[volt-nuts] LTZ1000 at higher currents
john at devereux.me.uk
Fri Oct 21 12:07:10 UTC 2011
Thank you for your thoughts!
> if you increase the zener current you have to increase temperature
> setpoint of the heater. Otherwise the temperature regulation will be
> bad. You need to have a temperature stability of 0.001 K on the zener
> to get the 0.05ppm/K temperature gradient.
I think you only have to increase the setpoint if there is a risk of
exceeding it during operation. So I could run at 20mA with a 10C extra
temperature rise. So as long as the set point is >>10C above my board
temperature this should be OK.
> The "Pickering patent" changes the temperature setpoint of the heater
> not the zener current.
The patent I meant does change the current (as I understand it). The
point of it is to allow (intermittent) high-current operation while
keeping the temperature set point low. It is US 6342780.
I suppose it does allow control of the temperature too but I thought the
benefit is the ability to run the diode at higher current than would
otherwise be possible.
By the way, I am not sure about a statement in the patent that "the
typical thermal time constant is many tens of seconds". It seems to me
this may be true of the entire device but the chip itself must have much
faster thermal time constants in the order of milliseconds. So with the
method described, the temperature will be changing even as the voltage
is being sampled.
> Although there is no direct prohibition, all diagrams are ending at
> 5mA. And if there would be a significant improvement at higher
> currents then HP would have used this in his 3458A. The only thing
> they did is changeing the 120 Ohms to 111 Ohms. On my reference I have
> about 3.8 mA with the 120 Ohms.
Yes I suppose it would be on the datasheet too. Perhaps it improves the
noise but worsens the drift. But I gather (from this list) that Fluke
used the "Pickering Patent" in their recent references.
"Andreas Jahn" <Andreas_-_Jahn at t-online.de> writes:
> Just some thoughts:
> With best regards
> ----- Original Message -----
> From: "John Devereux" <john at devereux.me.uk>
> To: <volt-nuts at febo.com>
> Sent: Wednesday, October 19, 2011 3:14 PM
> Subject: [volt-nuts] LTZ1000 at higher currents
>> Hello everyone,
>> Is there some reason why LTZ1000s are not run at a lot higher current
>> than 5mA?
>> Thermal resistance is 80'C/W, or ~0.5'C/mA. So it would seem OK to run
>> it at 10, 20, 30mA before seeing serious problems due to raising the
>> temperature too high.
>> The datasheet noise graph shows a big fall in noise level between 1 and
>> 5mA, so this ought to reduce the noise further. I am not sure how far
>> down in frequency this improvement continues. On the datasheet the
>> "noise gap" widens more and more with lower frequency, does this extend
>> to frequencies so low they would be regarded as long term drift?
>> There is no current limit for the zener that I can see (of course there
>> will be one eventually).
>> The "Pickering patent" uses pulsing to achieve a high current in what is
>> obviously a LTZ1000, the stated reason being to minimise "VLF and long
>> term instability".
>> But it seems like even a higher DC current could be of benefit.
>> John Devereux
>> volt-nuts mailing list -- volt-nuts at febo.com
>> To unsubscribe, go to
>> and follow the instructions there.
> volt-nuts mailing list -- volt-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
> and follow the instructions there.
More information about the volt-nuts