400d wrote:
Finally something interesting.. I am not an Electical/Electronics Eng, but here's my .02.
I own the ST-E2 and 580EX. ST-E2 AF beam sucks big time comparing to 580EX's, I did a shoot on a beach in the evening once, the ST-E2's beam won't go farther than 15m (if I remember correctly, the manual of ST-E2 may tell you the range of the AF assist light). Another thing about ST-E2 is the battery-2CR5..it doesn't last long, since every time you fire it, it fires the flash too (yes, ST-E2 has a small flash tube inside). 580EX is better, but it's not like we want to have the bulky hotshoe flash on our camera all the time..
Now, there are two ways to do it:
1) Get an IR flashlight, and hook it up to a pressure switch, turn on the IR when you are focusing. I need to do more test before I can confirm Canon's AF works with the IR LED I have..
2) Figuring out the pinout of the Canon hotshoe. In other words, once you identify the + and - for controlling the AF assist light, you can simply hook it up to a device with external power and a powerful LED (say, a red CREE LED @ 700mA + aspherical lens will blow the 580EX's AF LED away). Using an op amp like mpmendenhall said, will be the easiest way to implement it.
Thank you for the contribution! I think mpmendenhall has given us the pin information we need. That would be great if you could confirm that Canon's AF works with IR. (and would IR mess up the metering?)
Update: this P60 IR drop-in (http://www.dealextreme.com/details.dx/sku.12625) does nothing in my test. It doesn't even affect metering..unless you point it directly into the lenses. Note that the drop-in does produce a small amount of visible light.
So IR doesn't work, or someone may try another IR diode..
If we are going to use the op-amp, we will use pin 1 (upper left)
0
1 2
3 4
I am thinking about the P4/Q5 of CREE, which we will need a driver (LED runs at 3.7V, P4 @ 700mA or Q5 @ 1000mA)..I am wondering if we can create a circuit, so the power for the LED is coming for an external source? (like some sort of relay, maybe a NPN?)
400d wrote:
Update: this P60 IR drop-in (http://www.dealextreme.com/details.dx/sku.12625) does nothing in my test. It doesn't even affect metering..unless you point it directly into the lenses.
<cut> 400d wrote:
I am thinking about the P4/Q5 of CREE, which we will need a driver (LED runs at 3.7V, P4 @ 700mA or Q5 @ 1000mA)..I am wondering if we can create a circuit, so the power for the LED is coming for an external source? (like some sort of relay, maybe a NPN?)
How are you planning on providing power?
I have a flashlight that uses 2xAA batteries and puts out a lot of light. It uses a toggle/momentary switch to adjust its mode. You could just get one of those (or a 1xAA, or even a 1xAAA variant) and replace the switch with a relay/electronic switch that your circuit turns on and off. Of course, this may be larger/heavier/whiter than what you are envisioning.
Some examples, full light and head only. These may be over budget, but I don't think getting the parts separately will be significantly cheaper (ref,ref,ref,ref,ref,ref,ref).
400d wrote:
Update: this P60 IR drop-in (http://www.dealextreme.com/details.dx/sku.12625) does nothing in my test. It doesn't even affect metering..unless you point it directly into the lenses. Note that the drop-in does produce a small amount of visible light.
I assume you mean "does nothing" for both metering and AF? Thank you for confirming that, I guess visible light it is.
I am fine with the effect the light would have one metering, as under extreme low-light I will be shooting manual and chimping the histogram, but I wonder if we could make Av work. It would be simple to incorporate a momentary button on the unit that turns it off while you grab a metering and do an exposure lock, not the most convenient...
Yes, a simple way to do it without messing with whole lot of circuit design will be: http://www.dealextreme.com/details.dx/sku.1025
Most of the flashlights which are compatible with a P60 drop in, for example, Surefire G2, UltraFire WF-502B, Ultrafire C1...etc. will take this switch. Btw, try getting a flashlight with memory, if the light doesn't have one, it will switch to the next mode every time you turn it on (if the light has multi modes-strobe or SOS). Fenix's light will be an overkill (and overprice). Once you find the right match of flashlight and pressure switch, the only thing you have to worry about is to align the flashlight so it points to the direction of your lens.
Has anyone made a Low-light Autofocus assist yet? And if so, does it work & how did you make it.
I am a Pentax K100D owner (as well as a Nikon D70s owner) & I would like to get the white-light low-light autofocus assist (that my Nikon has) onto my Pentax K100D.
I stumbled across this thread while trying to see if someone has already done this.
I think mpmendenhall had a good idea when he mentioned taking a wired remote trigger and modify it to activate the white-light autofocus assist LED (or whatever light) when the remote trigger is in focus mode. (Because, I think the Pentax remote trigger connector is the same as the Canon. <grin>
The focus assist would work like this . . .
1) Plug in the system into the remote trigger connector.
2) When you press the button down half way on the remote trigger, it shorts out pins ground + focus on the connector to get the camera to go into autofocus mode, and it also enables the white-light autofocus assist light.
3) When the camera has aquired focus, you push the button in more & it will short out pins ground + trigger & trigger the camera. (but also, since the switch is not shorting out the focus pin, the white-light autofocus assist light would already be off.)
I found the pin-out for the Canon remote trigger here . . .
1) Some sort of white-light focus light set-up (that can be triggered by the system.)
2) The system (which might be a modified remote trigger)
3) Some creative rigging to get this system to sit firmly on the camera so that the light is stable and properly aimed & that the button that you push (half-way to focus & all the way to trigger) can be actuated in a convenient manner when you are holding / using the camera.
If anyone has built this system, I would be very interested to find out . . .
1) Does it work?
2) What does it look like?
3) How do I build one?
I remembered that I had an old Nikon F-401x that didn't work, so I tore it apart & I think I can use the two position trigger switch. I just have to de-solder it and test out how it works.
I also picked up a cheap ($1) closet light from the dollars store. (3 LEDs, 3 AAA batteries.) It's not bright enough to get lots of distance, but enough to test out this concept. I'll open it up tonight to see how the batteries & switch is hooked up. (If this thing works, I'll spend some more time to get really powerful LEDs, a proper reflector dish & maybe a focus screen of some sort to focus the light onto the center - so I'll have to keep the focus position on center.)
I just need to get a 2.5mm stereo plug. (I can't figure out a cheap place to get that. I might have to go to the Radio Shack (or The Source by Circuit City as it is known up here in Canada.) to pick that part up.
Does someone have a circuit diagram on how to get the closing of the ground to focus circuit to also trigger the LED so I can try to hook all this stuff up?
Does someone have a circuit diagram on how to get the closing of the ground to focus circuit to also trigger the LED so I can try to hook all this stuff up?
See my post earlier in this thread (#3), which shows a circuit which will activate an LED (providing a regulated current) when the switch is closed to ground. The "switch to ground" in that circuit is through the camera hot shoe, but it will work just as well with the switch in the remote release.
Why don't you just get a remote shutter cable, and rig that up so that when you push the button half way, the flashlight comes on momentarily? That way you don't have to worry about messing up your camera's electronics, and it will only be on while you are focusing.
(I'm just to lazy & cheap to get a perfectly good remote and cannibalize for this project . . . when I have a dead Nikon F-401x that I can get the switch from.)
Here is an update on where I am so far . . .
1) I got the 2.5mm jack (I bought a cellphone headset for $1 from the dollar store.) (I am going to cut the ear piece / microphone off tonight and measure the voltage & amperage from the +focus to ground & +trigger to ground on my Pentax K100D so I have an idea of what is coming out of those pins! (I'll probably be able to measure my buddies Canon XT later this week.)
2) I took apart the closet (stick-up) light and determined its a simple hook-up. Three AAA batteries hooked up in series, with a small resistor dropping the voltage for the LEDs.)
3) I still have to get the multi-stage switch out of the Nikon F-401x so I can check how to wire it. (I'll try and do that tonight as well.)
