[time-nuts] Making a HP 10811 better

[WarrenS]

Bob wrote: about some possible shortcoming of the 10811 layout and design.

Good points, If those high impedance points are directly on the PCB and 
unguarded,
I would certainly consider that a major design flaw, but easily fixed with a 
little rework.
The integrator is there, and if the cap is still OK, its not going to have 
enough leakeage to cause a negitive effect. (they were not not dumb)
If they screwed it up in the layout, Have to wonder what other basic things 
can also be easily fixed.
There are many ways to take care of high impedance points on a circuit board 
to keep them high enough to make that a perfect integrator.
If they did not use some of those , then All I can say is shame on the PC 
board designer, he should go back to basic layout 101.

Then again I'm betting that at the temperature it is running at, not a whole 
lot of moisture in there.
Dirt and flux yes,  that could cause major performance concerns, need to 
keep it clean, and sounds like it needs a little fixing up.
I'll add it to my list of things to look at and do on the inside the Osc if 
it is opened.

BTW
If leakage is a problem like you seem to think it can be, THEN the reduced 
gain of the integrator is the least of the issues.
Leakage is going to change the temperature setpoint all over the place long 
before the reduced performance of the integrator has any negative effects.

 ws

*******************
Bob wrote:
Hi

Here's the gotcha with the "integrator". The poor thing starts out with a 
closed loop gain of a bit over 4. That's *very* low by oven controller 
standards. The gain gets up to 40 or so by the time the gizmo labeled 2 uf 
gets up to 40 meg ohms. It's going to have a hard time going 10X above that. 
Getting up to 400 Meg on the pc board is going to be a challenge. You have 
to guess what the cap is made out of, so coming up with an exact mega ohm 
microfarad product for it is a challenge. It's likely that it's going ot cut 
in below 1 G ohm. That's when everything is new, clean and dry.

Then if you just happen to have dust / dirt / humidity / spider webs ... 
there goes your 400 Meg. Humidity in particular is nasty. It goes in easily 
and it's very hard to drive out.

The DC gain is unlikely to make it past a few hundred under the best of 
conditions. You are getting maybe a 10X boost over a normal controller when 
it all works right. When the insulation resistance starts to go down, you 
pay for it with shifts.

Bob


On Mar 30, 2010, at 5:26 PM, WarrenS wrote:

>
> Bob said: a lot of good points
> except for a couple of little things?
>
>> "With the large increase in gain, since it's a simple controller,
>> you will need to re-set the control point.
>> There's no handy integrator in there to crank the offset out for you."
>
> Does not seem to be true, at least on the schematics that I've seen for 
> the 10811.
> There is a low leakage cap in series with the Feedback gain resistor which 
> does provide the near infinite gain integrator function.
> What I see is a Full PI controller circuit, with integrator.
> It would also be easy enough to retune the "PID" values if one did make 
> something that caused it to become unstable.
> So Yes, certainly something to be aware of but No, does NOT need to be a 
> big problem.
>
> ws
>
> **************
> ----- Original Message ----- From: "Bob Camp" <lists at rtty.us>
> To: "'Discussion of precise time and frequency measurement'" <time-nuts at 
> febo.com>
> Sent: Tuesday, March 30, 2010 9:18 AM
> Subject: Re: [time-nuts] Making a HP 10811 better
>
>
> Hi
>
> There are a couple of problems you will run into.
>
> The first is that if you do a very good job with the dewar flask, the 
> "dead
> heat" (power of the oscillator, regulator etc) can indeed raise the
> temperature of the OCXO beyond the oven control temperature.
>
> The second is that the gain of the control loop is in part determined by 
> the
> thermal resistance. Increase the thermal resistance by a very large amount
> and the control loop gain goes way up. The control loop gain is probably
> pretty high already, so increasing it by a large amount is likely to make 
> it
> unstable.
>
> The next thing is that control loop gain and effective thermal gain to the
> crystal are not the same thing. Even if it is stable, a large increase in
> control loop gain probably will decrease the effective thermal gain. The
> reason is a bit complex. The simple answer is that the thermistor is not
> mounted on the crystal blank, thus it "sees" something different than the
> crystal.
>
> With the large increase in gain, since it's a simple controller, you will
> need to re-set the control point. There's no handy integrator in there to
> crank the offset out for you.
>
> Simply put, you design a dewar flask / vacuum bottle OCXO in a very
> different way than one that's conventionally heated. Miliwatts of
> dissipation do mater in a vacuum design.
>
> Reducing drafts is a good thing. Moderating ambient variation hour to hour
> is a good thing. Burying an OCXO in the back yard works, dunking it in a 
> big
> barrel of water also works. Both require you to remember the waterproof 
> bag
> before you take the final step :)....
>
> Bob
>
> ****************
> -----Original Message-----
> From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] 
> On
> Behalf Of Steve Rooke
> Sent: Tuesday, March 30, 2010 8:12 AM
> To: Discussion of precise time and frequency measurement
> Subject: Re: [time-nuts] Making a HP 10811 better
>
> Has anyone tried putting one of these in a thermos flask? I have seen
> wide necked flasks available for food storage which should allow the
> device to be inserted and it could be wedged in place with something
> like foam blocks inside. I appreciate that this does not let the item
> loose heat but it should not make itself too hot in the first place.
> At least this should protect it from external temperature swings.
>
> Yes, I do know that they actually make OCXOs like this and that's
> where I got the idea.
>
> Cheers,
> Steve (slowly recovering)
>
> *******************
> On 26 March 2010 08:10, WarrenS <warrensjmail-one at yahoo.com> wrote:
>>
>> I was wondering if there is any interest for a single oven HP 10811
>> improvement design kit of some sort?
>>
>> Background:
>> I am now using a double oven 10811 as the Freq reference for my own
>> projects.
>> Some recent off line conversations made me realize that there are many
>> advantages to using a single oven instead of a dual oven 10811 osc.
>> Among them are:
>> 1) probable does not need separate floating PS to work good, due to it
>> having better grounds
>> 2) has accessible freq offset adjustment, The EFC input can be set to 
>> near
>> zero for 10.000 MHz output
>> 3) Does not need an outer oven driver
>> 4) It can be opened up for simple internal mods
>> 5) And most important, they are much more common for the average time 
>> nut.
>>
>> The only disadvantages I know of is that they change freq more with room
>> temp changes. Up to 1e-10 is what I've seen.
>> My idea of a reference is something that one does not have to worry about
>> things like it's PS and room temp sensitivities.
>> I think where possible things like temp, PS. Load, Freq stability, etc
>> should be made to have effects on its freq that are below the noise 
>> level,
>> which is around 1e-12.
>>
>> Here is my thought;
>>
>> I know how to make major improvements to the 10811 so that the standard
>> stuff does not have ANY measurable effect on its freq.
>> Nothing really magic, mostly simple things like secondary PS regulators, 
>> an
>> outer oven heater wrap and controller, some internal span and reference
>> voltage adjustments,
>> tilt it on its Zero G axes, add an RF buffer, isolate or have a less
>> sensitive EFC input, a fine freq adjustment pot, and probable others once 
>> I
>> get into it more.
>> The goal being if I can measure something that causes it to change freq 
>> at
>> the a-12 range, fix it.
>>
>> I am only taking about basic superficial things, I'm not knowledgeable
>> enough to fiddle with its RF stuff which seems OK as is.
>> But I do know from my DVM design experience, the way they offset the EFC
>> has to have some major contributing factor to its 1/F freq noise, at 
>> least
>> on some units.
>> It has long been my suspicion, that part of their noise grading process 
>> was
>> actually selecting units that happen to have a low noise internal 6.3
>> reference, The one used to offset the EFC.
>>
>> AND YES I do know none of this takes care of aging. The best solution for
>> that is to select a good one and Let in run continuously.
>> In my case aging does not matter much because I have it disciplined to 
>> the
>> GPS thru a very slow loop.
>>
>> Any additions things to consider and comments welcome on or off line.
>> By the way, I already have a long list of thing one could do to it to 
>> make
>> it worse, so probable don't need any more of those ideas.
>>
>> ws
>>
>> ****************