Question about the 1/32,000 second electronic shutter.

Mack

Mu-43 All-Pro
Joined
Jan 14, 2018
Messages
1,937
Very interesting discussion.

I wonder if the "Lightning Mode" of the MIOPs trigger could work in lieu of the Arduino?

Is there any interest in the old school flash cubes? Demand for them for this kind of thing? The company I used to work for still had a bunch in inventory (if they didn't get tossed when they moved to the new warehouse back in May)
I agree! :thiagree:

Another post in the Lighting forum here mentioned at the very high speeds while in Electronic or Silent shutter mode, the E-M1 Mark II has an X-sync of 1/50 second up to ISO 8000, and then it goes even slower to 1/20 second. I accidentally found that out when I was switching the Godox from HSS to non-HSS sync while at the 1/32,000 camera speed which dropped way down to 1/50 second.

That said, today I got a bunch of FP-26B flashbulbs to play with. They have a much longer burn time as they were made for old focal-plane cameras that had X-sync on them. Oddly, the Olympus is using an even slower digital scan sync (i.e. 1/50 to 1/20 sec. @ 1/32,000 sec.) than the old 1/60 second film cameras did.

However, my first few FP-26B bulbs are showing a lot more promise than the M3B bulbs that have a shorter burn time. The M3B's can be put in a middle of the 1/32,000 second shutter speeds sensor scan, but they vignette on the top on bottom. Top is the sensor scan blocking, bottom is the fade of the bulb having the shorter duration. I think I may be able to sync the FP-26B to be a perfect fit at 1/32,000 second even with a high ISO 25,600 (Where I think the scan will drop to 1/20 second). The FP-26B appears to burn around 70ms at a t0.5 average, or 1/14 second, which is slower than the Olympus high-speed sensor scan so it should cover for a perfect X-sync if initial bulb triggering time is accurate. That I need to play with.

My understanding of the flash cubes is they were built on the AG1 flashbulb. The Magicubes were made with a percussion trigger (No voltage supplied via the camera, just a hammer pin that popped up and ignited the bulb.). The other FlashCubes did take a voltage. However, I think both are too fast though for a coverage of the digital sensor's slow X-sync scan. More of a job for older FP-type bulbs (FP stood for "Flat Burn" tops, not "Focal Plane" according to Wiki.).

Below is my first try with the FP-26B bulb.

Scope showing (in yellow) the FP-26B flat burn duration of ~70ms. I'm calling it a t0.5 for the effective exposure output:

FP26B 35ms 20ms scope..png
Subscribe to see EXIF info for this image (if available)


Digital SG ColorChecker chart from above firing with a hold-off delay of 35ms in the Arduino (Fwiw, There's some other delays going on other than the red trigger (me) pushing the shutter button on the Arduino that feeds into the camera's remote shutter release jack. It looks like ~65ms from red trigger to bulb in scope's image, but software's internal shutter open to bulb firing timing is 35ms.). I need to change it back to maybe 31ms as the top's darkness is part of the sensor scan going dark. Bottom is still lit well enough by the bulb so I suspect an ignition timing issue. Lowering the time seems to raise the burn's brightness in the image:

_9160554_DxO.jpg
Subscribe to see EXIF info for this image (if available)


Above: ISO 25,600 at 1/32,000 sec. f/4.5. FP-26B bulb was 10' so a Guide Number is 45. Color of flashbulbs always seemed better for me on skin tones, but they are a bit warmer than electronic flash and more of a flatter spectrum going up towards the infrared end. Electronic does the opposite (unfiltered) and fades down from the higher bluish UV end with less at the warmer IR end, and it just seems more cold (bluish) imho.

For the curious, my intent is to use a Noct type of lens (<f/0.95) which would also lower the ISO and use the flash bulb as a fill in bright sun at 1/32,000 sec., <f/0.95, and lowest ISO of maybe 200. No ND filter! Just a fun theoretical idea to see if it will work. I think it should, but dunno as I need to settle on which fast m43 Noct lens to get.

Bulbs may need to be resurrected. :laugh1:
_________________________________________________________

This AM I changed sync delay timing to 29ms:

FP-26B-29ms-Combo.jpg
Subscribe to see EXIF info for this image (if available)


Now where's the Noct lens to try this thing out? :dance2:
__________________________________________________________________

Outside in bright sun to play.

No Noct yet (ordered a f/0.8), but had a test with the slow Olympus 45mm f/1.2 Pro in sunlight at 1/32,000 sec. I used the M3B flashbulb since I have more of them than the FP-26B although it is weaker in output and a shorter burn time too.

Below. One without the bulb, and second fill-one with the M3B bulb. I also need to aim the flash holder better as it's tough to not get glare off the x-rite Digital SG card as it is sort of glossy.

