Cooling Unit Challenge & Advice

[John K]

Hi everyone,

I am trying to design some type of cooling unit for my scope to speed up how quickly my full thickness 12.5" f/5 mirror cools down.

I have challenge ahead of me as the scope is a JMI NGT 12.5".

For those not familiar with this type of telescope, it's a truss mounted Newtonian, on a horseshoe equatorial mount (see pics attached). It's an excellent design, but does not come with a cooling fan! however the design does allow for lots of cool air to circlate very easily around the mirror. Being of all steel construction, it looses heat very quickly as there is no tube and so far I have used a domestic fan to simply blast the mirror with cool air.

The mirror is enclosed in a steel tub and there are 3 large counter-weights bolted to the back of the mirror cell holding it to the tub. There is no room to have anything extend past the counterweights as this would hit the horseshoe drive/gear.

So the challange is how to get cool air behind the mirror without having to completely rebuild the rear end which would require extensive metal work and could possibly jeoperdise the design of the telescope.

The only two ways that I could think of, is to place 2-3 fans simply on the inside side walls of the steel tub so as to create circulation in the tub area. I would make these units removable in order to remove the mirror tub if required. I have seen this design used in some large truss dobsonians, with fans placed in the inside square corners of the rocker box.

The other radical idea I have is to use 1/2 inch tubing to somehow pump cool air under the mirror cell and surrounding area. So I would need a unit of some sort that generates air and that I can simply plug into these tubes that could get air around and behind the mirror cell. Has anyone heard of anything like this? A small compressor of some sort perhaps to generate high pressure air?

Lastly, I could drill a few holes in the steel tub in locations that would not jeoperdise the stuctural integrity of the scope, however, this is once again a passive cooling method.

Images attached, so if you can put your thinking hat on and offer some suggestions that would be great!

Look forward to comments.

John.

[davidpretorius]

John, first things is, where do you get one of those!!!

I would use peltiers if you can get a flat surface to attach the 1" x 2" cold plate to.

Is there any room behind the mirror, say 2"???

Also a flip hinge with computer fans blowing down onto the mirror and then moving the cold air from below around the mirror.

Lovely setup!

[John K]

Hi Dave,

JMI made this scope for me and customised it at f/5. See link:

http://www.jimsmobile.com/ngt125_data.htm#Top

There is no room behind the mirror unfortunately nor behind the tub holding the mirror cell.

[asimov]

I've got the same problem with my 12.5" 2" full thickness primary. Extensive mucking about with fans & testing this & that was...it never really got to ambient. I figured my only hope would be forget about getting it to ambient, but rather concentrate on expelling the barrier layer, which can be done with a minimum of fuss with 2 fans blowing across the mirror.

[astroboy]

I agree
I have the NGT18 which has a fan at the back but never made much diferance.
I did but some fans for the front surface but never got round to installing them ( too many toys )
Mind you the temp variation isn't bad with the fiberglass dome even in summer ( during the day ) its quite cool in there.
I guess what I'm saying is if your going to bother put them in front of the mirror , with the NGT " tub " there should be room . Thats what I will do with the 24" Dobzilla when it finaly gets here.

Zane

[Roger Davis]

A cold finger and peltier wouild help enormously. You could supplement with a fan that will push the warmed air away. S&T had an article on fans ABOVE the primary mirror, directing the airflow across the surface, this apparantly worked better.

[John K]

G'day Roger, thanks for that, I dont think I have spoken to you for at least 10 years since ASV demonstrator days!
Excuse my ignorance, but what is a cold finger?

Asi - I gather that to mount a fan or 2 fans to blow air across the surface of the mirror I will need a hole for mounting the fan itself and then one at the opposite end of the fan to get the air to exit across the mirror surface?
And the suggestion is for at least 2 fans and 2 exit holes on the opposite end? or is it simply one fan to blow air in and the other to blow air out? Lastly, for a 12.5" mirror, what is the rule for the size of fan and in terms of position, is it simply at the same level as the mirror surface?

[Roger Davis]

Hi John! Been a while!
You place an aluminium plate behind the mirror with some silicon heat transfer paste between the two. Running the peltier will suck the heat out of the mirror via the aluminium plate (cold finger). You may need a couple of them.
Trick is to get the mirror down to ambient, not form stalactites! There has been a thread on this somewhere on this forum I think? You can place a sensor on the mirror as well to act as a thermal switch when the mirror reaches ambient.

[John K]

thanks Roger, that seems to be the best way. The problem I have is that there is just no room at the rear of the telescope for me to do that.

So looks like getting rid of the boundry layer is the go.

In terms of "thinking outside the square", what would the effect be if I pumped cold air under and around the mirror?
I came across a unit today that racing car drivers use to cool their helmets when car racing. Essentially it's a mini air conditioning unit with small hoses. Do I risk getting dew build up? or will this still be insufficient to remove heat fast enough from the mirror as it's still an air based system.

[Roger Davis]

Air conditioning means that you are dehumidifying the air so there is no risk of any moisture from the airconditioning itself, just the surrounding air. If you can even stick a small sensor (they aren't big, half a postage stamp size) to the back that could switch the aircon, that may work.

[John K]

I have been having an e-mail exchange with Bryan Greer. His latest e-mail is below. Very interesting!!

--------------------------
Dear John,

It's good to hear from you again.

It is easy to let two different issues to blend in one's mind. If you ask
ten telescope owners what they're trying to do by adding fans, maybe eight
of them will say they want to bring their scope into equilibrium faster.
Fans certainly do hasten this initial equilibrium (see the image john01.gif
attached), but this aspect was not the surprising part of what I discovered
ten years ago. (Amateurs have know for a hundred years about this initial
cooldown problem, and that fans help.) If this were the only thermal
problem, it could be "solved" by simply having a bit more patience, and
waiting for the scope to reach equilibrium on its own.

The more important function of fans is to keep the surface temperature of
the mirror tracking the ever-falling ambient temperature more closely, and
this must be done throughout the night. This is at the heart of the issue.
For most locations on Earth, the nighttime temperature under clear skies
falls too fast for even small Pyrex mirrors to track it close enough. See
the second graphic (john02.gif) to see what I mean. The mirror on the left
will suffer from thermal problems all night long. This is why any solution
you come up with MUST be able to operate continuously, and not just switched
off after the initial equilibrium is reached.

What delta T is "close enough"? The mirror surface needs to be within about
1 degree Celcius (~ 2 F) to keep the boundary layer from forming with any
significant refracting strength. This value is based on the color schlieren
testing I did in the 1990s (see Sept. 2000 Sky & Telescope for more
details). Remember, the strength of the boundary layer is PURELY a function
of this delta T.

Cooling by refrigeration, or other sub-ambient techniques (Peltiers), would
certainly help with the first problem (i.e., initial cooldown), but they are
problematic, and largely unnecessary, for the more important objective of
tracking the falling ambient. For one thing, you now have a new challenge to
prevent the mirror's temperature from falling below ambient. If that
happens, dew will form quickly if the relative humidity is high enough.
Also, a mirror's surface that is below the ambient by a few degrees will
form a boundary layer exactly like a warm mirror will (yes, it flows
downward instead), and will cause the same optical image damage. You would
need a temperature sensor and feedback to control the refrigeration (or
Peltier) process. (Of course, fans are blowing air that is always at ambient
temperature, so in this sense they are "self-regulated".) Finally, if you
tried to use the refrigeration device during observing, it would create very
strong thermal gradients in the optical path. This, alone, would relegate
its use only during the initial cooldown phase, and then switched off.

For my location (Ohio, U.S.A.), fans can keep a mirror that is up to about
2-1/2" thick (75 mm) tracking our ambient close enough to keep the boundary
layer at bay (i.e., within a degree of ambient). This requires airflow over
the front and back surfaces. If the mirror is around 1-1/2" thick or less,
airflow only across one surface (usually the back) will do the job. Your
day/night temperature drops are larger than mine, so your scope could
benefit from airflow across both surfaces. Don't fret if you can only get to
the front. That will still be a big improvement over no fans at all.

I am glad to hear this is being discussed online. I will try to read the
links you gave me, but also feel free to post this information (and images)
to those sites, as well. If you can get your hands on them, I recommend
reading my three "Sky & Telescope" magazine articles on this subject (Sept
2000, May 2004, and June 2004).

Sincerely,
Bryan Greer

[iceman]

Excellent information!
Thanks for posting.

So what does he recommend in terms of fans at the front? Something blowing across or sucking out from the side of the tube?

[davidpretorius]

yup, fantastic detail int that one.

One thing is that the peltiers seem to remove a lot of moisture out of the air. I simply do not have fogging issues on the main mirror

[John K]

The recommendation so far is for a medium speed fan with speed adjustment and isolated from the tube to provide minimum vibration blowing accross the surface of the mirror and a temperature probe for the mirror and ambent. As a start for me will be simply a fan on top of the bucket blowing down towards and across the mirror as with my set up vibration will be a huge issue being an all steel truss telescope sitting on top of a "shakey" sun deck.
His previous email is below which also offers some other alternatives:

-------------
Dear John,

A 2" thick mirror (of any diameter) will certainly benefit from forced
airflow, and the most noticeable improvement will be for planetary
observing. Whatever fan configuration you come up with, it must be able to
run continuously during the observing session without inducing
microvibration. Remember, the main benefit of blowing air over the mirror is
that it keeps the mirror's temperature much closer to the ambient
temperature, and of course the ambient temperature is continuously falling.
This requirement will guide your design, and immediately eliminate some
possibilities (e.g., mounting fans directly to the mirror bucket). Keeping
this "delta T" low is what eliminates (or weakens) the boundary layer.

For your scope, the solution that would require the least hacking and
cutting would be to simply clip on two or three small fans up at the top rim
of the mirror bucket, and roughly aim them down into the bucket. The airflow
recommendations I worked out would not be applicable, as those are tested
for the more common situation of having the fans located near the mirror's
surface. However, it would not take much airflow to do the trick. Try 30-50
cfm as a start. (A few years ago when I was measuring outdoor mirror cooling
rates, I discovered that even a natural breeze would cool about as
effectively as fans if the wind was in the right direction. So, it doesn't
take much airflow at all, nor does it need to be directed particularly
carefully.)

A more involved, but cleaner looking, fix would be to install three small
(~25-30 mm diameter) fans into the bucket wall. These would be positioned
right above the face of the mirror. Of course, this would involve hole
cutting. You also shouldn't try to mount the fans directly to the bucket
wall, or it will certainly induce microvibration (see my May 2004 S&T
article for how to test for microvibration). Foam gaskets are also only
partially effective. The best solution I've come up with so far is to mount
the fans to a soft elastic sheet material, which is in turn mounted within a
plastic frame. The elastic will absorb and dampen vibration before it can
get transmitted to the bucket.

If you really want to get clever, you can try something like the photo I
have attached. I have not done this solution myself, but I know two people
who successfully implemented this "suspended fan" in their large Dobs. The
suspension wires are aligned with the spider vanes, and two of the wires
also conduct the electrical current. Of course, the fan housing cannot be
larger than the secondary mirror, but this usually isn't a problem since the
fan can be located in an almost perfect location.

It's hard to tell from your photos, but it doesn't look like you have room
behind the mirror for fans.

Finally, if you are really serious about this, consider attaching a thermal
probe to the back of the mirror (temporarily). This will let you quantify
the thermal cooling rates, and let you conclusively determine when/if you've
made an improvement. It isn't as hard to do as you might think (see
http://www.fpi-protostar.com/bgreer/miscpages/probe.htm (http://www.fpi-protostar.com/bgreer/miscpages/probe.htm)). The little
handheld device I use costs about $100 US, but even a decent $20
indoor/outdoor electronic thermometer might be good enough. It's not the
absolute accuracy of the device you care about as much as its night-to-night
repeatability.

Good luck with your project. The improvements will be worth your effort,
especially for planetary/lunar viewing.

Sincerely,
Bryan Greer

[iceman]

Excellent.

Looks like I need to cut a hole in the side of my tube like ol' Davo has done.

[davidpretorius]

now who needs some holesaws?????:D

or asi with a plasma cutter???

[asimov]

What, has the penny just dropped ? I've been saying this is the way to go for 12 months. Gee you guys are quick !:P

[gbeal]

Hey, great place for a fan though, suspended like that.
Good article too. Bryan wrote (as he suggested) a few articles in Sky & Tel, and they were good meaty ones.
As an aside he is also a great guy to deal with, I got some flocking paper (say that after a few Bundabergs Ponders) a while back.

[John K]

As a first step, I am looking to get a temperature logger to track the temperature of my mirror and ambient.
To do this, the solution I am looking at is a dual input temperature thermometer with k type thermocouples which I will simply attach to the back/and or side of my mirror using insulation tape and some insulation paste.
The unit is a D/Smith unit retailing for $98 and inlcudes one theromocoulple (Will get at least 1-2 more at $10 each).
Could not find a cheaper unit that takes 2 thermocouples and which displays at 0.1 degree increments. I can then play with various fan combinations to see what works best.


Any thoughts guys? Is this overkill or is there a cheaper way to do this?




p.s. have tried e-bay and Jaycar. These things are damn expensive!

