Low Power Mode Revisited (S08GT60)

[admin]

I know this has been cussed & discussed here before but I can't find a solution, so...

GT60 sittin all by itself, nothing connected but Vdd, GND and a 32KHz Xtal. All ins have pullups and all outs are static high or low (except 1 pin used as a strobe in the RTI routine, at 1pps.) No peripheral subsystems enabled except RTI; the low voltage detection system is off. The code inits everything and then we loop on a STOP and a BRA back to STOP. RealTime IRQ at 1pps pulses a pin. ICG mode is FEE.

Current drain is 800uA.

If I set OSCSTEN from 1 to 0 to stop the oscillator during STOP, it does exactly that (and no more RTI). Current STILL 800uA.

Pull off the 3v from the Vdd pin and the oscillator still runs but the current goes to near zero.

Pull off the 3v from the VddA pin and the osc still runs and current stays at 800u.

Pull off the 3v from the VrefH pin and the osc still runs and current stays at 800u.

In AppNote AN2493 they did essentially the same thing and they report like 14uA when stopped.

Any hints or suggestions would be most appreciated.

[admin]

I've got more info on my problem. Surely someone has seen this before, but it's a bit weird.

If I reduce Vdd from 3.3v to around 2.8v my current drops to around 15uA. Above 3v it rises quickly to 1.2ma at 3.3v. It seems as though the internal voltage regulator is never going into standby. I've got all the low voltage detection stuff turned off, and I'm running it from power up with no debugger attached (and the BKGND pin pulled high), so it shouldn't be keeping the regulator on due to the background debugger being active. I see where it will keep it alive if the ENBDM bit in BDCSCR is set, but from a power up with no Cyclone or anything else present this bit should never become set.... should it?

I've tried another GT60A with the same results. Even tried a couple of GT32As with the same results also.

Obe-Wan, where are you?