4) I did a little research and came up with the idea to use an Optocoupler? to isolate the camera's circuitry from the light-setup circuitry. (Either a 4N25 or a 4N28) [NOTE!!! I am not EE, so I'm just guessing and would Greatly appreciate some comment on this!!! <grin>
Again, since I am no EE, to me, the Optocoupler seems like a simpler device for me to figure out.
Here is my attempt at drawing a circuit diagram on how I would hook all this stuff up. (NOTE: I don't know if this is correct - and would greatly appreciate any input & I have no idea what values to make the resistors since I have not read the data sheet on the 4N25 or 4N28 yet.)
An optocoupler would work, but is probably overkill (and overcomplication) for this application. Optocouplers are useful when one side of the circuit might be at dangerously high voltages compared to the other --- the 4N25 spec sheet indicates protection against voltage surges up to 7500V(!). In the case of your circuit, the LED side of the circuit will never be more than ~4-6V from the camera ground (from the AAA batteries), so an ordinary transistor/opamp circuit offers perfectly sufficient isolation.
Note that you will still need drive current for the input side of the optocoupler (~60mA), which I doubt the camera body will supply through the switch as shown in your circuit diagram. You'd need to add your own current source to drive the LED in the optocoupler, with similar amounts of current to what the output LEDs will be using, which makes the optocoupler kind of redundant (you might as well then directly drive the LEDs). Also, I think you have the LEDs pointing backwards (reverse biased) in your diagram.
The opamp circuit I gave in post #3 on this thread is easily modified for your purposes. Just remove the two resistors and switch on the left and replace them with the focus signal coming from the camera.
The basic operating principle of opamps is pretty simple: the opamp senses the voltage at its (+) and (-) terminals, and adjusts the voltage on its output terminal until the (+) and (-) inputs are at the same voltage. If (+) is higher than (-), the opamp increases its output voltage; if (+) is lower than (-), the opamp decreases its output voltage. The (+) and (-) terminals are "high impedance" --- as if they had tens or hundred of megohms resistance in front of them --- so they draw practically no current of their own; they can be though of as passive "observers" of whatever they are attached to, not adding or subtracting from whatever voltage is already there.
So, here's what the opamp in post #3 is doing: consider if some voltage V relative to ground is applied to the (+) input terminal. The opamp will adjust its output until the (-) terminal is at the same voltage, V. In order to get voltage V at the (-) terminal at the top of resistor R_LED, there must be current I=V/R_LED flowing through the resistor. The (-) terminal doesn't provide any current of its own, so the only way to get current through R_LED is to supply it from the output of the opamp, and have it pass through the LED. Thus the LED is given a constant current of V/R_LED.
Where's the battery in my circuit? Along with the (+), (-), and output terminals, an opamp has two other connections that I didn't draw in the circuit diagram. These are the positive and negative power supply connections on the opamp, on which the opamp draws in order to set its output at the right level. Rather than being in series with the LED, the battery is connected across the opamp's power inputs (and there is no need for your 'R1', since the current is regulated by 'R2').
To trigger the light off the camera output, you'll first need to measure what kind of signal the camera is putting out on the focus switch. You can then use this voltage as the V on the (+) input of the opamp, choosing the value of R_LED to get the right current V/R_LED through the LEDs.
I can't tell you how much I appreciate your patience!
I read through your description. (I sorta understood it.)
What type of op-amp would you use? I did a quick search, and the only name that seemed familiar to me was the 741. Would this be the type of chip you would use?
I printed out the pin-out and will be looking at this pin-out while reading through your post 3 & latest post to really try to understand this.
I'll measure the voltage off the remote connector tonight.
The good thing about opamps for simple designs like this is they are pretty much interchangeable. Use whatever you can get easily (a 741 would be fine). The differences between opamps only come into play when you are trying to do "fancy" things with them --- high power, low noise, high frequency, low power dissipation, etc. --- where you need to worry about stretching the limits of the device. It would be difficult to find an opamp that wasn't suitable for this type of basic, low-speed (millisecond time scales), low-power switching application.
