[time-nuts] HP 10811A failure

GandalfG8 at aol.com GandalfG8 at aol.com
Sat Oct 13 20:58:17 EDT 2012


Hi Adrian
 
Glad to hear you're back on track and that all is now well
 
I can understand that very thin coax being quite  deceptive, it's certainly 
amongst the thinnest I've ever come across.
 
I've got a couple of double oven10811s bought from one of the usual Chinese 
 Ebay sellers a few years ago as potential spares for my Z3801As and both 
have  one of the coax connectors chopped off, can't remember now whether it's 
the  10MHz output or the EFC input.
Either way, should they ever be needed in anger I'm keeping my fingers  
crossed that the leads can be swapped over, cos fitting another connector sure  
don't look to be a very user friendly option:-)
 
Regards
 
Nigel
GM8PZR
 
 
In a message dated 14/10/2012 00:27:23 GMT Daylight Time, rfnuts at arcor.de  
writes:

A very  good point!

I checked the output amplitude with a spectrum analyzer, a  power meter 
and a scope, the latter with a 50 ohm load to the input.
I  re-checked the power supply connections and can confirm they are the 
same  as before.
I used three different power supplies for the oscillators and  two for 
the heaters, still no difference.
But...

To make it  short, your comments helped to get me back on track. Both 
beauties escaped  unnecessary dismantling and are now working as they 
should.

Actually,  the problem was caused by a long-term misunderstanding. I was 
always  wondering why 'HP used simple stranded wire' and not coaxial 
cable on the  10 MHz output and EFC.
As long as I connected them on the bench, I replaced  the 'missing' 
ground connection with a short wired croc clamp between coax  and ground. 
Today I realized that the thin blue wire IS actually coax  cable, so I 
should have connected the coax shield to the BNC ground on my  newly 
built 2x 10811 enclosure. With the new wiring, the output ground had  
just become much more inductive, up to a point where the nominal source  
impedance of 50 ohms had increased to over 120 ohms, causing the  
amplitude loss of some 3 dB and capturing noise.

Btw. the Sprague  6800 uF / 40 V from my R&S NGA power supply has indeed 
died, but  independently of the osc. problem.

Adrian


GandalfG8 at aol.com  schrieb:
> Perhaps a silly question, but as I get the impression that  both seem  to
> have failed simultaneously with the same fault I'm  just wondering  if 
you're
> sure it isn't what you're checking them  with that's  developed a problem?
>   
>  Regards
>   
> Nigel
> GM8PZR
>    
>   
> In a message dated 13/10/2012 13:52:46 GMT  Daylight Time, rfnuts at arcor.de
> writes:
>
> Hi   All,
>
> both of my double oven 10811A's have been running  flawlessly until  now
> when I noticed that the output power has  dropped by about 3 dB  (measures
> only 4...4.5 dBm at 50 Ohm  load), and the noise floor has gone  up by
> about 20 dB, while the  frequency is still spot on.
>
> I used  them for various  mesurements, so I can unfortunately not remember
> what  might have  caused the failure. Only the 12V linear lab power supply
> that   fed the heaters had developed a completely dead filter cap, so the
>  feeding  voltage was a 100 Hz sawtooth rather than DC. The heaters  
appear
> to be  still working though. I'm only using the internal  heater. They
> draw some  700 mA at power up and drop to around 150  mA  each when warm.
> So the  oscillators are unlikely to be  cooked inside. Something must have
> happened  to the 10 MHz  output, possibly an ESD issue?
>
> Before I start taking  them  apart, can anyone tell what has happened or
> what to look  for  first?
>
> Regards,
>  Adrian
>
>
>
>  _______________________________________________
> 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.
>
>  _______________________________________________
> 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.
>



_______________________________________________
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.




More information about the time-nuts mailing list