[time-nuts] Double vs single oven TBOLT

WarrenS warrensjmail-one at yahoo.com
Tue Feb 23 22:17:52 UTC 2010


Murry
First off, Thanks, and I do agree completely with your basic point,
which I think is; this may not be what other GOOD single ovens do or can be 
made to do.

But
Murry said
>I suggest your test of double vs. single oven OCXO may be unfair to the 
>single oven device.
IT is NOT unfair to the THIS OSC and may show what could happen to others if 
the second oven is not used as some have purposed in the past.


>Your inner-oven only device isn't necessarily a good representative of a 
>single-oven OCXO.
TOO true, everyone is different, Then again this one as a single oven is 
nearly 5 timers BETTER than the single Oven It replaced.

>Try the test again with a single-oven variant 10811A,
I did, and this is one is better as a single oven. Which only means this 
unit operating as a single oven is better than the other single oven 10811 
Unit.

>or with a modern high performance SOCXO  such as RAKON STP2602LF.
I'd be glad to if you send me one.

I did try to point out,  it is not a fair direct comparison, cause I did not 
try and retune everything to make it the best it could be as a single oven 
GPSDO.
BUT IT IS A FAIL TEST FOR THIS OSC.
What would be unfair is to make too broad of comments about how others may 
perform. That is why we test.

ws

*********************

[time-nuts] Double vs single oven TBOLT
Murray Greenman Murray.Greenman at rakon.com

Warren,
I take the point that you are testing the SAME oscillator as a means of
adding some consistency to the test. However, leaving aside this
consistency and issues relating to the disciplining device (your TBOLT),


A well designed high performance single-oven OCXO will include such
techniques as ambient compensation and (for example the HP E1938A)
incredibly complex high-gain temperature control. As a result,
performance can challenge the traditional DOCXO, including temperature
slope performance. Your inner-oven only device isn't necessarily a good
representative of a single-oven OCXO.

While we might expect inner-oven-only to operate like a single-oven
10811A, it may be that the DOCXO inner oven doesn't have the power or
thermal gain to handle more than the slight temperature variations (and
temp slope) it would see within the outer oven. Try the test again with
a single-oven variant 10811A, or with a modern high performance SOCXO
such as RAKON STP2602LF.

Regards,
Murray ZL1BPU


******************************
From: WarrenS

It is said a picture is worth a thousand words, A good graph even more.
Attached is a Lady Heather, 2 day long plot showing the difference in 
performance
of a TBolt GPSDO disciplining a Double Oven and a Single oven HP10811 Osc.
The Single oven Osc is the same exact Osc and setup with its outer oven 
turned off.

The single oven mode shows a Freq TC drift of about 6e-11 / deg C
As can be seen in the plot, the TBolt controller has to work much harder 
chasing the single oven's drift.
Still the effect of the LH reported Osc Noise is minimum  (4 vs. 5 ppt)
Where as the difference in phase noise is about 4 times worse using the 
singe oven  (1.75 vs. 7 ns)
Both configurations have the same Tbolt settings which where optimized for 
the double oven osc.
A lower time constant setting on the Tbolt would of helped the single oven 
Phase noise at and expense of the double oven Osc noise.

ws

******* 




More information about the time-nuts mailing list