Yahoo Groups archive
Thread
2006-04-25 by Steve Franks
So, Closer inspection reveals to me that the Minimum Voltage for the VccIO and VccA pins on the 2103 is 3.0V. How disappointing - a tiny chip and I have to run a split supply just to get it to boot? Anyone done the unthinkable and tried to run the whole works off of 1.8V? And here I wanted to replace all my 8-bit uC's with the 2103 ever after... Steve [Non-text portions of this message have been removed]
2006-04-25 by rtstofer
--- In lpc2000@yahoogroups.com, "Steve Franks" <stevefranks@...> wrote: > > So, > > Closer inspection reveals to me that the Minimum Voltage for the VccIO and > VccA pins on the 2103 is 3.0V. How disappointing - a tiny chip and I have > to run a split supply just to get it to boot? Anyone done the unthinkable > and tried to run the whole works off of 1.8V? And here I wanted to replace > all my 8-bit uC's with the 2103 ever after... > > Steve > > > [Non-text portions of this message have been removed] > The datasheet says no. However, the LPC213x series uses a single 3.3V supply. Unfortunately, it is a 64 pin device versus 48. Richard
2006-04-25 by Robert Wood
>> Closer inspection reveals to me that the Minimum Voltage for the VccIO and VccA pins on the 2103 is 3.0V. How disappointing - a tiny chip and I have to run a split supply just to get it to boot? Anyone done the unthinkable and tried to run the whole works off of 1.8V? And here I wanted to replace all my 8-bit uC's with the 2103 ever after... << The other thing that puts me off using the 210x rather than, say, AVRs is the fact that you only have one 32 bit port. I'd much rather use an ATMega that has multiple ports. Too much bit shifting going on with the ARM for me.
2006-04-25 by rtstofer
--- In lpc2000@yahoogroups.com, Robert Wood <robert.wood@...> wrote: > > >> Closer inspection reveals to me that the Minimum Voltage for the > VccIO and VccA pins on the 2103 is 3.0V. How disappointing - a tiny > chip and I have to run a split supply just to get it to boot? Anyone > done the unthinkable and tried to run the whole works off of 1.8V? And > here I wanted to replace all my 8-bit uC's with the 2103 ever after... << > > The other thing that puts me off using the 210x rather than, say, AVRs > is the fact that you only have one 32 bit port. I'd much rather use an > ATMega that has multiple ports. Too much bit shifting going on with the > ARM for me. > I certainly like the ATmega128 but it isn't even close in speed to the LPC2106. It doesn't have as much SRAM either. The ARM7TDMI-S has a 32 bit barrel shifter. I don't think shifting things around is much of an issue. One downside of the 32 bit instruction (ARM mode) is that it chews up the 128k bytes in a hurry. The other side is that many instructions are conditional so small branches aren't required. There's good and bad in every device. It's always a question when I start a new project; which device to use. I like the 5V inteface levels of the AVR, I like the speed of the LPC, I think I would really like the LPC2138 with the A/D converter and single supply, if I have the pins left over I like using Rowley CrossConnect to do JTAG on the LPC but I'm ok with ISP for either device and the LPC has more SRAM. Always choices... Richard
2006-04-25 by Xtian Xultz
Em Ter 25 Abr 2006 12:56, Steve Franks escreveu: > So, > > Closer inspection reveals to me that the Minimum Voltage for the VccIO and > VccA pins on the 2103 is 3.0V. How disappointing - a tiny chip and I have > to run a split supply just to get it to boot? Anyone done the unthinkable > and tried to run the whole works off of 1.8V? And here I wanted to replace > all my 8-bit uC's with the 2103 ever after... > > Steve Its too dangerous to generate 1.8V (more or less) from the 3.3V with two diodes in series? Someone tryed this?