**** SMITH DELUXE TEMPERATURE DUAL INPUT THERMOMETER
------------------------------------------------------------------
A compact, handheld dual-input digital thermometer with 4.5 digit LCD screen and both Celsius and Fahrenheit readings. Provides Data Hold, selectable dual temperature or temperature differential displays, as well as a Relative mode which displays the difference between a reading and a preset value. Also includes a Min⁄max with time record mode, a Hi⁄Lo limit mode which sounds a beeper when a preset limit is exceeded, and Auto Power Off after 30 minutes.
Supplied with a K-type thermocouple with 4ft lead and Teflon tape coating which is insulated to 260°C.
Specifications :
Temperature metering range: -200 to 1,370°C (K-Type thermocouples),
-200 to 1,050°C (J-Type thermocouples).
Temperature range (supplied thermocouple): -40°C to 260°C
Resolution: 0.1°C
Accuracy: -200 to -50°C: +⁄- (0.05% reading + 0.3°C)
-50 to 1,370°C: +⁄- (0.05% reading + 0.7°C)
(for operating temperatures of 18 to 28°C)
Probe Accuracy: +⁄- 2.2°C or +⁄- 0.75% of reading (whichever is greater)
Reading Rate: 1 time per second
Battery: Standard 9V 216 type battery
Battery Life: approx. 100 hours with carbon zinc


Features:

Compact, handheld dual input digital thermometer.
4.5 Digit LCD
Offers both Celsius and Fahrenheit readings
Data Hold
Selectable dual temperature or temperature differential displays
Min⁄max with time record mode
Hi⁄Lo limits mode sounds beeper when preset limit is exceeded
Relative mode displays difference between reading and preset value
K-Type thermocouples supplied: 4ft, teflon tape insulated to 260 degrees C.
Auto power off after 30 minutes
Range: -200 to 1,370 degrees C (K-Type) -200 to 1,050 degrees C (J-Type)
Resolution: 0.1 degree C
Accuracy: -200 to -50 degs C: +⁄- (0.05% rdg + 0.3 deg)
-50 to 1,370 degs C: +⁄- (0.05% rdg + 0.7 deg)
(for operating temperatures of 18 to 28 deg C)
Probe Accuracy: +⁄- 2.2 deg C or +⁄- 0.75% of reading (whichever is greater)
Reading Rate: 1 time per second
Battery: Standard 9V battery
Battery Life: approx. 100 hours with carbon zinc

[davidpretorius]

$30 gets you a dual sensor with digital readout. The question is: do you need to auto log the info in 1 sec increments OR simply come out every 1/2 hr and check it OR come out side to see when the mirror is at ambient.

Wombat in Space made a great logger and so did bird. Mike & I have simply put one end of the dual sensor on the mirror with heat paste and the other sensor in the body of the display.

Given that you may be able to cool this mirror at best 5 degrees and hour with peltiers, how accurate do you really need it???

Like I say, I simply have the peltiers on until the readout has the mirror within .5 of a degree of ambient. I do not care how it got there from a logging point of view.

[John K]

yep, all makes sense, dont really need this info every second, just need the dual temperaure logging, less than 1 degree accuracy is good.

Just googled "Wombat in Space" with no success, do you have a link or more info Dave?

[asimov]

Wombat in space (Darren) is an IIS member John.

[John K]

doh!.....mmm....making one of these things......looks like I'll be spending some cash at D/Smith electronics!

[davidpretorius]

have some pictures in this thread of the sensor

http://www.iceinspace.com.au/forum/showthread.php?t=10252

[asimov]

For a few bucks more, DSE have the temp. sensor unit & humidity sensor in one. Extremely handy for telling you when your likely to have a dew problem.

[John K]

Guys,

thanks for all the suggesrtions here, appreciate it.

The base of the problem here is that I am absolutely and totally hopeless when I comes to any electronic, techo stuff.

So, in terms if what I need, it will be a small basic LCD display with an attached sensor that can give me an accuracy of either 1 or idealy less than 1 degree (ie. 0.1). does not have to measure it every milliscend or track it etc etc.

Can a place like DSE or Jaycar provide a unit like this or do I have to build one?

The only thing I have seen like this is like an aquarium type digital thermometer that unfortunately has a steel rod as a sensor rather a k-type thermocouple.

More suggestions?


p.s. remember what I said in the second paragraph above! and nice pics Dave!

[iceman]

Just get the one from d.ick smith for $20.

Pics of it here (http://www.iceinspace.com.au/forum/showthread.php?t=9981)

It's great.

[davidpretorius]

yup, accurate to .1 degree and cheap from dse or jaycar