Beginner Trouble with fill flash

[Phil V]

OK, so I've just done a test where I've deliberately taken a subject massively outside of the flash range and on the back of the flash unit itself it did flash up -3EV so it appears that @Phil V is right, so the problem that the OP had was that he was outside the flash range by at least 3 stops. I must admit, I thought they looked closer than this to me. To be 3 stops outside the flash range they'd have had to be 9.7m away.

I am quite surprised, never having used a Nikon flashgun it was a complete guess.
But I don't know where you're getting the 9.7m from, are you taking the HSS into account, it eats a lot more flash power than we imagine it would.

Actually I've now seen the table, at 2.8 and 35mm and 200 ISO we only get 0.9m the way I read it, so the HSS would easily kill that

 

[snerkler]

I am quite surprised, never having used a Nikon flashgun it was a complete guess.
But I don't know where you're getting the 9.7m from, are you taking the HSS into account, it eats a lot more flash power than we imagine it would.

Actually I've now seen the table, at 2.8 and 35mm and 200 ISO we only get 0.9m the way I read it, so the HSS would easily kill that

Nikon give a range in which the flash exposes correctly so at f2.8 35mm ISO 200 the range is 0.9-13m. In other words you can be as close as 0.9m but as far as 13m and it will expose correctly. At F11 (4 stops reduced to take into account HSS) flash range is 0.6-3.4m, so the OP should have been able to expose correctly up to 3.4m. However, his flash was showing it's 3 stops under exposed and so if you go back 3 stops on the chart the flash range is 0.7-9.7m, so in order for the flash to show 3 stops under exposure the subject must be at least 9.7m away. Hope that makes sense?

 

[Phil V]

Nikon give a range in which the flash exposes correctly so at f2.8 35mm ISO 200 the range is 0.9-13m. In other words you can be as close as 0.9m but as far as 13m and it will expose correctly. At F11 (4 stops reduced to take into account HSS) flash range is 0.6-3.4m, so the OP should have been able to expose correctly up to 3.4m. However, his flash was showing it's 3 stops under exposed and so if you go back 3 stops on the chart the flash range is 0.7-9.7m, so in order for the flash to show 3 stops under exposure the subject must be at least 9.7m away. Hope that makes sense?

I misunderstood the chart then.

 

[sk66]

OK, so I've just done a test where I've deliberately taken a subject massively outside of the flash range and on the back of the flash unit itself it did flash up -3EV so it appears that @Phil V is right, so the problem that the OP had was that he was outside the flash range by at least 3 stops.

Well, I've learned something new about Nikon flash behavior...

 

[sk66]

But as I've said before this would change the power and so underexpose the shot, but not reduce the EV/exp comp. Plus matrix will expose the scene, the flash should expose the subject (assuming the focus point is on them) and is 'metered' so to speak via TTL.

Happy to be corrected if I misunderstood how it works.

It would not change an EV/EC setting... having never seen the -3EV underexposure alert before I went with the other option of the viewfinder meter reading -3. Nikon's show the exposure metering w/o flash contribution...but that actually makes no sense either as the meter should have been "happy" unless set manually.

I don't think the flash metering moves with the focus point...

Jeeze, you would think I haven't been using Nikon's for 2+ decades....

 

[snerkler]

It would not change an EV/EC setting... having never seen the -3EV underexposure alert before I went with the other option of the viewfinder meter reading -3. Nikon's show the exposure metering w/o flash contribution...but that actually makes no sense either as the meter should have been "happy" unless set manually.

I don't think the flash metering moves with the focus point...

Jeeze, you would think I haven't been using Nikon's for 2+ decades....

LOl, it is all very confusing The flash output doesn't change with the focus point per se, but when using TTL flash exposure is metered from the subject which is, in part, determined by focus distance. The flash needs to know what you're focussed on to know when enough flash has been outputted so that it turns off (or stays on long enough).

 

[PeetyPeet]

Hi guys, thanks for your input and investigative efforts! I've just about managed to follow the discussion. Forgive my layman's terms, but if I followed correctly - I'm shooting in AP mode at f.2.8. The bright background, wide aperture, and distance from the subject (only a couple of meters at a guess) has caused my camera to choose a shutter speed of 1/2000 which is too quick for the flash (or rather the flash does not have enough power to sufficiently light at that speed). Would this be more or less correct?

If so, how can I avoid this next time?