No flash. E-M1 Mark II at 1/32,000 sec., f/1.2, ISO 500:

No-Flash.jpg
Subscribe to see EXIF info for this image (if available)


With M3B flashbulb as a fill. E-M1 Mark II at 1/32,000 sec., f/1.2, ISO 500:

M3B-Flahbulb.jpg
Subscribe to see EXIF info for this image (if available)


Com'on Noct!
 
Last edited:

Mack

Mu-43 All-Pro
Joined
Jan 14, 2018
Messages
1,937
Bah! Got the 29mm f/0.8 Noct along with its CA issues maybe fixed using Exposure X7 which beats PS and DxO PhotoLab 4 both for fixing CA (DxO doesn't have the Voigtlander 29mm in their corrective database yet.).

The above shots show the left side darkening from lack of light from the bulb as my ignition timing was off by ~10ms. Moved it down to 16ms after playing with an LED strip light hung vertically in lieu of a bulb and my prior hit rate was too low. Much better edge coverage results with the bulbs as in this lighting thread on here: https://www.mu-43.com/threads/using-flashbulbs-and-electronic-shutter.112421/

There is a good article on mechanical verses electronic shutter speeds here on "Why an electronic high-speed shutter can be much slower than a mechanical one at the same speed": https://blog.kasson.com/d850/how-fast-is-the-nikon-850-electronic-shutter/

This stuff gives me a headache, but I think I have it sorted for now . . . somewhat. :dash2:
 
Joined
Jan 31, 2014
Messages
169
I agree! :thiagree:

Another post in the Lighting forum here mentioned at the very high speeds while in Electronic or Silent shutter mode, the E-M1 Mark II has an X-sync of 1/50 second up to ISO 8000, and then it goes even slower to 1/20 second. I accidentally found that out when I was switching the Godox from HSS to non-HSS sync while at the 1/32,000 camera speed which dropped way down to 1/50 second.

That said, today I got a bunch of FP-26B flashbulbs to play with. They have a much longer burn time as they were made for old focal-plane cameras that had X-sync on them. Oddly, the Olympus is using an even slower digital scan sync (i.e. 1/50 to 1/20 sec. @ 1/32,000 sec.) than the old 1/60 second film cameras did.

However, my first few FP-26B bulbs are showing a lot more promise than the M3B bulbs that have a shorter burn time. The M3B's can be put in a middle of the 1/32,000 second shutter speeds sensor scan, but they vignette on the top on bottom. Top is the sensor scan blocking, bottom is the fade of the bulb having the shorter duration. I think I may be able to sync the FP-26B to be a perfect fit at 1/32,000 second even with a high ISO 25,600 (Where I think the scan will drop to 1/20 second). The FP-26B appears to burn around 70ms at a t0.5 average, or 1/14 second, which is slower than the Olympus high-speed sensor scan so it should cover for a perfect X-sync if initial bulb triggering time is accurate. That I need to play with.

My understanding of the flash cubes is they were built on the AG1 flashbulb. The Magicubes were made with a percussion trigger (No voltage supplied via the camera, just a hammer pin that popped up and ignited the bulb.). The other FlashCubes did take a voltage. However, I think both are too fast though for a coverage of the digital sensor's slow X-sync scan. More of a job for older FP-type bulbs (FP stood for "Flat Burn" tops, not "Focal Plane" according to Wiki.).

Below is my first try with the FP-26B bulb.

Scope showing (in yellow) the FP-26B flat burn duration of ~70ms. I'm calling it a t0.5 for the effective exposure output:

View attachment 908083

Digital SG ColorChecker chart from above firing with a hold-off delay of 35ms in the Arduino (Fwiw, There's some other delays going on other than the red trigger (me) pushing the shutter button on the Arduino that feeds into the camera's remote shutter release jack. It looks like ~65ms from red trigger to bulb in scope's image, but software's internal shutter open to bulb firing timing is 35ms.). I need to change it back to maybe 31ms as the top's darkness is part of the sensor scan going dark. Bottom is still lit well enough by the bulb so I suspect an ignition timing issue. Lowering the time seems to raise the burn's brightness in the image:

View attachment 908082

Above: ISO 25,600 at 1/32,000 sec. f/4.5. FP-26B bulb was 10' so a Guide Number is 45. Color of flashbulbs always seemed better for me on skin tones, but they are a bit warmer than electronic flash and more of a flatter spectrum going up towards the infrared end. Electronic does the opposite (unfiltered) and fades down from the higher bluish UV end with less at the warmer IR end, and it just seems more cold (bluish) imho.

For the curious, my intent is to use a Noct type of lens (<f/0.95) which would also lower the ISO and use the flash bulb as a fill in bright sun at 1/32,000 sec., <f/0.95, and lowest ISO of maybe 200. No ND filter! Just a fun theoretical idea to see if it will work. I think it should, but dunno as I need to settle on which fast m43 Noct lens to get.

