[time-nuts] temperature control
WarrenS
warrensjmail-one at yahoo.com
Sun Sep 18 20:43:33 UTC 2011
Bert
I for one, like SLOW fans (as long as they are set up so WHEN the fan fails,
nothing really Bad happens)
LH's temperature controller using a small fan blowing on the Tbolt is a
great way to improve its performance.
I have found that for general cooling, 5 Volts on a small 12 V fan that is
made for cooling PC chips is enough air to do a good job of cooling when the
air is blowing at a heat sink.
For the best long term and medium term performance in a GPS Rb OCXO
combination.
I use a LPRO in which I've added a 12k, 1% resistor, connected to it's near
zero volt C field winding.
This is then used as an independent low sensitivity third EFC input by
connecting it to the DAC out of an TBOLT which has been modified to use an
external OSC. An "extended time constant" of 10K sec is near ideal, IF the
Tbolt has a real good antenna system. Still works fine (as long as the
TBolt NEVER goes into holdover) when using a crappy antenna indoors by
setting the TBolt's extended time constant much longer.
I have found that blowing air at the LPRO to keep its Heatsink at a constant
temperature gives a few undesirable and unpredictable results, so I removed
the fan and it's controller, turned the LPRO upside down so the heat sink
was pointing up, insulated what was the units top, which is now at the
bottom, and added a small temperature sensitive adjustable correction
voltage to its external EFC input.
In my room, this keeps the freq real constant to well under 1e-12 throughout
the day and night, even without the Tbolt's help.
ws
****************
EWKehren at aol.com EWKehren at aol.com
WS
I did follow your suggestion and installed the original ATT heat sink and
got a base plate temperature of 47 C. Prior to that I did lower the fan
speed by reducing the voltage to 6 V, that increased the base plate to 43 C
but all vibration and noise disappeared. The only way to find out what if
any
performance degradation occurs is with a D/M measuring A/V. Once I have
software for my D/M I will conduct tests with different Rb/OCXO
combinations.
My goal remains the same, the best in long term and short term performance
in a GPS Rb OCXO combination.
This test was more of what can be done and the amazing part to me was how
little moving air will make a difference. I will stay with my 5X5X1 cm
vibration mounted fan that sits to the side of the Rb moving air across an
array
of small heat sinks with circular fins unless A/V tests tells me I have a
problem.
Bert Kehren
**********************
warrensjmail-one at yahoo.com writes:
Advantages and Trade-offs both ways
For a simple "no moving parts" version,
Oversize the heat sink and attach a 10 Watt resistor to it whose voltage
is controlled to keep the Heat sink temperature constant
ws
****************
More information about the Time-nuts_lists.febo.com
mailing list