View previous topic :: View next topic |
Author |
Message |
cstan_02
Joined: 24 Jul 2005 Posts: 13
|
My 6722 died yesterday !! |
Posted: Thu Dec 15, 2005 7:57 pm |
|
|
Hello,
Can you tell me what are reasons for the PIC to fail ??
And when it fails, it grounds the +5v supply. I was field testing my working code and it suddenly failed. I had tested my unit for 24 hour runs before and never had any problems. The power supply is stable(from a battery). Everything was fine ?? I am using a PIC18LF6722 with external 24Mhz oscillator. This happened before atleast once, which was more than six months back. That time i suspected that it had something to do with my hardware. This time I have separate power supplies for different classes of components, and all are stable. I just wanted to know how can I prevent this from happening again.
Thanks |
|
|
PCM programmer
Joined: 06 Sep 2003 Posts: 21708
|
|
Posted: Thu Dec 15, 2005 9:43 pm |
|
|
It's possible that it's a CMOS latch-up problem. Describe the components
and circuits that are connected to the PIC pins. |
|
|
cstan_02
Joined: 24 Jul 2005 Posts: 13
|
|
Posted: Thu Dec 15, 2005 11:04 pm |
|
|
Well to say, the circuit is simple with a spi flash memory which uses the 3.3v rail, one transciever connected to one of the uart and a gps on the other. The power for the latter two components are isolated from that of the main 6722 power. That is, there are two +5v rail and another 3.3v rail. Talking to the flash is through some voltage dividers. Other than that there is a MAX232, an external 24 Mhz oscillator and some few transistors.
Anyway, is there something like overclocking the PIC. Well I don't think so. But Mr. PCM_Programmer i don't understand by what you said it can be a CMOS latchup problem. Can you explain a bit more. That would be so nice !!
Thanks. |
|
|
asmallri
Joined: 12 Aug 2004 Posts: 1634 Location: Perth, Australia
|
|
Posted: Thu Dec 15, 2005 11:33 pm |
|
|
Quote: | Anyway, is there something like overclocking the PIC. Well I don't think so. But Mr. PCM_Programmer i don't understand by what you said it can be a CMOS latchup problem. Can you explain a bit more. That would be so nice !! |
This is the condition that occurs if the input voltage on a pin exceeds the VDD of the device. High current flows can occur and it is non unusaul for the device to be destroyed. This often occures if a voltage is applied to an I/O pin before VDD has applied to the device or in cases when the VDD voltage rise time is slow.
Another very common problem occurs when two systems, running from independant supplies (like a PC connected in a PIC) are connected togther without first connecting a common ground between then. In this scenario the devices are floating with respect to each other and the latchu condition can easily occur when a signal voltage is applied to an inout before the grounds have been connected together leading to the destruction of the device. This is how a lot of ICD2 get destroyed by hobbyists :-) _________________ Regards, Andrew
http://www.brushelectronics.com/software
Home of Ethernet, SD card and Encrypted Serial Bootloaders for PICs!! |
|
|
cstan_02
Joined: 24 Jul 2005 Posts: 13
|
|
Posted: Fri Dec 16, 2005 12:37 am |
|
|
Thanks Andrew, I'll try to take care of the latchup problem next time. But is there any sure way of preventing this, like if i power everything in the circuit with a single supply, will it eleminate the problem ?? Or is it enough that I put capacitors to increase the VDD rise time of the components connected to the PIC I/O ??
Thanks. |
|
|
asmallri
Joined: 12 Aug 2004 Posts: 1634 Location: Perth, Australia
|
|
Posted: Fri Dec 16, 2005 2:39 am |
|
|
Generally if you have a common power supply with a siingle power rail (VDD and VSS are common) then latchup will not occur. It is possible for latchup to occur if you have two seperate power rails. For example, assume you have a 9 to 12 volt power source that you use to derive 5volts for a PIC and 3.3 volts for some other device. Assume you have two regulators, a 7805 derives the 5 volts for the PIC from the power source and a 3.3 volt regulator derives its output from the +5 volt regulator. If you have a large capacitor on the output of the 3.3 volt regulator then the 5 volt rail will rise before the 3.3 volt rail. If the PIC asserts a high level out to the 3.3volt device during this period then latchup can occur.
This problem can be solved in two ways. First you mimimise the difference in rise time between the two power supplies and secondly you incorporate a delay in your PICs startup to enable the power supplies to stabalise before asserting any outputs. _________________ Regards, Andrew
http://www.brushelectronics.com/software
Home of Ethernet, SD card and Encrypted Serial Bootloaders for PICs!! |
|
|
cstan_02
Joined: 24 Jul 2005 Posts: 13
|
|
Posted: Fri Dec 16, 2005 3:21 am |
|
|
Thanks Andrew, have a nice day !!
:-) |
|
|
|