Converted Nikon D750 and Internal Shutter Monitoring IR LED


Introduction

I bought a Full Spectrum D750 from Life Pixel last year. I am very happy with it, and was able to do a lot of testing and write some articles on my website that you can find here:

  - IR Hot Spot Database

  - IR Neutral Density Filters

Yesterday, on Life Pixel and Kolari Vision websites, I saw some strange warnings that weren't there 1 year ago when I bought the D750:

  • Screen Shot 2020-12-03 at 18 43 51
  • Screen Shot 2020-12-03 at 18 44 04
  • Screen Shot 2020-12-03 at 18 50 03

So I have contacted Life Pixel right away because I was quite confused... I bought a full Spectrum D750 to mainly do 2 things:

  - IR Long Exposures (i.e. 5min duration)
  - Astrophotography (not with telephoto but with wide angle lenses eventually on a motorized mount)

These 2 types of photography require to use long exposures at least 20-25 seconds long and/or ISO levels up to 3200.

So that's true that the camera is subject to some IR light "fog" that is impacting images. You can see a video and some articles in the references below explaining the problem.

Here is what is happening on my camera when I try to reproduce the issue:

This is apparently caused by an internal IR light used to check the shutter mechanism:

It seems that Nikon has introduced for their newest DSLRs (D3, D3x, D3s, D700, D7000) an intelligent "shutter control unit" using an IR emitting LED / photodiode which tells the camera exactly the time a shutter needs to open an close, regulates that if necessary and if to far outside that permissable zone, sends an error message to the camera display asking for service. A very good idea to extend the lifespan of these shutters which are supposed to live longer than 150.000 actuations.Dr Klaus Schmitt

This issue is affecting D750 models too (see reference links at the bottom of the page).

Now, to my understanding, Life Pixel users of converted D750 did not seem to complain... And in the articlesI  have written above (especially when testing 10, 15 and 20 stop filters), I had to use some long exposures at higher ISO and didn't see such issue.

So...

Why this problem didn't occur to me before ?

And consequently... Is there a workaround ?


Testing

The Internal IR light issue on a Full Spectrum Camera can be seen quite easily by taking some dark frames shot at high ISO and long exposure time, pushed in post.

For all the shots, these conditions will always remain the same:

  - No lens, cap on camera
  - Viewfinder covered by tape
  - Manual mode
  - ISO 12800
  - White Balance: 2500K (any WB would be fine, this is just to keep the output of the camera screen consistent)
  - High ISO NR: OFF
  - Long Exposure NR: OFF
  - RAW + JPEG (FINE)
  - Focus tracking with lock-on: OFF
  - Focus point illumination: OFF
  - Built-in AF-assist illuminator: OFF
  - Viewfinder grid display: OFF
  - LCD illumination: OFF
  - Remote control mode (IR Remote): OFF
  - Wifi: OFF

And here are the parameters that will change:

  1. Release mode (RM): Single frame (S), Quiet shutter-release (Q), Self-timer (T), Mirror up (M)
  2. Delay Mode (DM): OFF or 1 sec
  3. Live View (LV): ON or OFF
  4. Shutter Speed (S) : 30 sec (30) or Bulb Mode (B)

The goal here is to try to find some combinations that do not make use of this shutter monitor IR LED.

Combinations:

  01 - S=30, RM=S, LV=OFF, DM=OFF
  02 - S=30, RM=S, LV=OFF, DM=1
  03 - S=30, RM=S, LV=ON, DM=OFF
  04 - S=30, RM=S, LV=ON, DM=1

  05 - S=30, RM=Q, LV=OFF, DM=OFF
  06 - S=30, RM=Q, LV=OFF, DM=1
  07 - S=30 ,RM=Q, LV=ON, DM=OFF
  08 - S=30, RM=Q, LV=ON, DM=1

  09 - S=30 ,RM=T, LV=OFF, DM=OFF
  10 - S=30 ,RM=T, LV=OFF, DM=1
  11 - S=30 ,RM=T, LV=ON, DM=OFF
  12 - S=30, RM=T, LV=ON, DM=1

  13 - S=30 ,RM=M, LV=OFF, DM=OFF
  14 - S=30, RM=M, LV=OFF, DM=1
  15 - S=30, RM=M, LV=ON, DM=OFF
  16 - S=30, RM=M, LV=ON, DM=1

  17 - S=B, RM=S, LV=OFF, DM=OFF
  18 - S=B, RM=S, LV=OFF, DM=1
  19 - S=B, RM=S, LV=ON, DM=OFF
  20 - S=B, RM=S, LV=ON, DM=1

  21 - S=B, RM=Q, LV=OFF, DM=OFF
  22 - S=B, RM=Q, LV=OFF, DM=1
  23 - S=B, RM=Q, LV=ON, DM=OFF
  24 - S=B, RM=Q, LV=ON, DM=1

  25 - S=B, RM=T, LV=OFF, DM=OFF
  26 - S=B, RM=T, LV=OFF, DM=1
  27 - S=B, RM=T, LV=ON, DM=OFF
  28 - S=B, RM=T, LV=ON, DM=1

  29 - S=B, RM=M, LV=OFF, DM=OFF
  30 - S=B, RM=M, LV=OFF, DM=1
  31 - S=B, RM=M, LV=ON, DM=OFF
  32 - S=B, RM=M, LV=ON, DM=1

Post-processing

I will  post-process in Lightroom to make the symptoms more apparent:

  - I set the WB to Daylight
  - I will push the exposure to +3EV, which is a LOT considering the ISO gain is already 12800.

