I have two 580EX with CP-E4 battery packs and an ST-E2 for off camera TTL.
I have been having trouble with consistent exposures, even with freshly charged batteries.

The Canon documentation says to use the same type batteries for the pack and the Speedlite. I have taken that to mean all rechargable or all alky etc.

Yesterday I noticed that one pack's red indicator light was off while the other was on. Swapping the packs between Speedlites seemed to have no impact.

Any suggestions? It's frustrating to have tools and have proper operation seem so difficult to achieve.

Well, ETTL is automatic, so unless you can duplicate each time how the flash is being metered between shots, there will always be some variance, whether minimal or great.

It takes a long time to get a good grasp on the ETTL system, how it operates and such.

More information would be helpful, such as what batteries and chargers you use, and whether this is a new problem which just happened after a period of normal performance or something which has been happening consistently since you've owned the units.

If swapping CP-E4 packs between flashes resulted in the same pack being slower I'd suspect a battery problem first. One bad battery in a set will adversely affect performance of the entire set. Often a bad battery will test out OK with a volt meter, but will fail to operate correctly under full load. The best way to test battery related problems is to use with a fresh set of Alkalines as a benchmark for performance.

A good conditioning charger like a MAHA or LaCrosse which charges each cell individually makes a big difference in battery performance and life. Some brands of NiMH batteries such as MAHA PowerEX and Sanyo perform better than others such as Energizers. New NiMH batteries need 5-6 full charge cycles before reaching optimum performance.

Note sure if this is related to your problem with consistent exposures but its worth mentioning that Canon 580ex flashes have a quick shot mode when used in ETTL. They will fire again before the capacitors are fully charged which can result in under-exposure in a rapid series on shots. The pilot light will first turn green when ready for quick-flash (1/6 to 1/2 power), then red when fully recycled (actually about 80% charged) This is covered on page 10 of the 580ex manual. Its impossible when shooting to keep track of what the pilot lights are indicating so you need to run some tests where you fire the camera while watching the ready lights to get an idea of how rapidly you can actually shoot and maintain correct exposure. A full charge with internal batteries takes 6 sec. Even with the CP-E4 it will take 3 sec.

With ETTL there is no indication of how much power is being used, so its also informative to benchmark the flash recycling time in M mode at 1/4, 1/2 and full power to get a grasp on actual performance.

I was referring to what appeared to be inconsistant battery performance. I am painfully aware of the tendency for ETTL to provide unexpected results since it is using the focusing distance and metering system to set exposure.

I expect that more reliable results could be achieved using spot metering which I usually do.

I have 20 Sanyo Eneloop batteries and two Sanyo chargers. The batteries are rated 2000mAh.
I also have 16 Energizer batteries that are rated for 2500mAh.
I also have 4 Duracell batteries that are rated for 2500mAh.

I have been loading the flash and battery pack with all the same brand batteries.

I set up one flash with the Sanyo's and the other with the Energizers. Both seemed to be working OK, then I noticed that the red indicator light was on for one battery and not for the other. The red "ready" light on the flash unit and the ST-E2 seemed to be recycling about as it should. Yet when I would wait a long time between shots the exposure seemed to be almost too hot. When waiting for the indicator to show ready many exposures seemed to be at least 1/2 stop under. Some of this I attribute to metering, but a lot of it seemed to be the flash under performing.

I will do some testing with the two units and the packs, along with ETTL and manual to see if I can get a better handle on what exactly is happening.

The difference in performance is likely due to the fact you are using different brands / power ratings in the two units. Try putting identical batteries in both CP-E4 units and set the CF.n of the flash so the internal batteries are only used to power the controls.

The fact that when you wait for the units to fully recharge (which actually occurs a sec. or so after the red light appears) indicates that the variation in likely due to shooting faster than the flash can keep up. Keep in mind there is only one pathway for power to the condensers and if the power is heading out to the flash it prevents the flow of current from the batteries which is trying to recharge them. Its a bit like trying to fill up a bucket with a hole it in: the bigger the hole (power going to the flash) the longer it will take to fill the bucket and at some point the incoming flow just will not be able to keep up with the outflow.

The way the exposure control is engineered it will keep flashing to the point of depletion even if the camera knows the flash power isn't adequate for correct exposure (as evidenced by the confirmation lamp) because any flash is better than none. Bottom line is that if you want consistent exposures you just need to slow down your shooting to the pace the equipment can handle, or use more robust equipment


Spot metering is only as good as your ability to interpolate correct exposure needed for the tone measured. When metering a spot the resulting reading will try to reproduce that spot as a middle gray value. If the spot measured is actually textured white, reproducing it as white requires you the photographer knowing how much more exposure, in f/stop adjustments to EC, FEC, shutter or aperture (depending on shooting mode) is needed to move the reproduction of the metered area from gray to white.