2006-04-25 by Peter Jakacki
Yes, something similar in that I have powered MSP430s with a red led from +5V which gives me 3.2...3.4V. This is of course limited to the maximum forward current of the led. It is true that the forward voltage drop of diodes varies with current and temperature but this should be fine as long as it is adequately decoupled on the 1.8V rail with a small tantalum as well. The specs state that you can operate from 1.65V to 1.95V with an absolute max of 2.5V so worst case max voltage should be around 2.1V if the cpu is in reset etc but you can always put a resistive load across the rails to maintain a minimum current (increased diode forward voltage drop). Personally, if I really needed to save 50c and design a board to use diodes I would still design in an ldo regulator just in case the burn-in tests reveal some problems. But for a quick&dirty proto I wouldn't hesitate to use the diode approach. BTW, I have been using torex XC6201 regulators in the tiny sot-25 pack as they are inexpensive and readily available (Farnell sell them). How has everyone else fared with regulators? *Peter* Xtian Xultz wrote:
> Em Ter 25 Abr 2006 12:56, Steve Franks escreveu: >> So, >> >> Closer inspection reveals to me that the Minimum Voltage for the VccIO and >> VccA pins on the 2103 is 3.0V. How disappointing - a tiny chip and I have >> to run a split supply just to get it to boot? Anyone done the unthinkable >> and tried to run the whole works off of 1.8V? And here I wanted to replace >> all my 8-bit uC's with the 2103 ever after... >> >> Steve > Its too dangerous to generate 1.8V (more or less) from the 3.3V with two > diodes in series? > Someone tryed this?
2006-04-25 by Robert Adsett
At 08:04 PM 4/25/2006 -0300, Xtian Xultz wrote: > Its too dangerous to generate 1.8V (more or less) from the 3.3V with two >diodes in series? >Someone tryed this? Just put in an LDO. You can get them as small as a sot23 giving you about 90mA with a 30C rise dropping down from the 3V3 supply. Robert " 'Freedom' has no meaning of itself. There are always restrictions, be they legal, genetic, or physical. If you don't believe me, try to chew a radio signal. " -- Kelvin Throop, III http://www.aeolusdevelopment.com/
2006-04-26 by Steve Franks
Ah, thanks for all the feedback. I was not specific enough. I was more interested in an all 1.8V soution than just a single supply. Which is to say 3.3V is a killer from 2x AAA or from a single LiPo, then you have to run a dc-dc converter, and that does get expensive on your bom. Not to mention those fat inductors for a boost converter with reasonable efficiency. Pretty much rules out using the 2103 for ZigBee/motes when you can fire up an 8-bit uC with only a 1.8V regulator with an Iq of 10uA. Steve [Non-text portions of this message have been removed]
2006-04-26 by Robert Adsett
At 05:50 PM 4/25/2006 -0700, Steve Franks wrote: >Ah, thanks for all the feedback. I was not specific enough. I was more >interested in an all 1.8V soution than just a single supply. Which is to >say 3.3V is a killer from 2x AAA or from a single LiPo, then you have to run >a dc-dc converter, and that does get expensive on your bom. Not to mention >those fat inductors for a boost converter with reasonable efficiency That's what switched capacitor regulators are for (like the TPS60240). Generally low current but it looks like that is something you are watching anyway. >Pretty much rules out using the 2103 for ZigBee/motes when you can fire up >an 8-bit uC with only a 1.8V regulator with an Iq of 10uA. Well considering the 2103 claims a typical operating current on the 1V8 supply of from 7mA to 41mA you are already in a different league. Three orders of magnitude is not easily crossed. " 'Freedom' has no meaning of itself. There are always restrictions, be they legal, genetic, or physical. If you don't believe me, try to chew a radio signal. " -- Kelvin Throop, III http://www.aeolusdevelopment.com/
2006-04-26 by fordp2002
The ports on the LPC2103 can be treated as 4 x 8 bit ports or 2 x 16 bit ports. You can even mix 8 bit / 16 bit / 32 Bit access. The chip also supports setting and resetting an arbitary number of port bits at the same time. That makes the LPC2103 the most flexible GPIO's to date, unless you know different ;) I have made a header file for the LPC2103 based on the GNUARM LPC2106 one. It has a few more features but is not yet production quality. I can post it into files if anybody is iterested. All the best. --- In lpc2000@yahoogroups.com, Robert Wood <robert.wood@...> wrote:
> The other thing that puts me off using the 210x rather than, say, AVRs > is the fact that you only have one 32 bit port. I'd much rather use an > ATMega that has multiple ports. Too much bit shifting going on with the > ARM for me. >
2006-04-26 by Xtian Xultz
Em Ter 25 Abr 2006 20:55, Robert Adsett escreveu: > At 08:04 PM 4/25/2006 -0300, Xtian Xultz wrote: > > Its too dangerous to generate 1.8V (more or less) from the 3.3V with two > >diodes in series? > >Someone tryed this? > > Just put in an LDO. You can get them as small as a sot23 giving you about > 90mA with a 30C rise dropping down from the 3V3 supply. > > Robert Thats what I am using, LM317 with SOT23 footprint. What makes me sad is that both regulators (one for 3,3 and one for 1,8V) plus the resistors ocuppies more space on the board than the LPC itself. So, LPC is a really small microcontroller (looking for a ARM based wich has only 48 pins, the concorrents are 100 or 144 pins, the smallers!) and with a very small footprint (soldering 20 mils pins with soldering iron its not easy, I am doing, but its not funny. Its interesting that a 25 mils chip is much more easy to solder and to design the board) but the power circuit is bigger...
2006-04-26 by fordp2002
I have looked on Digikey and found some regulators which have both 3.3V and 1.8V in one package :- PQ2L3182MSPQ and LP2966IMMX-1833. I have not tried them but they may be work checking out. The sharp one looks both small and cheap, but I could not find a decent data sheet for it. All the best. --- In lpc2000@yahoogroups.com, Xtian Xultz <xultz@...> wrote: > Thats what I am using, LM317 with SOT23 footprint. What makes me sad is that > both regulators (one for 3,3 and one for 1,8V) plus the resistors ocuppies > more space on the board than the LPC itself.
2006-04-26 by Leon Heller
----- Original Message -----
From: "Xtian Xultz" <xultz@...> To: <lpc2000@yahoogroups.com> Sent: Wednesday, April 26, 2006 3:45 PM Subject: Re: [lpc2000] 2103 forces dual supply? > Em Ter 25 Abr 2006 20:55, Robert Adsett escreveu: >> At 08:04 PM 4/25/2006 -0300, Xtian Xultz wrote: >> > Its too dangerous to generate 1.8V (more or less) from the 3.3V with >> > two >> >diodes in series? >> >Someone tryed this? >> >> Just put in an LDO. You can get them as small as a sot23 giving you >> about >> 90mA with a 30C rise dropping down from the 3V3 supply. >> >> Robert > > Thats what I am using, LM317 with SOT23 footprint. What makes me sad is > that > both regulators (one for 3,3 and one for 1,8V) plus the resistors ocuppies > more space on the board than the LPC itself. So, LPC is a really small > microcontroller (looking for a ARM based wich has only 48 pins, the > concorrents are 100 or 144 pins, the smallers!) and with a very small > footprint (soldering 20 mils pins with soldering iron its not easy, I am > doing, but its not funny. Its interesting that a 25 mils chip is much more > easy to solder and to design the board) but the power circuit is bigger... There is a tiny little Maxim dual regulator (MAX8882) in SOT23-6 giving 1.8 V and 3.3 V. They used only to be available by the reel, but one can buy small quantities, now. Leon -- Leon Heller, G1HSM leon.heller@... http://www.geocities.com/leon_heller --- [This E-mail has been scanned for viruses but it is your responsibility to maintain up to date anti virus software on the device that you are currently using to read this email. ]
2006-04-26 by Herbert Demmel
At 15:24 26.04.2006 +0000, you wrote: >I have looked on Digikey and found some regulators which have both >3.3V and 1.8V in one package :- > >PQ2L3182MSPQ and LP2966IMMX-1833. > >I have not tried them but they may be work checking out. > >The sharp one looks both small and cheap, but I could not find a >decent data sheet for it. Did you have a look for the MOQ (2 or 3000 !!!) ??? Regards Herbert >All the best. > >--- In lpc2000@yahoogroups.com, Xtian Xultz <xultz@...> wrote: > > > Thats what I am using, LM317 with SOT23 footprint. What makes me sad >is that > > both regulators (one for 3,3 and one for 1,8V) plus the resistors >ocuppies > > more space on the board than the LPC itself. > > > > > >---------- >YAHOO! GROUPS LINKS > > * Visit your group "<http://groups.yahoo.com/group/lpc2000>lpc2000" > on the web. > * > * To unsubscribe from this group, send an email to: > * > <mailto:lpc2000-unsubscribe@yahoogroups.com?subject=Unsubscribe>lpc2000-unsubscribe@yahoogroups.com > > * > * Your use of Yahoo! Groups is subject to the > <http://docs.yahoo.com/info/terms/>Yahoo! Terms of Service. > > >---------- [Non-text portions of this message have been removed]
2006-04-26 by Matthew Kavalauskas
I have used that Maxim chip for another purpose (it is pretty low noise), but good luck getting small quantities of it and forget Digi-Key or Newark. Fortunately, there are other manufacturers that have similar parts. Matt ________________________________
