ZWO Peltier Cooler

[cjdawson]

Just now, Dr_Ju_ju said:

Somewhere on there, are some A4 stainless steel ones, together with some M4 dome head 30~40mm & matt black paint make wonderful 'Bob's Knobs'...

I deliberately want to stick with plastic screws.  The reasons for this are as follows.

1. Nylon bolts will not damage the threads on the back of the camera.

2. Nylon bolt will flex, when the aluminium expands and contracts.

3. The CPU cooler mount is plastic, so I want to keep with the same material for the mounting.

What I don't want is to end up in a position where the bolts can corrode - and after a number of years that can happen.  Which may have the effect of welding the bolts into the camera.

[cjdawson]

Just had a parts delivery today.

   

The laser cut aluminium parts have all arrived.  I've done a few quick checks and things seem to be working a bit differently so far.

I'm going to use the 6mm square block rather than the 8mm as it appears to be quite a tight fit when I dry fit everything together.

 

My changing out the wingnuts for the thumb screws has also had the desired effect, there's enough clearance for the USB3.0 plug to get into the camera.

I just need to do a little cleaning up of the parts, removing the burrs, and putting tapping the thread into the camera plate.   Think I'm going to have a practice first on another piece of aluminium that I have just to be sure that the tool is good for the job, really don't want to ruin that part.

 

The only parts that I'm waiting on now are the m4 alan head nylon bolts - turns out I'd ordered the wrong bits (had ordered m5 and they arrived today) and I'm still waiting on the little plastic box that will go onto the main frame.

 

I'm starting to get excited now as I can see this build coming together really fast.

[1CM69]

16 minutes ago, cjdawson said:

Just had a parts delivery today.

   

The laser cut aluminium parts have all arrived.  I've done a few quick checks and things seem to be working a bit differently so far.

I'm going to use the 6mm square block rather than the 8mm as it appears to be quite a tight fit when I dry fit everything together.

 

My changing out the wingnuts for the thumb screws has also had the desired effect, there's enough clearance for the USB3.0 plug to get into the camera.

I just need to do a little cleaning up of the parts, removing the burrs, and putting tapping the thread into the camera plate.   Think I'm going to have a practice first on another piece of aluminium that I have just to be sure that the tool is good for the job, really don't want to ruin that part.

 

The only parts that I'm waiting on now are the m4 alan head nylon bolts - turns out I'd ordered the wrong bits (had ordered m5 and they arrived today) and I'm still waiting on the little plastic box that will go onto the main frame.

 

I'm starting to get excited now as I can see this build coming together really fast.

Aah! Memories. 

It’s  an odd feeling for me seeing this coming together, good feeling though. 

Looking forward to your updates. 

[Gina]

I have my DIY cooled ASC up and running <CROW>  ? ?

[cjdawson]

Nice Gina, but was that posted in the wrong thread?

[Gina]

It shows that the simpler approach of just putting the Peltier TEC in contact with the back of the camera works fine.  With an exposure of 60s every 5 minutes (4m delay between exposures) the camera temperature gets down to -9°C and at the end of the exposure where the camera generates most heat itself its at -5.5°C.  This is well enough to stop noise and hot pixels with the uncooled version of the ASI185MC.  Jus' sayin' ?

[cjdawson]

that might help me with a decision that I'm thinking about....    The order of parts   It's not clear in the instructable, and I think there's some wiggle room here.

 

Here's my two options...

 

1. Camera -> backplate -> al block -> peliter ->  cpu heatsink.

2. Camera -> backplate -> peliter ->  al block  -> cpu heatsink.

 

The different between the two is quite sublte, but may have a big impact overall.

Putting the peliter on the backplate means that the cold source is much closer to the camera.   However, the heat has to pass through a block of metal before getting to the heatsink

it mean that the cold will all be inside the insulation.

 

 

[Gina]

The less metal the heat has to travel through to get from sensor to Peltier TEC the better - everything has thermal resistance.  Internally the ZWO cameras have a thermal path from the back of the sensor to the camera casing to cool the sensor.  Cooling, albeit just fins in the casing, is needed even for planetary imaging.  The camera generates more heat during the capture time than idling and this heat has to be got rid of.  Therefore, directly cooling the camera casing has to be the most efficient method. 

Having cooled the casing you don't want heat getting in from the surroundings, so thermal insulation is needed.  These are the principles I have used in my latest ASC and I'm hoping this will be the final model.  With that in mind I have started a new blog describing just the construction of this ASC without all the trials and tribulation in getting there. I hope this information will be useful to others (the reason for posting it).

 

[1CM69]

21 minutes ago, cjdawson said:

that might help me with a decision that I'm thinking about....    The order of parts   It's not clear in the instructable, and I think there's some wiggle room here.

 

Here's my two options...

 

1. Camera -> backplate -> al block -> peliter ->  cpu heatsink.

2. Camera -> backplate -> peliter ->  al block  -> cpu heatsink.

 

The different between the two is quite sublte, but may have a big impact overall.

Putting the peliter on the backplate means that the cold source is much closer to the camera.   However, the heat has to pass through a block of metal before getting to the heatsink

it mean that the cold will all be inside the insulation.

 

 

In my instructable it is #1 - read Step 5 Main Structure (1 to 11)

I did try the version #2 while in my design stage as you set out but found that it was more difficult to remove the heat from the hot side as effectively.

 

[Stub Mandrel]

2 hours ago, Gina said:

It shows that the simpler approach of just putting the Peltier TEC in contact with the back of the camera works fine.  With an exposure of 60s every 5 minutes (4m delay between exposures) the camera temperature gets down to -9°C and at the end of the exposure where the camera generates most heat itself its at -5.5°C.  This is well enough to stop noise and hot pixels with the uncooled version of the ASI185MC.  Jus' sayin' ?

What power are you using, I found passive cooling with a small heat sink/no forced air was pointless.

[Gina]

For the longer nighttime exposures I'm using 13.8v at 1.4A on the Peltier TEC which is 19W.  See my blog last entry at this time.  Hot side cooling is provided by a large (94mm x 94mm x 69mm) passive cooler.

 

[cjdawson]

 

1 hour ago, 1CM69 said:

In my instructable it is #1 - read Step 5 Main Structure (1 to 11)

I did try the version #2 while in my design stage as you set out but found that it was more difficult to remove the heat from the hot side as effectively.

 

I was wondering about the ability to get the heat from the hot side.  Having the peltier on top of the block looks like it will also make the cabling just that little bit easier.  Thanks for telling me about the test you did, that's pretty much sealed the build order.

 

12 minutes ago, Gina said:

For the longer nighttime exposures I'm using 13.8v at 1.4A on the Peltier TEC which is 19W.  See my blog last entry at this time.  Hot side cooling is provided by a large (94mm x 94mm x 69mm) passive cooler.

 

That's a much bigger heat sink that you are using.   At this point, I'm going to stick with the instructable, I've got nearly all the parts and assembly is imminent.

[Gina]

I shall be interested in your results - good luck with the build.

[cjdawson]

And we’re off.   Gasket made, aluminium cleaned up, and the centre hole tapped

[cjdawson]

Insulation jacket mostly made, just need to do the cut outs for the USB port and so the excess  can clear the main frame, that’s a job for later.

 

[cjdawson]

Started work on the control box

and here’s where two minor modifications

rather than a cable gland, I’m using a plug aviation plug and socket.   Also the dc power supply plug and sock has a locking thread

 

[cjdawson]

Finally have that blasted threaded copper bar sorted out  

 

 

in the end I bought a Vorkel 1/4 unc die and a die holde4 for it.   Not to mention another copper rod.   Chamfered the end and the thread went on with no problem.   I’m putting the problems that I was having down to a bad quality tool that I was trying to use.   My fault for buying it in the first place.   The difference is stark, this new die bit into the copper immediately.    Much happier with this now.   Next step it to turn it into the lug nut.

[Stub Mandrel]

Sorry not to have done the one you sent me yet!

I have to write an article on thread cutting in the lathe next week so I was planning to use it as a worked example!

[cjdawson]

@Stub Mandrelit’s all good.   It was frustrating me that I wasn’t able to do something that should be simple.   So figured that big out.

cutting a thread on the lathe should be “fun”.   

[cjdawson]

Finally making some progress on this, here’s the construction so far.  The cooler itself is together now.

 

abvove shows the first change, using an 8mm block instead of the 6mm.   Added some washers between the cpu cooler frame and the aluminium main plate.   This means I don’t need to modify the cooler in any way.

 

thr second change is that I’m using an aviation 6pin socket instead of a cable gland.   This will let me detach the cable between sessions and leave the cooler in the scope permently.

[cjdawson]

I turned my attention to the control box now.  However, I realised that I didn't have any wire that's rated for 5 amp.  So, put in an order on ebay, should be here by the weekend.

Next I turned my attention to wiring up the cable that will go between the cooler and the control box.   I've got to rethink that as well.  Turns out the 7 core cable that I orginially ordered, same as on the original project, is too big to fit into the GX16-6 plugs.   So I've got some new cable on order,  hopefully that won't take too long.

The next part of the project that I think will be worth doing will be to write down the wiring diagram, that way I won't make a mistake when it comes to wiring everything up.

[cjdawson]

The new 7 core cable showed up.   It's small enough to fit into the GX-16-6 plugs.  The equipment wire has also arrived.   Onwards and upwards.    Next steps will be to put together the control box, the power cable and the 7 core cable.

Lots of soldering and then it'll be time for a test run.

[cjdawson]

Finally making progress on the control box.

 

In the end I’m using 6 wires in total from control to cooler.     2 for the sensor, 2 for the fan and the other 2 for the cooler itself.

[cjdawson]

 

its alive!!!!!!!!

 

got if doing the first test run right now.    Starting temp was 21.6, so it’s headed in the right direction.

[cjdawson]

It's been running for a while now, but it's not cooling very fast, it's down to about 17.5 now.

I wonder if this is something to do with the TES1-12704 peltier module, or maybe the thicker aluminium block.

Maybe it's because my peltier module is only 30mm across.  If I can find one that's 40mm maybe that'll cool the alu block better.