Sunday, January 27, 2008

Off-Camera Speedlite Setup for Canon...


When putting together a kit for off-camera flash with my Canon 20D I found I had to pull information from several different places to understand connections, compatibility and limitations so I wanted to put the final result down on paper in case anyone else is in the same boat...


As an amateur photographer, if someone asked me what distinguishes a professional photographer’s work from the rest of us, my top two answers would be a trained eye for composition and knowing how to use lighting for effect.  Without a strong composition a photo will not convey an intended meaning or feeling.  Lighting is a part of composition that will not make a photo great by itself, but often distinguishes a really terrific photo from an average photo.  In landscape photography we often hear of the “golden hour” - one hour after sunrise and before sunset - where warm colours in the sky or long shadows provide strong impact to photos.  


Until recently, I used a speedlite for some indoor shots and macro (ETTL only), and thought that great lighting came from either natural lighting under the right conditions or in a studio where lighting would be incredibly complicated, expensive and not at all portable - enter Strobist.


Thanks to David Hobby, I learned that amazing lighting can be generated using only small speedlites.  Relative to the investment in good lenses, a couple good flashes, stands and umbrellas is quite inexpensive and can dramatically improve picture quality.  When I decided to put together a small off-camera lighting kit I wanted it to meet three criteria:

  1. I wanted to be able to use Canon ETTL with my flashes under some conditions.  I am a macro shooter and ETTL allows quick adaptability that might not be possible with manual flash settings.  I have a MT-24EX and plan to add a ST-E2 to my bag for firing Canon wireless ETTL flashes.
  2. I am interested in high-speed flash photography, using speedlites set to low manual power.  For this reason I wanted to be able to trigger them manually with trigger kits such as those available at www.hiviz.com.  Some applications are too fast to use the Canon wireless to fire the flashes so synch cords are essential.
  3. If I should purchase radio triggers down the road (Pocket Wizard or Radio Popper), I would like my flashes to already be compatible with them.

Based on these criteria, I purchased a Canon 580EX and 430EX, which are each capable of manual control, and manual flash head zoom.  In order to fire them manually with both speedlites off-camera or using a trigger kit, I also purchased two Paramount hotshoe-to-miniplug (3.5mm) adapters and a pc-to-miniplug cord.  I happened to have some stereo miniphone cables and a mini-phone Y-cord to use as an improvised synch cord.    The high-speed trigger circuits are easily adapted to connect to the flash with a mini-phone plug.  This allows me to cover all the shooting scenarios above, for a minimal setup cost, and my setup is ready for radio triggers at any time in the future.  Now I have the best of both worlds... full ETTL or manual as I need them.

Speedlite flashes fire by shorting a trigger circuit, and on some older flashes the trigger circuit is very high voltage, which can damage newer digital cameras.  This can easily be measured using a multimeter.  Canon's newer cameras are limited to 5V trigger circuits.  On the Canon speedlites, the center pin on the base is +5V, and the rail contact (hidden on the top side of the hotshoe) is ground.  As an added bonus, the 430EX can be fired wirelessly by the 580EX, when the 580EX is fired using a synch cable.  This means you only need a synch cable to one flash (or one set of pocket wizards) to fire two flashes  - as long as the two flashes are close enough together.  A couple limitations that I’ve found when firing flashes manually are:

  1. When the MT-24EX is fired using the pc cable with hotshoe adapter it will not fire a slave flash set to manual power.
  2. When the 580EX is fired using the pc cable it will fire the 430EX in slave mode wirelessly, but the max synch speed is limited to 1/160.  This is presumably due to the time required for the master to communicate with the slave.
  3. From what I've read (but have no firsthand experience) the newer Canon 580EX II, which has a PC connection, will not fire a wireless slave if fired from the PC jack.  The workaround for this is to fire it using a PC to hotshoe adapter.
  4. When firing the flashes using the high-speed kits a silicon controlled rectifier acts as a switch to short the trigger circuit.  The Canon trigger circuits have some residual current flow that won't allow the SCR to reset after it fires, making the flash seem to freeze after one flash.  A solution that worked for me was placing a resistor in series with the SCR, thus allowing the SCR to reset - it took some trial and error to find the right resistance.  Similar problems have been encountered by some people using various types of wireless triggers on their Canon speedlites presumably because they also use an SCR to trigger the flash.

It's also worth mentioning that there is a more elegant solution to connect Canon speedlites to synch cables or Pocket Wizards than what I've done.  Michael Bass has been modifying flashes with sub-minijacks instead of using a hotshoe adapter or the crippled PC jack on the 580EX II.  This allows you to plug a sub-miniphone cord (2.5mm) right into the flash - cool!


If you have any questions or comments, (or see any errors) please drop me a line.  In later posts I'll talk about using high-speed kits to fire the camera, modifying my RS-80N3 and building an extension cable and trigger switch for my RS-80N3.


