PLX 9030 compatibility problem with STMicroelectronics 93S56

[See_si]

Hello all, it seems that we are unable to program a STMicroelectronics
93S56W6 2kbit EEPROM using the SDK from PLX technology (PLXmon version
4.10/PLX API version 4.10). Has anyone seen any compatibility problems
with the STMicroelectronics part? For future reference, how about any
other manufacturers of these EEPROMS? FWIW, we *have* used Fairchild
parts with success.

--
Jay Walling
Diagnostic Engineer
jwalling@comarkcorp.com
Comark Corporation
93 West St.
Medfield, MA 02052
http://www.comarkcorp.com

 

[See_si]

On Tue, 20 Jan 2004 11:19:38 -0500 See_sig<none@dev.nul> wrote in
Message id: <vjkq00p9c64hceem7ggmq2h84qt892i3e3@4ax.com>:

Hello all, it seems that we are unable to program a STMicroelectronics
93S56W6 2kbit EEPROM using the SDK from PLX technology (PLXmon version
4.10/PLX API version 4.10). Has anyone seen any compatibility problems
with the STMicroelectronics part? For future reference, how about any
other manufacturers of these EEPROMS? FWIW, we *have* used Fairchild
parts with success.

Follow-up - the STM 93S56 parts are in fact *not* compatible with the
PLX 9030, even though they're listed as an alternative device. We are
rather wary of using Fairchild parts after reading
http://www.plxtech.com/download/other/eeprom_v4.0.pdf where it states:

13. Note regarding Fairchild EEPROMs: In a letter from Fairchild Semiconductor dated Feb. 27, 2001,
Fairchild states that a problem may exist in writing multiple times to all FM93CS46, FM93CS56, FM93CS66
EEPROMs, and to National Semiconductor NM93CS66 EEPROMs beginning with date code B9942.
These EEPROMs are recommended only for write-once applications. A new device revision is announced
to fix this problem. Contact your Fairchild Sales Representative for further information.

14. Note regarding Fairchild EEPROMs: In a letter from Fairchild Semiconductor dated June 19, 2002,
Fairchild states that in FM93CSxx EEPROMs starting with date code B14xx, a WRITE command could fail
to write data to the EEPROM, if during the previous write the SK clock pin and the CS input pin were both
high at the time that the previous write completed internally. This erratum precludes use of the PCI Vital
Product Data (VPD) feature of the PCI 9656, PCI 9056, PCI 9054, PCI 9030, and IOP 480 to write to the
EEPROM. VPD reads are not affected. Software can alternatively use the CNTRL register bits for read/write
access to affected Fairchild EEPROMs.

In addition, none of the parts listed in the above .pdf as having "no
restrictions" are currently available from our distributors in
quantities of less than 200, and lead times of less than 10 weeks.

Ya just Gotta love this industry...

--
Jay Walling
Diagnostic Engineer
jwalling@comarkcorp.com
Comark Corporation
93 West St.
Medfield, MA 02052
http://www.comarkcorp.com

 

[Carter Buck]

Hi Jay,

Our listing as an "Alternative" device means that we reviewed the data
sheet and believed the part will work with the PCI9030, however we
have not tested it. Only devices that we listed as "Recommended" have
been tested here at PLX. At the time of our testing, the
manufacturer's rep was not able to provide the ST 93S56 or 93S66 to
us. The rep did provide the ST 93S46 (1Kbit) EEPROM, which tested
successfully with our 5V chips. The EEPROM signaling protocol in our
3.3V devices including the PCI9030 is slightly different from that in
our 5V devices that use the 1Kbit EEPROM.

Whether our API (called by PLXMon) works with this EEPROM is a
different issue from hardware compatibility. The software has no way
of knowing which manufacturer's EEPROM is being accessed. Various
EEPROM devices have different timings and errata. Our API is tested on
various EEPROMs but not all, and the software timings work for most
EEPROMs. I have written a new algorithm that accounts for all of the
various timing and errata issues that I know of, however the
programmers have not yet had time to code it. This task is on our
schedule for a future SDK release.

Our EEPROM Guidelines document is due for update (I've started but
haven't finished yet). Fairchild EEPROMs have been End-of-Life'd, and
recommendation will be removed from our list.

As to availability, Microchip 93LC56B is a recommended device, and the
Digikey website shows approximately 1000 in stock available for
immediate shipment, and they accept low volume orders. Microchip
recommends that new designs use the 93LC56C instead of the 93LC56B,
however we have not tested this new revision and Digikey does not list
it in their catalog.

Make sure that you are powering the EEPROM with at least 3.3V supply.

Regards,

Carter Buck
Technical Applications Engineer
PLX Technology, Inc.
http://www.plxtech.com

See_sig<none@dev.nul> wrote in message news:<baf420t9qg5i38mot77a8m0de8e4p0gt7g@4ax.com>...

On Tue, 20 Jan 2004 11:19:38 -0500 See_sig<none@dev.nul> wrote in
Message id: <vjkq00p9c64hceem7ggmq2h84qt892i3e3@4ax.com>:

Hello all, it seems that we are unable to program a STMicroelectronics
93S56W6 2kbit EEPROM using the SDK from PLX technology (PLXmon version
4.10/PLX API version 4.10). Has anyone seen any compatibility problems
with the STMicroelectronics part? For future reference, how about any
other manufacturers of these EEPROMS? FWIW, we *have* used Fairchild
parts with success.

Follow-up - the STM 93S56 parts are in fact *not* compatible with the
PLX 9030, even though they're listed as an alternative device. We are
rather wary of using Fairchild parts after reading
http://www.plxtech.com/download/other/eeprom_v4.0.pdf where it states:

13. Note regarding Fairchild EEPROMs: In a letter from Fairchild Semiconductor dated Feb. 27, 2001,
Fairchild states that a problem may exist in writing multiple times to all FM93CS46, FM93CS56, FM93CS66
EEPROMs, and to National Semiconductor NM93CS66 EEPROMs beginning with date code B9942.
These EEPROMs are recommended only for write-once applications. A new device revision is announced
to fix this problem. Contact your Fairchild Sales Representative for further information.

14. Note regarding Fairchild EEPROMs: In a letter from Fairchild Semiconductor dated June 19, 2002,
Fairchild states that in FM93CSxx EEPROMs starting with date code B14xx, a WRITE command could fail
to write data to the EEPROM, if during the previous write the SK clock pin and the CS input pin were both
high at the time that the previous write completed internally. This erratum precludes use of the PCI Vital
Product Data (VPD) feature of the PCI 9656, PCI 9056, PCI 9054, PCI 9030, and IOP 480 to write to the
EEPROM. VPD reads are not affected. Software can alternatively use the CNTRL register bits for read/write
access to affected Fairchild EEPROMs.

In addition, none of the parts listed in the above .pdf as having "no
restrictions" are currently available from our distributors in
quantities of less than 200, and lead times of less than 10 weeks.

Ya just Gotta love this industry...