Hi SGI'ers,
as part of Dallas DS1742W chip initialisation/swapping Khral
posted on how he used TL866 MiniPro programmer to read a valid Tezro's DS1742W chip and and then used this dump to restore/initialise a new Dallas chip.
This helps bypass the entire Dallas initialisation problem, as you can now plug in the Dallas with restored configuration and then use l1 interface (either via l1cmd if your machines boots or l2 controller) to reset the machines serial back to the correct machine specific value.
I have now done some additional testing with the TL866A on a number of Dallas chips.
For those who want to test this stuff first be aware that TL866 is officially made by: XGecu (
www.autoelectric.cn).
On their web site they indicate that the TL866A/CS variation is now "End of Life" and that the device to get is "TL866II Plus". This supports Dallas (as DS1220 which is NVRAM version only without RTC like the DS1742W). The new version also supports many more additional chips than the TL866A/CS version.
I did test reads of DS1742W-120 chips taken out of real SGI equipment and these all read/write ok, with caveat that they will get verification failures at address: 07F8 . This is in range where the RTC stores its values (07F8-07FF), so the values can change between write / read. So when doing write probably best to leave this range out.
I also tested the potentially "fake" ebay sourced DS1742W-120+ chips (see this
thread), these fail to verify at different address ranges, which indicates that they are likely manufacturing faulty chips, not total fakes.
Having the programmer means it is easy to test any sourced Dallas chips to verify if they are ok or not and if not working get immediate refund. It is also easy to take readings of RTC to see if this is operating or not by just reading the correct address range.
My next step in testing will be to pull Dallas from Fuel and copy it to good SGI Dallas and then run the Fuel eeprom init on the backup chip to see the result. Now I know I am using backup chip and still have known good chip I think it is safe to try this to see if this helps with L1
issues I have been having with my Fuel.
Hope this is helpful info to others with Dallas DS1742W issues and thank you Khral for the testing with the TL866 EEPROM programmer.
Cheers from Oz,
John.
