EOS 1000D extreme modding

[yesyes]

Inspired by Gina's 1100D cooling thread I took the plunge and bought another used EOS 1000D (not arrived yet though). The idea is to mod that for pure astro imaging.

What I want to do:

- remove IR cut filter

- remove mirror assembly

- make new housing for the camera (not too sure about that one yet)

- peltier cooling the sensor

I want to strip out everything I don't need for astro imaging. I will not be using the view finder any more, so I can get rid of the flip mirror and thus avoid vibrations. I might also leave out the LCD screen but I'm not sure about that yet. It might be good to know what the camera is doing. On the other hand I want to reduce weight as well. I'll also be adding weight with the peltier cooling so any parts that I can get rid of will reduce the overall weight after the mod.

A few questions...

- Filter

I've read a few tutorials and watched youtube videos. What I learned is that I should remove the blue-ish IR cut filter. Do I *have* to replace this with some other filter or glass or can I just get rid of it without replacing it?

What about the other clear-looking filter with the cleaning piezo attached? Do I need to keep that or would it be advisable to remove that as well?

- Cooling

What power should the peltier have?

Should I make it as cold as possible or is there such a thing as over-cooling the sensor?

Has anyone taken apart a 1000D (or similar) further than removing the sensor? I'm interested in whether the lens attachment mechanism is attached to the outer housing or the "mirror box". And whether it is actually possible to remove the flip mirror mechanism (I'm a bit worried that there might be a feedback switch that tells the electronics "yes, the mirror is now up").

Any other thoughts and suggestions are also welcome, of course.

[Spikey]

- Filter

I've read a few tutorials and watched youtube videos. What I learned is that I should remove the blue-ish IR cut filter. Do I *have* to replace this with some other filter or glass or can I just get rid of it without replacing it?

What about the other clear-looking filter with the cleaning piezo attached? Do I need to keep that or would it be advisable to remove that as well?

To my knowledge it doesn't have to be replaced.

[Islander]

I'm fairly sure there will be some sort of feedback from the mirror to the control electronics to verify whether it's fully up or fully down. I reckon you'll have to fool the camera into thinking that it's in the locked up position.

I think you could lose the anti aliasing filter if you wanted it's only there to prevent moire patterning and that isn't likely to be a big problem with astro imaging.

[Horwig]

I took my 450d apart last year, with the intention of cooling it, the body can all be safely removed, as can the pentaprism. The lens mount is part of the mirror box, and the mirror itself can be removed, but the actuating cam, a piece of black plastic on the right side (looking from the front) has a sensor on it I seem to recall, so this was cut short, but the actuating mechanism was left in place. On the 450 the battery compartment lid has a power switch which needs to be bodged, and the piece of flexible pcb going to the mode dial has to have contacts shorted. The camera can then be remoted via usb, and a psu used to replace the battery.

Hope this helps

Huw

[yesyes]

Thanks for the replies so far! :-)

I guess I need to have a closer look at the mechanics once I have taken mine apart...

Any thoughts on the cooling? How cold should I make it and what wattage peltier would you recommend?

[Merlin66]

Chris,

If you're only interested in Astrophotography and not out into the NIR, then the colour balance filter (blue) is the only one that needs to be removed. It doesn't have to be replaced by anything if you use the body on the scope...

The front (clear?) filter is actually a UV-IR cut-off filter as well as the ant-alias/ dust shake.

(If this is removed you can access down into the UV and out into the NIR, but you'll probably need to add back a UV-IR filter in the optical path for "normal" astroimaging.)

HTH

[yesyes]

Yes, this will be purely a AP camera.

So if I wanted more of the Ha (which is in near IR IIRC) I would need to remove both filters?

Could you please explain why I would need a UV-IR filter again for "normal" astro imaging? What kind of objects do you mean by "normal" AP?

[Merlin66]

No, the removal of the colour correction filter gives the enhanced response to Ha.

Without the UV-IR cut filter you tend to get blue halos and star "bloat" in astroimages of DSO's.

[yesyes]

Ah, right, so the IR cut part of the clear-ish filter lets enough of the near IR through to get more of the Ha.

When removing both filters the range of wavelengths is to great for the image to focus at the same point?

[Merlin66]

basically that's it!

I have to re-focus the spectroscope collimator lens when moving from the UV - green- red.

[Gina]

Ha is deep red and within the pass-band of the IR/UV filter from what I've read.