Cheers, Brad

Wednesday, January 23, 2008

Soft copy of a lens...??? (Pixel Peeping 101)

When researching lenses over the past couple years I have read a lot of reviews from great websites such as Fred Miranda, The Digital Picture and Photozone. Often, when discussing the Canon 24-70 f/2.8L or 70-200 f/2.8L IS, people mention that they got a "soft" copy of a lens. After dealing with an autofocus (AF) calibration issue on my Canon 70-200 f/2.8 IS I thought I'd share my experiences and offer a technique for checking your AF to determine if you have a problem.  I don't think that there are "soft" copies of lenses floating around, except perhaps in rare cases where there are manufacturing defects.  I think that there are sometimes AF accuracy issues for bodies and lenses, and this is easily rectified by sending gear back to Canon for a calibration.


The following image is a shot of my dog with the 70-200 and can be seen at 100% crop by clicking the image to go to my Flickr gallery (all shots here are downsized slightly by Flickr for posting). After the second trip to Canon I am finally happy with the AF performance. The detail in the eyes is excellent at f/2.8, and the AF is consistently giving good results.


Eyes


Many times I have seen people post questions in a forum asking about lens sharpness and calibration.  Many, many responses blame user error and seem skeptical that these problems even exist since they may never have had a gear issue themselves in many years of photography.  I have no doubt that technical problems arise.  Under certain conditions the gear has to work very hard, and this is where my problems arose... shooting close to minimum focus distance and maximum aperture (f/2.8) for portraits I was not happy with the detail, particularly in the eyes.  I didn't need to do lens tests to be dissatisfied with my lens, especially for the price.


I sent my lens to Canon for a calibration shortly after I purchased it.  It definitely worked better after the calibration but I was still not satisfied with it.  In fact I got so fed up that I was going to sell it - so I decided to do one last test.  I reasoned that if IQ was improving faster by stopping down than IQ was degrading by bumping ISO, then I might just as well shoot at high ISO with an f/4 lens, since the lens was so "soft" at f/2.8.  What I learned was that the "softness" at f/2.8 as not the optics, but in fact the focus.  Of course as you stop down the lens the DOF increases, thereby masking an AF error.


To conduct the test I used the tripod, remote shutter release and mirror lockup.  The center AF point was used on my Canon 20D, which is a high-precision AF point on a f/2.8 lens.  I set up a test target with good detail (UPC) which is shown below.  The full frame is shown at left, and gives you an idea how demanding this test is when pixel peeping at 100%.  I also found that shooting like this there is some slight variation shot-to-shot in the AF lock, which unfortunately means some photos are less sharp than others, but I don't think there's anything that can be done about this.


70mm-crop comparison


I tooks photos using AF and compared these with photos taken using AF, but with small manual tweaks to the focus after the AF lock was established.  What I found was that the lens is excellent at f/2.8, when it's in focus - I was able to improve the focus by manually making small adjustments.  It should not be possible to manually improve upon the auto focus.  The shot below shows a manual tweak on the left, and pure AF on the right.  The lens looks pretty darn sharp to me at f/2.8, when the focus is accurate.


70mm-AF-MF comparison


So based on this test I decided to send the lens back to Canon for a second AF calibration.  It came back shooting much better, although under tough conditions still has some shot-to-shot variation as I mentioned previously.  I have chosen not to send my camera body in for calibration since the AF is now good enough for me to be happy.  Unfortunately for me I bought the lens used so had to pay for the first calibration at about $165 Cdn, but the second calibration was done under warranty.  Incidentally when I tested my 70-200, I realized that my 50mm f/1.4 had an AF calibration issue also.  It was calibrated by Canon under warranty.


The paperwork from Canon that accompanied the 50mm upon return did recommend that if the issue continued I should send my camera body to Canon for calibration also.  I think that ideally the lens and body should be sent in together for Canon to calibrate them each independently, and then together.


Hope that some people find this useful information.  And if anybody has a soft copy of a lens that they just want to get rid of, you can send it my way... ;)  Any questions or comments, please drop me a line.

Tuesday, January 22, 2008

Blogging...

Well, it's been almost 6 months since I signed up for a blog of my own.  Pretty skinny on content so far... but I decided I'm going to change that.  I have a few ideas that I've wanted to put down on paper for a while, so I'll be making time shortly to get started.

I've gotten some inspiration from a couple photography blogs that I read regularly.  There is so much shared information now available on the world wide web that it amazes me.  I would like to make my own small contribution to this continuum of shared knowledge that is now growing exponentially.

Cheers, Brad

Saturday, September 15, 2007

My own custom domain name...

Well, I decided to take the plunge and register my own domain name, www.bradsteels.com. It is set up to automatically forward to my Flickr gallery for now, and I must give credit for the idea to the Blog at the No Cropping Zone. Now my friends and family will easily be able to remember my web address.

Saturday, August 25, 2007

First Entry...

Finally started my own blog. More to follow...