Using the BasicStamps at 3.3 volts for low power consumption
LoopyByteloose
Posts: 12,537
First, I know this is a pretty advanced question.·
{I am trying to build remote location nodes that would use 3.3 Volt RS-485 drivers with a BasicStamp.· I would like to extend battery life to the extreme.}
Second,I do understand that to use the Stamp at a lower voltage would require that ALL the logic that interfaces it would require the ability to perform at the lower voltage I/0.
Having said all that, it seems to me that by not using the on-board regulator and providing a 3.3 volt supply, I could significantly reduce the power consumption of the BasicStamp [noparse][[/noparse]any BasicStamp?].
Will it work?
Is there something I have overlooked?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)
······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
{I am trying to build remote location nodes that would use 3.3 Volt RS-485 drivers with a BasicStamp.· I would like to extend battery life to the extreme.}
Second,I do understand that to use the Stamp at a lower voltage would require that ALL the logic that interfaces it would require the ability to perform at the lower voltage I/0.
Having said all that, it seems to me that by not using the on-board regulator and providing a 3.3 volt supply, I could significantly reduce the power consumption of the BasicStamp [noparse][[/noparse]any BasicStamp?].
Will it work?
Is there something I have overlooked?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)
······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
Comments
Yes, I fear there·is a good deal that you have overlooked. Rather than go into the details, which might well fill a small book, please find attached the MicroChip Applications Note Number 522 which deals with designing for low power consumption. There should be enough reasons in there to show that it can't be done easily, since we, the end users, have no control over how the underlying microprocessor (MicroChip PIC or Ubicom SX) was programmed by Parallax, nor should we have such control. Although I didn't attempt to seek it out, I'm sure there is a similar Ubicom Applications Note dealing with designing for low power
operation.
Since I don't think the following two items are covered in the text cited above, and one also needs to consider these following two COMMON microprocessor support circuits when varying the operational voltage being used, I include the following below.·I won't go into specific detail however, and leave that for the reader to determine:
POR: power-on reset - Keeps the microprocessor in a stable and controlled reset state, unless or until the design voltage is met and stabilizes.
B-O: brown-out protection - Causes the microprocessor to cease operation, or causes it to enter the reset state, if or when the supply voltage falls below design levels.
Note: The POR and B-O circuitry may be internal or external to the microprocessor, depending on the particular microprocessor platform being used.
Suffice to say also, the there is _always_ a finite trade-off between higher processor speeds and lower power consumption. One can't have one's cake and eat it too, in most cases! The hardware thrust for any interpreter based language is to p-u-m-p the underlying processor as hard and as fast as one possibly can, within the appropriate design limitations.·This push-for-speed (vice minimal power consumption considerations) is to re-gain some throughput performance which is always lost, and which is always inherent, when using any interpretive language. That's "just the nature of the beast", so to speak. PBASIC is such an interpretive language.
Even beyond that mentioned here and in the attached applications note, there are quite a few more considerations as well, but those noted above should be enough to shy you away from such a 5.0 VDC --> 3.3 VDC plan. Sorry if I ruined your day.
Just as a brief aside, there are some processors which support a CORE voltage and a separate operational voltage. The later Intel Pentium chips provide such an example. In those cases, the CPU CORE voltage (the greatest consumer of power) is separated from the (shall I say) "peripheral" voltage(s), and they can and often are, vastly different.
Regards,
Bruce Bates
I took a look at the PIC notes. From what I understand of the SX-Ubicom, it will handle 3.3 and much lower fine. It seems that the Interpreter chip would be the primary problem, if any at all. The oscillator will function and consume less power.
Most of what you mention {Power on reset and Brown Out] are well presented in the SX-Ubicom literature. Brown-out is an option provided for devices that become unreliable at a high specification than the SX, thus creating garbage in - garbage out.
From looking at the PDF, I see that most batteries are charted down to 0.8 volts. 4 ni-cads start at 4.8 volts and are pretty dead at 3.2volts. Why should I be using a regulator that dumps 30% of the battery power in heat and requires and additional 2 ni-cads to become functional.
Additionally, I can trickle charge from a solar cell in that range quite nicely.
It seems to me that if the hardware design of BasicStamps will not currently do it, there is a possible demand for creation of a low-voltage, lower power BasicStamp.
BTW, that is Application note 606, not 522.
I need to figure out the interpreter chips power specs.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)
······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
I looked at the PROM for the BS-24p and it is a 24CW128J. While I cannot locate a particular spec sheet on that, the generic 24C128 is good down to 2.7volts from 5.5 volts max.
I don't expect to program the BasicStamp at these voltages, I just want to see if I can reliably use it in an application.
Maxim has a 3.3volt RS-232 driver and a 3.3 RS-485 driver.· Either will allow transmit and receive to reach futher at low voltage.· Obivously, somethings just cannot work with variable power supply [noparse][[/noparse]like RC timing], but somethings can adapt.
There are 3.3volt regulators too.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)
······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
Post Edited (Kramer) : 11/13/2005 2:10:38 PM GMT
People have use the PIC16C57 programmed by Parallax as an Interpreter chip for OEM boards, and purchased their own external components. Parallax might have to change some fuse settings though.
And you could always burn your own SX-28 chip to use the lower voltages, using the SX/Basic. Testing might be problematic -- I think the current environment requires 5 volts. But you could build a 5-volt version, use that to test with, then create a 3.3 volt version.
Bean.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"SX-Video·Module" Now available from Parallax for only $28.95
http://www.parallax.com/detail.asp?product_id=30012
"SX-Video OSD module" Now available from Parallax for only·$49.95
http://www.parallax.com/detail.asp?product_id=30015
Product web site: www.sxvm.com
Those that would give up freedom for security will have neither.
·
The SX based chips use the on-chip reset and brownout detection circuit, and in the BASIC Stamp 2sx, 2e, 2p, 2pe and 2px, that fuse is set to put the chip into reset at 4.2 volts. You would have to ask for a special version from Parallax to have it set at the 2.7 volt level or turned off. But, the SX chip runs at its full rated frequency all the way down to the minimum voltage.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
So, it implies an avenue of future product development.
I am not all that surprised, but I suspect the FUSE in the SX chip is programmed for the higher Brown out voltage . It may be done that way to assure stability as the BasicStamps tend to be robust and conservative in their engineering. Or they just never considered the option.
The current BasicStamp specifications seem to indicate 4.5volts as the advertised bottom line.
The Brown Out detector is SX-Ubicom factory default at 4.2volts, but can be lowered down to either 2.6V or 2.2V by resetting the FUSE.
In the interim, this sends me back to SX/B and SX-Key for such battery powered schemes. I was hoping to get a quicker development track with the Stamp.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)
······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
Since this 3.3 volt operation can only become available on the 'unregulated' side of the BasicStamp, it seems that the regulator will always be available to protect the BasicStamp when it is properly used.
So, it really seems to me that you could just reprogram the FUSE to 2.6V Brown Out and have a wonderful product improvement - extended battery range. The clock would have to remain the same.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)
······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan