Jump to content


Photo

Instrumenting a Dob with Temperature Probes

  • Please log in to reply
40 replies to this topic

#1 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1910
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 26 June 2013 - 07:50 PM

The "visualizing airflow" thread got me to thinking about collecting data directly from Dobs about how temperatures are distributed internally and externally.

What you would like is a temperature readout mounted on the scope that can switch among at least three probes, and have the whole kit and caboodle neither weigh nor cost very much. One probe attached to the mirror and two for air temperatures at different locations (which could be moved around to investigate air flow) wouls be the minimum set up for collecting data.

I am not an electronics hobbyist and do not have a ready knowledge base of what is available these days for putting something like this together. A little Googling for kits, or ready-built monitors, turned up little or nothing for the first, and systems costing hundreds of dollars for the second.

The absolute temperature precision is not as important as accurately measuring the differentials between the probes. A low incremental per-probe cost would be good.

Any suggestions?

#2 richard7

richard7

    Not Quite

  • *****
  • Moderators
  • Posts: 5914
  • Joined: 02 Nov 2007
  • Loc: Sacramento

Posted 26 June 2013 - 08:13 PM

1. Circuit Specialists.
2. Trailtech.
Or if you're up to some diy. Electronics DIY.

#3 Meep_Esq

Meep_Esq

    Explorer 1

  • -----
  • Posts: 77
  • Joined: 02 Jan 2007
  • Loc: Dunedin, New Zealand

Posted 26 June 2013 - 08:13 PM

y'know I'm sure I read a website where a chappy found a cheap garden thermometer that had an inside and outside temperature probe which he used to compare tube temp with ambient.

#4 highertheflyer

highertheflyer

    Viking 1

  • *****
  • Posts: 712
  • Joined: 08 Jul 2005
  • Loc: Aledo, Texas

Posted 26 June 2013 - 09:29 PM

Each of the home made reflectors here is equipped with this very simple setup and fun to monitor.
It's pretty easy to find small indoor/outdoor digital thermometers for this use.
With one registering in the cooling fan flow, and one attached/buried/surface glued against the back of the mirror.
The mirror monitor probe should be encased in a cotton lined plastic housing to insulate the mirror temp probe from the outside temperature affects.

Jim

#5 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1910
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 27 June 2013 - 01:34 AM

Thanks, this is exactly what I was looking for.

The first site has several interesting solutions.

Simply buying several of these LCD temp units (http://www.circuitsp...om/dtm0503.html), and making a simple switching mechanism to connect to different probes (patch panel maybe?) would be an inexpensive solution for a non-electronically adept individual.

(I build computers from parts, but have never become familiar with building circuits which requires considerably more expertise.)

#6 don clement

don clement

    Vendor (Clement Focuser)

  • -----
  • Vendors
  • Posts: 812
  • Joined: 02 Feb 2011
  • Loc: Running Springs, California

Posted 27 June 2013 - 01:43 AM

I would suggest using an inexpensive temperature sensor such as the LM35 http://www.ti.com/li...mlink/lm35.pdf. Band gap temperature sensors such as the LM35 don't require much signal conditioning and have a linear voltage output proportional to temperature that can be read by a voltmeter. There are probably inexpensive temperature USB modules that would make it easy to collect data with a laptop PC and if one had Labview http://www.ni.com/labview/ would be, for me, really easy and quick to put a program together to collect and analyze the results.

Don Clement

#7 derangedhermit

derangedhermit

    Apollo

  • -----
  • Posts: 1158
  • Joined: 07 Oct 2009
  • Loc: USA

Posted 27 June 2013 - 02:04 AM

Simply buying several of these LCD temp units (http://www.circuitsp...om/dtm0503.html), and making a simple switching mechanism to connect to different probes (patch panel maybe?) would be an inexpensive solution for a non-electronically adept individual.

That looks like one complete unit - temp probe and display. Buy as many as you need - I'm not sure what the patch panel would be for.

It doesn't give accuracy or precision, though, that I could find. It may not do better than 1 degC.

I'd like to do the same thing you're asking about - I want better than 1F accuracy.

#8 don clement

don clement

    Vendor (Clement Focuser)

  • -----
  • Vendors
  • Posts: 812
  • Joined: 02 Feb 2011
  • Loc: Running Springs, California

Posted 27 June 2013 - 03:20 AM

It doesn't give accuracy or precision, though, that I could find. It may not do better than 1 degC.

I'd like to do the same thing you're asking about - I want better than 1F accuracy.


Since we are looking for differential measurements wouldn't it be resolution less than a degree and not accuracy or precision? Also linearity.

Don Clement

#9 freestar8n

freestar8n

    Skylab

  • *****
  • Posts: 4048
  • Joined: 12 Oct 2007

Posted 27 June 2013 - 04:08 AM

Here is my write up of measuring temperature at various points in an SCT exposed to the sky:
SCT Temperature measurements.

I used ds1820's and custom code in a microcontroller. These devices are simple and serially addressable individually - so you can have as many as you want on a single two-wire connection. The temperature precision and accuracy were adequate for this study - and could be improved by calibrating them carefully.

