Pentax *ist D IR Mod

I have been using my Pentax *ist D DSLR for astrophotography in the last few years. However, for some object in the sky like emission nebulas, the far red of the H-alpha (656.28 nm) line are blocked up by the standard IR filter in front of the CCD. The IR filter in this camera is also combined to an antialiasing filter and both contribute to the correct rendition of everyday photos.

It is possible to remove these filters to unleash the full spectral sensitivity of the Sony CCD, or to replace them to select particular wavelenghts. I wanted to replace the standard filter assembly with a Baader ACF filter (#245 9212) that enlarges the spectral window in to the deep reds including the H-alpha. These Baader filters are designed for Canon cameras, and this particular one for the 350D/20D.

See http://www.astrosurf.com/buil/baader/eval_us.htm for further tests on this filter. In the graph below is a comparison of a Canon IR and the Baader filter. Unfortunately I don't have any mean to scientifically evaluate the Pentax IR filter. As you can see below the H-alpha transmission of the Baader is very good. 

Fig.1 - Transmission: Baader vs. Canon

There was no guarantee that the Baader filter would fit in the Pentax but I wanted to give it a shot as the general dimensions matched (Baader is 27x19.6x2.7mm). I knew that the Baader filter thickness of 2.7mm was larger than the Pentax own filter (I did see a report of *ist DS filter thinkness of 2.2mm and I assumed it to be the same on the*ist D). So a focus plane position change would occurr and the focus position calculated by the autofocus system or the viewfinder focus would have been off from the real focus. In practice the actual focus plane will move towards the back of the camera by an amount that can be calculated knowing the thickness difference between the old and new filter:

Focus Plane Shift = ThicknessDifference * ( n - 1 ) / n     with n being the refractive index of the filter glass

If the camera is to be used only for astrophoto this wouldn't be a problem, however by shimming the sensor mount plate by the required amount  everything will continue to work as expected. Given the different spectral response, the modified camera is more sensitive to the red giving a red cast to pictures. In order to use the camera for regular photography, a custom white balance would be necessary. The camera is capable of that intrinsecally and this can also be done in software during post processing, even more easily if shooting raw.

So, I completed the modification achieving all the desired results and in this page I'll show you how to disassemble the Pentax *ist D.

What tools are needed:

1) Small Philips screwdriver.

2) Tweezers: I used one made of bamboo to avoid static electricity.

3) Lint free tissues.

4) Gloves to avoid unwanted fingerprints.

5) Small plastic holders for the screws. Do not mix them. I labeled the various groups and kept them separated.

6) Rocket Blower or similar to blow dust off.

NOTE: I left a 50/1.7 lens mounted as it easier to hold the camera or have it standing on it during work on the back side.

Start with removing the backup battery.

Remove the 7 screws from the bottom plate. Note that a shorter screw is used in the recessed hole (see picture).

Fig. 2 Bottom plate removal

After removing the screws you can gently lift the bottom plate. It is still connected with the blue wire for the button battery connection. Unscrew the connector from the screw showed in the picture below so the bottom plate can be completely removed.

Fig. 3 Removal of the button battery connection

Fig. 4 Bottom plate removed

Removal of the top cover: 8 screws in total, there is a small one underneath the flash, two long ones on each side of the eyepiece, and of the four screws signalled by the yellow arrows there is a short one on the side of the ISO dial. See Picture below for the top 7 screws. The eighth screw is inside the battery holder, at the bottom, silver coloured and long.

Once all the screws are removed, the top plate can be lifted up. You can pop the flash out so you get a better grip. The reason to remove the top plate is to get access to the eyepiece that must be removed to take off the back LCD assembly.

Fig. 5 Top plate removal

The top plate is connected to the main body with cables for flash and dials, so you won't remove it completely. NOTE: The flash circuitry may have high voltage capacitors holding charge so you may get a shock if you touch the wrong spot :-). I didn't receive any shock but it is better to be aware. See below for the picture of the top plate removed.

Fig. 6 Top plate removed

Now proceed with the removal of the eyepiece. See Fig. 7. You need to remove two screws at the side of the eypiece and this will easily slide out. You will see that back plate (the one holding the picture display) has two protrusion that would be blocked if the eyepiece is not removed.

