TI E2E Community
8255.sequencer block diagram.doc
I'm trying to implement a voltage sequencing and monitoring machine for our new project and I’m facing some implementation difficulties..
In my machine i have several voltage rails. The rails are enabled one by one with 100ms delay between each step.
Also before each step is being enabled, the voltage which was enabled in the previos step is being checked, in case the voltage is in bound the sequence is proceed up to the last step.
in case one of the previously enabled voltages outgoing the bounds all the enables to all the voltages is need to be de-asserted and the "step up" sequnce need to be started again from the beginning. (in case of one of the voltages supply is malfunctioned - the sequencer must continuously keep trying to step up the rails any way)
In the Fusion Design GUI i can't figure how to configure the sequencer that in both cases of under or over voltage to perform continuous try's of stepping up the voltages .
In the Fusion Design GUI , in the "Fault Responses and Limits" menu, inside the "Edit" tab there is a option of " Restart Continuously" but this option is available only for the "Vout over Voltage" for the "Vout Under Voltage" this option is not available...
As i already mentioned, In case of voltage fault , i need that the sequencer will shut down all the rails that was already enabled and will not try to enable the next ones.
So for each rail i configured it dependencies - i.e previous voltage rails and also in the "Fault shutdown Slaves" menu i marked all the other rails. But still when i emulate a voltage failure on one of the rails, i see that for a short period of time the sequencer enables all the rails - even those which is located in the power up sequence after the failed one ( i monitor the power supplies enables with oscilloscope) .
In addition to described power up and monitoring functionality, we also intend to use the "System Watch Dog" functionality integrated in the UCD90120A, but i need that WD timer expiration will cause to the sequencer resequence all the voltage rails - i.e. disable all the rails and after small delay step up according to the sequence.
So is it possible ? and if so, how it should be done?
For making it easyto understand of what I’m trying to do, iIattached my project file and the sequencer functional block diagram.
Thanks in advance ,
Hi Slave, please see our response inline.
VOUT_UV_FAULT is detected only after the voltage on the rail crosses the Power Good On => once VOUT_UV_FAULT is detect, if Fault Response is set to “shut down” and then “restart” => the number of restart is 1 time only. After that, since the voltage on the rail never goes above Power Good On => VOUT_UV_FAULT is not detected. Try the Resequence option with the “Max Re-Sequence” set from 1 to 4, or unlimited time (on Other Config tab).
It can check your project tomorrow on the EVM to see if I can understand.
How about wiring the WDO pin back to the CTRL pin which can be designed to sequence up and down?
Hi Eric and thanks for the quick replay,
Regarding the VOUT_UV_FAULT response, i understood your explanation and i changed the "Max Re-Sequence" option to "unlimited" (as you can see from my attached project file ) but still when i simulated it on my EVM the sequencer didn't preformed resequence of the rails as you can see in the attached oscilloscope screenshot (screenshot1).
I also attached screenshot (screenshot2) of the sequencer reaction to VOUT_OV_FAULT , and as you see the "failed rail" enable pin is being toggled while the next rails are enabled in spite of the fact i configured the sequencer to shut down all the rails when voltage fault is detected. (you also can see it from my attached project file ) .
Regarding the implementation of the functionality i need in case of WD expired , i don't fully understood your suggestion, especially regarding how to configure the CTRL pin to enforce the sequencer to perform a shutdown of all rails and a resequence right after.. I'll appreciate it, if you could be more specific .
Slava, I will run the test and report back tomorrow. Thank you for your patience.
After testing on the EVM, I also cannot get an unlimited resequence behavior for UV faults (only restart up to 1 time). Our docs only show max sequences up to 4 times but the GUI allows unlimited. I will have to check why that is there.
For OV fault response, since the choice is retry continuously, the OV rail never shuts down and as such will not shut down the fault slaves. The OV faulting rail must shut down in order to take down the slave rails which excludes retry continuously.
For the WDO to CNTRL pin connnection, here is more explanation. The CNTRL pin can be used for rail ON/OFF control and sequencing. If you want to set rail 2,3,4 (which have enable pins) to turn ON/OFF (per rail on off config) with the WDO pin, just physically wire the WDO pin to the CNTRL pin and set up the pin polarity accordingly. This way, the WDO pin provides ON/OFF for rail 2,3,4 assuming that is your intent.
Also set the time between restarts (in fault response) to something other than a zero value.
All content and materials on this site are provided "as is". TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with regard to these materials, including but not limited to all implied warranties and conditions of merchantability, fitness for a particular purpose, title and non-infringement of any third party intellectual property right. TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with respect to these materials. No license, either express or implied, by estoppel or otherwise, is granted by TI. Use of the information on this site may require a license from a third party, or a license from TI.
TI is a global semiconductor design and manufacturing company. Innovate with 100,000+ analog ICs andembedded processors, along with software, tools and the industry’s largest sales/support staff.