Many thanks

 

[J Veitch]

When shooting High Speed sync the flash basically has multiple flashes (because its not continuous light, and the shutter curtains act like a letterbox moving across the sensor). You need to lower the shutter speed to "sync" speed or below (turn off HSS), where the shutter reveals the whole sensor when the flash pops. However with the large aperture and lower shutter speed you might end up overexposed... so your other choices are smaller aperture, lower ISO, ND filter (make the metering behave as if its a smaller aperture while retaining the shallow depth of field).

In fact, if you have a little time, have a read of "strobist.blogspot.co.uk" and the "Lighting 101" as there are sections on Blending flash and ambient.

 

[snerkler]

When shooting High Speed sync the flash basically has multiple flashes (because its not continuous light, and the shutter curtains act like a letterbox moving across the sensor). You need to lower the shutter speed to "sync" speed or below (turn off HSS), where the shutter reveals the whole sensor when the flash pops. However with the large aperture and lower shutter speed you might end up overexposed... so your other choices are smaller aperture, lower ISO, ND filter (make the metering behave as if its a smaller aperture while retaining the shallow depth of field).

In fact, if you have a little time, have a read of "strobist.blogspot.co.uk" and the "Lighting 101" as there are sections on Blending flash and ambient.

This is incorrect. Modern flashes fire many bursts of light during 'one flash' up to the sync speed allowing a 'build up of light to the correct exposure, but in high speed sync there is only one burst of light. This is because during HSS sync the second curtain starts to close before the first curtain has fully opened causing a 'slit' between the curtains to move across the sensor (good demonstration of this in the link below). As the shutter speed increases the flash doesn't have enough time to expose the sensor which is why we lose stops of light during HSS.

https://fstoppers.com/originals/demystifying-high-speed-sync-68527

Edit: Ignore my first comment, it is incorrect.

 

[snerkler]

Hi guys, thanks for your input and investigative efforts! I've just about managed to follow the discussion. Forgive my layman's terms, but if I followed correctly - I'm shooting in AP mode at f.2.8. The bright background, wide aperture, and distance from the subject (only a couple of meters at a guess) has caused my camera to choose a shutter speed of 1/2000 which is too quick for the flash (or rather the flash does not have enough power to sufficiently light at that speed). Would this be more or less correct?

If so, how can I avoid this next time?

Many thanks

Yes, this is more or less correct. The strange thing is why the flash didn't have enough power. According to the charts I posted on the previous page your flash should have had enough power to light subjects a couple of meters away.

In order to prevent this in the future there are three main things that influence the effective flash power, aperture, ISO and subject distance. The wider the aperture the more light you will get from the flash, the higher the ISO the more light you will get from the flash, and the closer the subject the more light the flash will create on the subject. Shutter speed does not effect flash output up to the sync speed. In HSS flash power is reduced as already discussed. So if there's not enough power then your options are, choose a wider aperture, slower shutter, higher ISO (although all of these will obviously affect the background), or move closer to the subject.

However, as I've already said based on the table your flash should have had enough power. It does start to make me think that sk66 is also partially right, in that it's that it's not metered correctly, but it's the flash TTL that's not metered right rather than than camera metering. It may be due to the unique scene and that I think that TTL BL means that it tries to balance flash with the ambient (I think that's what BL means). There's two things I would try next time. The obvious one is to shoot everything in manual, take a test shot and then adjust flash power/compensation as necessary (or you could calculate it beforehand using the table). Or have your camera in manual mode and set the exposure as per the camera meter and then leave your exposure settings alone. Then swap to spot metering which changes the flash to TTL (no BL) which I believe means that the flash will not try to balance the ambient. Providing your focus point is on the subject (or at least the same distance as the subject if you've focussed recomposed) I think that the flash will expose for the subject only. Having not tried this for myself I cannot absolutely guarantee the latter, but I'm fairly sure that this is how it works, and what the difference is between TTL and TTL BL.

Another method that could work (although again I'd need to try it) is to increase the flash exposure compensation for scenes like this where the poorly lit subjects are heavily backlit. Obviously the TTL BL is metering in a way that is causing underexposure of the image in these unique circumstances and so just boost the flash exp compensation to , well, compensate. It's the same as your camera meter in unique scenes such as shooting a polar bear in the snow. We all know that the camera will underexpose these scenes and so we dial in +ve exposure comp.