Fig. 7 Eyepiece removal

Next is the removal of the back plate. There are six screws total, two on each side of the plate (two long on the battery side, short ones on the opposite side) and two screws inside the compact flash compartment. Just open the door and remove the two screws.

NOTE you also need to open the rubber cover for the I/O ports and power supply  as it will interfere with the removal of the back blate. Once all the screws and removed and rubber door opened, start lifting the back plate from the ISO dial side and once it is lifted it will also come off from the battery holder side.

Fig. 8 Back plate removal 

Once the back plate is removed, it can be completely disconnected from the body by disconnecting the flat cable and two-wires power supply  signalled by the two yellow ARROWS in the picture below. NOTE if you want, you can also disconnect the flat cable that runs from the CCD to the CF electronics (yellow circles below), but it is not at all necessary. If you do remove it remember its orientation (see photo below) as you can also mount it backwards but the memory card would be inaccessible :-).

Fig. 9 Back plate removed. It can be disconnected now.

Before addressing the CCD, we need to disconnect (use a soldering iron) the ground cable (black) shown in Fig. 10. That's the only cable connecting the sensor plate to the frame on the lower side of the sensor.

Fig. 10 Desoldering the grounding wire

Removal of the sensor and electronic boards: yellow arrows can be taken out, on the right is the CF assembly, on the left is the power supply, in the center board the yellow arrow will have to be removed with the red arrow ones as they (6 in total) will free up the sensor plate..

NOTE: The electronics behind the sensor is fixed to the CCD plate so consider that as a single module.

NOTE: The red arrows keep the CCD plate attached to the steel frame, the yellow once only support the electronics.

Fig. 11 Screws holding the electronics (yellow) and the CCD plate (red).

Once all the electronics and CCD plate are free to move, lift them and disconnect the connector that goes from the power supply to the CCD electronics (see long arrow in Fig. 12).  So you'll be able to lift the CCD/CF asseblies and have access to the CCD. Four screws keep the IR filter holder as shown in the picture below.

Fig. 12 CCD assembly is now accessible

Removing the four screws shown in Fig. 12, you can remove the metal frame with its elastic flaps that pushes dows the IR filter and then the thin frame that keeps the IR filter in place above the CCD.

Fig. 13 IR filter holders removed.

Now you can remove the IR filter. it is made of two glass plates one larger on the bottom and one smaller on top. These are IR and antialiasing filter. Total thickness is 1.2mm. This thin glass is then replaced with the thick Baader filter and the frames are put back, see Fig. 14. The Baader filter is smaller (shorter and narrower) than the original filter so I slightly stretched the flat springs holding the sensor from the top.

NOTE: The baader filter has one side pink coloured. That side goes on top facing out.

Fig. 14 Baader filter mounted

Once the filter is assembled everything can be mounted following this procedure backwards. Hopefully you have all the screws separated appropriately.  Use the rocket blower to get rid of dust on the CCD, Filter, eyepiece, prism, and on the top plate LCD.

NOTE: In order to maintain the autofocus accuracy, I used 3 washers 0.7mm thick that I used to shim the CCD plate. Putting the washers in between the plate and the frame mountings (see red arrows in Fig .11).

Total time for this was about four hours for my first camera. My second camera only took 1 hour.

After everything is put back together, get the batteries and a memory card and check that everything works as expected. If you put washers to offset the CCD you can do a front focus/back focus test. Here is mine:

Fig. 15 Focus test - with D FA 100 at 2.8

As a quick comparison I took a picture of my MS mouse with the modified camera and with my other *ist D. You can see a little red cast and higher red sensitivity:

Fig. 16 Left: Modded *ist D, Right: regular *ist D

Results are pretty good given the major surgery ;-). I want to comment on the quality of the assembly especially the CCD filter mounting. In fact I have seen web sites about Canon DSLR being modified where the IR filter was held with glue and where some irreversible action was required to dismantle it. On  the Pentax *ist D everything is done with screws, nothing is glued even around the sensor and I was happy to discover such a fine engineering. I don't expect this to still be the case for newer model where the cost must have been kept under control, maybe I'll have a look sometimes into a K20D...

Vincenzo Miceli

08-Feb-2009

contact me at:

vincenzomiceli at hotmail.com

vincenzo.miceli at gmail.com

Visitors since Feb 2009: