ZWO Peltier Cooler

[Gina]

?

[cjdawson]

First of the needed tools have arrived.   That's the imperial tap and die set.  Already had a metric one, but needed the 1/4" thread.

waiting for payday, then I'll be getting the alumium parts ordered,  have checked on the website and it seems that 7mm plate doesn't exist, so I'll most likely order both a 6mm and 8mm 40x40 blocks, so that I have the choice when assembling.

Will be best to order both at the same time, as it'll mean one lot of postage

[Stub Mandrel]

A few random thoughts from my DSLR experience, using a cold finger.

ice (rather than condensation) on the sensor demanded that I fit a heater to the cover glass, after gradually reducing the in-line resistance the dew/condensation stopped and I haven't had any problems. ZWO sell affordable heater straps for their cooled cameras, I'm sure one would fit.

I don't currently use temperature control or PWM.  Using a nominally a 40W peltier with a CPU fan and the dew heater my kit draws about 4.4A on startup which drops to 3.6A after about half a minute and appears to stabilise.

I run it off a 60 Ah leisure battery, but usually attach a charger, especially on long (~6 hour) sessions.

I don't know the temperature but it pretty much kills noise stone dead, so much so that unless my control frames are cooled as well they ADD more fine-grained noise than they remove.

If any part of the cold finger is exposed to the air, it rapidly develops a layer of frost so is well below zero.

The Exifs only show a relatively modest temperature drop (in January it was reporting 1 and-1), but the temperature sensor in the 450D is remote from the imaging sensor and is not directly cooled.

I have it in mind to fit a temperature sensor and manually adjustable PWM (but probably NOT bother with active control) as a way of extending battery life for remote use at a dark site.

[1CM69]

21 minutes ago, Stub Mandrel said:

A few random thoughts from my DSLR experience, using a cold finger.

ice (rather than condensation) on the sensor demanded that I fit a heater to the cover glass, after gradually reducing the in-line resistance the dew/condensation stopped and I haven't had any problems. ZWO sell affordable heater straps for their cooled cameras, I'm sure one would fit.

I don't currently use temperature control or PWM.  Using a nominally a 40W peltier with a CPU fan and the dew heater my kit draws about 4.4A on startup which drops to 3.6A after about half a minute and appears to stabilise.

I run it off a 60 Ah leisure battery, but usually attach a charger, especially on long (~6 hour) sessions.

I don't know the temperature but it pretty much kills noise stone dead, so much so that unless my control frames are cooled as well they ADD more fine-grained noise than they remove.

If any part of the cold finger is exposed to the air, it rapidly develops a layer of frost so is well below zero.

The Exifs only show a relatively modest temperature drop (in January it was reporting 1 and-1), but the temperature sensor in the 450D is remote from the imaging sensor and is not directly cooled.

I have it in mind to fit a temperature sensor and manually adjustable PWM (but probably NOT bother with active control) as a way of extending battery life for remote use at a dark site.

While testing my setup I did find that condensation formed on very few occasions on the front AR glass but never on the sensor itself. 

Through testing I could stop this happening altogether by running the camera without cooling until sensor temp stabilised & then turning on cooling with the camera located in its final location, i.e. not moving it around from one place to another.

Remember that removing the camera from the stable temp of the scope while cooling or while cooler than ambient will cause the AR glass to flash over with dew.

Also the neoprene jacket is a must as the bare camera exposed to the air dews up heavily before completely freezing. The insulation stops any dew forming on exposed parts. 

[Stub Mandrel]

my experience is with a DSLR where the cover glass is a mm or two from the sensor, on the ZWO cameras there's a much bigger air space.

[1CM69]

19 minutes ago, Stub Mandrel said:

my experience is with a DSLR where the cover glass is a mm or two from the sensor, on the ZWO cameras there's a much bigger air space.

Very true ?

[cjdawson]

The ZWO ASI290-MM it rated like this Working Temperature -5°C—45°C.    My plan is to try and get the camera down to -3 using the cooler. that leaves 2 degrees safety margin and should stop noise dead.

In addition, my camera will have the neoprene jacket like in the instructable.   This is why I'm following this design rather than making up my own from scratch, it looks very very well thought out.

I do get where you are coming from with Dew forming, or more to the point ICE.  Have had many an observing session ended because of it in the past.

As for power, then I'm at home, I'm now using mains to power my kit, when I'm in the field I've got a leisure battery.  This cooler is going to be the biggest current draw, and once it's down to temp, that should be reduced because of the temperature gauge.

 

On my setup, the camera is attached permently to the scope, and since it's an SCT, we're talking about a sealed (though not airtight) unit.  I'm going to be making an adjustment to the plans, so that the cable is detachable. This will means that I can happily leave the whole device in place on the scope whilst it cools, warms, and everything else.   The only time it'll be detached is for maintenance.

I've seen the Anti-Dew heater for ZWO cameras...

https://www.firstlightoptics.com/zwo-accessories/zwo-anti-dew-heater-strip-for-asi-cooled-cameras.html

 

This is something that I might consider adding sometime, and I'm sure that it will happily fit on my ZWO. That's a something that I'll think about adding, if my testing shows that it's a worthwhile upgrade ?

[Stub Mandrel]

1 hour ago, cjdawson said:

Working Temperature -5°C—45°C.

I suspect that's over cautious, given the ZWO cooled cameras are capable of 40-45 degrees below ambient...

I estimate my camera manages about -15C, I think I measured -10 pointing an IR thermometer down its throat under not very controlled conditions.

[cjdawson]

Yeah it probably is over cautious, but it's ZWO's figure.  I copied it from the FLO website.

[cjdawson]

Just had a delivery of several parts today.   Now have the EDPN 2mm rubber for the gasket.   The copper rod arrived, and so have the project box for the controller.

This weekend, I'm going to start construction, but putting the thread on the copper rod, and maybe mounting the control panel into the box.   I'll make sure to leave room for the fuse holder, sockets.

[cjdawson]

Here's a quick update on this build.

Have got a choice of aluminium or copper rod now.    However I'm having trouble getting the thread cut.  The problem is that I can't get my 1/4 die to cut a thread, it does bite, but only takes off such a small amount that it is stripping the copper bear.

Starting trying the same with the aluminium, but I'm getting the same result.

This leads me to the part about using a file to get the thread started.   is it sufficient to simple make a flat spot, or should it be something that's some other shape?

I really don't want to wreck the whole 100mm of copper rod, and i'm sure that I'll only need about 10mm to make the lug nut.

Actually, I'm thinking about maybe making another alteration to the design, and cutting the lug nut so that it's long enough to touch the peltier, then putting a thread into the 40 * 40 plate, so that the copper finger is cooled directly from the cooler itself.     But before I can make decisions like that, I need to have a thread on the rod.

 

 

In other news, I've got a price for getting the parts cut, so that's likely to happen next month, when I'm allowed to spend a few pennies.

[1CM69]

41 minutes ago, cjdawson said:

Here's a quick update on this build.

Have got a choice of aluminium or copper rod now.    However I'm having trouble getting the thread cut.  The problem is that I can't get my 1/4 die to cut a thread, it does bite, but only takes off such a small amount that it is stripping the copper bear.

Starting trying the same with the aluminium, but I'm getting the same result.

This leads me to the part about using a file to get the thread started.   is it sufficient to simple make a flat spot, or should it be something that's some other shape?

I really don't want to wreck the whole 100mm of copper rod, and i'm sure that I'll only need about 10mm to make the lug nut.

Actually, I'm thinking about maybe making another alteration to the design, and cutting the lug nut so that it's long enough to touch the peltier, then putting a thread into the 40 * 40 plate, so that the copper finger is cooled directly from the cooler itself.     But before I can make decisions like that, I need to have a thread on the rod.

 

 

In other news, I've got a price for getting the parts cut, so that's likely to happen next month, when I'm allowed to spend a few pennies.

OK, difficult to say what the problem might be other than;

• file a chamfer around the top of the rod to ease starting the thread cutting
• be very gentle as the copper is soft, I did 1/2 turn then turn back each time to clear the swarf.
• obvious but make sure it is a 1/4" 20 UNC die (fairly coarse thread) & also the rod is 1/4" diameter rod

hope this helps

[cjdawson]

13 minutes ago, 1CM69 said:

OK, difficult to say what the problem might be other than;

• file a chamfer around the top of the rod to ease starting the thread cutting
• be very gentle as the copper is soft, I did 1/2 turn then turn back each time to clear the swarf.
• obvious but make sure it is a 1/4" 20 UNC die (fairly coarse thread) & also the rod is 1/4" diameter rod

hope this helps

Wish I could say that helped, however, I've so far managed to take about 4mm off the end of the copper rod, trying to get started.

Think I might try filing a bit of thread manually, to see if I can get enough bite, to help the tool get going properly.

[Gina]

I've found the end of a rod needs filing to a taper to get a die started for threading.  This applies with any metal and even wood though I haven't tried copper.

[Stub Mandrel]

Copper is an evil metal to machine as it is very sticky, as is 'pure' aluminium.

I would try using 6mm rod with your 1/4" die, more likely to get a good result.

Lubricate - light oil for copper, paraffin is best for aluminium and its alloys, but even spit is better than nothing.

As Gina says, a tapered lead on the end of the rod will help.

[cjdawson]

guess I just need more of a taper on the end, it's not all the way through the die just yet, so maybe that'll make the difference

 

[Gina]

I would say so - yes.

[Stub Mandrel]

I haven't got any copper rod, but if you post me an overlength bit I could cut the thread on my lathe for you.

[JamesF]

I had been wondering if it might be easier to do on a lathe.

James

[Gina]

Even if using a die, it's easier on a lathe.  It's important to get the die at right-angles to the work piece.

[1CM69]

I guess that I must have been lucky because I didn’t have an issue at all when I did it. Rod was threaded within minutes & I only put a slight chamfer around the top. 

Have you measured the rod with calipers just to check that it is in fact 1/4”, just a thought because all the struggling you’re having seems very odd & something is definitely not right somewhere. 

[cjdawson]

I've not measured it yet.  All I know for sure is that it does sit into the die ok I think.

I've also got some aluminium rod and am having the same problem with that.  I think (not opened it yet) I've just received a tub of cutting paste.  I know when I was cutting a small thread that using cutting lube made the difference.  maybe this will make the difference on a larger thread.    not quite ready to give up just yet.

[1CM69]

No, definitely do not give up, I’m sure you’ll figure it out. 

[cjdawson]

btw, when I say give up.   I don't mean with the build, I mean trying to put the thread on the rod myself.  If I give up, it'll be a case of taking up @Stub Mandrel on the offer of thread cutting      I'm most definatly going to have the 1/4" threaded code finger, it just makes too much sense to miss.

[Stub Mandrel]

Try with the cutting paste, but if that fails drop me a PM.