Crossover time 500us not 5us - DLP4500

Hello Carl,

Welcome to the DLP Advanced Light Control section of the TI-E2E community.

I would like to find out a little bit about your sensor, how it is connected to the scope, and where it is in relation to the projector?

1. What detector are you using?
2. How is it connected to the scope. 
   
   1. Are you using a trans-impedance amplifier or some kind of detector bias?
   2. What kind of cable are you using to connect to the scope and what is its impedance?
   3. What is the impedance of the detector?
   4. What impedance is the scope input set to?
3. Is the detector at the image plane of the projector or near the lens?

The waveform suggests that something is going on besides the DMD crossover.  It is suggestive of some kind of capacitive coupling of the detector system input to the scope.

Fizix

Thank you for your quick response. My detector is a biased Thorlabs DET36A/M. The output is connected directly to the scope (TDS2002B (60MHz)) using a 50 Ohms BNC-cable. The impedance is 50 Ohms on both. The detector is very fast (down to nanoseconds if I remember correctly). I have also tried another detector with the same result.

I have tested 2 units DLP4500 with the same result.

I have a OD-filter in front of the detector so a suitable amount of light is reaching the detector. Max voltage output of the detector is 10V.

Carl

The sensor is placed approximately 20 cm in front of the projector.

Carl, 

I am familiar with that detector.  You are correct it is a fast detector and your connections seem fine.

The remaining question I have is whether the detector is in the near field or the far field of the DMD? I.e. at the focal plane or near the lens?

Also be aware that the sequences turn the LED's off and on also, but they should not be that slow either.

Fizix

Carl,

Clearly our posts crossed each other.  Please make sure that the detector is as the same distance as you have set the focus of the projector (i.e. a sharp image at 20 cm.  If you cannot focus that close then move the sensor to the screen.

I will ask my colleague who is more familiar with this EVM to step in.

Fizix

The Image is fully visible on a wall 30 cm away. The detector position is about 20 cm away from the projector lens.
The LEDs should not be the problem. I didn't mention this above but I have also tested focusing a laser on a DLP4500 with the light engine removed. Focusing the laser as a small dot in the middle of the DMD. Measuring the transision time by placing the detector about 30 cm away at 24 degree.
Using the laser gives a similar result.

Hi Carl,

Cross-Over time: This time is referred to the mirror state change from 1 to 0 or 0 to 1 (cross-over); there is other possibility, 1 to 1 or 0 to 0 called as same-side transition time.

Both cross-over or same-side transition starts whenever the Mirror goes through the RESET cycle; let's assume time t = 0 when RESET cycle starts, the mirrors settle down to the requested state here the cross-over or same-side time is ~5us i.e., t+5us.

This time can be visualized on the oscilloscope by keeping a photodector and zoom in For example: Black to White image change, you can notice that at the beginning of intensity change low to high, if you zoom-in at point of high intensity change, you will see a damping signal for about < 5us duration, after that the intensity settle to the nominal level.

Now, there is other thing, the pattern display time,  this time can be defined as the minimum time required to display a pattern (load all the mirrors and then display), in case of binary pattern is it around  235us. So, if you are displaying alternating black and white 1-bit patterns, when captures on the photodector you a see 235us square wave, in your case  you may have set pattern exposure as 500us, so 500us square wave is seen, on the o-scope you can zoom in at the rising edge of the square wave, you should see a damping signal which settle down within 5us. 

Regards,
Sanjeev