > vcc5 (5 volts) and vcc3.3(3.3 volts) > and maybe also vcc1.8 ? > > also additional grounds > minimum of 2 , maybe have more ? > maybe > pwr gnd > signal gnd > I toyed with adding extra power/gnd to the pin-out when I put together the proposal. I chose instead to add the seperate SI2 and SO2 pins so as to map as many of the ATmega128's 64pins as possible. But these really don't have to be seperate pins. We could simply make the proposed B0 and B1 pins become B0/SI2 and B1/SO2 pins instead and use the proposed SI2 and SO2 pins for an added GND and VDD. > pci should be around for a while in one form or another. > What connector is Intel going to use for pci x > or whatever they are calling the next iteration? > (PCI-X is 64 bit, so 90 or 120 pin connector ?) > The 64-bit PCI uses the 92-pin connector mentioned in the proposal. Bear in mind that the 92-pin connector is 5.04" long, which significantly enlarges the footprint. I would recommend that the DTMB-2 bus use the 60-pin connector as it maintains the current footprint, and use the 92-pin connector for a future expansion of the DTMB-2 which simply adds 32 new pins to the DTMB-2 bus. Or maybe specify that the DTMB-2 could be either 60-pin or 92-pin and give both pinouts and footprints in the definition. This works fine as long as pins 1-60 are the same on both connectors. This allows a 60-pin module to plug into the 92-pin connector and still work properly. > Another option is ide connectors. > I had trouble finding IDE connectors larger than 50 pins, which is not enough for the 64-pin CPUs. The 64-pin Atmel ATmega128, with two VCC pins, three GND pins, and two XTAL pins, works OK with a 60-pin bus. > What about the pc/104 connectors ? > An existing standard. > Going to be around for a while. > Designed for industrial uses. > pc/104 uses 40 or 60 pin connectors > http://www.pc104.org/technology/reg_info.html > pc/104 plus uses 120 pin connectors > http://www.pc104.org/technology/plus_info.html > scroll down for comiparison of pci form factors. > This is a good idea for those who want stackable boards. PC/104 is stackable only, no card-edge definition. However, the 64-pin PC/104 connector with standoffs could easily be used for those who prefer a stackable solution rather than a card-edge solution. The proposal allows for both stacking and card-edge solutions. The 64-pin PC/104 connectors appear to have a footprint similar to the dual-row 64-pin 0.100" header like Don had mentioned on his web-site. Also, the stackable PC/104 standoffs eliminate the need for the optional 30-pin single row header at the top of the board mentioned in the proposal. I had added that simply to make a stackable solution more stable. > Another thing to look at > is having builtin or pins/connections reserved for usb > making it easier for when parallel and serial > ports finally get removed from pc's. > > Starting to get > increasing numbers of new pc's and > laptops with no parallel or serial ports > having usb or easily added option would eliminate > the need for multiple adaptors. > > Are you proposing adding the USB signals to the DTMB-2 so that you can have the interface on the module and the connector on the motherboard ? Or are you saying you want to add the USB signals to the DTMB-2 so each module can have its own USB interface and talk to other modules via USB ? It's easy to redefine B4-B7 so that they could also be used for USB, but I really don't encourage adding a USB specification to the DTMB-2 definition. The SimmStick(TM) bus definition is a simple one, and I think the DTMB-2 definition should be, too, and not deal with the USB specs or even what version of USB is used, simply keep it to which pins are OK to use to route USB signals to the motherboard connector. That way the user is free to choose a USB1.0 module or a USB 2.0 module, etc, without having it defined by the DTMB-2 definition. And we don't have to worry about redefining the DTMB-2 bus when the industry adds a new USB spec. > Few links of interest maybe > http://www.meritec.com/Pages/products/pci/ (scrolldown for specs) > http://www.pcisig.com/home > http://www.pcisig.com/specifications > http://sunsolve.sun.com/handbook_pub/General/PCI_Info.html > http://www.nullmodem.com/PCI.htm > http://www.nullmodem.com/USB.htm a few nice usb links > http://www.howstuffworks.com/pci2.htm > http://www.douglas.com/hardware/pcbs/pc104/ (those flash disc > adaptors > look > good) > http://www.pc104.org/technology/pc104_tech.html > > > Alex > Great links, thanks ! Carl
From: al <alxx@ihug.com.au> Date: Tue Oct 8, 2002 1:29pm Subject: RE: Re: [dtdealer] A Proposal for a New DTMB-2Bus (DonTronics MicroBus)
