QHY5L-II Mono Options for Passive / Active Cooling

[tonedeaf]

so the ice age has begun!

fixed the camera to the scope for a test run

but had (maybe) stray light or whatelse...

need some more thest runs

gain 90% 1200s fan on fullpower (had ice at the beginning on the coverglas of the chip )

[SnakeyJ]

Tonedeaf, if you can get a couple of temp probes and get or make up a Peltier controller you should be able to setup some setpoint temp control on the camera - think the freezing up/ice build up will be much worse once outside in a colder environment with higher humidity.

[tonedeaf]

you´re right...

this will be the next step...

have made some test runs, and the ice build-up just happens shortly ofter getting the camera down from the scope -> to take a peak on the chip if ice build up

regarding the stray light i found out that an 1,25->2" adapter had some guilt in this matter...

precautionary i temporary installed a tape on the led, as also "light" sealed the back of the camera

[tonedeaf]

so after some testruns with darks (it´s raining like "§$%§ here in austria)

i´ll give a summary for anyone who is crazy enough to mod this camera this way:

with the "voltage" controller for the fan, it´s best to run it at ~1/2 "power" @ ~22°C room ambient.

(due to dew/icing problems on the chip)

on the dark testruns i´ve found out the best gain seeting with not to high noise build up would be ~80-90% gain

(even better with "line noise reduction <- have to try this on some DSO objects)

the 2"-1,25" SW eyepiece adapter has to be tightened hard not just by hand, otherwise it will hold only on two points.

the bright areas on the darks are resulting from heat from some pcd elements (seems to be)

tried to "insulate" some resitors near the cmos, but without big effect

the insulation (as seen on the photo further above) will freeze everything inside (with fan at 2000rpm full power)

so it´s only there to get the "needed" rpm down (-> forces through unbalance are squared -> 1/2 of rpm will result to 1/4th of the forces, therefore lowering the vibrations!)

further you have to screw the extension to the camera to get the 55mm back-focus of the 0,85ff/reducer

oh yes, on the darks testrun i´ve seen a big increase in noise from 600s (acceptable) to 1200s (whooa don´t do this) exposures

so at first would recommend the max exposure at around 600s

i think thats all...

i hope for clear skies...hoping to give some test dso´s

[SnakeyJ]

Look forward to seeing some real images back on this once your weather improves. I had a quick go at capturing M13 last night as a prior to mod comparison, and although I'm having some issues with DSS and stacking the frames the camera certainly seems sensitive enough, yielding 70% histogram @ F5 on the Exp 200P (red filter, 5 sec exposure and 99% gain). I certainly had a few hot pixels, but nothing in comparison to my previous attempts with the Pt Grey Firefly mono - look forward to testing post mod and see if I can get a reasonable LRGB image.

Unstacked single frame from capture:

[SnakeyJ]

Just found an interesting alternative for the Thermal Pad, when searching for reviews of different TIMs - Phobyas self adhesive XT Thermal pad, with a claimed 7W/mK which is close to traditional thermal paste and 7 times better than the previous pads I was using. I believe Arctic Silver 5 is rated at 8.9W/mK It's also non conductive and nice and thin:

Phobya XT 120 x 20 x 0.5mm Self Adhesive Thermal Pad available from http://www.specialtech.co.uk/spshop/customer/Phobya-XT-120-x-20-x-05mm-Self-Adhesive-Thermal-Pad-pid-14140.html

I've ordered a sheet and will do a quick comparison against the standard thermal pads that come with the QHY5L-IIs.

[tonedeaf]

oh that seems a nice info...

btw nice picture for 5s...

...looking forward for more

[Gonzo]

excellent, I've been looking for something like this as I also have a mono QHY5L-II.

[SnakeyJ]

Cheers Tonedeaf - meant to add that the Arctic Silver 5 is electrically non conductive (however they do advise keeping it away from sensitive electrical contacts due to capacitance properties). However the original Arctic Silver is conductive.

Cheers on the image - quite noisy, but better sensitivity than I expected on a relatively short exposure - I finally managed to get my capture to stack in DSS and got a surprisingly good image

[tonedeaf]

http://stargazerslou...ii-hypercooled/

so here my "first" light with the cam...

added a temperature controller

yes and it went from +23 to nearly -12°C

[Gonzo]

Have you done any more testing with the cooled camera? I'm tempted to do a similar mod, but a little scared about the icing issue.

