No warmup needed if the counter switches to the external reference. If
the counter uses the external reference to lock the internal one then
it is better to wait until the internal one is stable, maybe the
counter has a standby mode where it appears powered off but the
internal reference is active (usually an ovenized reference). For the
best performance (stability of trigger levels, input amplifiers and so
on) it is better to warmup the whole counter always.

On Thu, Feb 8, 2018 at 8:33 AM, Chris Wilson chris@chriswilson.tv wrote:

Hi,

On 02/08/2018 08:33 AM, Chris Wilson wrote:

Well, while the internal reference is not directly steering, if it is
being locked to the external reference, at least some warming up is
needed before it is very stable.

Also, depending on the details of the counters, high resolution
interpolators may need warmup to perform well and for the autocal to
perform meaningfull values.

However, usually it's not very long times and often shaddowed by the
internal referece anyway.

Just to get a rough reading, no. If you want to trust it and trust the
noise, let it warm up for half an hour or so.

Cheers,
Magnus

Hi

It depends a lot on the frequency counter. An old style “just count the number of
edges” device should be good to go pretty fast. One of the “fry an egg on it” interpolating
counters that get into the 20 ps range may well need some time to stabilize. If you
are doing ADEV runs, a couple hours of warmup would be a good idea.

Bob

Yes,  You have removed the reference from the warmup but the input circuits still need time to reach thermal equilibrium for most accurate results

On Feb 8, 2018, at 3:34 AM, Azelio Boriani azelio.boriani@gmail.com wrote:

No warmup needed if the counter switches to the external reference. If
the counter uses the external reference to lock the internal one then
it is better to wait until the internal one is stable, maybe the
counter has a standby mode where it appears powered off but the
internal reference is active (usually an ovenized reference). For the
best performance (stability of trigger levels, input amplifiers and so
on) it is better to warmup the whole counter always.

time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Hi,

Well, actually there is two methods:

1. Switch reference clock - a switch or mux changes which clock is being
   used.

2. Lock internal reference to external reference - a PLL lock of the
   internal reference to the external reference is enabled.

For both, the counter core may need to heat up.

For the first, the internal reference heatup can be ignored completely.

For the second, the internal reference heatup cannot completely be
ignored, it takes time for it to be within lock range and the lock-in
behavior stabilizes. Still relatively quick, but an actual effect.

At the end of the day, it depends on how precise you attempt to measure.

Cheers,
Magnus

On 02/08/2018 03:16 PM, Scott McGrath wrote:

Exactly what I have run into.
Some systems lock an internal reference to to the incoming system. So that
oven has to come up to temp. Then the trigger and interpolators need to
warm up and stabilize.

So the really right answer is what accuracy are you looking for? Because
what I mention above may simply not matter.
Regards
Paul
WB8TSL

On Thu, Feb 8, 2018 at 9:16 AM, Scott McGrath scmcgrath@gmail.com wrote: