Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

LED signal to open 7VDC CKT?

Status
Not open for further replies.

motleypixel

Newbie level 5
Joined
Apr 26, 2012
Messages
9
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
Austin, TX
Activity points
1,397
Hello everyone,

Pretty cool how the new post title generates a possible "search" for the user...nice forum module to implement. I've lurked here the past two days, so I've read a bit and feel I needed to register to get some dialog going on this project.

First post, looking forward to leveraging this community. Background: In a nutshell, I have a business degree, but was 6 years US Navy Submarines (reactor operator). I know fundamentals with regards to electricity and some solid state stuff, but really zero experience in design.

I have a one-off requirement and was wondering if this community could help.

First of all, the system: I have built a DIY Time-Lapse Photography Turn Table: **broken link removed**

I'm controlling the camera with a smart device called a Little Bramper: **broken link removed** Essentially this device can bulb-ramp the shutter.

I'm panning the camera with a worm-gearbox that has a 30:1 gear ratio, powered by a 7.4VDC 2-cell lipo battery, speed controlled by a simple PWM controller. The controller is variable from 0-100% modulation and will recall last setting upon pwr interupt.

Problem: As the turn table rotates very slowly (1 rev. per hour) and the camera begins taking pictures (starting out as a shutter open duration of 40ms or 1/40th sec and ~ 3 second delay before the next shot and then after time goes on the shutter open duration increases to the tune of 2+ seconds). I have found that shutter speeds slower than 2 seconds, with the moving turn table, produces motion blur that is too excessive.

Best solution: Stop the power to the speed controller which will stop the power to the DC 1rpm motor, which will stop rotation of the camera.

How to implement the solution?: The good news is that the Little Bramper has a Green LED that energizes for the exact duration of the shutter open cycle, so if the shutter is at 1/40th second, then the illumination period of the LED is 1/40th second, same with a 2 second shutter speed, etc.

I would like to find a way to harness this LED signal (either light from the LED or Vd across the LED) to electronically open the pwr CKT to the motor speed controller.

Any ideas here?

6965151606_ea8bbe7d31_c.jpg


Thanks!
 
Last edited:

Here's what I would do. Connect a comparator to the LED so that when the LED is on, the comparator will trip. Use the output of this comparator to drive a solid-state relay connected to the motor. I would stay away from optical stuff.

I might be a bit concerned with how the whole system works when the motor is abruptly started and stopped (backlash and bouncing, etc.) , but the gear reducer might help mitigate that.
 

Barry is correct in that the signal to the LED can be used directly along with a driver circuit.He also correctly points out the issues with a motor.
A gear reducer may work, however you should also consider using a stepper motor that will much better control than a normal motor.
 

Thanks guys for the input. Focusing on the motor and gearbox and the interrupt of power...this particular system is VERY tolerant of this...I should post a video. The DC motor is a 1rpm motor (3-12VDC) and I feed it a pwm signal at ~7VDC (2-cell lipo). The worm and gear mesh is tight and I've manually interrupted power to the speed controller in quick fashion and there isn't any noticeable lashing or jarring going on...it's just to small and slow to see these effects.

So I want to start working on a schematic and getting some components to start testing. Again, I have zero experience designing CKTs on my own.

I will start looking for comparators and how they operate. My assumption at this point in time is that my IP signal will be a square wave from the Vd of the green LED ~ 3VDC amplitude with the period being what ever the shutter is set on the Little Bramper. Then I want to create an open CKT on the pwr supply side of my speed controller so that pwr is interrupted for only the duration that the LED has pwr.

Thanks!
 

This is going to be tough going solo, I just haven't the experience. I suppose I could use a voltage comparator like this? **broken link removed**

Again, my signal is ~ 3.5VDC with a varying period of 40ms to 30,000ms. When I have a positive 3VDC I want to open my 7VDC PS to my speed controller. Can anyone give me a good start here?

Thanks.
 

Use a LM311 or similar comparator. Power it with 7VDC to pin 8 and ground to pin 4.

Tie the + input pin 3 to your 0 to 3V shutter signal. ( assuming signal goes high when shutter is open )

Wire a 18K resistor in series with a 10K resistor. Tie the 18K to 7VDC and the 10K to ground. tie the junction to the - input pin pin 2 of the LM311. (this should set trip point of the comparator to 2.5V)

Wire the output pin 7 to the base of a 2N3904 NPN transistor using a series resistor of 1K.
Wire the emitter to ground. Wire the collector to the minus side of the relay.