I have removed both filters from my modified sensor assembly but I'm debating with myself whether to put the IR/UV clear looking filter back in. I have an Astronomik CLS-CCD clip filter which both stops LP and replaces the IR/UV filter. Apparently, the Canon (1100D) IR/UV filter is as good as or better than any other filter so if you haven't already bought a CLS-CCD filter and aren't interested in the rather limited field of NIR imaging, you may be able save yourself some money by just replacing the Canon IR/UV filter. Here I'm assuming the 1000D IR/UV filter is as good as the 1100D. I know there are other differences between the two models such as the much improved digic 4 processor and lower noise level. Also, the 1100D filter doesn't have a shaker to shake off loose dust. So it could be different in its filtering properties - I don't know.

I've read that you can over cool the sensor and get frost or condensation problems. I do not intend cooling my sensor below 10C or ambient whichever is lower. Currently the sensor runs at around 20C above ambient. The Peltier TEC I have in mind for mine is about 20W and 25mm square. That's if I use one. I shall be running tests to see if natural cooling with fins is sufficient.

AFAIK the 1000D doesn't have the amp-off circuitry that the digic 4 versions have so you might want to consider adding that. There's some info on it somewhere that I read a while back but didn't keep the reference once I discovered that the 1100D already has it built in.

From my experience,you want to be very careful how far you go with removing seemingly unwanted parts. I managed to stop my first 1100D body from working by doing just that. It was fine after the filter removal and reassembly. I've pretty much given up on getting that first body working and concentrating on using the second body - probably with the modded sensor assembly from the first body (if I haven't managed to destroy that!).

[yesyes]

OK, the 1000D has just arrived, so the fun can commence

I think I'm all clear on the filters now. As for the cooling I've ordered a 12V 60W (20W cooling power) Peltier. That should hopefully do the trick. I've been reading a bit more about this and it seems I need to measure sensor temperature and ambient temperature + humidity, then control the Peltier via PWM and a n-channel logic level MOSFET and make sure that the sensor temperature does not go below the dew point. This will be done using an Arduino Mini. This will also allow me to display the temperatures on a small display and to add computer control to the cooler controller.

I'm not so clear on the dew point. Do I calculate that from the ambient or sensor temperature?

[yesyes]

Gina, do you have any further info on that amp-off mod you mentioned for the 1000D? I can't seem to find anything on that....

[SniffTheGlove]

I'm not so clear on the dew point. Do I calculate that from the ambient or sensor temperature?

I would calculate it on the ambient temp.

If this helps here is my code for calculation of dewpoint from humidity and temp.

It php code but easily converted into what ever code you need

function Calc_Dewpoint($rel_humid, $temp_c) {
    $KC_Offset = 237.3;
    $l = log($rel_humid / 100.0);
    $m = 17.27 * $temp_c;
    $n = $temp_c + $KC_Offset;
    $b = ($l + ($m / $n)) / 17.27;
    $dew_c = ($KC_Offset * $ / (1 - $;
    return $dew_c;
}

[yesyes]

Thanks for that, Sean. There is already a dew point calculation function in the DHT11/22 Arduino library I'm going to use. But I'll compare that to your code.

I just wasn't sure if I shouldn't use the "local" dew point near the sensor where temperature is colder. If I can use ambient temperature and humidity I might look into reading that off my DIY weather station that already calculates the dew point instead of having yet another sensor. ;-) On the other hand, it might be better to have this all self-contained.

[Gina]

Gina, do you have any further info on that amp-off mod you mentioned for the 1000D? I can't seem to find anything on that....

Done a Google and I believe this what I was thinking of.

EOS 300D Amp Off

Only it's for the 300D so changes may be needed for the 1000D.

[yesyes]

Oh dear, I didn't want to go as far as cutting traces. (but then, I did say "extreme")

Thanks for that, Gina. I think I'll wait with that and see if amp glow becomes an issue...

[nytecam]

No, the removal of the colour correction filter gives the enhanced response to Ha. Without the UV-IR cut filter you tend to get blue halos and star "bloat" in astroimages of DSO's.

No 'bloat' with pure reflectors which bring all radiation to a common focus but yes with [silly] refractors even so called APOs

[Merlin66]

Agreed!!

[RAC]

Gina, do you have any further info on that amp-off mod you mentioned for the 1000D? I can't seem to find anything on that....

The 1000d does not have an amp glow problem.

[yesyes]

That's good to know, thanks RAC.

[Islander]

That was an almost audible sigh of relief, Chris

[yesyes]

did you hear that all the way over there?

[Gina]

Great I'm glad to see that.

I did wonder, Chris, why you went for the 1000D rather than the 1100D. Was it very much cheaper?

[yesyes]

Yes, I got a used 1000D for £150. Couldn't find a used 1100D's anywhere near that price. Also, I already have a 1000D with mains PSU and modded battery adaptor and wanted to use that and other accessories...