Film-Tech Forum ARCHIVE: Checking illumination of LED reverse scan

This is topic Checking illumination of LED reverse scan in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=006272

Posted by Tom Wienholt (Member # 1510) on 06-15-2004, 01:01 AM:

The installation instructions for a Kelmar reverse scan reader require the use of illumination test film. What is the proper procedure for using the film to test the uniformity of the LED output and what should be done if is not uniform? Thanks.

Posted by Randy Stankey (Member # 64) on 06-15-2004, 09:29 AM:

Do you have some green, "Dolby Tone" test film? Make a loop of that and run it.

Set your oscilloscope for "dual trace" so that the left channel is on top and the right is on the bottom. Carefully move the LED up, down, left or right until you get the strongest signal that gives you the most uniform picture between the top and bottom traces on the scope. When you have the best signal, lock it down. You can move on with the rest of your A-Chain process.

If you don't have a scope you can look at the level meters on the processor and get a good approximation. It won't be as accurate as using a scope but it will be good enough until you do get one.

All of this assumes that you have the reader and LED assemblies installed correctly. Also, I assume you have checked the adjustments on the sound processor and put them back to the "starting point" as outlined in the A-Chain procedure in the manual.

For instance, if the trim pots for your preamp gain aren't set right to start with you'll never be able to get equal signal in both the left and right unless you go back and "zero out". IF your reader and LED aren't installed so they are "plumb, level and square" you'll have a hell of a time getting a good A-Chain, no matter how hard you try.

I imagine you're good on those points but it doesn't hurt to double check... Measure twice! Cut once! [Wink]

Posted by Gordon McLeod (Member # 33) on 06-15-2004, 09:49 AM:

The dolby illumination alignment film has 6 tones recorded in 6 positons on the soundtrack 3 on the left and 3 on the right channel
As such when connected to an RTA show the relative level at 6 points across the width of the track
For low distortion it is more important that the illumination be level across the wisth of the track rather than peaked out

Posted by Tom Wienholt (Member # 1510) on 06-15-2004, 10:30 AM:

So what happens if the signal is not level across the width of the track?

Posted by Rick Long Jr (Member # 541) on 06-15-2004, 08:40 PM:

You will get distortion. Dolby tone is fine for up and down position of the L.E.D., but as it is only 50% modulation, you need to use the Cat (566 I think) illumination uniformity test film to be sure you are illuminating the entire track laterally. Cat 97 and buzz track will help if you pay attention to the wave forms on a scope.

Posted by Tom Wienholt (Member # 1510) on 06-16-2004, 12:09 AM:

What adjustments have to be made if the track is not uniformly lit?

Posted by Sam D. Chavez (Member # 1841) on 06-16-2004, 01:12 AM:

The following is my method for setting the aforementioned.

The easiest way to set peak illunination with LEDs is to run a loop of Cat. No. 97 with your scope set on X/Y mode and the Dolby preamp gain controls set all the way to max or min.

You will get a Plus sign representation as Cat. No. 97 is alternating 100% modulation tones. All four legs of the Plus sign or vector should be perfectly straight and equal in length if your scope is set to the same gain on both channels. It is very simple easy to tell when you have hit the peak without the scope losing the trigger.

This is also useful to see if the LED bracket is all the way tight as you can bend the bracket up and down and see if returns to the same place every time.

Once you've hit the peak, run Cat. No. 566 and use the RTA with both channels shorted together to see the 6 bars Gordon mentioned. You should be within 1 dB if everything is cool. The film is only accurate to 1 dB bar to bar so don't expect perfection.

With CE readers you can move the LED back and forth until you make yurself crazy, eventually get it perfect, or give up. With Kelmar and BACP the LED lateral position is fixed and the solar cell moves instead so the options are a bit more limited, but you will still have your sanity.

Posted by Tom Wienholt (Member # 1510) on 06-16-2004, 02:16 AM:

If you move the lateral position of the cell then wouldn't it mess up the left/right alignment that was previously set? I still don't understand how you can make adjustments for illumination without messing up the other settings.

Posted by Steve Guttag (Member # 268) on 06-16-2004, 07:40 AM:

You can't really use the solar cell to compentsate for misaligned LED. You would have to move the film with the lateral guide roller, then move the cell to compensate.

If you go crazy moving the LED with the CE reader, you aren't very good. Cat. 566 is also the easiest film to set LED height for maximum gain. If the LED isn't in that sweet spot, you are apt to get microphonics created by the vibration of the LED.

Generally, you won't have problems with Simplex's LED postion since they had decent consistancy. With Century, watch out...no two were the same but they were adequate in the mono era.

If you have a fixed LED bracket system, on a Century, you may find yourself shimming the LED bracket to get good alignment.

With the CE system, you can adjust the LED to match the film and wit the 30 series, you can also adjust the solar cell (laterally) if you need to compensate for a poorly machined soundhead.

The only time I drag out Cat. 97 anymore is to reposition the cell and that is rarely.

While I've never been a fan of multi-loops (some splice two or more test films together)...I can generally do a combination analog and digital basement reader in four loop passes with reliable and precise calibration for both systems.

Steve

Posted by Gordon McLeod (Member # 33) on 06-16-2004, 02:48 PM:

Many older century and RCA soundheads don't have enough of a recess in the scanner to keep the LED from rubbing and need to be turned down a bit

Posted by Sam D. Chavez (Member # 1841) on 06-16-2004, 09:23 PM:

The prcodure I described finds the absoulute peak of the light into the cell and since it is definitively peaked it gives the most headroom for longest LED life.

The problem with using the Cat. No. 566 has mainly to do with LED display analyzers where the resolution can be 1 or 2dB between segments. A slight move of the LED and some of the display moves up to the next level before others. Since the film is not perfect and has divots of it's own you are left with a feeling of uncertainty as to whether you got it perfect.

I've had no particular problem getting even illumination with a fixed LEd as the LED is .144 wide and the film is .084. Leaves room for a little slop.