Wire the other side of the relay to +7VDC. Also wire a diode (1N914) across the relay reversed biased ie the negative side of the diode to +7VDC side of relay. ( diode is to protect comparator output from inductive kick of relay )

If your shutter signal is reversed (high to low) swap input pins on the comparator.
I'd attach a image but my work restricts access to do that.
You can increase the 18K resistor to get a lower turn off voltage from the shutter LED.

Theory of operation... Shutter signal goes from 0 to 3V on pin 3 of comparator. As the voltage on pin 3 exceeds the 2.5V reference set on pin 2 the output will go low De-energizing the transistor and relay, Removing the 7V from the speed controller.
When shutter closes the signal on pin 2 goes below the reference on pin 3 and the output goes high energizing the relay applying power to the speed controller.
If you PM me with your email address I can send you a PDF of the schematic to help clarify things.
 

Sounds good, a schematic would be nice. I'm working towards the design in the attached file (both wiring and schematic) it uses an opto-coupler and a relay, what do you think of this design? I figure about a 4ms response time for the ckt to cut power. I would like to speed this up if possible and I would like to consider adding other logic to the ckt later in the future such as only enabling the ckt for green LED signals > 1 second, i.e. only enable the c/o ckt when shutter speeds are 1 second or slower.

Thanks,
Roy
 

Attachments

  • wiring_diagram.jpg
    wiring_diagram.jpg
    155.3 KB · Views: 172

Hi Roy

I figure about a 4ms response time for the ckt to cut power. I would like to speed this up if possible and I would like to consider adding other logic to the ckt later in the future such as only enabling the ckt for green LED signals > 1 second, i.e. only enable the c/o ckt when shutter speeds are 1 second or slower.
That sounds like the sort of thing that could be done with a couple of (dirt cheap) 555 chips and a MOSFET to switch the motor power.

A couple of questions:
You mentioned that the turntable rotates one revolution per hour and the camera takes a photograph once every 3 seconds. So you're taking 1200 photos per revolution?! What is the purpose, if you don't mind me asking?

I presume the short exposure times are during the day and the long exposures are at night? One concern is that if you stop the motor for 1 or 2 seconds once every 3 seconds, the turntable will slow down to 1 revolution every 1.5 to 3 hours. Will that not be a problem?

The idea of stopping the motor only for LED signals > 1 second may have an wanted side-effect. For a 2 second exposure, the motor will stop half way through the exposure,resulting in a half-fuzzy image.

Perhaps it would be better to use a slightly more complex circuit that uses the length of one exposure to decide whether to stop the motor for the next exposure. That way the motor will either be stopped for the entire exposure, or not at all.

Regards - Godfrey
 

That sounds like the sort of thing that could be done with a couple of (dirt cheap) 555 chips and a MOSFET to switch the motor power.

Would like to see a working schematic :)

A couple of questions:
You mentioned that the turntable rotates one revolution per hour and the camera takes a photograph once every 3 seconds. So you're taking 1200 photos per revolution?! What is the purpose, if you don't mind me asking?

Yep, for time-lapse photography, those 1200 shots would produce a 40 second video clip at 30fps.

I presume the short exposure times are during the day and the long exposures are at night? One concern is that if you stop the motor for 1 or 2 seconds once every 3 seconds, the turntable will slow down to 1 revolution every 1.5 to 3 hours. Will that not be a problem?

Yep and I frankly haven't had any test runs, but my thinking is that it will basically give a slowing movement effect of the time-lapse...for the beginning of the footage the turn table will effectively move faster than towards the end. I would think I could compensate this by adjusting the PWM speed controller to slowly increase the speed of the turn table and also in post editing with velocity envelopes.

The idea of stopping the motor only for LED signals > 1 second may have an wanted side-effect. For a 2 second exposure, the motor will stop half way through the exposure, resulting in a half-fuzzy image.

You thinking is correct, but my explanation simply didn't come across correctly. When shutter speeds are faster than 1 second I really don't need the CKT to cut power to the turn table motor because the shutter speed is fast enough where motion blur will not occur, but once shutter speeds start to get longer than 1 second I need that motor de-energized. So when the LED has an energized period of > 1 second, then cut-in CKT to de-energize motor, which happens the very moment the shutter opens, well, close, about 4ms after the green LED signal happens. Get it?

Perhaps it would be better to use a slightly more complex circuit that uses the length of one exposure to decide whether to stop the motor for the next exposure. That way the motor will either be stopped for the entire exposure, or not at all.

If I was the engineer behind the Little Bramper, I think I could leverage many other, better, signals, but for now this seems like the easiest approach.
 


Okay skydiverc, looks interesting. I'm attaching a copy of it with some scribbles on it. So what would the wiring look like connected to my system, as I did with my attachment above with the CKT I'm working with and how it wires up to my little bramper and the motor system?

So with the design I'm using I will have to lift the existing green LED and move it and then land an opto-coupler to sense the voltage change when it would normally be energized. So I'm not using a comparator. Any design considerations that I'm not aware of here?

Thanks,
Roy
 

Attachments

  • new_ckt_comparator.jpg
    new_ckt_comparator.jpg
    95.9 KB · Views: 169

Would like to see a working schematic :)
This is the sort of circuit I'm thinking of now (having given up on the 555 idea).

The idea is that during the day the motor runs continuously. In the evening, when the average exposure time (for the last dozen or so exposures) rises to about 1 second, the circuit switches to "night mode" and stops the motor for each exposure.

In the morning, when the average esposure time drops to about 3/4 of a second, the circuit switches back to "day mode" and the motor runs continuously again.

I'm still fiddling with the details so it's not quite a working circuit yet, but does look promising!

The reason for the delay and different thresholds is to prevent it switching back and forth at night every time there's a flash of light or a cloud passes in front of the moon.

 

Thanks godfreyl!

Just a quick summary of what happens when I use this system. The usual total time I use the system is about 1-2 hours, never more than 2 hours. The system starts usually at 40ms shutter open and generally ends up at shutter open for about 20 seconds. The number of shots taken between 40ms and 1 second is roughly 500. the more I think about it I think just enabling the the c/o ckt all the time will be fine. The effect that will be noticed is the effective speed of the turn-table. If the turn-table is set as a rotational speed of 1 rev per hour and we start at a 40ms shot with about 3 seconds between shots that means that about ever 3 seconds the table stops rotating for a mere 40ms...not much stoppage, but towards the end of the time lapse capture there will be a 20 second stop of the turn table and only about 3 seconds of run time. So the result of the pan when the time-lapse is played back is a faster moving sequence at the beginning and very slow at the end. Now I can compensate for this a bit if I gradually speed up the motor using the pwd speed controller.

Last, I have to be up-front, I don't like the photo-cell design because it's too hard to isolate the light from the LED. This is the design you are showing right?

Thanks,
Roy
 

The system starts usually at 40ms shutter open and generally ends up at shutter open for about 20 seconds.
Oops. I thought you were bulb-ramping with a fixed interval, but you're obviously interval ramping as well. btw, I've belatedly looked at your website and followed the links so I now have a bit better understanding of the terminology and what you're doing.

the more I think about it I think just enabling the the c/o ckt all the time will be fine.
That would make it much easier. There's another benefit too: When the video is replayed, the rotational speed will reduce smoothly and continuously.

If the motor is stopped only for exposures longer than one second, then when replayed the speed would reduce suddenly at that point, creating a sort of jerk effect.

Now I can compensate for this a bit if I gradually speed up the motor using the pwd speed controller.
Even better!

Last, I have to be up-front, I don't like the photo-cell design because it's too hard to isolate the light from the LED.
No problem. An opto-isolator can be used, with it's internal LED connected to the Little Bramper instead of it's green LED. I think that's what you showed in one of your earlier pictures.

Looking at the user manual for the speed controller, I see that it switches the negative rail of the motor's supply, so the cutoff circuit will need to switch the positive rail.

The optocoupler can't do the job on it's own; it will need one or two transistors to help, but the circuit will still be a lot simpler with the cutoff permanently enabled.

I was concerned about the amount of current drawn by the motor, that has to be switched. However the datasheet for the 1 RPM motor says the current is only 90mA at max efficiency, with quite a hefty torque at that current. IIRC, you indirectly measured an average of 35mA over an hour. So it seems to be a really light load. I was envisioning having to use a large power device, but it seems not. Any thoughts on the maximum current to cater for?

Earlier you mentioned a 2 cell 7.4V battery but on your website you say you're using a 3 cell 11.1V battery. I guess the smart option is to cater for a wide range of voltages.
 

Yes, the ckt I provided in my OP uses an opto-coupler. I plan to stay with a 7.4VDC P/S.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top