A key issue in studies like these is the actual thermal/radiative environment where the 'scope is used. The cooling behavior in a wide open field with very clear skies would be very different from a tree-shaded area, or in an observatory dome. The radiative losses don't just happen on the front lens, but on the entire system including the mount.

Frank

#10 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1910
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 27 June 2013 - 08:11 AM

Simply buying several of these LCD temp units (http://www.circuitsp...om/dtm0503.html), and making a simple switching mechanism to connect to different probes (patch panel maybe?) would be an inexpensive solution for a non-electronically adept individual.

That looks like one complete unit - temp probe and display. Buy as many as you need - I'm not sure what the patch panel would be for.o do the same thing you're asking about - I want better than 1F accuracy.


To just have one readout mounted on the scope for multiple probes, not a whole collection of them. Yes, it would involve chopping off the probes and ending up with surplus LCD panels, but they are cheap.

Getting two units would allow direct comparison for consistency between the units - much more important than absolute accuracy since it is really the temperature differentials we are interested in.

#11 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1910
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 27 June 2013 - 08:16 AM


It doesn't give accuracy or precision, though, that I could find. It may not do better than 1 degC.

I'd like to do the same thing you're asking about - I want better than 1F accuracy.


Since we are looking for differential measurements wouldn't it be resolution less than a degree and not accuracy or precision? Also linearity.

Don Clement


Precision, but not accuracy (the absolute temperatures are really not of interest).

The LCD units I was looking at actually read-out to 0.1F, but testing would be required to show how reproducible the measurements are between units.

#12 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1910
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 27 June 2013 - 08:42 AM

I would suggest using an inexpensive temperature sensor such as the LM35 http://www.ti.com/li...mlink/lm35.pdf. Band gap temperature sensors such as the LM35 don't require much signal conditioning and have a linear voltage output proportional to temperature that can be read by a voltmeter. There are probably inexpensive temperature USB modules that would make it easy to collect data with a laptop PC and if one had Labview http://www.ni.com/labview/ would be, for me, really easy and quick to put a program together to collect and analyze the results.

Don Clement


Thanks, this the sensor you recommend looks like a very good candidate for the sort of low cost of entry monitoring.

With the 0C + 10 mV per C output I would just need a millivolt meter reading to 400 mV or so (to reach 40 C, as high as I would be likely to ever measure). Unfortunately I note that inexpensive ubiquitous digital LCD panel voltmeters seem to be limited to 200 mV (must be a widely used chip spec).

For a really serious monitoring approach a digital system would clearly be best, but the total "cost of entry" (money plus learning time) is higher.

Would the trial version of Labview work? Otherwise a license is $999, a non-starter unless I was getting into really serious data collection for other projects.

#13 don clement

don clement

    Vendor (Clement Focuser)

  • -----
  • Vendors
  • Posts: 812
  • Joined: 02 Feb 2011
  • Loc: Running Springs, California

Posted 27 June 2013 - 09:16 AM

Carry,

I mentioned the LM 35, http://www.amazon.co...e-Sensor/dp/... an analog temperature sensor that has a linear output of 10mV/*C . The LM35 is pretty easy to set up, even for the electronics challenged. All you need is an unregulated DC voltage source of 4V to 20V and a way to read the analog output voltage. A 9V battery would do fine. An inexpensive digital voltmeter can be used for readout such as http://www.harborfre...eter-98025.html A simple analog switch would be used between sensors. Since you are interested in differences between sensors, the analog switch allows the same DAC (the digital voltmeter) to be used for all sensors. I believe this scheme has the simplicity, precision, and resolution you are looking for.

Don Clement

#14 Pinbout

Pinbout

    Cosmos

  • *****
  • Posts: 8246
  • Joined: 22 Feb 2010
  • Loc: nj

Posted 27 June 2013 - 09:31 AM

With the 0C + 10 mV per C output I would just need a millivolt meter reading to 400 mV or so (to reach 40 C, as high as I would be likely to ever measure). Unfortunately I note that inexpensive ubiquitous digital LCD panel voltmeters seem to be limited to 200 mV (must be a widely used chip spec).

For a really serious monitoring approach a digital system would clearly be best, but the total "cost of entry" (money plus learning time) is higher.



on the spec sheet it states if there is a high difference in air temp and surface temp you have to epoxy the leads to the surface so the device won't read air temp.

also there is a way to turn it to F°. they show a circuit for the change on the spec sheet.

#15 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1910
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 27 June 2013 - 03:21 PM

I think this slightly more expensive multimeter from HF would be needed:
http://www.harborfre...eter-37772.html

It gives (says the manual) 0.1 mV accuracy in the 200 mV range (20C/68F) but is still 1 mV in the 2000 mV range (+/- 0.1C, taking the temp up the maximum of interest).

The 7-function one lists 1% accuracy in the 2000 mV range (i.e. 20 mV or 2C, or maybe 40 mV/4C).

