Go Back   IceInSpace > Equipment > ATM and DIY Projects
Register FAQ Calendar Today's Posts Search

Reply
 
Thread Tools Rate Thread
  #41  
Old 15-11-2015, 11:19 AM
ZeroID's Avatar
ZeroID (Brent)
Lost in Space ....

ZeroID is offline
 
Join Date: May 2010
Location: Auckland, NZ
Posts: 4,949
OK, here's a cardboard template cut to fit. No need to modify the frame, just carve around it. BUT... where the yellow line is I will double the thickness of copper, solder sweat another piece on top to increase the thermal path. Also the outer locating lug will lose it's bent over top so just require a small hole to penetrate the finger.

That will drop the whole finger down slightly beneath the larger lug and will also mean it is just a straight piece of copper, no bending required. It also means that I will support it from an external bar underneath the camera and the finger will therefore 'float' inside the sensor assembly and reduce any tendency to lift or twist it. Also means you could refit the vid\remote board I think.

Insulation and earthing was always in the plan Glen. Just got to decide whether to leave the finger straight or kick a 90* bend in it.

Part of my reason to not remove the lugs is that the larger one is quite robust and removal may either distort the frame or end up with metal filings over the internals causing shorts etc..

Right, off down to the workshop now ... see ya soon.
Attached Thumbnails
Click for full-size image (Cardboard.jpg)
144.2 KB23 views
Reply With Quote
  #42  
Old 15-11-2015, 05:10 PM
Camelopardalis's Avatar
Camelopardalis (Dunk)
Drifting from the pole

Camelopardalis is offline
 
Join Date: Feb 2013
Location: Brisbane
Posts: 5,429
One thing I'd suggest Brent is to make it shorter i.e. to leave as little copper outside the case as possible. With my experiment last week, I had cold dew forming on the outside of the insulating tape where my cold finger exited the box. Also, whether you choose to bend or not, leave only enough copper square at the end to cover the Peltier, otherwise you'll get an iceberg forming on any exposed copper!
Reply With Quote
  #43  
Old 15-11-2015, 07:31 PM
glend (Glen)
Registered User

glend is offline
 
Join Date: Jun 2013
Location: Lake Macquarie
Posts: 7,051
I sprayed mine with four coats of Plasti-dip rubber paint, and insulated external part with closed cell foam (cut from an old camping or yoga mat.

The only exposed copper was where the TEC attached.
I know Rowland has always been in favour of leaving a patch where frost can form but I don't have any. I do get some condensation on the outside of the foam insulation over a period of time but inside It's seems fine. Everything inside the camera is in Argon anyway.
Works for me. I am experimenting with GoPro camera dessicant strips inside the camera as well.
Reply With Quote
  #44  
Old 16-11-2015, 06:02 AM
ZeroID's Avatar
ZeroID (Brent)
Lost in Space ....

ZeroID is offline
 
Join Date: May 2010
Location: Auckland, NZ
Posts: 4,949
My intention is to only get down to the +5* mark, not subzero temperatures. Icing should only be a minor issue at that level I think. Easy enough for me to insulate any exposed surfaces. I've currently got extra as I was not sure how I was going to support the finger externally so left width to add securing hardware. Easy enough to attack with the hacksaw and file.

Longer term I am seriously considering stripping the camera to basics. Read a thread where it was done, the real gain being to chuck away the rear screen or at least offset it elsewhere. The camera doesn't need it. With an external powr supply and permanent USB connection the TEC unit could then be hang off the back and the whole thing built into a cold box, sealed and argon filled.

Too many ideas, need to get this version working first. I keep watching out for cheap DSLRs to try other ideas and options.
Reply With Quote
  #45  
Old 16-11-2015, 09:29 AM
Camelopardalis's Avatar
Camelopardalis (Dunk)
Drifting from the pole

Camelopardalis is offline
 
Join Date: Feb 2013
Location: Brisbane
Posts: 5,429
I stripped down my 1100D, it's pretty straightforward although a fair amount of screws

Suffice to say, at this time I don't have a working 1100D as it failed at a dark site last week, and I haven't found the culprit yet

Even cooling down to get EXIF temps of 12C I was still getting an iceberg on the little excess exposed copper I had next to the Peltier.
Reply With Quote
  #46  
Old 17-11-2015, 07:49 AM
ZeroID's Avatar
ZeroID (Brent)
Lost in Space ....