[tonedeaf]

the weather and work was a bit in my way to do more testing

this is a sample of 15x300s with -2°C cooling guided with my C270 logitech cam

M101 15x300s gain90 -2°C log.tif

had some issues with a damage i´ve had (where i screwed a screw to deep , now i know where the saying "you´re screwed is comming from)

had to repair it...at least it went good

(you can see the damage on the lower right side)

but with the "protection" glas cap on i haven´t any icing on the chip itself...

have to do some more testing

[SnakeyJ]

Nice M101 tonedeaf, the guiding is definitely working very nicely with nice star shape and the cameras sensitivity looks good with quite a lot of detail comming through on M101. There is far less noise to the top left of the image and it seems to increase on a gradient down to bottom right - not sure if this represents local LP, or perhaps uneven cooling of the sensor, or possibly some ice? Did you take any dark/flat frames for this - would be interesting to see if this gradient is also present on the calibration frames.

I've got all my parts ready now - just waiting for a friend to machine some heatsinks for me - but hopefully I can be built and testing for next weekend.

[tonedeaf]

there was deffinitly ice or dew...

(wanted to peak, but i´ve got the cam focused and wanted some results )

here´s a dark that i made afterwards i had to re-repair the pcb (had connection errors at the end of the session, as i´ve found out the silver paint stick which i used for repair for one connection, had peeled of a bit and caused it)

there is only a bit of amp-glow...

D_gain 90 -2°C ohne line_Bin1x1_300s_na_2013-06-09_15-27-20.tif

oh btw the amp glow was a bit "better" after i "sealed" the pcb with hot glue

[tonedeaf]

oops forgot to say (can´t edit, but i´m only 70 posts away from it )

it´s a streched dark 0-255 linear...with fits liberator...

[SnakeyJ]

That dark is really very good, assuming the same 300s exposure and 90% gain @ -2C - certainly manageable amounts of noise that should callibrate out. I'm really looking forward to having a proper go with this camera and hopefully getting some LRGB data on one of the brighter messiers, but don't yet have a guiding solution so will probably be limited to shorter subs. Also using the 200P on HEQ5 it will need to be very calm conditions (I had about 65% subs ruined by wind/shake on my first attempt at M13).

Hope you can get that damaged track repaired successfully. As a future alternative to hotglue for waterproofing, I got a Silicon based Conformal Coating spray from my local electronics shop. Whilst not imersion waterproof this provides sufficient protection against condensation build up and is a relatively thin/insulating layer so shouldn't trap too much heat in non heatsinked components. As its a spray you need to mask up any off board electrical contacts before applying. If you can seal the cmos enclosure a small sachet of dessicant should prevent any internal condensation, but I favour the belt and braces approach. Here's one that Cath recommended to me in my firefly cooling thread, http://www.plastidip.co.uk/eStore/index.cfm?type=Home_Solutions/PlastiDip&stage=1

[tonedeaf]

regarding guiding...the C270 (from Logitech) debayered makes a wonderfull job...

...a lot of noise, but at 3sec exposures i´ve got plenty of guidestars to choose from. (not supirior to the qhy5II, but for ~ € 20,- mod casing included it´s great!)

the damaged track is allready repaired...i think that the lights will be also better, as i´ve got already troubles at short exposures before i´ve taken the lights.

the reason for the track was perhaps that i´ve applied the hotglue to early, and as it shrunk -> connectivity trbl!

also have to admit that the lights were taken without an re-adjusted tripod (bumpy cats), so polaralingment was poor, i just wanted some lights to see if it works...

...as it looks like mabye in the next days i´ve got a chance...but for now the humidity is realy high here...

[SnakeyJ]

Fingers crossed - we've got 6 days of cloud forecast in southern UK, though on the plus side this is perfect modding weather

Hope to get the modded heatsink back tomorrow and rumours of firecapture 2.3beta11 for the end of the week, which may fix the LX hangs with the QHY5L-II.