 

[J Veitch]

This is incorrect. Modern flashes fire many bursts of light during 'one flash' up to the sync speed allowing a 'build up of light to the correct exposure, but in high speed sync there is only one burst of light. This is because during HSS sync the second curtain starts to close before the first curtain has fully opened causing a 'slit' between the curtains to move across the sensor (good demonstration of this in the link below). As the shutter speed increases the flash doesn't have enough time to expose the sensor which is why we lose stops of light during HSS.

https://fstoppers.com/originals/demystifying-high-speed-sync-68527

I really don't believe I am incorrect and reading that article I don't see where I am incorrect. Remember we are talking speedlight flash guns here. Not hypersync long duration studio heads. But the differences between High Speed Sync (mulitple flashes each with exposing only a slit of light. The rest of each "pop" is wasted against the shutter) and a single "normal" pop, where the whole sensor gets the whole flash power output. Please do highlight where I am wrong though, as it did all make sense to me until you said it was wrong...

 

[sk66]

This is incorrect. Modern flashes fire many bursts of light during 'one flash' up to the sync speed allowing a 'build up of light to the correct exposure, but in high speed sync there is only one burst of light. This is because during HSS sync the second curtain starts to close before the first curtain has fully opened causing a 'slit' between the curtains to move across the sensor (good demonstration of this in the link below). As the shutter speed increases the flash doesn't have enough time to expose the sensor which is why we lose stops of light during HSS.

https://fstoppers.com/originals/demystifying-high-speed-sync-68527

Speedlight("True") HSS is multiple flashes at reduced power (~1/4 max) because there's no time to recharge the caps. And flash up to x-sync is one pop at up to full power.

"Hyper sync," "tail sync," and what some strobe manufacturers call "high speed sync" is a single pop of long duration.

 

[sk66]

If so, how can I avoid this next time?

Many thanks

Going with the lack of power required, let's look at the settings: f2.8, exp time: 1/2000 sec, ISO-250

The most obvious one would be to get the SS below x-sync (1/250) so that you gain two stops of power. Unfortunately, any factor that reduces the SS also increases the power requirement equally. So, while aperture and ISO could (should?) have been set differently, they wouldn't solve the problem. You can often get somewhat better performance at the limits by using an ND as opposed to HSS... ~1/2-1 stop gain... but still not enough.

This leaves flash spread and flash distance. You could have taken the flash out of auto zoom and zoomed it to a longer/tighter pattern which effectively increases the power. Looking at the chart, going from the 35mm zoom position to the 70mm position would be a gain of 3 meters (~1 stop). You don't really need to fill the frame with flash, so this is a viable option.
The other option is to change the distance, and a change by a factor of 2 (2x or 1/2) is two stops of power. Placing the flash half as far away at 70mm zoom would have gained you the 3 stops it was reporting (although I don't think it looks 3 stops under).

The biggest thing you could do would be to get the flash off camera, so that you can make these types of changes without having to compromise the composition.

 

[snerkler]

I really don't believe I am incorrect and reading that article I don't see where I am incorrect. Remember we are talking speedlight flash guns here. Not hypersync long duration studio heads. But the differences between High Speed Sync (mulitple flashes each with exposing only a slit of light. The rest of each "pop" is wasted against the shutter) and a single "normal" pop, where the whole sensor gets the whole flash power output. Please do highlight where I am wrong though, as it did all make sense to me until you said it was wrong...

Nope it is me who is incorrect thanks to an incorrect article, so I apologise for this. It is indeed as you say, one large pulse for shutter speeds up to sync speed and pulses in HSS. That'll trust me to believe the internet (well that's assuming that the latest articles I've read are correct of course )

 

[snerkler]

Going with the lack of power required, let's look at the settings: f2.8, exp time: 1/2000 sec, ISO-250

The most obvious one would be to get the SS below x-sync (1/250) so that you gain two stops of power. Unfortunately, any factor that reduces the SS also increases the power requirement equally. So, while aperture and ISO could (should?) have been set differently, they wouldn't solve the problem. You can often get somewhat better performance at the limits by using an ND as opposed to HSS... ~1/2-1 stop gain... but still not enough.

