Jump to content


Photo

Real life experiment with fan configurations...

  • Please log in to reply
31 replies to this topic

#1 sixela

sixela

    Voyager 1

  • *****
  • Posts: 14318
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 17 October 2010 - 07:16 PM

Since a year or so I've had a self-built truss Dob (a 16" f/4.5), and I was always wondering what fan set-up would actually work best at scrubbing the layer of warm air above the mirror...

Well, tonight I had an ideal opportunity to experiment. Temperatures were dropping like a brick and the outside temperature was a lot lower than where the scope was stored.

Jupiter was relatively high, with an Io transit in front of Jupiter roughly where the now-not-so-clear SEB used to be, so it was relatively easy to determine how good the views were (cues: how easy was it to see Io itself --not its shadow-- on Jupiter's disk, how round and how stable did the shadow appear, could you still see remnants of the SEB, could you still see a fairly faint blue festoon just reaching for Io, and how well delineated were two nice white storms and a barge in the NEB; there was also some detail closer to the North pole that helped a lot, but I need to brush up on knowing my nomenclature to name these.)

Observation without Paracorr, with 11mm TV Plössl in 2x Baader/Zeiss barlow, 324x. Obviously with the scope not colled properly on purpose (although fans had been running during the scope buildup routine and collimation, for roughly 10 minutes).

The scope's "official fan configuration": a mirror box top roughly 16cm above the mirror face, and 6 80mm fans on the sides (three at the bottom pushing air, three at the top pulling air). At the back there's a central 120mm fan that can be reconfigured to either push air or pull air. The side fans and back fan all have controllable speed.

Because of recent threads, I made a baffle that could be installed just on top of where the side fans are located (actually, the baffle used the part of the fans sticking out for support).

1) Side fans on, moderate speed (at full speed the mirror cools fast, but during observing running these at full speed gives an unstable view with momentary good images for very small intervals and mush at all other times).

1a) no back fan.
1b) back fan pulling
1c) back fan pushing

1b beats 1a beats 1c. If the side fans are switched off, 1c beats 1a, and 1b still beats 1c.

As the back fan pulling clearly wins, I decided to determine what side fan speed won. It turned out that very gentle side fan air movement won, at least with only the mirror box determining where the air comes from if you don't run the side fans.

2) Mauro da Lio baffle: 7cm over the mirror, with a hole just as large as the mirror (40cm); side fan orifices covered so that the airflow is forced to pass over the mirror's sides.

2a) fan pushing
2b) fan pulling

2b) beats absolutely everything else (and by quite a margin). Unlike the side fan arrangements, you don't get good "seeing" for brief moments and then mush, but a fairly stable view.

You did need more power fed to the fan to get the best results when it was pulling air.

Just to make sure I wasn't dreaming, I reverted to the other set-ups. The views had improved (the mirror had been cooling), but all the views with side fan arrangements still had more trouble with the mirror thermals than the baffle plus single fan pulling at the back.

Moving the baffle up (using some very crude spacers) made it less effective. Moving it down was impossible due to the fans still being installed.

So it looks my scope is going to become a lot more simple (thanks to Mauro da Lio).

Anyone close to Belgium interested in 6 good 80mm fans? ;)

#2 backwoody

backwoody

    Surveyor 1

  • -----
  • Posts: 1554
  • Joined: 08 Jan 2007
  • Loc: Idaho USA

Posted 17 October 2010 - 09:35 PM

Sixela, thanks, great report. I've often wondered about the various fan options.

c/s,

#3 vliegnerd

vliegnerd

    Sputnik

  • -----
  • Posts: 34
  • Joined: 26 Jan 2009
  • Loc: Gouda, NL

Posted 18 October 2010 - 12:17 AM

Thanks, for the hard work, sixela.

That settles it for me. I will use a pulling fan and the "mauro da lio baffle" in front of the mirror in my 12" tube to truss conversion.

#4 Olivier Biot

Olivier Biot

    Amused

  • *****
  • Posts: 24946
  • Joined: 25 Apr 2005
  • Loc: 51°N (Belgium)

Posted 18 October 2010 - 02:05 AM

Thanks for reporting your test results Alexis!

I think I am set now with my fan configuration: 1 big fan at the back + a couple cheapies to the side for initial cooling.

Cheers!

Olivier

#5 Víctor Martínez

Víctor Martínez

    Messenger

  • -----
  • Posts: 491
  • Joined: 05 Feb 2007
  • Loc: Cadiz - Spain

Posted 18 October 2010 - 02:58 AM

Sixela,is this design isn't it?. And then the side fans don't help very much. The most effective would be a good rear 120 mm fan pushing air and a deflector ring
http://autocostrutto...ng-mirrors.html

#6 sixela

sixela

    Voyager 1

  • *****
  • Posts: 14318
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 18 October 2010 - 03:47 AM

Thanks for reporting your test results Alexis!

I think I am set now with my fan configuration: 1 big fan at the back + a couple cheapies to the side for initial cooling.

The problem with the cheapies on the side is that when they're not working the big fan will pull air through their blades (instead of from between the mirror and the baffle, from the top of the mirror).

Unless of course you put them on top of the baffle, but then they simply don't do much, at least not if they're not well above the baffle angled down (what could work for initial cooling is blowing from the corners of the top of the mirror box, with fans angled downward).

Looking a the airflow with just the "deflector" baffle, though, I doubt all these top fans are actually necessary for anything.

I'm getting rid of mine.

#7 sixela

sixela

    Voyager 1

  • *****
  • Posts: 14318
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 18 October 2010 - 03:48 AM

Yes, that's the design, but pulling air works better than pushing air.

#8 magic612

magic612

    Soyuz

  • *****
  • Posts: 3798
  • Joined: 30 Sep 2008

Posted 18 October 2010 - 08:19 AM

Interesting results. I wonder if this would have a similar effect on a tube reflector as well.

One of the benefits of a refractor is that the lens is up quite high above thermal issues occuring on the ground - rocks, pavement, concrete, body heat, etc. So pulling air from the same general location as where a refractor lens would be (from the top of the tube down to the mirror) I would think might create better thermal consistency. The other thing this potentially does for a reflector is stabilize the air flow into the tube; pushing air towards the back of a mirror and then up and around it would creates turbulence. The airflow coming towards the face of the mirror should eliminate a good deal of that.

Of course, it also means mirrors might need to be cleaned more often, and the secondary may need a heater to avoid dew problems. But if it stabilizes the image, then this would be a worthwhile idea.

Thanks for sharing your results, sixela.

#9 sixela

sixela

    Voyager 1

  • *****
  • Posts: 14318
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 18 October 2010 - 08:27 AM

and the secondary may need a heater to avoid dew problems.


Actually, I found that if you use a shroud (which forces the air to come from the top of the UTA) it also helps in preventing dew just as much as a set-up that blows air into the tube (and it keeps the spider vanes close to ambient, too). Granted, my sceondary isn't that prone to dewing either, not being installed in a shroud: I've found that a shrouded secondary holder and the vanes act as a very efficient heat sink, so my secondary is just glued with three dabs of silicone to the (plastic) secondary holder.


I expect it works even better in tube scopes (those are radially symmetric).

#10 Fireball

Fireball

    Viking 1

  • -----
  • Posts: 533
  • Joined: 24 Mar 2006

Posted 18 October 2010 - 08:45 AM

Sixela, I assume your dob has a square layout of the rocker box similar to Mauro's, hasn't it?.
Do you think it will be any different when you have a round one (e.g. like the Lightbridge)?
I was thinking about introducing it to my scope but still do not know what the optimum distance to the mirror will be.
BTW, I use a shroud and have one fan sucking the air out of the tube. The back of the tube is isolated with an acrylic glass very similar to the one Sky Captain is using.

#11 sixela

sixela

    Voyager 1

  • *****
  • Posts: 14318
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 18 October 2010 - 09:24 AM

Sixela, I assume your dob has a square layout of the rocker box similar to Mauro's, hasn't it?.

Yes, but that's actually worse than a round one.

Do you think it will be any different when you have a round one (e.g. like the Lightbridge)?


This thread suggest that it works. Note that he actually blows air from the back; I tested that and at least in a square mirror box with a baffle at the top it doesn't work as well (the air over the mirror is evidently a lot more turbulent), though it might actually indeed work well in a Lightbridge plus one ring baffle (the Lightbridge has no baffle at the top of the mirror box and has a cylindrical mirror box).

#12 Olivier Biot

Olivier Biot

    Amused

  • *****
  • Posts: 24946
  • Joined: 25 Apr 2005
  • Loc: 51°N (Belgium)

Posted 18 October 2010 - 10:52 AM

For my 14" I plan on using one 120mm box fan. The unit I purchased is silent (20.5Db) and has a 44.7cfm air flow at 12VDC. If I install a baffle ~5cm above the top of the primary, I understand this should work for my setup, right?

#13 sixela

sixela

    Voyager 1

  • *****
  • Posts: 14318
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 18 October 2010 - 12:06 PM

Yup. Mine is capable of 73 cfm, but I'm only driving it at a speed that generates roughly 40 cfm. You do need to baffle well so that all the airflow is useful airflow, though.


#14 Olivier Biot

Olivier Biot

    Amused

  • *****
  • Posts: 24946
  • Joined: 25 Apr 2005
  • Loc: 51°N (Belgium)

Posted 18 October 2010 - 01:47 PM

Okay :)

I just cut a 120mm x 120mm square hole in the mirror cell board.

I was planning to use rubber spacers mounted in small O bolts in the 4 corners of the hole to secure the box fan and a grid to the back of the cell. The rubber spacers serve as shock absorbers.

#15 sixela

sixela

    Voyager 1

  • *****
  • Posts: 14318
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 18 October 2010 - 06:18 PM

I use a Noiseblocker multiframe fan - rubbers already integrated in the corners of the fan.

#16 cheapersleeper

cheapersleeper

    Gemini

  • *****
  • Posts: 3357
  • Joined: 22 Jan 2010
  • Loc: Sachse TX

Posted 19 October 2010 - 02:09 AM

Do you know anyone with a tubed scope that you can try that on?

#17 sixela

sixela

    Voyager 1

  • *****
  • Posts: 14318
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 19 October 2010 - 04:51 AM

I'll let others decide, but you could always ask SteveG (thread details above) to reverse fan flow direction and report results for a second smoke test.

I already did ask, but SteveG must've missed it.

#18 jtpowers

jtpowers

    more clever in the first place

  • *****
  • Posts: 1093
  • Joined: 03 Nov 2005
  • Loc: Cambridge, MA

Posted 19 October 2010 - 07:35 AM

Thanks for well reasoned report, Sixela. With a mirror as large as yours perhaps it's a small detail, but the exactly-mirror-sized baffle will cause some degree of vignetting. Won't it allow only strictly parallel lines to hit the edge of the mirror.

#19 RossSackett

RossSackett

    Surveyor 1

  • *****
  • Posts: 1718
  • Joined: 17 Aug 2007
  • Loc: Memphis, TN

Posted 19 October 2010 - 08:22 AM

This looks like a nice clear result: a back pulling fan does a better job of sweeping up convecting air in the light path, which in retrospect seems obvious since the back pushing fan is delivering a lot of warmed air from the back and sides of the mirror to right where we don't want it to be.

But what about initial cooldown? My guess is that the higher-velocity blast from a back-pusher would remove heat from the mirror much faster than the gentler pull of the pull-fan at the same cfm.

If this is the case, we might want a back-pusher for launch and a back-puller for cruise. How about a fan mounted on magnets that we can easily reverse?

Ross

#20 sixela

sixela

    Voyager 1

  • *****
  • Posts: 14318
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 19 October 2010 - 10:06 AM

But what about initial cooldown?

My experience is that you sometimes need a stronger fan, but once you've generated enough airflow (very little, actually, less than what is necessary to "scrub the boundary layer") it makes little difference. All you want is to refresh the air around the mirror, and how fast you do it has little influence as long as it's fast enough (that's also the results than Bryan Greer arrived at in his S&T articles).

My guess is that the higher-velocity blast from a back-pusher would remove heat from the mirror much faster than the gentler pull of the pull-fan at the same cfm.

Once you have baffles, the same cfm will probably do the same thing, but you do need a stronger fan to arrive at the same real cfm (those mirror boxes do have quite a bit of impedance as far as moving air is concerned, especially when you direct the airflow via baffles).

In my case, pushing and pulling is the difference between a Noiseblocker M12-S2 and an M12-S3HS.

How about a fan mounted on magnets that we can easily reverse?

As you can see from the test, mine can actually be reversed (it's held in a hole just big enough for it to be held through the pressure of the Noiseblocker vibration suppression rubbers in the corners, so you can just pull it out, reverse it, and stick it back in). But I don't think I will ever reverse it, since the airflow at 12V is enough to make pulling displace quite some air...

#21 sixela

sixela

    Voyager 1

  • *****
  • Posts: 14318
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 19 October 2010 - 10:08 AM

Thanks for well reasoned report, Sixela. With a mirror as large as yours perhaps it's a small detail, but the exactly-mirror-sized baffle will cause some degree of vignetting.


