Running Basic Stamp 2pOEM at 3.3V

glysy

Hi:

I am doing an OEM version of the Basic Stamp2p40 module where all of the discrete components will be placed on my custom printed circuit board.

Since all of my circuitry is running at 3.3V, I need to know if there any issues with running all of the OEM components (such as the SX48BD processor chip and the EEROM) at 3.3V. The data sheet for the Parallal SX48BD chip allows the VDD voltage to be from 3.0 to 5.5Volts.

Thanks


George Lysy

Phil Pilgrim (PhiPi)

George,

It depends on where the internal BASIC interpreter firmware has set the brownout voltage. If it's 2.2V or 2.4V, you're safe. If it's 4.1V, you won't be able to run at 3.3V. Parallax can give you the specs for that. You will also need to make sure the EEPROM you've chosen can run at the 2p's SCK clock speed at 3.3V.

-Phil

Jeff Martin

Hi,

Phil is correct. Both the internal brown-out (which is set to 4.2 volts) and the external EEPROM's capabilities at 3.3 volts are obsticles.

There is some good news on the horizon, however... we will likely be releasing BS2p, and above, modules where the firmware in the interpreter disables the internal brownout chip (in order to use a more reliable external brown-out), and in that case, it's just the EEPROM to be worried about. We are doing this to increase the temperature range at which they work reliably and we're nearly done with our prototyping.

I can't promise when this change will occur, but I believe it will be pretty soon.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
--Jeff Martin

· Sr. Software Engineer
· Parallax, Inc.

Tracy Allen

Since you brought it up, I will point out that the problem with the brownout in relation to temperature is unique to the 4.2 volt setting. That problem disappears when the brownout is set to 2.6 volts. It is for that reason that the OEM BS2pe40 chips that go into my OWL2pe loggers have the 2.6 volts as a custom setting. I regularly test them down to -55 and up to +85 degrees C in an environmental chamber.

When the brownout is set to 4.2 volts and with a resonator drive set for 8mhz , the SX48 chip stops operating at a temperature individually somewhere in the range -5 to -22 degC. A lower temperature is possible with the drive set for 20mhz, as it is in the BS2p, because there seems to be an interaction between the brownout voltage and the drive setting. That observation is based on a much smaller sample. The point is, the problem empirically disappears when the setting is BOR26 instead of BOR42. Of course, a lower setting also means other components in the system and in particular the eeprom, have to perform at the lower voltage too.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com

Phil Pilgrim (PhiPi)

Tracy,

Are your environment chamber tests for the 2.6V chips done at a Vdd of 3.3V, or 5V?

Thanks,
Phil

Tracy Allen

Phil,

My tests used a variable power supply. The brownout setting at BOR26 worked fine at either 3.0, 3.3 or 5 volts power and at temperatures down to -40 degC and below. You use 3.3 volt power for your products like the MOBO, right? So you should be fine. I use 5 volts power for my loggers, but I cannot tolerate failures of data logging at subzero temperatures, and that's why I asked for the OEM BOR26 setting.

I want to emphasize that the issue with BOR42 is due to the SX48 chip that underlies the multibank Stamps, not to the Stamps themselves. I did a lot of these tests with plain SX48 chips with the external eeprom and the watchdog timer out of the picture. With BOR46 there are weird interactions between brownout, power supply voltage and crystal drive, so the low temperature effect is more evident on the BS2pe and BS2e than it is on the 'sx and 'p, but the limitations all disappear with the lower BORxx settings. There is no such problem with the original PIC based BS2, and in the small sample I tested, even the non-industrial versions of the original BS2 have operated fine over the full range.

Another way around the limitation is with an external reset chip, and from what Jeff says, that is the route Parallax will take in the next generation. That will open the possibility for the OEM user to swap reset chips.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com

Post Edited (Tracy Allen) : 8/13/2007 5:42:24 PM GMT

Phil Pilgrim (PhiPi)

Tracy,

Oh good! I was concerned for a moment that the more reliable performance at BOR26 might be due solely to the wider differential between Vdd and V_BOR at Vdd = 5V. The MoBoSTAMP Vdd is user-selectable for either 3.3V or 5V. Hence my initial concern.

Given that the BS2pe chip is reliable at BOR26, I hope Parallax continues to provide one programmed that way. There's no room on the MoBo for an external reset circuit! Of course, if they were all programmed for BOR26, an external reset circuit could still reset the chip at a higher threshold voltage.

Phil