Does VSTBY need to be connected on MCF52258 if we're not using it?

[davebryant]

If we're not using the RTC and we don't need to retain SRAM through a short-term power-down, do we need to connect VSTBY to anything?  In our last design, we just tied VSTBY to Vcc, which probably didn't accomplish much since VSTBY will go away at the same time as Vcc.  On this design, we want to lay it out for an RTC, in case we decide to use it, but if not, I'd like to leave VSBTY floating.  I measured some voltage on VSTBY when it was disconnected, so it appears that the micro is powering it internally.  In case someone tells me that they want RAM power through a Vcc glitch, is it ok to use a 1N914 didoe between that pin and Vcc, with a storage cap between the diode and VSTBY?  I assume so, since we'll still be above the 1.8V min voltage spec for the pin.  The main thing I'm looking for is whether we'll have problems with the RAM if we leave VSTBY unconnected.  I also heard that some datasheets for other ColdFire parts recommend tying VSTBY to ground.  That seems counterintuitive to me, but if that's what you say needs to be done, so be it.

I found a post from 2008 on this subject, but noone ever responded to it. https://community.freescale.com/message/46422#46422

[TomE · Accepted solution]

You've probably read through the 28 articles that are found for a search for "VSTBY" in this forum.

This one details that the clock doesn't seem to work at 1.8V - there are others comparing the contradictions in quoted VSTBY specifications in the data sheets you should also check. Also the ones static the part won't even start with a "high battery voltage" (3.6V).

Re: MCF52259 RTC details

> I'd like to leave VSBTY floating.

i think that would be a bad idea. Leaving pins floating always causes problems. I would suggest you design the circuit with "optionally installed zero ohm resistors" so you can connect it to anything (ground, Vdd, diode, battery) by installing the right links. Previous posts indicate that on some chips (not this one) an unused VSTBY should be (or can be) connected to ground, but you'll have to perform "archaeology" on all of the Reference Manuals to try and work out yourself if this is a "feature" of the part or another bug/omission in the documentation.

The Data Sheet and previous posts notes the VSTBY current changes when Vdd is on or off, and that indicates there is a "power switch" in the part. The Reference Manual also implies a switch:

7.3.2 RTC Mode

In normal operation, this oscillator is powered by the VDDPLL and VSSPLL pins.

When the part is shut down, this oscillator is powered by the VSTBY pin.

When it doubt, always copy the Reference Hardware Design. If they have jumpers on VSTBY to somewhere to install when there's no battery, then that's probably the right way to do it.

Tom

[davebryant]

Tom,

Thanks for the response.  Yes, I did go through alot of the other posts, but most of them dealt with how that pin works with the RTC or with a backup battery, niether of which I'm using.  Even the eval board design won't help out much, since it's configured to use a battery when in standby, but we don't have a battery to use (we were thinking of just using a big cap fed by Vcc through a diode back when we were planning on using the RTC, just to have a few seconds to log what happends on a power cycle).

Yeah, I saw the one about tying it to ground, but I'm leery of that, just in case they changed things on the part I'm using.  Tying a power pin to ground just doesn't sound right, even if it's supposed to be strictly an input.  I'm leaning toward just putting the diode in between Vcc and VSTBY and leaving it at that.  Should give the required voltage and will keep me from wishing I had done the sensible thing and tie it (more or less) high.

Thanks again,

Dave