> -----Original Message----- > From: Don McKenzie [mailto:support2003@d...] > Sent: Tuesday, October 08, 2002 8:52 AM > To: SimmStick Group; dietzer@a... > Subject: [Fwd: Re: [Fwd: RE: [simmstick] Re: [dtdealer] A Proposal for a > New DTMB-2Bus (DonTronics MicroBus)]] > > Please consider what I posted previously as only a few ideas / suggestions for options not as proposals etc. I've only worked with the simm stick and DTMB-1 for a short while. And any way I would not presume to have enough electronics design experience to have any major input. Spent 10 years fixing aircraft (Avionics modifications and maintenance) currently at uni (struggling with Applied Physics and Electrical Engineering). (once I've finished writing this, back to the maths study) :-( > > > I toyed with adding extra power/gnd to the pin-out when I put together > the proposal. I chose instead to add the separate SI2 and SO2 pins so > as to map as many of the ATmega128's 64pins as possible. But these > really don't have to be seperate pins. We could simply make the > proposed B0 and B1 pins become B0/SI2 and B1/SO2 pins instead and use > the proposed SI2 and SO2 pins for an added GND and VDD. I'd argue for keeping the DMTB-2 as generic as possible so can be used with as many micros as possible. But the option to have connector compatibility (size and number of pins not actual bus) with a standard like pc104 plus (not straight pc104) would give a few advantages. > > pci should be around for a while in one form or another. > > What connector is Intel going to use for pci x > > or whatever they are calling the next iteration? > > (PCI-X is 64 bit, so 90 or 120 pin connector ?) > > > The 64-bit PCI uses the 92-pin connector mentioned in the proposal. > Bear in mind that the 92-pin connector is 5.04" long, which > significantly enlarges the footprint. > > I would recommend that the DTMB-2 bus use the 60-pin connector as it > maintains the current footprint, and use the 92-pin connector for a > future expansion of the DTMB-2 which simply adds 32 new pins to the > DTMB-2 bus. Or maybe specify that the DTMB-2 could be either 60-pin or > 92-pin and give both pinouts and footprints in the definition. This > works fine as long as pins 1-60 are the same on both connectors. This > allows a 60-pin module to plug into the 92-pin connector and still work > properly. One other thing is there another or different 30 (or 60) pin connectors around that would allow double or triple stacking of connectors to give the expansion required and maintain existing bus. I would like to see the existing 30 pin to be maintained for as long as possible. But could use now both the 60 and larger(92, 120 or 128) pin versions. And a stackable option would be very handy. But we don't want too many variants. and with the existing boards it is easy enough to make either edge or stackable versions by just changing the connectors. Need the same for whatever format connector is chosen. Another thing is parallel connections the way to go? Things to consider: pcb costs due to added complexity connector costs grounding of boards and ground planes / interference pwr and gnds over thinner traces - enough current ? > > Another option is ide connectors. > > > I had trouble finding IDE connectors larger than 50 pins, which is not > enough for the 64-pin CPUs. The 64-pin Atmel ATmega128, with two VCC > pins, three GND pins, and two XTAL pins, works OK with a 60-pin bus. And with ide moving towards serial ide with much smaller connectors availability would be a problem. Scsi connectors have enough pins but would be more expensive ?. Also rather bulky. Don't know if there are edge connectors ? scsi 1 50 pin Centronics 50 scsi 2 50 pin Micro-D 50 Scsi 3 68 pins newer sca connectors 80 pin with all scsi connectors and cables have option of sheilded or unshielded. > > What about the pc/104 connectors ? > > An existing standard. > > Going to be around for a while. > > Designed for industrial uses. > > pc/104 uses 40 or 60 pin connectors > > http://www.pc104.org/technology/reg_info.html > > pc/104 plus uses 120 pin connectors > > http://www.pc104.org/technology/plus_info.html > > scroll down for comiparison of pci form factors. > > > This is a good idea for those who want stackable boards. PC/104 is > stackable only, no card-edge definition. However, the 64-pin PC/104 > connector with standoffs could easily be used for those who prefer a > stackable solution rather than a card-edge solution. The proposal > allows for both stacking and card-edge solutions. The 64-pin PC/104 > connectors appear to have a footprint similar to the dual-row 64-pin > 0.100" header like Don had mentioned on his web-site. Also, the > stackable PC/104 standoffs eliminate the need for the optional 30-pin > single row header at the top of the board mentioned in the proposal. I > had added that simply to make a stackable solution more stable. > > > Another thing to look at > > is having builtin or pins/connections reserved for usb > > making it easier for when parallel and serial > > ports finally get removed from pc's. > > > > Starting to get > > increasing numbers of new pc's and > > laptops with no parallel or serial ports > > having usb or easily added option would eliminate > > the need for multiple adaptors. > > > Are you proposing adding the USB signals to the DTMB-2 so that you can > have the interface on the module and the connector on the motherboard ? > Or are you saying you want to add the USB signals to the DTMB-2 so each > module can have its own USB interface and talk to other modules via USB Having the interface on a module for those who want it rather than taking up space on the motherboard. But then again if the db9 or db25 connectors go there would be plenty of run for both interface and connector or 2. Or have a new usb only motherboard. Hey Don when is one going to be available ? I'm willing to have a go desgining one (or two, one for DTMB-1 and one for DTMB-2) once I get some free time (mid November). And share the design (FreeBSD license). > It's easy to redefine B4-B7 so that they could also be used for USB, but > I really don't encourage adding a USB specification to the DTMB-2 > definition. The SimmStick(TM) bus definition is a simple one, and I > think the DTMB-2 definition should be, too, and not deal with the USB > specs or even what version of USB is used, simply keep it to which pins > are OK to use to route USB signals to the motherboard connector. That > way the user is free to choose a USB1.0 module or a USB 2.0 module, etc, > without having it defined by the DTMB-2 definition. And we don't have > to worry about redefining the DTMB-2 bus when the industry adds a new > USB spec. No I wasn't thinking of having usb in DTMB - 2. If someone wanted usb interconnections it would be better to have a seperate connector like the usb connectors on computer motherboards. (8 or 10 pin ?) Alex
From: Ben Hitchcock <ben@wollongong.apana.org.au> Date: Wed Oct 9, 2002 0:51am Subject: Re: Re: dumb questions - gcc?
Hi, On Tuesday, October 8, 2002, at 02:38 AM, dshesnicky wrote: > Can the SP12 be used to program with any compiler's output ie. > you just need the hex or bin file? Yep. You only need the hex file for SP12, so if you used AVRBASIC then you could use SP12 as well. > Does gcc have an assembler as well? I think it has to if I > remember correctly. I've downloaded it, sp12,and bascom-avr > for tests tonight. Yep, gcc does the lot. Good luck, Ben
From: Antti Lukats <antti@case2000.com> Date: Wed Oct 9, 2002 4:24am Subject: DIY PCI Interface board for < $15 USD
Hi funny the PCI connector is/was recently discussed here :) I was just having some idea(s) of how to make cheap interfaces for todays computers where ISA bus is diminishing and well COM/LPT ports are not so funny either. Choices are 1) USB, here we can use FTDI chip thats real easy on the software side, but we need the hardware to build und to get the chip 2) PCI, PLX chips are pretty easy but again PCB needs to made.. so there is no easy way of wire wrapping or real low cost add-on board, or is it? well answer is closer as one might think - Network card! yes any RTL8139 (or compatible) based network card can be used as general purpose IO interface. The ROM socket pins can be accessed by software so we have one 8 Bit In/Out port and 19 Output pins. All the tools and information needed to build this and to write the software is available for the internet but it takes some time to find it all and make it happen. So hardware ingredients: 1) any RTL8139 based old network card with ROM socket 2) small PCB (sized as DIP28) 3) addon LPT port cable DB25 with bracket to 26 pin IDC header solder a pin header onto PCB connect dip pins to it, place it into ROM socket install the network card and the bracket with DB25 now you are ready to use it :) I am doing it under windows 2000 so special driver is needed to access the IO ports - driverlinx is free and can be used. -- I am planning to make some better project description and to release some software drivers to use this homebrew PCI interface if anyone is interested to participate please email me directly comments are also welcome :) Antti Lukats
From: Antti Lukats <antti@case2000.com> Date: Wed Oct 9, 2002 5:28am Subject: Re: DIY PCI Interface board for < $15 USD
= Owner
= Moderator
= Online