Actually, at the edge of a 1 degree field it will clip slightly more than a millimetre. All it's going to hide is part of the bevel ;). That's what's good about a baffle just above the mirror: the closer it is to the mirror the less wide you have to make it to avoid vignetting.

#22 sixela

sixela

    Voyager 1

  • *****
  • Posts: 14318
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 19 October 2010 - 10:16 AM

This looks like a nice clear result: a back pulling fan does a better job of sweeping up convecting air in the light path,


Only if there's a baffle of some sort to force the air flow to come from the top of the mirror. In my case just the mirror box cover already starts to work (though it's less efficient), but it's only 17cm above the mirror as my mirror box is fairly low.

If I remove that mirror box top baffle, then a pulling fan has little effect, i.e. even less effect than a pushing one. The air seems to come straight from the sides of the box and there's little movement of the air over the mirror.

I think that a pulling fan installed in a K&R/Webster etc. high mirror box wouldn't work without an extra baffle somewhere.

#23 jOe~

jOe~

    Lift Off

  • -----
  • Posts: 6
  • Joined: 24 Jul 2010
  • Loc: Spokane, Wa

Posted 20 October 2010 - 01:07 PM

Would anyone like to speculate on the thermal dynamics of a naked scope, i.e. tubeless, boxless, shroudless, a minimalist or low rider style? It would seem that any form of enclosure would only complicate thermal problems. Could a couple fans, above and below the mirror blowing toward the observer be the best solution?

#24 gatorengineer

gatorengineer

    Mercury-Atlas

  • *****
  • Posts: 2553
  • Joined: 28 Feb 2005
  • Loc: Hellertown, PA

Posted 20 October 2010 - 09:08 PM

Sixela,

I read your post with great interest. You have a very shallow mirror box 16CM.... I am in the design phase of starting rebuild number 4 or 5 on my 16" F4.5...... Each time learning more and more...

One of the design decisions I have been struggling with is baffling, for light that comes in at a shallow angle to the optical axes of the scope, which easily makes its way to the eyepiece.

My plans for the latest rebuild, to address this as they stand now, include a circular mirror box liner 1.5" off of the mirror (19 inch ID) to eliminate mirror box currents, and to extend this circular liner up about 500mm above the mirror to cut down on stray light from this source... Do you think the Mario di Lauro baffle would work in this configuration? The other thought was to over width the cage rings to accomplish this, and try a shorter mirror baffle....

#25 sixela

sixela

    Voyager 1

  • *****
  • Posts: 14318
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 21 October 2010 - 02:11 AM

Sixela,

I read your post with great interest. You have a very shallow mirror box 16CM.... I am in the design phase of starting rebuild number 4 or 5 on my 16" F4.5...... Each time learning more and more...

One of the design decisions I have been struggling with is baffling, for light that comes in at a shallow angle to the optical axes of the scope, which easily makes its way to the eyepiece.

My plans for the latest rebuild, to address this as they stand now, include a circular mirror box liner 1.5" off of the mirror (19 inch ID) to eliminate mirror box currents, and to extend this circular liner up about 500mm above the mirror to cut down on stray light from this source... Do you think the Mario di Lauro baffle would work in this configuration?


Sure. It may even work better because the flow will be radially symmetric (it isn't on my scope, given there's more space around the mirror in the corners of the mirror box than in the middle of the sides). For that size of mirror, you should probably ask Mauro da Lio how high the baffle should sit, but this is the image from another thread of his setup for the 24" scope:

Posted Image



If you're worried that 1.5" is going to be very narrow: yes, perhaps, although you'd need to do a CFD analysis of that precise configuration to know.

Mauro's got a simulation on his page that's similar to a very narrow tube plus very narrow baffle around the mirror, where the centre of the mirror is left a bit unswept --but partly because the simulation assumes the rest of the mirror box is humongous--, and another simulation that's better, in which the width of the baffle is roughly a quarter of the mirror radius (in your case that would be roughly 2.5").

My personal baffle is actually 20% of the mirror radius wide even at the most narrow point (I have 4cm of clearance from the mirror to the mirror box side) - a lot more than Mauro's baffle at the centre of the sides. That would be very close to your 1.5" (but obviously you'd have it all aorund the mirror while my "average" baffle width is higher).

Even narrow baffles can direct airflow pretty well:

http://www.youtube.c...h?v=2LWOEbQQEsI

but here the fan is pushing.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics