Monday, February 11, 2013

Optimizing TEC

I've been using a Tellurex C2-30-1505 device (Qc 46 Watt at 15V) in the new configuration.  After spending some time at this page I found that for the load of 5-10 Watts and the voltages I am using, a better match would be a device with a Qc of 20 Watts.  Luckily I had one in the Tellurex C2-25-1502.

What did this mean for the configuration?

  1. Had to trim the size of the copper finger to better fit the smaller TEC.  I didn't make the fit exactly 25mm, as I wanted to have the option to go to a bigger device in the summer.
  2. I also got better thermal compound at Fry's.  I got some paste with pretty decent numbers and used that in the place of the cheap stuff I had.
  3. I also replaced the support pad that was behind the chip.  I had been using more thermal gel to support the finger below the device.  Instead, I put a thick pad of paper below it.  In the future, I'll probably get more plastic from TAP.  I wanted something easy to construct for now.  I'm thinking that the paper may help hold any condensation and prevent heat leakage.  The pad can be seen in the picture above before I trimmed it smaller.
  4. I also redid the weather stripping around the heat sink.  This was too firm at first and then once I trimmed it, the device did a better job of making connection.
  5. I also put some insulation on the bottom of the camera to prevent leakage there.
  6. I also put a tiny pad of thermal gel on the sensor to improve the reading.
What did all this do?

I was able to see that the device did a better faster job of cooling with much less power waste.  Instead of a max draw of 3 or more amps, the smaller device pulls 2 amps.  This will mean less wasted power in the field.

I'll see how fast it pulls the camera down at night.

Still to be done:

  1. Replace thermal compound on the finger inside the camera
  2. Put thermal gel pad inside the camera closer to the CCD.
So far what I've seen is that the darks from the last two nights are giving weird results in calibration in Maxim.  Nothing I can't fix in PI with DBE.  Just strange that there's a halo.

I'll do another few more nights of testing to see what I can get.

With the system running in the back yard, I've been rather pleased with the results so far.  It's been actually getting results below 0C (although the camera EXIF temp records 5C).  When it nears the set temp, the power requirements drop from 2.08 A to 1.6 A or lower.  Not bad and it's much better than what I was getting with the other setup.

This should allow me to run the dew heaters a little harder to help fight the moisture problem here in at home.

I've run the numbers on the flexure from last night and there's not much to say other than that it's still there.  I was hoping that some firm tightening of the rings would make a difference.  It appears that this did not make any change.

What this means is that there is still a sag in the rings someplace.  Given the way the stars appear to run "fast" when the scope is tracking from East to West it leads me back to the model I was using before.  I'm not sure what the problem may be.  If I think about the way the guidescope might sag relative to the main scope, then the stars should move in the opposite direction than they are.  This makes me consider the sagging of the main scope since the rings were suspect before.  However, the way I consider this problem is that the main scope should sag more at lower altitudes rather than at zenith.