[SnakeyJ]

Quite a long delay in progressing this thread - partly due to the excellent run of weather that allowed some real astronomy and partly due to asking a friend to bore out a heat sink to fit around the outside of the QHY5L-II. They took an awful long time and didn't quite manage the required diameter, which left me trying to widen it out with minimal tools. I finally managed with a mandrell of broom stick, wrapped in tape and then fine wet & dry. This being held in a vice, with the heatsink free to rotate around it - after several hours getting nowhere fast, I added some power with a cordless drill and wrapped flap wheel providing an automated spinner, which worked surprisingly well. The finished heat sink is a snug fit to the outside of the camera and should provide some additional ambient cooling.

The other factor delaying testing is bugs in the qhy5l-ii drivers which caused frequent hangs in LX modes and considerable frustration. QHY have just released an updated version of the API to developers, that works around the issue by increasing timeouts - its still clunky and not entirely stable, though much improved on the standard. Hopefully the next versions of Firecapture and Sharpcap betas will ship with this - though I'm really waiting for updated drivers to cure the problem properly.

I've done four runs tonight to compare dark frames - all at 12bit full frame, 30s exposure and 264 gain.

1 run with an unmodded qhy5l-ii mono

1 run with an unmodded qhy5l-ii mono + passive heatsink

I've then modded the camera by replacing the stock thermal pads with the Phobyas self adhesive XT Thermal pad 7W/mK (previously mentioned) and repeated both runs with and without the heatsink.

Not convinced there's any obvious difference in the results, but these were done inside with an Ambient temp of 22.5C, which is not going to help greatly.

I'll post the dark frames tomorrow after having a proper look - otherwise this now leaves me ready to try with the small 2V Peltier(s) (TEC101706).

[SnakeyJ]

Hmm - not convinced by the passive heatsink, this actually seems to increase noise levels (pre and post thermal pad mod) - somewhat unexpected, but I will try grounding it to make sure I'm not introducing any RF noise....

[SnakeyJ]

Quick pics of the heatsink modded cam fitted to my scope for last nights dark frame tests:

Front view:

Rear view::

May not be terribly effective, but certainly looks the part

[SnakeyJ]

Light leakage:

Whilst taking the dark frames last night, I noticed that I was getting quite a lot of light leakage in through the rear end of the scope. I brought this s/h with Bob's knobs and no cover plate fitted - I haven't noticed the effect visually before and had left it naked to aid cooling of the primary. Shooting the dark frames at 30 second exposure has revealed quite a lot of light getting in. I bodged a temporary fix - but need to manufacture a neater solution with some cooling fans.

Not pretty, but did the job! - There's a dew tape/band fitted around the rear mirror cell.

[tonedeaf]

hi,

nice cooling...

but i think with such small chips a passive cooling would only be effective at taking pictures of the sun, where i heard of chip temperatures of +50°C.

the delta t isn´t high enough for passive cooling...(i think so)

oh have seen this recently....

http://www.conrad.at/ce/de/product/235442/Reely-Kuehlkoerper-fuer-540er-Motor-mit-Ventilator-Ausfuehrung-Mittigsitzender-Ventilator-Farbe-Rot

with a small mod it should be attacheable to a the qhy5II....

[SnakeyJ]

Cheers Tonedeaf,

Just trying to eliminate the passive cooling options and get some useful comparison data in before making any more serious mods to the camera.

I agree that the passive cooling, will make very little difference with low gain LX images. But I think it could provide a useful reduction for high gain planetary and certainly solar imaging, though in both these areas we are generally stacking large numbers of frames which is also removing a lot of noise. It also needs a strong negative thermal gradient, low ambient temperature to be effective.

The RC 540 motor coolers look quite a good fit for 1.25" diameter and are a good price - I had looked at these, but thought a bigger surface area and uniform cross secion would dissipate heat more effectively/evenly.

I'm firmly convinced that active cooling is the way ahead with this sensor and your results look very promising - just hoping that QHY can resolve the driver issues by the time I get this together, or perhaps I'll have to try the ascom driver route. It would also be nice to get the sensor temperature, which I believe is now available for this sensor in the ZWO SDK, but not currently implemented in the QHY's SDK.

[tonedeaf]

yeah...the driver issues... -.-

and maybe a temperature readout of the chip!

somewhere here was a discussion regarding low gains on these chips...

...there seem to be a problem under 9% gain...but i haven´t tried it yet