Another note about the circuit design:
The input voltage V to the (+) terminal of the opamp is also the voltage drop across the current-setting resistor R_LED. The total voltage available to the circuit comes from the batteries (e.g. ~4.5V for 3xAAA). You don't want to "use up" all that voltage drop on the resistor R_LED --- if the input signal voltage were, e.g., 3.5V, that would require 3.5V across R_LED and only leave 4.5-3.5=1V for the LEDs (and losses in the opamp) --- far too little. That's why my original circuit in post #3 has the voltage divider (400K and 100K resistors) on the input voltage: it reduces the input voltage signal by a factor of 5, so not too much is "wasted" on R_LED. You'll probably want to do something similar with the signal from the camera --- use a fairly high impedance voltage divider so the signal at the opamp input is reduced to well under 1V.
I still don't understand how could the LEDs automatically blank out during exposure. I have also found that voltage of pin 2 and 4 change once you press the focus button (I assigned it to the * button on my camera).
Voltage change:
0
1 2
3 4
1:4.79V
2: 2.18V
3: 5.7mV (unchanged when focusing)
4: 1.243V
For the experiment I am trying to put together, I am not using any signal off the hot-shoe.
I am going to put a separate shutter button on my camera that will plug into where a Canon RS60-E3 would plug into. (The 2.5mm connector on the side.)
That remote trigger connector (that the Canon RS60-E3 would plug into) has three contacts. (1) Ground, (2) Focus, and (3) Trigger.
I'm doing this on my Pentax K100D, which has a similar connector.
So, when I plug in a 2.5mm stereo connector into the plug & short out the ground & focus, my camera starts trying to focus. When I stop shorting out ground & focus & short out trigger & ground instead, my camera fires and takes a shot.
So, the idea is . . . I'll make my own shutter button (which I am pulling out of my dead Nikon F-401x camera) and rig it up so that . . .
1) When I press down half way, it (in essence) shorts out the Ground + Focus to the remote trigger connector (putting my camera into focus mode), while at the same time triggering the OpAmp to light up an LED that my camera can use to help it lock onto something to focus on.
2) When I "hear / see" that the camera has acquired focus, I press the trigger down all the way, which cancels the shorting of Ground + Focus (so the camera stops focusing & "hopefully" my LED is turned off) and then shorts out Ground + Trigger, which then makes my camera take the shot.
If this works, it will work for both the Canon & Pentax dSLRs (and probably most dSLRs that use the same 2.5mm connector.) & It will leave the hot-shoe open to plug in flash modifiers, such as the Gary Fong Puffer.)
For both the Focus to Ground & Trigger to Ground, I measure 3.65V. (It seemed to fluctuate between 3.65V & 3.67V, but I couldn't tell if that was me just not being able to hold the leads still.)
I read through your descriptions again and drew up this . . . my 3rd attempt. I know it is probably still wrong, so your advice is greatly appreciated!
Now, if my diagram is kinda right . . . I still don't understand this . . .
1) Do I have to connect V+ to the inverting input (instead of the non-inverting input like in your diagram) [because I am not using the signal from the hot-shoe to "cancel" the LED light, but using the signal from the half-pushed trigger button to "enable" the LED light?]
2) The Focus to Ground will always be shorted out through the 400K + 100K resistors (so 500K total). Won't the camera see the voltage that does get through and put itself into focus mode? (I guess I can use a 400K + 100K resistor and put them across the focus & ground in series and try it out?)
3) If the 500K of resistance is enough to prevent the camera from going into focus mode, when I short out the Focus trigger, it sends V+/5 (3.65V/5 = 0.73V) through, would this been enough to trigger the focus mode? (I guess I can use a 400K resistor and put it across the focus & ground and try it out?)
4) If I decide to only use one LED, I probably won't need 3 AAAs? Right? But still, when the circuit is activated, the power comes from the battery and goes through the LED, how does it not go back through R_LED into the camera?
5) What is operating current for the LED. I looked up some specs on LEDs & they have two current readings.
I am not an Electical/Electronics Eng, but here's my .02.
