Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Small, fast solenoid - or alternative. External shutter actuation.

Status
Not open for further replies.

astronomerroyal

New Member
Does anyone know of (or has used) a small, fast pull-type solenoid? By fast, I mean the pin is drawn in (0.5") in <10mS.

I'm currently using

Guardian Linear Solenoid-The Electronic Goldmine

which at 36v (max rating of solenoid is unknown) draws in fully in about 20mS. Not too bad, but I'd like something even faster, <10mS. At 24v is takes about 30mS. Extrapolating to 48v presumably gives a time of 10-15mS, assuming inductive effects etc. don't hamper things terribly. I'm reluctant to go to these high voltages, because of my general lack of experience.

in case you're wondering, I'm using the solenoid to open an external camera shutter so I can photograph flying insects. Inspired by this chap,
Equipment insects in flight 2009 Photo Gallery by fotoopa at pbase.com

I'd also be interested in any other fast actuators that can pull a lever really quickly, over a linear distance of 0.5"

I also wonder if a smaller solenoid would be faster, because the inertia of the pin is smaller...

Many thanks.
 
You should be okay with a higher voltage just don't let the pulse last very long. It's not so much about voltage as it is about current, you just don't want to melt the coil.

Are you using a digital or a film camera? Because if it's digital you can usually attach a circuit directly to the button to actuate the shutter instantly, no delay at all except what the camera introduces.

Generally speaking the smaller the pin/coil the faster it will move IF it can handle the current, but if it's too small it will go slower because it's harder for the small pint to tug on the cable.
 
Hi,

I'm using a Canon digital SLR. From previous experience I find that between activating the shutter button and the camera lifting the mirror and clearing the shutter, is about 0.1seconds (100mS). In that time a flying insect at full magnification can easily fly out of the frame.

The amazing Mr Footopa, using somewhat different hardware, achieves a 6mS delay between trigger and photo taken. I'd be very happy with 10mS. The shutter mechanism can handle this sort of speed.

(0.1s delay is also too long to capture the *first* lightning strike, which is another anticipated use of this device. )

With the solenoid I'm using, at 12v it can't even open the shutter - fairly pathetic. I'm only activating the solenoid for <<50mS, just long enough for it to open the shutter, so perhaps I could whack it with a huge voltage >>24v, using the strobe circuitry from a small disposable camera? Any thoughts?

(Interestingly, the camera I gutted for the shutter mechanism - dropped it into a lake - already had tiny electromagnet type things, I think for holding open the shutter during long exposures.)
 
If you want fast responce time you need to directly electrically trigger the shutter, then the only delay is the time it takes the camera's shutter to actually open. The remote trigger you're using, I'm guessing it's not an old wire type since this is a digital SLR, so that remote is probably just something that closes electrical contacts inside. Can you take a good photo of the connector at the end of the remote trigger? And also what's the exact model of your camera? I'm not sure what alterations you've made to this camera already.
 
Last edited:
I have both the canon digital rebel and the 40D. I looked up the shutter lag on the 40D and it is only 59mS, much shorter than on the Digital Rebel (>100mS). Not quite fast enough I feel, from what Footopa has written. since the camera's Xsync speed is 0.04mS (1/250s), I think that means the shutter is opened fully in less than 4mS. The rest of the 59mS is presumably mirror and metering...

I build my own microcontroller triggers. I simply use optocouplers to fire the pretrigger and shutter lines. I suppose one of my gadgets could actually be used to measure the camera's shutter lag under various operational conditions, since it independently fires the camera and flash with a programmable delay between them. Yes, I think I've done this before on a different camera, using a different gadget, and I found that the shutter lag is the absolute shortest delay the camera permits (e.g. even if using mirror lock-up, manual metering mode etc.).

Mr Footopa uses the Nikon equivalent camera and I think if he knew of an in-camera way to get <10mS delay, he would've used it.

... in conclusion I think I need to use the external shutter.

Attached a picture. Solenoid to upper right. pulls a level that opens the shutter curtain.
 

Attachments

  • external_shutter.jpg
    external_shutter.jpg
    100.7 KB · Views: 1,619
hmm guess you really need to use the solenoid then. If it's really short you can usually send a LOT of current through a solenoid. But aside from faster you're also making it stronger, you might actually start breaking things at some point. Try gradually stepping up the voltage 5 volts at a time or so, and measure the responce increase, if it doesn't get much faster soon or the solenoid fails then you know you need a different solenoid anyways.

I've seen a feature on some consumer digital cameras that allows you to actually take pictures of the past, it basically keeps constantly taking photo's and saving them to memory, and when you press the shutter button it saves the last three photo's from BEFORE the shutter was pressed. I'm not sure if something like that is practical or possible on your camera but I've websites that have alternate firmwares that you can load into the camera to get features.
 
Yes, I might need to add rubber washers to absorb some of the impact.

I've seen that buffering here and there. I have it in a Marantz solid state audio recorder. Allows me to record what I heard 4 seconds ago. Very nice for recording birdsongs.

As for this project, the insect will need to be lit with a camera flash (to freeze the action), so it's not a feature I could use.

thanks for your replies.
 
I build my own microcontroller triggers. I simply use optocouplers to fire the pretrigger and shutter lines.

I am trying to do the same thing in order to syncronize two cameras by using a PIC to set a delay on the fasted one. I have a Canon 1DMK2 and a Canon 20D.

I use a radio remote transmitter with a receiver in each of the cameras remote sockets. Both cameras are pre focused on my computer screen and set in maual focus. I then run an online stopwatch which counts in milli seconds - Press the remote trigger and then check the image in both cameras to see the difference in lag times.