#16 Arjan

Arjan

    Mariner 2

  • -----
  • Posts: 220
  • Joined: 21 Jan 2009
  • Loc: Netherlands

Posted 27 June 2013 - 03:27 PM

also there is a way to turn it to F°. they show a circuit for the change on the spec sheet.


Who would want this? :grin:

#17 Pinbout

Pinbout

    Cosmos

  • *****
  • Posts: 8246
  • Joined: 22 Feb 2010
  • Loc: nj

Posted 27 June 2013 - 03:28 PM

couldn't you just shoot the mirror with IR thermometer that way you can measure the edge, the back middle, front middle... +/-2°, do you really need to read more?

#18 Pinbout

Pinbout

    Cosmos

  • *****
  • Posts: 8246
  • Joined: 22 Feb 2010
  • Loc: nj

Posted 27 June 2013 - 03:31 PM

also there is a way to turn it to F°. they show a circuit for the change on the spec sheet.


Who would want this? :grin:


you know who else would like to do that...My Mom

#19 don clement

don clement

    Vendor (Clement Focuser)

  • -----
  • Vendors
  • Posts: 812
  • Joined: 02 Feb 2011
  • Loc: Running Springs, California

Posted 27 June 2013 - 04:11 PM

Carry,

There is always the old standard (and still is)HP3458 8.5 digit DMM.

http://www.youtube.c...h?v=yBb24pCUDpk


When I was a teenager I built a Heathkit IM13 VTVM
http://www.ohio.edu/...bapix/vtvms.htm

Don Clement

#20 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1910
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 27 June 2013 - 07:09 PM

couldn't you just shoot the mirror with IR thermometer that way you can measure the edge, the back middle, front middle... +/-2°, do you really need to read more?


You can do the back and side of the mirror with an IR thermometer, but you can't do the front unless you put an opaque patch on the mirror surface, and you can't take air temperatures. A patch stuck temporarily in the Newtonian central "blind spot" woud be an option for getting mirror front surface temperatures.

Putting a probe on the mirror front surface as similar issues - probe covered with a bit of foam or cork help down by a clip.

Consider the back exhaust fan in a closed mirror box with a front baffle, which has been recommended for all night cooling (ensuring mirror tracks ambient air, scrubbing boundary layers, preventing dew formation, etc.).

In a closed box taking IR measurements is difficult anyway, but we want to track the air flow, using the air temperature as a proxy, and a physical feature of primary interest which the IR thermometer cannot easily measure. A temperature probe above and outside of the mirror box gives ambient, a probe just inside the baffle where the air flows off the mirror surface, an insulated probe on the top side edge and on the center back of the mirror, and perhaps one just in front of the exhaust fan could give a pretty good view of the air circulation and cooling/warming that is going on.

#21 Tom and Beth

Tom and Beth

    Soyuz

  • -----
  • Posts: 3777
  • Joined: 08 Jan 2007
  • Loc: Tucson, AZ

Posted 27 June 2013 - 08:17 PM

Would something like THIS work for what you have in mind? If nothing else, the same error would be in each of your readings.

I use these for rough temp readings, such as the underside of the roof in the observatory, drive motor temp, heat issues with Observatory Computer ...

#22 Dale Eason

Dale Eason

    Explorer 1

  • -----
  • Posts: 91
  • Joined: 24 Nov 2009
  • Loc: Roseville,Mn.

Posted 27 June 2013 - 09:37 PM

Would something like THIS work for what you have in mind? If nothing else, the same error would be in each of your readings.

I use these for rough temp readings, such as the underside of the roof in the observatory, drive motor temp, heat issues with Observatory Computer ...



The problem to watch out for with those non-contact thermometers is that glass even uncoated is reflective in IR. So they don't read the temp of the glass but what is reflected by it. I solved that problem by putting a piece of tape on the back of the mirror and pointed it at that.

Dale Eason

#23 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1910
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 28 June 2013 - 12:55 AM

IR thermometers are useful, I have one that I have used to monitor open back Dob mirror temps.

They do not however take air temperatures, an important aspect of monitoring cooling design; air temps also allows monitoring the air circulation - it is not only directly relevant measurement, it is also a proxy for monitoring the air flow through the system.

Also, not all parts of telescope systems are directly accessible to IR monitoring (closed Dob boxes, telescope tubes).

#24 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1910
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 28 June 2013 - 12:59 AM

I am sure there are five figure lab units that also make good voltmeters.

Ah the late lamented HeathKit! The website assets "the news of my death is greatly exaggerated" but that is so far the only sign of life.

#25 FHarry

FHarry

    Sputnik

  • -----
  • Posts: 43
  • Joined: 20 Apr 2007
  • Loc: Bavaria

Posted 28 June 2013 - 01:38 AM

why not use the one-wire sensors? Super-small in size and cheap as well. If you would not like to put the necessary sensors to a bus-cable (actually, a simple cable like the Cat5 or so and all sensors connected as a bus to this cable) then you even could use the one-wire buttons called iButton. Sensor with battery in a small housing, like a coin. So you could equipp your scope with this "buttons" wherever you like. In the end, you pick them out and put them into a little reader unit and read the data. Buttons can be configured to take a measurement all n-minutes or so.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics