Building A Professional Grade Shutter Tester

[ic-racer]

Oh yes, a closed loop system would be great, but the main drawback for any modification of this project is the lack of open source code.

I was hoping someone might offer up some open source code for the device, but from this thread and the on on GitHub, everyone is using the Hex code firmware (that probably won't ever get any updates or changes).

The Kyoritsu testers, on which this project is based, use a series of trimmer pots for the intensity calibration; so it is not closed loop either.

BTW When I analyzed the X-rite sensitometer, it appeared to me to have a closed loop system for monitoring the brightness.

X-Rite 334 Sensitometer: Inspection and Analysis

With the demise of film medical radiography, sensitometers have hit the used market. Currently on ebay there are about 80 sensitometers and about 80 transmission densitometers. But the sensitometers are about ten times less expensive. I found a seller with six working X-Rite 334 sensitometers...

www.photrio.com

 

[bernard_L]

but the main drawback for any modification of this project is the lack of open source code.

That is if you close the loop in software. With my background, I just assumed that the loop would be closed in hardware. The PWM output from the µC (properly filtered) would just provide the reference value that the loop strives to match. With proper scaling (in hardware, again) that voltage might just be the exact same one used to drive the LED in the base version.
In my previous post, among the issues that a closed-loop circuit addresses, besides thermal drift after turn-on I should have mentioned device ageing. Power LEDs have a finite lifetime (though a shutter tester logs fewer hours than a street lamp). The sensor photodiode can more safely be assumed not to age.

 

[ic-racer]

That is if you close the loop in software. With my background, I just assumed that the loop would be closed in hardware. The PWM output from the µC (properly filtered) would just provide the reference value that the loop strives to match. With proper scaling (in hardware, again) that voltage might just be the exact same one used to drive the LED in the base version.
In my previous post, among the issues that a closed-loop circuit addresses, besides thermal drift after turn-on I should have mentioned device ageing. Power LEDs have a finite lifetime (though a shutter tester logs fewer hours than a street lamp). The sensor photodiode can more safely be assumed not to age.

Clever idea! Do you think you will build one of these testers?

After re-reading my X-Rite 334 thread above I thinking out loud about pros and cons of using an electroluminescent lamp as a testing light source since they have constant color temperature when dimmed and a near perfect even illumination.

 

[RMN]

Assembled the tester but still need some cable management and the sensors. Luckily the darker yellow LED does have linear output! My rough calibration needed 12.9 V to get there. Hopefully I installed the fresnel lens the correct way because it was hard to tell which way the ribs and smooth sides are. Not sure if it matters but light is bleeding through the reflector.

 

[bernard_L]

Clever idea! Do you think you will build one of these testers?

It's a nice instrument, fit for a professional repair person. I get by with a quick-and-dirty circuit and a 30-year old Tek TDS320. For meter calibration, I found that the limitation is often the difference in wavelength response more than the global sensitivity.

 

[aconbere]

Have a camera that needs curtain tension adjustment so I'm back at the tester.

First a quick question: What button caps are folks buying? I've just been poking my finger through the hole like a scrub.

Second: Some advice on assembling the sensor units. I had enough trouble getting the diode aligned with sensor #1 that I took a different approach for sensor #2. I bent the legs and drilled the holes using the template on the PCB like usual. But then positioned and tacked the diode to the countersunk bit with a small dot of hot glue. The hot glue holds the diode well, but is viscous enough that it doesn't flow into the hole (and can be removed easily in the future if needed).

Then it was just a matter of getting everything all lined up (hot tip cut the legs at an angle from one end of the sensor to the other so you can do one at a time).

This made assembly an absolute breeze, can't recommend it enough.

 

[vandergus]

Have a camera that needs curtain tension adjustment so I'm back at the tester.

First a quick question: What button caps are folks buying? I've just been poking my finger through the hole like a scrub.

Second: Some advice on assembling the sensor units. I had enough trouble getting the diode aligned with sensor #1 that I took a different approach for sensor #2. I bent the legs and drilled the holes using the template on the PCB like usual. But then positioned and tacked the diode to the countersunk bit with a small dot of hot glue. The hot glue holds the diode well, but is viscous enough that it doesn't flow into the hole (and can be removed easily in the future if needed).

Then it was just a matter of getting everything all lined up (hot tip cut the legs at an angle from one end of the sensor to the other so you can do one at a time).

This made assembly an absolute breeze, can't recommend it enough.

I ended up doing a similar type of assembly for the sensor. I glued the diodes in place to make sure that the silicon sensors were centered in the holes. But instead of trying to line up the leads with the holes in the PCB, I just snipped them then connected them with jumper wires.

 

[aconbere]

I ended up doing a similar type of assembly for the sensor. I glued the diodes in place to make sure that the silicon sensors were centered in the holes. But instead of trying to line up the leads with the holes in the PCB, I just snipped them then connected them with jumper wires.

That would have been my next attempt if this hadn’t worked. I had considered it and then figured I’d at least try just fitting the leads. Save me a lot of soldering and wires. Happy that I’m not alone in considering the blind alignment to be a cursed procedure.

 

[ic-racer]

I finally finished the flash synch tester cable. First thing I had to do was to redraw the schematic to match the connectors that I'm using:

 

[ic-racer]

This synch cord was one of the cheapest I could find. I guess because the odd connector on the other end and the length (everyone is wireless these days? No one wants a 33ft cable).

 

[ic-racer]

That cable contained a white and black wire.

 

[ic-racer]

I bent the resistors like this so they would fit on the multi-connector:

 

[ic-racer]

Then solder the leads to the correct terminals. Make sure the Vcc goes to the correct side of the resistor. When the synch terminals in the camera are closed, it is shorting the 5v Vcc to ground. Without the 10k limiting resistor the Arduino can fry.

 

[ic-racer]

Works great!

 

[ic-racer]

Of note, when using it with leaf shutters, the measurement needs to be at the rim of the blades. You want to know the synchronization with respect to when the blades are fully open.