How to Get Less Noise

[Dragon_Astro]

Having been looking at cooling my camera, I've found those TEC units too. I also found the KKMoon temperature relay units on Amazon, which enable you to actually set a target temperature to cool down to and cut off the cooling when it hits target.

As for cool boxes, the other day I experimented putting my Hypercam 183 in an ordinary cool bag for lunches, along with some frozen chicken fillets (in bags). The Hypercam has a fan inside, which sucks in air from the side and blows it out of the back after passing around the sensor.

From room temp of 23C, I got the cam down to 5C in half hour. It made a big difference to noise in the test dark subs.

Not bad from just chicken fillets lol

[JohnSadlerAstro]

17 minutes ago, Dragon_Astro said:

Having been looking at cooling my camera, I've found those TEC units too. I also found the KKMoon temperature relay units on Amazon, which enable you to actually set a target temperature to cool down to and cut off the cooling when it hits target.

As for cool boxes, the other day I experimented putting my Hypercam 183 in an ordinary cool bag for lunches, along with some frozen chicken fillets (in bags). The Hypercam has a fan inside, which sucks in air from the side and blows it out of the back after passing around the sensor.

From room temp of 23C, I got the cam down to 5C in half hour. It made a big difference to noise in the test dark subs.

Not bad from just chicken fillets lol

Now all I need to do when I want some nuggets for supper is put them in to cook with the sensor!

John

[Dragon_Astro]

Poultry based cooling could catch on lol Jokes aside, it does show what could be done with a bit of diy.

[david_taurus83]

Cold finger is the way to go but if you don't want to butcher your camera or faff with a cool box you could try the cold plate on the back if you have a flip out screen. I have just made one of these like below for mine. Tested indoors the other night and it can get to approx 15°C below ambient. I done a 900s dark with no cooler and EXIF temp was 28°C. Ran the cooler for an hour and done another 900s and temp was 16°C. So does work. Should work better outside. As it happens, there is a promised gap in the clouds tonight (fingers crossed!) So assuming I can get an old battery charged in time (Needs circa 60w/5amps per hour) I will be testing it out in anger!

 

 

[JohnSadlerAstro]

Hi,

Thank you so much for all the advice, suggestions and ideas, everyone!

As I wont be able to do any mods within the next few weeks, I will probably just go down to ISO 400 for now, and use the brightness histogram to make sure the peak is separated by about 10% from the left hand edge. I'm guessing this will probably mean about 5-6 min subs, but that should be ok. With AG, theres not much difference. ISO 200 just seems a bit too low to me, although I will do a test sub and just see if it's practical.

So all I need to look for is the histogram peak leaving the left hand edge a small way? Should I use the overall luminance histogram, or make sure the individual RGB peaks have all moved away? With a modded cam the Red always goes to the moon before the blue and green even start!  

John

[Whistlin Bob]

2 hours ago, david_taurus83 said:

Cold finger is the way to go but if you don't want to butcher your camera or faff with a cool box you could try the cold plate on the back if you have a flip out screen. I have just made one of these like below for mine. Tested indoors the other night and it can get to approx 15°C below ambient. I done a 900s dark with no cooler and EXIF temp was 28°C. Ran the cooler for an hour and done another 900s and temp was 16°C. So does work. Should work better outside. As it happens, there is a promised gap in the clouds tonight (fingers crossed!) So assuming I can get an old battery charged in time (Needs circa 60w/5amps per hour) I will be testing it out in anger!

 

 

Please let us know how you get on! I've just picked up a modded 600d (first light tonight?) and was planning to do the same. I'm not brave enough to try @Stub Mandrel 's approach. I suspect I'd brick my camera!

[Scooot]

8 hours ago, JohnSadlerAstro said:

So all I need to look for is the histogram peak leaving the left hand edge a small way? Should I use the overall luminance histogram, or make sure the individual RGB peaks have all moved away? With a modded cam the Red always goes to the moon before the blue and green even start!  

John

I’d guess the former John, make sure all the channels are exposed. Otherwise a colour would be clipped?

I’d also like to know for sure so hopefully someone will put me right if this is not the case.  

[david_taurus83]

18 hours ago, Whistlin Bob said:

Please let us know how you get on! I've just picked up a modded 600d (first light tonight?) and was planning to do the same. I'm not brave enough to try @Stub Mandrel 's approach. I suspect I'd brick my camera!

