[time-nuts] Z380XA The saga of the aging 10811

[Bob Camp]

Hi

The “stuff’ that should contribute to aging on an OCXO is all packed away with
the crystal inside the oven. The frequency is sensitive to it so you heat it for 
stability. 

The temperature of the oven that the crystal is in, is a function of the crystal it’s self. 
It does not matter if it’s a double oven, single oven, or triple oven. What ever the final 
oven in the chain, it will be at the crystal’s magically determined sweet spot temperature. 

So the simple answer is - no, a double oven (or boosted single oven) will not age 
worse than a single oven. The only case that’s not true is when either of the ovens 
fails and runs away. That result is not generally called aging, it’s usually called melt down …

Bob


> On Sep 5, 2015, at 2:19 PM, Bob Benward <rbenward at verizon.net> wrote:
> 
> Hi All,
> 
> I would like to convey the saga of my Z3801A and my Z3805A.
> 
> 
> 
> I purchased a Z3801A a few years ago at the Dayton Hamfest.  I got it
> running and a few months later the unit lost it's GPS lock.  I did some
> trouble shooting and finally realized the unit ran out of EFC range.  I
> opened the double oven 10811 and retuned the unit to the lower end.  The
> unit ran for another year or so, and then the oscillator died.  The
> frequency was close, but the amplitude dropped too low for the circuitry to
> lock on to.  I put the Z3801A aside and purchased a Z3805A from China (I
> think from Yixunhk).
> 
> 
> 
> I plugged in the Z3805A and it ran for a few years.  One day, a few days
> after a relatively active thunderstorm, I notice the alert light was on and
> I had lost GPS lock.  I thought a close strike wiped out the GPS receiver.
> After a lot of screwing around, I figured out that the 10811 had also lost
> EFC range.  The output voltage sat at the neg2.0V level and would not budge.
> I opened up the 10811 and readjusted the tuning to put the EFC counts at
> below 100K.  I never reassembled the oscillator, I just left it sprawled out
> on the table.  Checking the EFC stats, it started with a relatively high
> 1E-8 per day of drift.  It has since calmed down to 5E-9, and you can see a
> slight flattening of the curve in the plot.  I am also running about a
> positive 5K counts/day.  The 5E-9 figure is still 10X higher than the basic
> 10811 spec, and with 5K counts/day, I have only 160 days left ( I am at
> about 200K counts now, max=1000K).  It always seems the drift is up in EFC
> counts, negative going voltage, and down in frequency.
> 
> 
> 
> An interesting observation, I would notice a periodic dip in the time
> interval error.  I would come home from work, turn on the air conditioning,
> and I after an hour or so, I would see the time interval error take a large
> dip.  I didn't think the drift vs temperature was that sensitive.
> 
> 
> 
> I also wonder if the excessive heat of the double oven shortens the life or
> accelerates the aging of a typical 10811.  The heat certainly dries out the
> polyester tape and silicon rubber foam inside. Could it accelerate the aging
> of the oscillator, shorting the life of the oscillator and the useful life
> of the GPSDO?  It's no wonder the Morions on Ebay are sold by the
> manufacturing date.  Once you run out of EFC, that's all she wrote..  It
> makes me question the wisdom of buying a used 10811 double oven oscillator,
> how much life is left?
> 
> 
> 
> So, does anyone have any comments on the excessive drift of the 10811?
> 
> 
> 
> On another tangent, I tried hooking up a Morion to the Z3805A.  I used an
> inverting amplifier for the EFC since the coefficient is positive and on the
> 10811 is negative.  I never achieved good control before the Morion itself
> died.  I will probably continue working on marrying a new OCXO to the Z380XA
> control board.  I need an alternate source for the OCXO without depending on
> used double oven 10811s.
> 
> 
> 
> 
> 
> Bob
> 
> AC2AZ
> 
> 
> 
> 
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.