ZeroID is offline
 
Join Date: May 2010
Location: Auckland, NZ
Posts: 4,949
Quote:
Originally Posted by Camelopardalis View Post
I stripped down my 1100D, it's pretty straightforward although a fair amount of screws

Suffice to say, at this time I don't have a working 1100D as it failed at a dark site last week, and I haven't found the culprit yet

Even cooling down to get EXIF temps of 12C I was still getting an iceberg on the little excess exposed copper I had next to the Peltier.
Ice situation noted, I guess it has to get a lot colder to get a thermal gradient compared to the sensor end. I'm still working on a simple support system for the whole thing. I don't want all the weight hanging off the camera and possibly moving the sensor assembly. Glens carbon fibre bracket or similar is the answer but I want it so it is an easy dissasembly and access process. We'll get there.

Opening another thread in this area for another idea I've had re cooling DSLRs. Comments welcome of course.
Reply With Quote
  #47  
Old 17-11-2015, 11:36 AM
Camelopardalis's Avatar
Camelopardalis (Dunk)
Drifting from the pole

Camelopardalis is offline
 
Join Date: Feb 2013
Location: Brisbane
Posts: 5,429
As if by magic, I unpacked my camera, removed it from the case and plugged it in and it's working fine
Reply With Quote
  #48  
Old 17-11-2015, 11:47 AM
glend (Glen)
Registered User

glend is offline
 
Join Date: Jun 2013
Location: Lake Macquarie
Posts: 7,051
Quote:
Originally Posted by Camelopardalis View Post
As if by magic, I unpacked my camera, removed it from the case and plugged it in and it's working fine
Hmmm, moisture related short perhaps.
Reply With Quote
  #49  
Old 18-11-2015, 12:31 PM
Camelopardalis's Avatar
Camelopardalis (Dunk)
Drifting from the pole

Camelopardalis is offline
 
Join Date: Feb 2013
Location: Brisbane
Posts: 5,429
Maybe. Fired it up yesterday and it was imaging fine again last night
Reply With Quote
  #50  
Old 19-11-2015, 07:47 AM
ZeroID's Avatar
ZeroID (Brent)
Lost in Space ....

ZeroID is offline
 
Join Date: May 2010
Location: Auckland, NZ
Posts: 4,949
Quote:
Originally Posted by Camelopardalis View Post
Maybe. Fired it up yesterday and it was imaging fine again last night
Don't you just LOVE astronomy .. ??




And on another note I've designed my TEC support bracket, apparently it requires a trip to Bunnings, need some 50mm alum angle as the starting point. It's pretty lightweight but I'll lighten it further by filling it full of holes. Get the mill out and ready this is going to look pretty as well as functional.
How do I get that nice satin black anodised look
Reply With Quote
  #51  
Old 29-11-2015, 04:45 AM
ZeroID's Avatar
ZeroID (Brent)
Lost in Space ....

ZeroID is offline
 
Join Date: May 2010
Location: Auckland, NZ
Posts: 4,949
Big Progress !
Got my Alum angle and got stuck in. Pictures should tell the story. You'll see the finger changes shape a bit to clear the USB port. My sensor must have been manufactured with a minimal gap between it and the board because thin taped it only just got the copper in between, a tight fit. I guess good contact will make for good cooling though.

The finger is sandwiched between two foam blocks on exit and the rear of the finger will be also covered ( see last pic dummied )

The support bracket ( alum angle ) is still under development. It's a bit dependent on which cooling fin arrangement I come up with although I am trying to keep it flexible for further upgrades. A lot of mass will be milled away and access to ports etc sorted. I'll post [pix of the finished bracket later but right now I'm wrapped, it's all working again !!

For any wannabe DIY folks out there who want to give anything like this a go here's good tip. Work in a 'box'. All the bits etc removed or awaiting reassembly all stay together in the largish flat box. If I drop any screws etc they just roll around in the box and don't get lost. I was lucky with having some black boxes from stuff at work and metallic bits show up easily on that. No parts were lost in making this cooled DSLR
Attached Thumbnails
Click for full-size image (Finger04297.jpg)
138.9 KB17 views
Click for full-size image (TapedFinished04301.jpg)
138.8 KB27 views
Click for full-size image (RearInternal04302.jpg)
131.4 KB18 views
Click for full-size image (RearFoam04303.jpg)
141.0 KB20 views
Click for full-size image (Working4309.jpg)
127.1 KB19 views
Click for full-size image (Bracket04305.jpg)
168.5 KB21 views
Click for full-size image (BracketTop04307.jpg)
141.8 KB18 views
Reply With Quote
  #52  
Old 29-11-2015, 07:02 AM
glend (Glen)
Registered User

glend is offline
 
Join Date: Jun 2013
Location: Lake Macquarie
Posts: 7,051
Looking good. Certainly a robust bracket, should support any heatsink/fan combo. If you are not going to use the little connector board you could 'remove' the two hole bracket and support post; probably too late now and it would be messy to do, the examples I have seen removed the posts with the frame removed from the camera. There was one guy that bent them down flush with the frame.

Last edited by glend; 29-11-2015 at 07:56 AM.
Reply With Quote
  #53  
Old 29-11-2015, 12:41 PM
ZeroID's Avatar
ZeroID (Brent)
Lost in Space ....

ZeroID is offline
 
Join Date: May 2010
Location: Auckland, NZ
Posts: 4,949
OK, here's a slightly 'Heath Robinson' ready to go version. Small heat sink and fan for a starter, prove operation. I'll go chuck 5volts at it and monitor the sensor temp.

Re the pin and bracket. All good to stay actually. I could have refitted the remote board over top of it all but didn't bother. What I did do was cut the bent bit off the pin and used the upright bit left as a locator for the finger which also hooks over the larger bracket and keeps it all in place

Still a lot of Aluminium I can remove to reduce weight but it's not bad as is.
Attached Thumbnails
Click for full-size image (TECFrt04311.jpg)
140.5 KB24 views
Click for full-size image (TECBottom04312.jpg)
144.8 KB30 views
Click for full-size image (TECBack04313.jpg)
123.4 KB24 views
Reply With Quote
  #54  
Old 29-11-2015, 01:30 PM
Camelopardalis's Avatar
Camelopardalis (Dunk)
Drifting from the pole

Camelopardalis is offline
 
Join Date: Feb 2013
Location: Brisbane
Posts: 5,429
Good stuff Brent looks very heavy duty.

Next stop: temperature readings eh?
Reply With Quote
  #55  
Old 29-11-2015, 01:43 PM
glend (Glen)
Registered User

glend is offline
 
Join Date: Jun 2013
Location: Lake Macquarie
Posts: 7,051
I realise it's just a test rig right now; a few things to consider in later versions:

Isolation of the heatsink from the cold finger - try to use nylon screws and nuts as they will not transfer heat the way those steel ones will. Most RC Hobby online sellers have nylon fasteners available. This will improve efficiency. The only direct contact should be through the TEC plate, everything else should be isolated.

Isolate the fan from the heatsink by the use of silicon pads (which most good CPU heatsink/fan combos have these days) this will prevent vibration transfer to the sensor. Even bedding the fan onto silicon blobs on the heatsink works pretty well as isolators. Those screws shown will transfer vibration to the coldfinger.

Where have you put your temperature sensor (or did I ask this before)?

Will be interesting to see how your test data looks. If it's a 12V TEC I'd be running it at that, or pushing it through a PWM where it will still get 12V but current gets varied, I don't know if you will get much out at a lower voltage. Rowland would know.
Reply With Quote
  #56  
Old 29-11-2015, 02:52 PM
ZeroID's Avatar
ZeroID (Brent)
Lost in Space ....