I have another receiver connected to the PIC. It's outputs like yours go through optocouplers to the an N3 connector.

I find that when I use the PIC, even with the delay turned off, it still takes about 50ms longer that when I'm just using the receiver.

The optocouplers I am using are 4n25. Can you show a schematic of your input / PIC /output stages ?

Stopwatch can be found here - **broken link removed**

Cheers
Adam

**broken link removed**
 
I build my own microcontroller triggers. I simply use optocouplers to fire the pretrigger and shutter lines. I suppose one of my gadgets could actually be used to measure the camera's shutter lag under various operational conditions, since it independently fires the camera and flash with a programmable delay between them.

I am trying to build a similar system to yourself. I have 2 cameras, Canon 1Dmk2 and a 20D. I want to try and syncronise them by stlightly delaying the fastest.

I am using a radio remote receiver, conneted to a PIC, then plugged into the 1D. I have another receiver plugged into the 20D. Like yours, my PIC also outputs via 2 x optocouplers, in my case 4n25's

The problem I have is that when I use my PIC delay board, even with the delay set at zero, it takes an extra 47 milli seconds to fire the shutter than it does with the receiver plugged straight into the camera.

Can you post details of your PIC input / output circuit - I'm not sure if using the 4n25 is fast enough. I am running the whole system on the 3v supplied from the radio receiver.

Cheers
Adam
 
Dear Atronomerroyal,

searching in google to solve a similar problem, I've found your post.

Did you find a solution? What type of shutter are you using?

Regards.
 
Last edited:
Did you find a solution? What type of shutter are you using?

I didn't find a better solution. If I were to try it again I might look at liquid crystal shutters - some are very fast (millisecond shuttering time) but may not be entirely opaque/transparent.
 
You could charge a cap to a higher voltage, then dump the cap into the solenoid coil to give much faster operation.

Also the solenoid will move faster if the throw (distance the armature moves) is reduced. That gives a higher field strength and more linear force at the armature at the instant of turn on. If the throw distance is long the field starts weak and will slowly accelerate the armature, and the whole process will be much slower.
 
I've seen a feature on some consumer digital cameras that allows you to actually take pictures of the past, it basically keeps constantly taking photo's and saving them to memory, and when you press the shutter button it saves the last three photo's from BEFORE the shutter was pressed. I'm not sure if something like that is practical or possible on your camera but I've websites that have alternate firmwares that you can load into the camera to get features.

Think for a moment about what you said there ;)

Any images the sensor was able to image before the shutter was opened would just be black blank images. There is no light entering the sensor to be recorded.

The shutter MUST be opened before any images are recorded

Dave
 
Thank's a lot for answering. After a few research, I've found this: **broken link removed** referring to Ledex solenoids. They have high speed solenoids but didn't answer to my email request.

Another option I'm thinking is to dismantle a film camera, they had to use solenoid to move the shutter... Astronomerroyal, could you please tell me what shutter you've used? Maybe we can find a reflex camera with a shutter who don't need to be armed by a lever. For example, Canon about EOS 1N RS say: "Vertical-travel, double-blind, focal-plane electronic shutter with carbon blades and metal curtains. First and second curtains controlled electronically independently."

Maybe I can investigate on this.
 
If you can design a rotary actuator into your shutter linkage, you might consider using an analog servo-motor. The advantage would be an extremely fast time constant ... can't give specifics though. These motors have a maximum torque at zero rpm ... at take-off, so to speak. They can usually take quite a few amps ... probably just a matter of dialing it in.
The control scheme might consist of pulsing the motor in one direction to open, and the opposite direction to close ... if that is how the shutter works.
You would have to come up with a functional mechanical linkage between the rotational motion of the motor and the linear motion of the shutter.

Although DC servo-motors usually require a custom power supply to accommodate proportional control, it may be possible in this case to utilize an H Bridge configuration to achieve an on-off, reversible current pulse mode.
 
Last edited:
It's hard to imagine the mechanics, but you might consider using the actuator and magnets out of a disk drive. You could easily drive it with 24 volts. See if you can work the crash stops into your mechanics.:D
 
Rotary solenoids are pretty quick. The only problem is if you need them to kick hard and stay open, you have to reduce to a smaller holding current. They were used as the pen lift mechanism on HP plotters. I used them for mechanical shutter actuators.
 
It seems like solenoids ... of whatever type ... have a restoring spring, which returns the moving pin to its original position after the control pulse has ended. This return spring opposes the electrically induced primary force. If you have any significant pin travel length, your primary actuating force will consequently decrease as you extend the pin ... Remember F=kx? ... The more the restoring spring is extended, the more the negative acting force. The decreasing actuating force will decelerate the pin, and increase its travel time.

Getting an electric motor to operate the shutter would require some time, effort, and modification to adjust the pulse strength and pin travel distance, but would ultimately produce a faster shutter speed cycle, as compared to a solenoid.
 
I have a similar project in mind. However, looking at the fotopia page mentioned in the first post, the clear aperture of the Compur 0 shutter is only 25mm. It seems to me that would cause the image to vignette if it is in front of the lens. [In the camera, the shutter is behind the lens.]

I do believe that the only way to do such photography is to put the camera in B and use an electronic shutter before the lens. But I believe the clear aperture would have to be much higher. A good lens typically has a filter thread in the 55mm and up range.

Incidentally, the advantage to putting the camera in B is it bypasses all the power saving features. Otherwise most DSLRs would go into shutdown if not "touched" every few minutes.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top