Hi. I tested this out last night. Ambient temp outside was around 7°C and I got the cooler down to 0.3°C. I did expect a bit more but unfortunately the leisure battery I rescued from the recycle bin at work wasn't fully charged so the cooler was only running off 11v. More tests needed! EXIF temp on my subs were around 10°C after I realised I still had the LCD screen on and switched it off! Hopefully with a full battery and starting off with the screen off I can get colder subs. Also worth noting, next time I'm going to try and focus without using live view as this increases the temp substantially. 10s exposures behind a Bahtinov mask will be the way forward I think.

[Stub Mandrel]

I ordered one of these KK Moon things, I got tricked into signing up to a free trial of Prime, so I had to cancel that ASAP but it looks like I might still get a month free.

The box came this morning - argh! 220V! I shoudl have checked.

Just opened it up and it's got a sealed transformer to a diode bridge and standard 7805 regulator.  I can just remove the transformer (add to bits box) and wire the 12V straight to the bridge rectifier and I'll even be reverse polarity protected ?

If I attach the sensor to the cold finger just outside the camera case  it shoudl read within a few degrees of the temperature at the sensor.

More wires!

[tooth_dr]

38 minutes ago, Stub Mandrel said:

I ordered one of these KK Moon things, I got tricked into signing up to a free trial of Prime, so I had to cancel that ASAP but it looks like I might still get a month free.

The box came this morning - argh! 220V! I shoudl have checked.

Just opened it up and it's got a sealed transformer to a diode bridge and standard 7805 regulator.  I can just remove the transformer (add to bits box) and wire the 12V straight to the bridge rectifier and I'll even be reverse polarity protected ?

If I attach the sensor to the cold finger just outside the camera case  it shoudl read within a few degrees of the temperature at the sensor.

More wires!

Can you send me a link please?  I've a spare 40d that I was going to sell, but I might just take it apart, mod it and cool it.

[Dragon_Astro]

I bought one of those KKMoon Tec units too, arrived this morning lol The instructions look "interesting" ?

Saves on the chicken though lol

[tooth_dr]

9 minutes ago, Dragon_Astro said:

I bought one of those KKMoon Tec units too, arrived this morning lol The instructions look "interesting" ?

Saves on the chicken though lol

Have you the link handy?

[JohnSadlerAstro]

Hi,

Well I tried imaging last night at ISO 400, it really seems to have helped. The galaxies stand out a lot better despite the fact that I only got 4 subs in, although the subs had to be 4 mins long to get the same brightness as a 30sec ISO 1600 one. I'm happy with it as a temporary fix, until I can work out this cooling mod.

Unfortunately there was a periodical dec catastrophe, I think the worm gear isn't quite straight, ill need to sort that out. Do tell me how you get on with those cooler units, I have a £10 amazon git voucher!  

John

[Stub Mandrel]

1 hour ago, tooth_dr said:

Have you the link handy?

Just google KK Moon temperature

Here's the 12V version - cheaper too!

https://www.amazon.co.uk/KKmoon-Temperature-Controller-Thermocouple-Protection/dp/B00HC6PDAW

[Dragon_Astro]

7 hours ago, tooth_dr said:

Have you the link handy?

This is the unit I bought for the cooler I have planned....

https://www.amazon.co.uk/KKmoon-Thermoelectric-Refrigeration-Semiconductor-Conduction

Only had a 9v battery to try it with, but it all worked and the Tec unit turned cold/hot.

[JamesF]

5 hours ago, Stub Mandrel said:

Just google KK Moon temperature

Here's the 12V version - cheaper too!

https://www.amazon.co.uk/KKmoon-Temperature-Controller-Thermocouple-Protection/dp/B00HC6PDAW

Ah, now.  These I know a little about...

There are lots of knock-offs kicking about and if you just want a straight set-point cooling type thing probably any of them will do.  However, some of them (and I can't recall off the top of my head how to tell the difference, but I have some of both types) have the hardware left in a programmable state so you can upload your own code to them if you wish.

I know this because they're also used for temperature control by homebrewers, and there's a new firmware image created for that purpose that allows stepped temperature control and all sorts of other good stuff.

I use one for controlling the temperature of my fermentation cabinet.