Bulbs may need to be resurrected. :laugh1:
_________________________________________________________

This AM I changed sync delay timing to 29ms:

View attachment 908112

Now where's the Noct lens to try this thing out? :dance2:
__________________________________________________________________

Outside in bright sun to play.

No Noct yet (ordered a f/0.8), but had a test with the slow Olympus 45mm f/1.2 Pro in sunlight at 1/32,000 sec. I used the M3B flashbulb since I have more of them than the FP-26B although it is weaker in output and a shorter burn time too.

Below. One without the bulb, and second fill-one with the M3B bulb. I also need to aim the flash holder better as it's tough to not get glare off the x-rite Digital SG card as it is sort of glossy.

No flash. E-M1 Mark II at 1/32,000 sec., f/1.2, ISO 500:

View attachment 908152

With M3B flashbulb as a fill. E-M1 Mark II at 1/32,000 sec., f/1.2, ISO 500:

View attachment 908155

Com'on Noct!

Interesting discussion regarding timings and I surely learned somethings and as an engineer I understand the technical interest in making this work, but regarding the results if it's that sunny out wouldn't it be possible to fill with a reflector rather than flash?
Simple curiosity.
 

Mack

Mu-43 All-Pro
Joined
Jan 14, 2018
Messages
1,937
Interesting discussion regarding timings and I surely learned somethings and as an engineer I understand the technical interest in making this work, but regarding the results if it's that sunny out wouldn't it be possible to fill with a reflector rather than flash?
Simple curiosity.
I have one of those large triangular Lastolite Trigrip reflectors. It has the white and silver stripes on one side, and the gold and white stripes on the other (Their Sunlite and Soft-Silver model.). I have a friend who owned the silver one and it was like a signal mirror and way too bright so I went with their Sunlite/Soft-Silver one thinking it would be better and not as harsh.

When I have tried to use it as a fill, I had to move it maybe 20 feet away from the subjects if they were standing and bow the frame a bit. It was still too bright to the point the poor subjects squinted and some complained about it as if looking into a sunlit mirror. Holding it steady in afternoon desert wind and keeping it on target was an issue too.

It does make a good fan though if you need wind in someone's hair, or using it for macro in the field, still subjects, or adding shade or subtractive lighting to something. I have bounced flash (and flashbulbs) into it which also works and that doesn't blind a person as using it outside with direct sunlight.
 

Mack

Mu-43 All-Pro
Joined
Jan 14, 2018
Messages
1,937
Aside, I began playing with the long-duration PF-330 bulbs for waterfalls where I needed their longer burn time to blur the water out and not create the "Frozen-in-Time" water droplets as with current electronic flash units.

I inquired to a xenon flashtube maker about a longer burning tube. Their response was "The xenon tubes, to burn that long, would likely crack from the heat much as old glass flashbulbs sometimes did and why they began coating them with a clear or blue lacquer (daylight balanced) to keep any glass shards from flying off." Option would be to go to a water-cooled xenon tube (Surprise, I did run across some back then for scientific purposes at military spec prices!) to deal with the cooling. Then there was the need for the cooling system and a large capacitor bank much like a large studio strobe's power unit as it would be a throttled discharge.

Given the old folding Honeywell Tilt-a-Mite and a bulb weighs much less than a speedlight, might be time to bring back some FP type of bulb and a hotshoe setup timed to a specific camera brand's hotshoe since they can throw out a trigger pulse via their hotshoe's connection pins to pre-fire the flash for TTL where the actual exposing flash follows the pre-flash. The flash holder and bulbs could be made a lot smaller to carry than the current gear.

A newer designed bulb could be made much brighter to run with the current high-speed electronic shutters. Sell the special hotshoe adapter/reflector as an option, and maybe a dozen bulbs for $15 or so. Maybe a Bowens adapter for adding softboxes too. The portability would be welcomed as it might fit into a small pocket or corner of a bag and produce far more light than some carrying a 600ws studio unit and dealing with its size/weight and a heavy-weight supporting stand.

Oddly, Olympus allows for X-sync with their electronic shutter even though it kicks down to 1/50 or 1/20 sec. now, but Nikon doesn't even allow for x-sync in their Z7 II 1/8,000 sec. (fastest) electronic shutter mode. That likely can be changed via the hotshoe info/firmware to allow these things to run to full speed and allow a bulb to fire. I might attack the Nikon next with the Arduino Uno sync gizmo since I first made it for the Nikon for using longer-burning flashbulbs, but changed the timing to work with Olympus up to 1/32,000 sec. It's all in the timing, and likely the entire thing could be small and built inside some hotshoe or bulb reflector.
 
Links on this page may be to our affiliates. Sales through affiliate links may benefit this site.
Mu-43 is a fan site and not associated with Olympus, Panasonic, or other manufacturers mentioned on this site.
Forum post reactions by Twemoji: https://github.com/twitter/twemoji
Forum GIFs powered by GIPHY: https://giphy.com/
Copyright © Amin Forums, LLC
Top Bottom