Just curious about this. Does anyone have a perspective on why the stock Arduino boards are 5v instead of 3.3? It seems like it would have simplified much with peripherals.
It's easy to go down, it's harder to go up 
A 5V board can drop its outputs to 3.3V if necessary with a simple voltage divider, and can generally recognize 3.3V inputs as "high".
A 3.3V board would need a 5V-tolerant I/O input (which the AVR is not, but newer microcontrollers are).
--
The Ruggeduino: compatible with Arduino UNO, 24V operation, all I/O's fused and protected
There are probably historical reasons as well. For example, if you're trying to program an AVR processor using a PC parallel port it is more convenient to run the processor at 5 V.
I would also like to mention the fact that there is so much cheap and surplus TTL glue out there, as well...
The first logic I studied was RTL (resistor/transistor/logic) which If I recall correctly ran at 3.6 VDC. So what what's new is old.
Seems that Don Lancaster has posted the book I started on my logic journey:
Lefty
Thanks for the fascinating insight, everyone! Very interesting.
retrolefty:
The first logic I studied was RTL (resistor/transistor/logic) which If I recall correctly ran at 3.6 VDC. So what what's new is old.Seems that Don Lancaster has posted the book I started on my logic journey:
http://www.tinaja.com/ebooks/RTLcb.pdf
Lefty
Hey - thanks for posting that; it'll go in my file of "old stuff I'll probably never use or read" - seriously, I have yet to find any RTL parts or anything (I would expect to see something at Apache Reclamation) - I think it would be neat to have a few for my budding "museum" my wife calls "junk". You rarely hear it mentioned much any more. I think I was browsing wikipedia when I first read about it (the way it seems to be kept hush on, its like the ugly step-brother to TTL who's locked in the attic or something).
I think I was browsing wikipedia when I first read about it (the way it seems to be kept hush on, its like the ugly step-brother to TTL who's locked in the attic or something).
Not ugly, just a short life run as the advancement from RTL to DTL (diode/transistor/logic) to TTL happened in over such a few short years. Once all the advantages of TTL were understood it became the dominate logic for a long time until various CMOS logic favors came along.
Lefty
retrolefty:
I think I was browsing wikipedia when I first read about it (the way it seems to be kept hush on, its like the ugly step-brother to TTL who's locked in the attic or something).
Not ugly, just a short life run as the advancement from RTL to DTL (diode/transistor/logic) to TTL happened in over such a few short years. Once all the advantages of TTL were understood it became the dominate logic for a long time until various CMOS logic favors came along.
Yeah - I suppose so; there was actually a lot of churn during those years (and immediately prior) from what I have gathered from various books and magazine articles; I'm currently reading old back-issues of Popular Science on Google Books (currently reading 1963) - just about every month there is mention of some "miniaturization" advancement (mostly to transistors, or weird devices like complex circuits packed into small lucite cubes for missiles - precursors to integrated circuits). I'm sure as I get closer to 1970, there will be more of these rapid "changes" in electronics. Based on books I have read from the era, there were all kinds of strange technologies and ideas tried (for instance, in the last issue I read, there was an idea on a home video taping technology that used a strange electron beam arrangement to scan each line, as a magnetic field on the tape - using a weird read/write head of titanium wires fed into a vaccuum tube where an electron beam scanned them to set up the voltages - this was supposed to be a better, but cheaper, alternative to a spinning read/write head, as was used in commercial machines of the era).
Cool. The "TVT Typerwriter Cookbook" is there too (http://www.tinaja.com/ebooks/TVTcb.pdf ), which is a good introduction to video technology (The actual circuitry might be a bit obsolete, but video principles remain the same.)
There is also a "Machine Language Programming" cookbook. I'm not specifically familiar with this one (I actually BUILT stuff from Lancaster's CMOS Cookbook and TVT Cookbook, back in the day), but at a quick glance it looks to have an intro to microprocessor architectures and assembly language that covers several different CPUs, and might therefore be much better philosophically than trying to figure things out from a single vendor's datasheets. (It talks about 8048, 8080, and 6502 architectures. All mostly long gone, but similar enough in principle to modern micros that I think this would be a pretty good thing to read if you're interested in how things work "under the covers." It looks like it defines terms like "address space", "registers", and etc. that programmer types are apt to throw around to the confusion of many.)
I think i sfill have some rtl somewhere. Thanks for dons book link.
maniacbug:
Just curious about this. Does anyone have a perspective on why the stock Arduino boards are 5v instead of 3.3? It seems like it would have simplified much with peripherals.
Back to the OP's question... I think it is important to note that if you are running an Atmega328 at 16MHz and 3.3V you are running it out of spec. I don't think the Arduino team would want to mass produce a board that was not within the specs of the processor. The only way around this would be to use some sort of level translator which is probably not worth the additional cost for most users.
Although running it at 3.3V is out of spec for 16MHz, I have never heard of anyone having a problem doing so. Ladyada has a great tutorial on how to convert your 5.0V Arduino to 3.3V. You just need to replace the 5.0V regulator with a 3.3V one.
I have also designed a board for this type of situation that keeps the Atmega328 running within spec. The Freeduino Hybrid has a built in 3.3V regulator and an 8-bit level translator to allow you to use up to 8 I/O's at 3.3V. There are also some other additions that a regular Arduino is lacking.
You other option is to just use something like a 74AHC125N tri-state buffer or an CD4050 level shifter. Both of those parts are through hole components and will work well on a breadboard.
Don't forget also that the original arduino used an ATmega8, which were not spec'ed for the wide voltage-range operation of the current picopower devices. You got to pick either the atmega8 that was good from 4.5 to 5.5V, or the more-expensive, harder-to-find, and slower ATmega8L...
Another reason for sticking with 5v is that if you need to drive logic level power MOSFETs such as the STP40NF10L, then 3.3v isn't really enough.
Because,
A 5V board can drop its outputs to 3.3V if necessary with a simple voltage divider, and can generally recognize 3.3V inputs as "high". but 3.3V cant.
if it drop out to below this level (3.3V) then Digital level becomes LOW !! i.e, 0.
another reason is that the USB-power-line carries 5V, so why not to use it ^^