This leves flash spread and flash distance. You could have taken the flash out of auto zoom and zoomed it to a longer/tighter pattern which effectively increases the power. Looking at the chart, going from the 35mm zoom position to the 70mm position would be a gain of 3 meters (~1 stop). You don't really need to fill the frame with flash, so this is a viable option.
The other option is to change the distance and a change by a factor of 2 (2x or 1/2) is two stops of power. Placing the flash half as far away at 70mm zoom would have gained you the 3 stops it was reporting (although I don't think it looks 3 stops under).

The biggest thing you could do would be to get the flash off camera, so that you can make these types of changes without having to compromise the composition.

Changing the zoom is a good option actually. Whilst the edges may be underexposed by the flash I don't think it would matter in this instance.

But as you say, and I said earlier it is strange that it was 3 stops under. As I said in post 50 I think you might have been right about the scene not been metered correctly. I think it's most likely due to it being in TTL BL rather than TTL as I said. If I ever get a scene like this I'll try it out for myself.

 

[sk66]

In TTL BL it tries to underexpose for a brighter BG which does increase the flash requirement... but it's also "what you want" (BL is the default behavior now for this reason). In TTL it would be more likely to overexpose the BG and need less flash, and perhaps not go to HSS (auto FP).

TTL and other automated features are great when you need them... but they add a layer/layers of complexity/difficulty to keeping control over what's happening. Manual modes are my "fall back" for when I get lost, and I'm getting results I don't want that I can't resolve at the moment (it's usually something small/simple I just fail to think of at the time).

And that brings up the last point which I purposely left out before. There really wasn't a need to be using auto anything in this situation. If it was taken in full manual (camera and flash) the metering would have been irrelevant... a couple of test shots to tune the settings and you're good to go... and there's a lot less potential "confusion" in getting to the settings/results you want.

 

[sk66]

For all I know "-3EV" might be the "out of range" warning regardless of how many stops it is actually underexposed. I just realized how rarely I ever actually look at a flash once it's setup. I even have the audios turned off...

 

[snerkler]

In TTL BL it tries to underexpose for a brighter BG which does increase the flash requirement... but it's also "what you want" (BL is the default behavior now for this reason). In TTL it would be more likely to overexpose the BG and need less flash, and perhaps not go to HSS (auto FP).

TTL and other automated features are great when you need them... but they add a layer/layers of complexity/difficulty to keeping control over what's happening. Manual modes are my "fall back" for when I get lost, and I'm getting results I don't want that I can't resolve at the moment (it's usually something small/simple I just fail to think of at the time).

And that brings up the last point which I purposely left out before. There really wasn't a need to be using auto anything in this situation. If it was taken in full manual (camera and flash) the metering would have been irrelevant... a couple of test shots to tune the settings and you're good to go... and there's a lot less potential "confusion" in getting to the settings/results you want.

Agreed I would have used manual too.

 

[Phil V]

Nope it is me who is incorrect thanks to an incorrect article, so I apologise for this. It is indeed as you say, one large pulse for shutter speeds up to sync speed and pulses in HSS. That'll trust me to believe the internet (well that's assuming that the latest articles I've read are correct of course )

The article you linked was correct though. It was your misunderstanding of it that led to the confusion.

 

[snerkler]

The article you linked was correct though. It was your misunderstanding of it that led to the confusion.

Yes and no. I linked to that thread purely for the diagram, I didn't read the article. The mis-information was from another article I have double checked on Nikon USA (has much better info than Nikon Europe imo) and that confirms that HSS is short pulses.

 

[mr_pants]

I've skimmed through this thread, so forgive me if I've missed something and got it completely wrong. Disclaimer: I've never used Nikon flashes, only Canon

If you can get your shutter speed below the flash sync speed, things will get easier. If not, use the high-speed sync feature of the flashgun.

The Canons have a flash exposure lock feature. I usually point the camera directly at the subject and hit the FEL button, then recompose with the subject off-centre and everything is mostly OK. I presume that the Nikon system has the same sort of thing. If, for instance, the flash system is trying to get itself set up while looking almost directly at a bright sky or scene (as is pretty much in the centre of your posted pic), it will completely dial down the flash output, as the flash system does a quick test-fire before the main exposure, just to measure how much light to put out.

 

[sk66]

The Canons have a flash exposure lock feature. I usually point the camera directly at the subject and hit the FEL button, then recompose with the subject off-centre and everything is mostly OK. I presume that the Nikon system has the same sort of thing.

Nikon has the same (FV lock)... it just has to be assigned to a button.