James

[JohnSadlerAstro]

11 hours ago, JamesF said:

I use one for controlling the temperature of my fermentation cabinet.

Yes please.

The power supply is the main bit im concerned about now, I don't know how power hungry these would be? Otherwise I could make a bank of my rechargeable D cells and power it off that.

John

[david_taurus83]

1 hour ago, JohnSadlerAstro said:

Yes please.

The power supply is the main bit im concerned about now, I don't know how power hungry these would be? Otherwise I could make a bank of my rechargeable D cells and power it off that.

John

They use about 5 amps an hour. So a 60ah leisure battery will run it for approx 10/12 hours.

[Stub Mandrel]

1 hour ago, david_taurus83 said:

They use about 5 amps an hour. So a 60ah leisure battery will run it for approx 10/12 hours.

I suspect mine uses more than that!

[david_taurus83]

I've made 2! I made a coolbox using the 1 linked to Amazon above with the 2 heatsinks and fans. The coolbox is purely for maintaining an ambient temp while creating darks to match the EXIF temp of my imaging sessions. I bought some 5.5mm x 2.1mm power plugs to connect power and a 6amp 12v dc converter. Works a treat though I reckon I will have to put it outside the back door to get to those colder temps in winter! I ran it of my Tracer 12ah power bank and it ran for 3 hours easily but voltage dropped off to 11°C. I will have to post up some pics later.

[dph1nm]

On 11/09/2018 at 08:33, JohnSadlerAstro said:

I've noticed quite high noise levels at ISO 800, is this the optimal speed to use? Would it be better to capture subs at ISO 400, or would they then be really underexposed? Would I be better doing 3 min subs at ISO 1600?

There seems to be some confusion in this thread.  So the only real difference between ISO400 and ISO1600 is that the read noise is lower at ISO1600 (for the Canon 1000D) and the dynamic range is greater at ISO400. The amount of light you collect is the same at either ISO. So to overcome read noise you will need somewhat longer subs at ISO400 than at ISO1600. However, If your subs are already long enough to overcome read noise then you may as well use ISO400 to get the higher dynamic range - but the final signal to noise will be the same for either ISO, if your total exposure time is the same.

NIgelM

[Scooot]

I've been reading about the Pixinsight Calibration process (which I use) and there's been a lot of discussion about cooling to reduce thermal noise in this thread. The Pixingsight Aquisition notes seem to suggest to me that the temperature of the lights and darks won't affect the noise in the integrated image because the calibration process deals with it? I've probably misunderstood but wonder if someone could put me right.  

https://www.pixinsight.com/tutorials/master-frames/

[Stub Mandrel]

22 minutes ago, Scooot said:

I've been reading about the Pixinsight Calibration process (which I use) and there's been a lot of discussion about cooling to reduce thermal noise in this thread. The Pixingsight Aquisition notes seem to suggest to me that the temperature of the lights and darks won't affect the noise in the integrated image because the calibration process deals with it? I've probably misunderstood but wonder if someone could put me right.  

https://www.pixinsight.com/tutorials/master-frames/

Personally, I think approximate temperature matching makes a big difference. PI may have magic wand to compensate for temperature differences, but I doubt it.

[michaelmorris]

1 hour ago, Scooot said:

I've been reading about the Pixinsight Calibration process (which I use) and there's been a lot of discussion about cooling to reduce thermal noise in this thread. The Pixingsight Aquisition notes seem to suggest to me that the temperature of the lights and darks won't affect the noise in the integrated image because the calibration process deals with it? I've probably misunderstood but wonder if someone could put me right.  

https://www.pixinsight.com/tutorials/master-frames/

Image calibration will help take out hot and cold pixels and other fixed pattern noise and the signal generated by IR photons caused by the internal electronics of the camera (i.e. amp glow), but as far as I understand it, not the relatively random noise generated by ambient heat (i.e. IR photons).  That is why astro cameras are cooled.

[Scooot]

16 minutes ago, michaelmorris said:

Image calibration will help take out hot and cold pixels and other fixed pattern noise and the signal generated by IR photons caused by the internal electronics of the camera (i.e. amp glow), but as far as I understand it, not the relatively random noise generated by ambient heat (i.e. IR photons).  That is why astro cameras are cooled.

Thanks Michael, there had to be a good explanation.