030 exxos RTC issues.
Posted: Mon Sep 03, 2018 10:02 pm
There have been a couple reports of my RTC modules not working in some people's machines last year. It was generally thought that the previous batch of my boards could have been faulty, or had a fault. However, the current batch I did were definitely 100% tested and working, but I have had two reports of the modules failing to store data this past week. Both were machines I had in for repair, and the module was working fine in both machines originally (and my test machine).
I don't know if any of the people I am emailing are on here or not.. But one chap chap fitted a CT60 with pico psu, and the machine did not boot, upon reverting to a stock machine, it also did not boot like the NVRAM settings were corrupted. Basically the settings are no longer been stored in the RTC chip, resulting in corruption, and the machine not booting.
I just had another report that a similar thing is also happened on another machine... Only this machine has one of my own power supplies.. So it should not really be a power issue. I could understand it if a bad power supply spiking and killing the chip.. Or it could be possible that there is simply bad grounding on the Falcon causing the module to malfunction or even kill it.
The overall consensus is that there is a write issue to the module, as to why this is, and why it only happens seemingly random on people's machines it is very difficult to tell. I did originally find that there was no decoupling capacitor on the Falcon which was causing write issues, however that capacitor is now fitted as standard on the later batches, so this should really not be a problem.. But this does actually point into a power problem again.. And the same issue as before..
https://www.exxosforum.co.uk/forum/viewt ... t=37#p2479
The first idea here simply the power supply.. But obviously in today's report, does not seem to be the case.. but does not rule out a power issue.. If the modules are working in my own Falcon, then I would lean towards the thought that there may be some noise on the RW line to the RTC module. It could explain write issues.. but not why some modules are being killed off.
The modules are not shipped with the battery, I don't think it is possible to blow the module by putting in the battery wrong.. It is also possible the metal shielding could come in contact with the top of the module and maybe short the battery out.. I lost my shielding a while ago so not really sure about that.. But it could explain some issues of the module storing time battery shorted out.
So for the time being I am pulling these modules from my store until this issue can be resolved.
I don't know if any of the people I am emailing are on here or not.. But one chap chap fitted a CT60 with pico psu, and the machine did not boot, upon reverting to a stock machine, it also did not boot like the NVRAM settings were corrupted. Basically the settings are no longer been stored in the RTC chip, resulting in corruption, and the machine not booting.
I just had another report that a similar thing is also happened on another machine... Only this machine has one of my own power supplies.. So it should not really be a power issue. I could understand it if a bad power supply spiking and killing the chip.. Or it could be possible that there is simply bad grounding on the Falcon causing the module to malfunction or even kill it.
The overall consensus is that there is a write issue to the module, as to why this is, and why it only happens seemingly random on people's machines it is very difficult to tell. I did originally find that there was no decoupling capacitor on the Falcon which was causing write issues, however that capacitor is now fitted as standard on the later batches, so this should really not be a problem.. But this does actually point into a power problem again.. And the same issue as before..
https://www.exxosforum.co.uk/forum/viewt ... t=37#p2479
The first idea here simply the power supply.. But obviously in today's report, does not seem to be the case.. but does not rule out a power issue.. If the modules are working in my own Falcon, then I would lean towards the thought that there may be some noise on the RW line to the RTC module. It could explain write issues.. but not why some modules are being killed off.
The modules are not shipped with the battery, I don't think it is possible to blow the module by putting in the battery wrong.. It is also possible the metal shielding could come in contact with the top of the module and maybe short the battery out.. I lost my shielding a while ago so not really sure about that.. But it could explain some issues of the module storing time battery shorted out.
So for the time being I am pulling these modules from my store until this issue can be resolved.