From: lpc2000@yahoogroups.com [mailto:lpc2000@yahoogroups.com] On Behalf Of Leon Heller Sent: Wednesday, April 26, 2006 11:29 AM To: lpc2000@yahoogroups.com Subject: Re: [lpc2000] 2103 forces dual supply? ----- Original Message ----- From: "Xtian Xultz" <xultz@...> To: <lpc2000@yahoogroups.com> Sent: Wednesday, April 26, 2006 3:45 PM Subject: Re: [lpc2000] 2103 forces dual supply? > Em Ter 25 Abr 2006 20:55, Robert Adsett escreveu: >> At 08:04 PM 4/25/2006 -0300, Xtian Xultz wrote: >> > Its too dangerous to generate 1.8V (more or less) from the 3.3V with >> > two >> >diodes in series? >> >Someone tryed this? >> >> Just put in an LDO. You can get them as small as a sot23 giving you >> about >> 90mA with a 30C rise dropping down from the 3V3 supply. >> >> Robert > > Thats what I am using, LM317 with SOT23 footprint. What makes me sad is > that > both regulators (one for 3,3 and one for 1,8V) plus the resistors ocuppies > more space on the board than the LPC itself. So, LPC is a really small > microcontroller (looking for a ARM based wich has only 48 pins, the > concorrents are 100 or 144 pins, the smallers!) and with a very small > footprint (soldering 20 mils pins with soldering iron its not easy, I am > doing, but its not funny. Its interesting that a 25 mils chip is much more > easy to solder and to design the board) but the power circuit is bigger... There is a tiny little Maxim dual regulator (MAX8882) in SOT23-6 giving 1.8 V and 3.3 V. They used only to be available by the reel, but one can buy small quantities, now. Leon -- Leon Heller, G1HSM leon.heller@... http://www.geocities.com/leon_heller --- [This E-mail has been scanned for viruses but it is your responsibility to maintain up to date anti virus software on the device that you are currently using to read this email. ] SPONSORED LINKS Microcontrollers <http://groups.yahoo.com/gads?t=ms&k=Microcontrollers&w1=Microcontroller s&w2=Microprocessor&w3=Intel+microprocessors&c=3&s=69&.sig=c-HXthtbZy4TZ bI3ib0PMg> Microprocessor <http://groups.yahoo.com/gads?t=ms&k=Microprocessor&w1=Microcontrollers& w2=Microprocessor&w3=Intel+microprocessors&c=3&s=69&.sig=ijt0SspWtjogcHC uFD0lUQ> Intel microprocessors <http://groups.yahoo.com/gads?t=ms&k=Intel+microprocessors&w1=Microcontr ollers&w2=Microprocessor&w3=Intel+microprocessors&c=3&s=69&.sig=WOZdpklk gHbXR5quAgrl5w> ________________________________ YAHOO! GROUPS LINKS * Visit your group "lpc2000 <http://groups.yahoo.com/group/lpc2000> " on the web. * To unsubscribe from this group, send an email to: lpc2000-unsubscribe@yahoogroups.com <mailto:lpc2000-unsubscribe@yahoogroups.com?subject=Unsubscribe> * Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service <http://docs.yahoo.com/info/terms/> . ________________________________ [Non-text portions of this message have been removed]
2006-04-26 by rtstofer
--- In lpc2000@yahoogroups.com, "Matthew Kavalauskas" <mkavalauskas@...> wrote: > > I have used that Maxim chip for another purpose (it is pretty low > noise), but good luck getting small quantities of it and forget Digi-Key > or Newark. Fortunately, there are other manufacturers that have similar > parts. > > Matt > The TI TPS767D318PWP is in stock at Mouser $4.8. Unfortunately, it comes in a 28 pin power pad package about 7x10 mm. Richard
2006-04-26 by Steve Franks
>Well considering the 2103 claims a typical operating current on the 1V8 supply of from 7mA to 41mA you are already in a different league. Three orders of magnitude is not easily crossed. Not necissarially - the AtMega128 8-bit Avr sucks a healthy 10mA when operating at 8MHz. I did the calculations and when a mote is operating (sample sensors, tx data) for 10mS every 60s, you have a 6000:1 duty cycle, so your 10mA looks like 1.6uA. Therefore the sleep current is the dominant drain on your batteries. TI's vey best LDO's eat 10uA Iq, so you're looking at two LDO's doubling your dominant current consumption. Can take a mote's effective deployed life down accordingly. As people mentioned, having as much board eaten by regulators as the processor is quite anoying as well. I suppose the ideal solution is a discrete watchdog chip which would pull up the power supplies on interrupt long enough for the chip to boot. Adds yet more mm-sq to your design, and I doubt very much I can find one of those to run at 10uA. Well, I'm out. Maybe Philips will take notice and let VccIO go down to 1.8Von their next effort. I think it's silly to run a 3.3V rail to get a 1.8V core processor to talk to a 1.8V core FPGA which is another thing I do.... Steve [Non-text portions of this message have been removed]
2006-04-27 by Robert Adsett