I started out in the early 1970s with spot metered zone system B&W and done quite a bit of testing with the various metering combinations of my Canon 20D and concluded the quickest path to correct exposure of the highlights is to point and shoot a baseline shot in evaluative, look at the results in the playback over-exposure warning (OEW) and histogram and then adjust as necessary. Evaluative compares zones, finds the brightest and darkest, and makes a guess at an exposure that will reproduce the highlights correctly without blowing any detail. It automates the manual interpolation process necessary with spot metering. Getting correctly exposed files is simple: just increase until the point were textured whites clip in the OEW and then reduce exposure by 1/3 stop. The OEW reveals both when and exactly where detail is clipping.

I did some quick tests with Alky batteries in the flash and packs. I used the motor drive function to indicate speed of recycle to the point where flash would fire. (knowing that was only about 80% of complete.) I also set the CFn on each flash to use only the external battery for recycle.

Results indicate that one flash unit preforms worse than the other. I used the same batteries for both flash units and both battery packs to minimize variable.

The under-performing Flash also does not activate the red indicator light on the battery pack when switched on. Using the motor drive function on a 5D the lesser performing Flash, with either battery pack attached, had 5 black frames between each exposed frame. The exposure for the first and subsequent "flashed" frames was essentially identical on the histogram.

The other flash, again using same batteries in flash as well as packs, had 2 black frames between each exposed frame. And again the exposures had essentially identical histograms.

Chuck, I also did some basic testing with exposure reading options and arrived at essentially the same conclusion. Back in the film days I used a combination of incident and spot metering depending on subject and film being used. It appears that ETTL with evaluative metering offers a base line similar to incident from which to make exposure compensation based on subject and intent.

Now that I have confirmed that there is an equipment issue, I will have to decide if its worth sending the flash in for repair or just live with it.

Good that you've isolated the problem to the flash. I've read of a couple instances of people having two flashes with different recycle performance getting them fixed successfully by Canon. Canon has a flat rate fee for flash repair, about $125.

If you opt to live with the problem using the slower of the two flash units as your fill source, which typically is lower in power than the key light, would be the logical approach.

I now have two 580EXII's (and two manual SunPak 422D's, one or both of which I may throw in for back lighting, etc.) and am wondering if the ST-E2 has any place in my camera bag.

I have also seen some irregular lighting using the ST-E2 but cannot determine if it is the weak link or am just battling the battery/charge time issue from above which also makes me wonder if I should keep the ST-E2.

I’m following this thread with great interest!

-rce-

On the plus side the ST-E2 is very light and compact and its pre-flash signals are in the deep red, near-IR part of the spectrum human eyes can't see well so subjects are less likely to blink due to pre-flash.

On the negative side, in comparison to a 580ex/580exII/550ex used as Master, the ST-E2: 1) has a much smaller effective range, especially outdoors; 2) is only designed to control two groups (A:B) and only in ETTL mode. The the 580ex/580exII/550ex as Master has a much greater signaling range can remotely control three groups (A:B C) in either ETTL or Manual power modes. The ST-E2 can be used to fire any number of Canon slaves set to M mode, but changes to power level must be done on each slave flash.

I've used dual flash using one flash on a bracket for fill + one off camera flash as key light since the early 1970s when I learned the dual flash technique with manual flash from the guy who introduced and popularized the dual-flash technique for wedding candids. It allows seamlessly switching from single flash ideally positioned over the camera on the bracket, to dual flash in situations where two flashes are practical. The simple expedient of putting the off camera flash on a compact stand with wheels (I use a converted IV stand) makes the use of two flashes practical is nearly any situation. If the off camera flash can't be aimed precisely for short lighting simply parking it behind the action will produce 3D modeling via back/rim lighting which is far better than any single flash solution.

When switching to Canon I never even considered buying an ST-E2. Sure it saves a weight on the camera, but I see needing two stands for the flashes makes the overall set-up far more cumbersome than the bracket / single-off-camera-flash approach. A fill source kept near the camera axis creates few shadows visible to the camera. If there are no visible shadows there is no need for a huge modifier to soften them. Also with fill over the camera the apparent softness of the key light can be modulated with the lighting ratio. So its actually possible to get soft looking lighting in the overall perceptual sense with a pair of flashes in an overlapping neutral fill / key light configuration with relatively small modifiers.

If you want to mix Canon wireless with third-party manual flashes the best approach is to use the Canon flash in the hot shoe in either M or ETTL mode as Master to a Canon slave, then trigger the third-party flash via radio trigger connected to the camera PC outlet. The Canon will do its normal pre-flash sequence, then all the flashes will fire at the same time for the main exposure. The "fire main flash" signal is sent via the large center pin of the hot shoe and the PC connector at the same time. The pre-flash is controlled via the other 4 pins in the hot shoe.

Click the WWW button below for tutorials on Canon flash use...

Chuck

Thanks a bunch, Chuck. I ALWAYS enjoy your explanations and suggestions.

Two things.

1) anyone want to buy a St-E2?

2) Where can I get a converted IV stand?

Rich

So Chuck...

In short you don't recommend STE-2?.... is this STE-2 doesn't really work at all?