Which one can you recommend?
Just stick a Zener in series with the existing fan. I think I used a 5.6V 1W part.
Look at my 34410A thread, I posted a picture of the new fan here:
https://www.eevblog.com/forum/repair/agilent-34410a-repair-attempt-of-2-dead-meters/msg820491/#msg820491
Interesting, your old fan is a Matsushita, while mine is a Delta Electronics Inc. AFB03512MA.
Your new one (Multicomp) is a special low noise type?
In addition to a new fan, I have a 100 Ohm resistor in series with the fan, in all my 34410A and 34411A DMMs.
Just stick a Zener in series with the existing fan. I think I used a 5.6V 1W part.
I don't want to change air flow too much...
Interesting, your old fan is a Matsushita, while mine is a Delta Electronics Inc. AFB03512MA.
Your new one (Multicomp) is a special low noise type?
The multicomp was praised as a low noise fan but since it is so small and mounted directly flat at the frame, it will also make noise.
One way of reducing the noise would be to cut out all the metal from the frame that covers the fan flow area and add a mesh or wireframe on the outside. A few mm distance to the fan make a huge difference.
But I did not want to cut out the metal from the frame and choose to add a 100 Ohm resistor (1 W) and that slowed down the fan enough to take almost all nasty noise away.
The airflow is still there and as far as I have measured, the slower fan has no influence on the measurements. I think the fan was only added for hot work places.
OK, I will try it that way...
Where did you buy the Multicomp fan?
I would not expect miracles from a low noise fan. A large part of the noise is the result of turbulence, and without changing the amount of air flow there is little a fan can do about it. Cutting out the stamped metal grill and replacing it with a finger guard can reduce turbulence. Some fans can have noise due to bearings, but this is mostly audible at lower fan speeds. Especially if the fan is worn and the bearings are not well lubricated. A new fan can reduce bearing noise.
As for lowering the fan speed, keep in mind that changing the internal temperature (and temperature differentials) can affect calibration accuracy, even though the components may still be well within their thermal specifications.
As for lowering the fan speed, keep in mind that changing the internal temperature (and temperature differentials) can affect calibration accuracy, even though the components may still be well within their thermal specifications.
The 34410A specifies full rated accuracy in
ambient temperatures up to 55C. If you know you're not going to be using it within 25C of that temperature, there is no harm at all in slowing down the fan, as Agilent should have done. A variable-speed drive would have cost next to nothing and made the DMM
more stable over temperature, not less.
The 34410A is one of the examples I point to when I rant about how important it is for T&M engineers to eat their own dog food. Agilent's developers clearly never used the 34410A in-house. If they had, then (a) the fan noise would be tolerable; (b) the numeric display format would make it easier to distinguish between 120 mV and 120V; and (c) it wouldn't occasionally power up in the 1000V DC range for no reason.
Fortunately these gripes were all addressed in the 34461A.
The 34410A specifies full rated accuracy in ambient temperatures up to 55C.
Do you have a citation for that? The
User's manual I found (specs start on page 125) states that the 24 hour accuracy spec are for T
CAL ± 1°C, and the 90 days and 1 year specs for T
CAL ± 5°C. If the device is calibrated at 25°C, then running it at 55°C would degrade the 1 year accuracy by 125 ppm in its basic 10 VDC range. That is not exactly within the 30 ppm basic specification. In addition, lowering the fan speed is necessarily the same as increasing the ambient temperature (though this probably does not matter if the differences are small).
Of course performing a new calibration after changing the fan will fix most of that.
The 34410A specifies full rated accuracy in ambient temperatures up to 55C.
Do you have a citation for that? The User's manual I found (specs start on page 125) states that the 24 hour accuracy spec are for TCAL ± 1°C, and the 90 days and 1 year specs for TCAL ± 5°C. If the device is calibrated at 25°C, then running it at 55°C would degrade the 1 year accuracy by 125 ppm in its basic 10 VDC range. That is not exactly within the 30 ppm basic specification. In addition, lowering the fan speed is necessarily the same as increasing the ambient temperature (though this probably does not matter if the differences are small).
Of course performing a new calibration after changing the fan will fix most of that.
I was looking at page 136 of 34410-90001 ("Full accuracy for 0C to 55C"). A bit hand-wavy, but then this isn't a 3458A...
Using the 34410A at 55°C, I would not slow down the fan !
But, whoever will use such a meter at this temperature, will probably not worry about fan noise.
But in normal laboratory use (22°C to 26°C ) slowing down the fan will not make any difference.
I have tested this on my stable 10V reference on three 34410A and one 34411A.