[time-nuts] 10811 changing frequency when load impedance changes ?

Bruce Griffiths bruce.griffiths at xtra.co.nz
Sat Dec 25 08:16:26 UTC 2010


Mark Spencer wrote:
> Thanks.    I'll have to do some more experimenting, in one case I was feeding
> the time base output of the 5328A into a 1 meg scope input, and using the scope
> to compare the frequency to my thunderbolt, in the other case I was feeding the
> time base output into my 5370B set to 50 ohms and using the time interval
> function of the counter to compare the frquency to the thunderbolt.
>
> I couldn't figure out why I was seeing different frequency results on the scope
> and the 5370B untill i put a 50 ohm terminator in paralel with the scope input
> then things made sense.   
>
>
> I'll try altering the load impedance along the lines shown in the specs and see
> what the results are.    I also noticed that the wave form on the scope was very
> dirty (ie the peaks were very hard to see), and it cleaned up to a more or less
> sine wave when I added the 50 ohm terminator, perhaps the 5328A is not properly
> terminating the 10811.   
>
>
> The 10811 seems stable so long as the load impedance does not change so it may
> still be useable.
>
> I'm also wondering if I was getting some injection locking when I was comparing
> the frequency using the scope, the results seemed very very good, almost to good
> and the cable between the scope and the 5328A was not seeing 50 ohms at either
> end which probably wasn't helping either.   My elcheapo pre made RG58 patch
> cords of un known shielding probably aren't helping either.  I'll have to make
> up some proper patch cords using some double shieled cable I've stashed away.
>
>
>   
>
>
> ----- Original Message ----
> From: Bruce Griffiths<bruce.griffiths at xtra.co.nz>
> To: Discussion of precise time and frequency measurement<time-nuts at febo.com>
> Sent: Fri, December 24, 2010 10:42:46 PM
> Subject: Re: [time-nuts] 10811 changing frequency when load impedance changes ?
>
> Bruce Griffiths wrote:
>    
>> Mark Spencer wrote:
>>      
>>> Seasons greetings.
>>> I was pleasantly surprised to find that a new to me HP 5328A had an 10811 in
>>> it.    (I was expecting the older OCXO..)    After leaving it running for a
>>> week I adjusted the frequency and while doing so I noticed that the out put
>>> frequency changes by a third of a hz or so if I add or remove a 50 ohm load. 
>>> I'm measuring the frequency at the timebase output connection of the
>>> 5328A.  Is this to be expected ?
>>>
>>> Many thanks
>>> Mark Spencer
>>>
>>>
>>>
>>> _______________________________________________
>>> 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.
>>>
>>>        
>> No, that seems to be a couple of orders of magnitude larger than the specs for
>> the 10811A would lead one to believe.
>> Maybe there's a problem with the output buffer amp in the 10811A?
>>
>> Bruce
>>
>>
>>      
> The spec for the 10811A actually states that the shift is<  5E-10 (5mHz) for a
> 10% (5 ohm) change in a 50 ohm load and a for a 25% change (250 ohm) in a 1K
> load
> A naive extrapolation would suggest a shift of around 20mHz for a 50 ohm to 1K
> load change your frequency shift (333mHz) is about 15x this.
>
> However the 10811A is notorious for injection locking if the reverse isolation
> of the external circuitry is inadequate.
>
> Bruce
>
>
>    
The 5328 terminates the OCXO output in 1200 ohms so that its compatible 
with the predecessor OCXO a 10544 (I believe ) which had an emitter 
follower output and could only drive a 1K load.
There is a multiplexer chip after the 1k2 load in the 5328.

A poor waveform with a 1k2 load implies that either the multiplexer 
input is clamping the signal or maybe the 511 ohm resistor across the 
output transformer in the 1811A has gone open circuit?
Alternatively the power supply for the 10811A output buffer is too low?
The output buffer device could be entering saturation if the 511 ohm 
resistor is open circuit or the output buffer stage collector supply is 
too low.
Repairing the 10811A isn't too difficult unlike the 10544A (although 
even that is possible).

Bruce




More information about the time-nuts mailing list