At 03:09 PM 4/26/2006 -0700, Steve Franks wrote: > >Well considering the 2103 claims a typical operating current on the 1V8 >supply of from 7mA to 41mA you are already in a different league. Three >orders of magnitude is not easily crossed. > >Not necissarially - the AtMega128 8-bit Avr sucks a healthy 10mA when >operating at 8MHz. I had misread your quiescent current as operating current. Makes a bit of difference :) >I did the calculations and when a mote is operating (sample sensors, tx >data) for 10mS every 60s, you have a 6000:1 duty cycle, so your 10mA looks >like 1.6uA. Therefore the sleep current is the dominant drain on your >batteries. TI's vey best LDO's eat 10uA Iq, so you're looking at two LDO's >doubling your dominant current consumption. Can take a mote's effective >deployed life down accordingly. Not quite, you still have to add your sleep current that brings you back up to almost 10uA again. Your increase in Iq from the extra regulator is now 'only' 50%. I don't suppose any of the dual regulators help? >I suppose the ideal solution is a discrete watchdog chip which would pull up >the power supplies on interrupt long enough for the chip to boot. Adds yet >more mm-sq to your design, and I doubt very much I can find one of those to >run at 10uA. One of the 6 pin PICs maybe? I haven't checked the current draw on one of them. >Well, I'm out. Maybe Philips will take notice and let VccIO go down >to 1.8Von their next effort. I think it's silly to run a >3.3V rail to get a 1.8V core processor to talk to a 1.8V core FPGA which is >another thing I do.... Now there's a matter of perspective. I'd like to see true 5V I/O rather than merely 5V tolerant. I'd like the extra noise tolerance. Robert " 'Freedom' has no meaning of itself. There are always restrictions, be they legal, genetic, or physical. If you don't believe me, try to chew a radio signal. " -- Kelvin Throop, III http://www.aeolusdevelopment.com/
2006-04-27 by Robert Adsett
At 11:45 AM 4/26/2006 -0300, Xtian Xultz wrote: >Em Ter 25 Abr 2006 20:55, Robert Adsett escreveu: > > At 08:04 PM 4/25/2006 -0300, Xtian Xultz wrote: > > > Its too dangerous to generate 1.8V (more or less) from the 3.3V with two > > >diodes in series? > > >Someone tryed this? > > > > Just put in an LDO. You can get them as small as a sot23 giving you about > > 90mA with a 30C rise dropping down from the 3V3 supply. >Thats what I am using, LM317 with SOT23 footprint. What makes me sad is that >both regulators (one for 3,3 and one for 1,8V) plus the resistors ocuppies >more space on the board than the LPC itself. Others have pointed out the use of dual regulators to save space but just as important is the use of a fixed voltage regulator. Unless you are using 0203's your resistors probably take more room than the regulators. With a sot23 dual regulator you are down to what? a sot23 and 3 caps? Reflecting on the past embedded systems's I've built, the power supply has almost always been comparable in size to the microcrontroller. Of course most of those has external memory but even so the PS was a significant piece. Robert " 'Freedom' has no meaning of itself. There are always restrictions, be they legal, genetic, or physical. If you don't believe me, try to chew a radio signal. " -- Kelvin Throop, III http://www.aeolusdevelopment.com/
2006-04-27 by Ed Schlunder
--- In lpc2000@yahoogroups.com, Herbert Demmel <dh2@...> wrote: > > At 15:24 26.04.2006 +0000, you wrote: > >I have looked on Digikey and found some regulators which have both > >3.3V and 1.8V in one package :- > > > >PQ2L3182MSPQ and LP2966IMMX-1833. > > > >I have not tried them but they may be work checking out. Quick heads up for all you guys looking at fixed voltage regulators for the LPC2103... There's an errata on the 1.8V pin that says you should provide 1.95V instead because there will be too much internal voltage drop to operate properly at full speed with all peripherals. Doh! As a result, I ended up choosing an adjustable regulator on my BBMICRO development board for the 1.8V pin, even though it meant using more board space than a fixed regulator would have used. Here's the LPC2103 Errata: http://www.standardics.philips.com/support/documents/microcontrollers/pdf/errata.lpc2103.pdf --- K9spud Microcontroller Electronics http://www.k9spud.com/