Increasing voltage

[Rob Northen]

I recently recapped some Pentium SBC boards.
When testing one today, it halted after 8 hours. When checking the voltages,
I saw 3.3v was 3.6v. Turned off power, turn on again, now it was 3.4v.
Then keeping an eye on it, it was very slowly increasing..
I then tried replacing the 3.3v regulator, no dice, still the same.
The board gets power (5v only) from a 12v-5v step down module.
It just sits calm at 4.9 volts ALL the time, so powersupply is no problem.
Any idea, why the 3.3v is slowly increasing ???

[PCBONEZ]

Re: Increasing voltage

Is there a zener diode between 3.3v and ground or between 5v and 3.3v?
It may be drifting w/temp.

[Rob Northen]

Re: Increasing voltage

Thanks for the reply!

Just checked, and no zener.

I come to think of a thing. I replaced an old Panasonic 10v/1000mf cap
GE series. I could not find any info on GE series, so I went for Panasonic FM.
As it turns out, I measured old cap, it had ESR 0.20 Ohms. New is ESR 0.08 Ohms.
The 3.3v regulator output, is connected directly to the positive leg of the cap.
So, can the lower ESR be a problem ?

[PCBONEZ]

Re: Increasing voltage

In that application I can't see how ESR would do that.
ESR affects AC and ripple [noise] and not DC.
I can't see how it would cause a static DC voltage to slowly drift.
.

[PCBONEZ]

Re: Increasing voltage

Maybe leaky cap if Vref to VR chip?

Adding a zener would be cheesy but should fix it.

[Rob Northen]

Re: Increasing voltage

Maybe leaky cap if Vref to VR chip?

Adding a zener would be cheesy but should fix it.

I replaced all electrolytics, there are only some VERY small SMD caps left.
The zener sounds like an interesting idea.
Thanks for the advice!