View Full Version : Motion Blur

Nigel G
31-08-11, 22:06
Having made a bit of a thing of steam locomotives recently and got images I'm reasonably happy with for most of the regular vistors to my part of the country I thought this morning I would try and add a bit more movement to the image as Dominion of New Zealand left Poole on the run to Bath.

As they start this trip from Poole they have to pick up momentum and puff nicely. My spot is about 1/2 mile out of the station and you can stand at the fence right by the side of the track so they pass about 5yds away.

The effect I hoped to achieve was a clean image of the loco but with motion blur on the runnning gear. I set to full manual - opted for 160th at f5 and an appropriate iso. IS on and pre-focussed on the track at a point where the second (small) wheel is on the image and then keeping the camera still and fired a sequence as the train passed.

There is clearly motion blur on the front of the loco suggesting my guess of 160th was too slow but what surprises me is that the main drive wheels are, if anything, sharper than the front of the engine. Surely the wheels must be going round faster than the body is going forward so why is this so.

Any ideas on settings to achieve the desired effect.

01-09-11, 08:44
I don't know how fast the train would be going at this point but let make some assumptions.

Assume 30 mph = 1/2 mile/min = 44 ft/sec so with your 1/160 shutter speed the train will move ~3 inch while the shutter is open hence the blur of the train. (60 mph = 6" moved etc) Notice further back down the train, the blur appears less because the angle that 3" creates is smaller because of the greater length of the sides of the triangle so the apparent movement is less.

So why do the wheels look sharper? Well the rims of the wheels are in contact with the track and as the train is moving along the track at 30 mph (our assumption) the rim of the wheels is also move at 30 mph. If the wheels are say 4' in diameter, at 2' from the centre the spokes are only going to have moved 1.5" ... the nearer the centre you get, the less movement. So that coupled with the fact that they are some way back from the front of the loco, makes them look sharper than the front :eek:

The only way to get the effect you are after is panning. Making another assumption, say its a 24 spoke wheel, I would suggest that to get the sort of blur you are after you might want the wheel to turn say 2 spokes worth while the shutter is open. So 4' wheel has a rim diameter of ~12' (diameter x Pi = D x 3.14) so the spokes are 6" apart at the rim so to move 2 spokes worth ie 12" would suggest a shutter speed of ~1/40 sec or more.

Look at this image (http://tinyurl.com/3sgxa8z) in Foxy's gallery taken at 1/160 ... & I bet that prop was spinning much faster than your train wheels :)

01-09-11, 09:39
Great explanation I think. Lost me in first few lines.
My explanation the spokes moving up are going backwards in relation to the train. :D

01-09-11, 10:11
Very exhaustive analysis here....

Nigel G
01-09-11, 15:51
Thanks Clive - thats going to make for a very interesting challenge to get the composition right. I think your assesment of 30mph is probably about right. This one was picking up speed faster than Britannia did a couple of weeks ago but even so they are on a bit of a climb.

The trouble with panning will be were to stand as there always a fence in the way and for this shot I lean over/rest on the top strand of barbed wire to get a clean image and this will inhibit a smooth pan. Prehaps a set of steps is called for.:D

Next oportunity to test the theory will be next Tue afternoon although that one will be passing through, rather than starting from, Poole and may be going a bit quicker.:rolleyes:

05-09-11, 10:06
Great explanation I think. Lost me in first few lines.
My explanation the spokes moving up are going backwards in relation to the train. :D

Hahah, I got lost completely as well. Started reading and I felt like being back in a school's math class, when I needed a lot more time to think over something the others were saying. Or listening to my father's explanations, which are absolutely this exhaustive and I can't seem to stay focused for this long lol :D

05-09-11, 21:56
Panning takes a good deal of practice (even without the need for steps!) I've been doing it at motorracing for years & still only get it right on a small proportion of shots :(

If you have the chance to take practice shots with less interesting trains at the same spot it will drastically improve your chances of getting reasonable settings.

Nigel G
05-09-11, 23:36
Thanks for the imput Mike. I've tried panning before with usually very mediocre result although one biker got me my only ever POTW!

Certainly aim to give it a go as Sir Laminel is due through Poole at 17:22 tomorrow evening and if I can get away from work in good time that will let me practise on a couple of commuters first. However the steam locos tend to be a good bit slower than the modern trains and and although Clive's assesment of 30mph is probably about right I am trying to shoot a 45 degree angle which means the crossing speed will be slower still and I think slow panning is maybe even harder than quick panning:(

Must remember to set IS mode 2 to at least take that out of the equation.

06-09-11, 00:27
If you want to practice panning on things moving at 30 mph about 5m away go and stand on the street corner and photograph the cars gong past. I use one of the focus points in the viewfinder as a sight and try to lock it onto a specific point on my subject leaving space for the subject to move into.... but not as easy as it sounds :)

06-09-11, 06:23
Only the wheel hub goes at the speed of the train. The bottom of the wheel, where it's in contact with the rail, has no forward speed; whilst the wheel rim at the top is moving twice as fast as the train. But I don't think that helps with your problem though.

Nigel G
06-09-11, 09:43
Hmm - I am happy that the wheel hub will be moving forward at the same pace as the train so in a panned shot it should be sharp (assuming I get the panning right).

But it was rotational blur in the wheels I was hoping to achieve.
Clive seems to indicte that the wheel rim is moving round at the same speed as the train goes forward with the spokes (which are the bits that will actually blur) traveling less distance as they get closer to the hub.
Mike - not sure I see how a wheel can be going at differrent speeds at the top and bottom ?

Somewher RPM has to come into this doesn't it? The 4 smaller wheels at the front of the engine must have a faster rpm than the 6 big ones?

06-09-11, 09:59
It's a difficult one to get your head around. Imagine a vehicle with a caterpillar track: where the track is in contact with the ground it is stationary and the vehicle moves along it, but the top of the track is moving (at twice the speed of the vehicle - giving an average speed of the track equal to that of the vehicle). Now imagine the track getting shorter: the same principal applies even when the track is so short that it is circular, like a wheel. Still doesn't help with your problem though.

06-09-11, 12:58
At least you shouldn't have the problem I encountered on many of my recent racecar shots at Knockhill, where one end of the car is sharp & the other shows considerable vertical movement.
I guess I must have picked a spot where the racetrack was rather bumpy. Train tracks are usually fairly smooth.

Nigel G
06-09-11, 23:38
It takes 2 Steam Locos to haul a train up the hill out of Weymouth and for some reason Sir Lamiel wasn't available to help Oliver Cromwell so this weeks Weymouth Express was being diesel hauled Weymouth to Southampton.

So - in the 2 images below the leading diesel was the easier to pick up and get a consistant aiming mark on the front. Oliver Cromwell was at the back and travelling backwards so the target focus point was rather more hasty and was somewhere in the middle of the engine.

There's certainly lots of movement going on - probably too much - but I think that is because I am so close to a pretty big subject. Getting further away will cause fence problems again although at trackside a small pair of steps was great. At least the gloomy wet afternoon made getting a 1/50th easy.

Interested in your thoughts and observations.

07-09-11, 22:09
Interesting ... Looking at the diesel loco the front it's pretty sharp where you were tracking ... bit difficult to tell what's happening with the wheels.

For the Oli Cromwell, its nice & sharp in the middle of the engine particularly around the centre of the three large wheels .... where you were tracking and there's some great motion blur in the wheels.

So why isn't more of the image sharp? In this case the subject is tracking at an angle to where you are stood. Looking at the Oli Cromwell shot ... both the front (in advance of the tracking point) and the back (behind the tracking point) show blur. Because of the angle, parts of the subject in front of the tracking point are moving (relative to where you are stood) faster than the tracking point and those to the rear slower ... so both are blurred relative to the tracking point.

I would say your panning is pretty much spot on relative to the points you were tracking ... to get more of the image sharp - you need to press the button when the line of motion is at a 90 degree tangent to where you are standing - them the points in front & behind the tracking point are moving at the same (relative) angular speed.

The challenge is then that this is the fastest point of panning ... and you don't get any lead lines for your composition so you need to allow space for the train to move into. Practice, practice, practice - get on a street corner & photograph the fiestas & mondeos - lol