[volt-nuts] HP-3458A questions
frank.stellmach at freenet.de
Tue Jun 2 17:50:10 EDT 2015
in advance, this change only improves the stability of the instrument,
when the 3458A is powered.
The LTZ1000 will not age (nearly), when the instrument is switched off.
Therefore, it makes sense mainly, if your instrument is running for an
extended period of time.
(Reference: Spreadbury: "Ultra Zener..")
If you change the oven temperature, you will loose the 10V calibration,
as the LTZ1000A will change its reference voltage.
Although my instrument is strictly switched off, when not doing
measurements, I reduced the oven temp. by connecting a 100k precision
wirewound (3ppm/K) in parallel to the 15k BMF resistor. That gives 13k
in total, yielding about 65°C like in the LT datasheet.
An alternative placement is already foreseen on the PCB; in the CLIP,
it's component designator is X411.
I'll try to link in a photo of my change.
lymex on the bbs35.hot forum obviously was more agressive, and paralled
75k of Vitrohm Thin Film, giving 12.5k for about 55°C.
I would recommend my solution above the one of lymex, although his one
gives theoretically more stability.
Also, the lower stability of TF reistors may affect the stability (T, t)
of the 15k BMF.
With 100k/65°C, the oven runs about 30°C lower.
Therefore, the ambient temperature has to be limited to max. 35°C, the
fan has ALWAYS to be kept clean, giving a maximum temperature rise of
15°C, and the instrument may not be put in a rack (which would give
additional 10..20°C temperature rise). Under these conditions, (35°C
ambient + 15°C 3458A inner heating + 10°C self heating of LTZ1000A + 5°C
regulation margin), the oven will always be stable.
These 75k / 55°C would be possible, only if you limit the ambient
temperature further, to <= 25°C, and precisely monitor the 3458A inner
Although my instrument is kept at 20..23°C throughout the year, i
wouldn't feel comfortable with 55°C.
The only possibility would be the exchange of the reference to the non-A
version, as the self heating effect may vanish!
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 76191 bytes
Desc: not available
More information about the volt-nuts