So keep in mind that I am trying hard to worsen the conditions!

We can clearly see that Mirror Up and Exposure Delay have a positive impact on the IR led.

BULB Mode 1, 2, 3 minutes

Here is now a matrix of several long exposures - dark frames pushed at +5EV in post, all taken with Bulb Mode + Mirror Up + Exposure Delay Mode set to 3 seconds, LENR = OFF.

  - Shutter speeds: 2min, 3min, 4min
  - ISO: 100, 200, 400, 800, 1600, 3200

We can see that bulb long exposures up to ISO 400 are quite nice, but at ISO 800, long exposures start to be impacted by the internal IR light at the bottom of the frame (purple glow). This is very low though, knowing that these long exposures were at ISO 800 and have been pushed to +5 EV in post.

At ISO 1600, we can see a stronger glow at the bottom of the pictures plus a sort of cloud at the bottom left. But considering that these exposures are quite long and have been pushed in post, I think ISO 1600 could still lead to good results.

At ISO 3200, we start to see the problem even more. I think the pictures could still be usable but we are reaching a limit.

30 seconds

Here is now a matrix of several 30" exposures - dark frames pushed at +5EV in post, all taken with Shutter speed 30", Mirror Up + Exposure Delay Mode set to 3 seconds, LENR = OFF.

  - Shutter speeds: 30s
  - ISO: 100, 200, 400, 800, 1600, 3200, 6400

Exactly like with bulb mode, we see a purple effect starting at ISO 800, but really small at the bottom. Overall, until ISO 3200, this is quite acceptable.


Conclusion

  To reduce (or even cancel) the impact of the internal shutter monitoring LED light of the D750, here are my suggestions in order of priority that you can combine, whatever the mode you use:

  1. Use Mirror Up

  2. Use Exposure Delay Mode 3 seconds. Switching from 1 second delay to a 3 seconds delay has definitively reduced the purple glow. So I would suggest you test with different combinations to see. It seems that the IR LED light has lower impact when the delay increases.

  3. Use Live View (in last resort if you cannot use the other ones)


For Bulb mode (2, 3, 4 mins)

  1. Stay at ISO between 100 and 1600

  2. For longer than 3-4 min exposures, stay at ISO between 100 and 800

  3. Use with Mirror Up with Exposure Delay Mode 3 sec


For High ISO shots (<= 30 sec)

  1. Use Mirror Up

  2. Use Exposure Delay Mode 3 sec

  3. If not possible, use Live View


For Daylight Long Exposures

  Since you can use long exposures with bulb mode, there is no issue, just look at the bulb mode recommendations.


For astrophotography (tracked mount)

  Since you can use long exposures with bulb mode, there is no issue, just look at the bulb mode recommendations.


For astrophotography (standard mount - untracked)

  For night scapes or landscape photography at night, where you want to avoid star trails, your exposure time will probably varie between 20-30 seconds. But you will have to raise your ISO. This is where it hurts. My testing show that sticking with an ISO <= 3200 seems OK. And this value would in theory be enough to reach 7 EV with a wide angle lens at f/2.8.
  Use Mirror Up with Exposure Delay Mode 3 seconds. 


Whatever you use, also think about the fact that the Nikon D750 is a camera that is ISO invariant and has a higher dynamic range at ISO 100-200 and can be pushed in post after. So it's better to avoid high ISO, under expose and increase EV in post.


Updates

  - December 3rd 2020: Initial version

  - December 4th 2020: I had an email discussion with Dan Llewellyn from LDP LLC (MaxMax) - quite a reference for me and a lot of people! He suggested I also test with bulb mode. Also, to  avoid any effect from light leaks I did again all the testing in a dark environment, and also ensured all the following settings are set: Focus tracking with lock-on: OFF, Focus point illumination: OFF, Built-in AF-assist illuminator: OFF, Viewfinder grid display: OFF, LCD illumination: OFF. These comes from a recommendation seen in the UV Photography Forum.

  - December 7th 2020: Continuing the discussion with Daniel Malkin from Life Pixel, he found that on his D750 modded, the effect wasn't like me. It was less apparent. He tested with different value of Exposure Delay, which made me think that this delay could probably be the solution. So I did again all my tests, this time with a delay of 3 seconds. I added  dark frames pushed at +5EV for all combinations of 30s, 2min, 3min, 4min exposure times.


  • No Comments

Original text