ZeroID is offline
 
Join Date: May 2010
Location: Auckland, NZ
Posts: 4,949
Think I'm gonna need a bigger heat sink. Although I seem to be putting quite a bit of heat out through the existing small sink it does not seem to be making any difference at the sensor although it may be actually holding ambient (26*-27*) and not climbing up into the 30's.
Finger doesn't feel all that cold to the tough, just cool. Ran at 5 volts first but no noticeable effect so went up to 12 volts on the TEC and fan. Still nothing.

Slightly disappointed so far. I'll switch over to my copper pipe stack used previously, it's quite lightweight and can really soak it up. Back to the drawing board.

The one screw that does go through the finger is protected and has clearance, no contact. The rest of the finger is in a foam sandwich, it doesn't even touch the camera chassis. First hardware contact is the sensor gap. Temp reading is coming from BYE output.
I'll get to the fan isolation later although this wee thing is pretty good especially at 5 v.
Reply With Quote
  #57  
Old 29-11-2015, 03:51 PM
ZeroID's Avatar
ZeroID (Brent)
Lost in Space ....

ZeroID is offline
 
Join Date: May 2010
Location: Auckland, NZ
Posts: 4,949
Version II.
Bigger Heat sink, bigger fan. Also cut away some alum at back so I can see the screen.
The finger is that little show of red sandwiched between the TEC and the foam in the last pic. After the TEC the next thing it physically touches is the back of the sensor.
Attached Thumbnails
Click for full-size image (HTSK204314.jpg)
131.1 KB24 views
Click for full-size image (HTSK204315.jpg)
161.8 KB19 views
Click for full-size image (HTSK204316.jpg)
139.2 KB17 views
Reply With Quote
  #58  
Old 29-11-2015, 04:06 PM
glend (Glen)
Registered User

glend is offline
 
Join Date: Jun 2013
Location: Lake Macquarie
Posts: 7,051
Now that is a serious heatsink/fan combo! I don't trust the EXIF temp info because I have heard the sensor used for that is located away from the sensor itself. If your using a 12706 TEC your going to need 6 amps to get it really going. A few seconds at 6amps and you shoud have good cooling on the cold side immeadiately provided the heatsink fan is running to extract. Where is Rowland, he knows all about the voltage and current requirements of these things. Even throttled way back that fan/heatsink shoud be giving you good cooling that you can feel straight away.
Reply With Quote
  #59  
Old 29-11-2015, 06:14 PM
ZeroID's Avatar
ZeroID (Brent)
Lost in Space ....

ZeroID is offline
 
Join Date: May 2010
Location: Auckland, NZ
Posts: 4,949
Ok, confused but I think you may be right about the sensor not being associated with the sensor. That got the finger down seriously cold ! I could feel it with a finger tip on a top edge but yet the image temps coming back from BYE stayed around ambient .. !!

So how come the reported temps go up with long exposures. I did a 10 x 60 sec darks run, reported temp, 27* for all images which was about ambient. I'm going to have to find an independent temp sensor and get it in there.

That radiator fan combo is what I used to do my earlier cold box experiments with. It also can drop the ZWO to about -20* very quickly. I've got plenty of amps and volts to play with, the Ob is on mains supply and I have a couple of grunty home built power supplies.
Reply With Quote
  #60  
Old 29-11-2015, 07:31 PM
ZeroID's Avatar
ZeroID (Brent)
Lost in Space ....

ZeroID is offline
 
Join Date: May 2010
Location: Auckland, NZ
Posts: 4,949
OK, ordered one of these and 3 spare probes
http://www.dx.com/p/tm902c-1-9-lcd-t...8#.VlqyL03ou70

Got it because the supplied probe is small. Got 3 extra probes so I can build them into future projects and just swap the display unit to whatever requires it.
Reply With Quote
Reply

Bookmarks


Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 06:11 AM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement
Testar
Advertisement