About time-shift circuit

Thread Starter

hwoarang

Joined Dec 28, 2015
18
Hello, guys.

I am a physicist using the laser system and am stuck in circuit problems during the experiment, so I beg your advices.
As the picture is shown below, the laser pulses with 10 khz are separated into two photodiodes with a given RC decay time. Odd laser pulses go to one photodiode and even pulses go to the other photodiode. So, the temporal delay of each pulse between two photodiodes is 0.1 milliseconds.
The problem is that I have to make this time delay zero. How can I shift the time delay of one photodiode ? Is it possible to combine this with a kind of timer circuit ?
Any help will be appreciated. Thank you in advance.

upload_2016-2-26_21-11-51.png
 

crutschow

Joined Mar 14, 2008
34,282
I don't understand your problem. :rolleyes:
You introduce a deliberate 0.1ms shift between pulses and then say you don't want that shift.
If you don't have the 0.1ms shift and make the time delay zero then both pulses will occur at the same time. Is that what you want?
 

#12

Joined Nov 30, 2010
18,224
The idea that light goes into, "some electronics" and emerges as light escapes me. Even if that is possible, the contents of, "some electronics" is exactly what needs to be revealed in order to modify them.

Edit: Of course, I am terribly ignorant about changing the speed of light. Is there any such thing as a 100 microsecond light delay device? 18.6 miles in a vacuum? Certain transparent substances with a different internal speed of light?
 
Last edited:
As #12 says, we need to know what's in "some electronics" to know how the output pulses are generated in order to figure out how to add the delay. It's far from obvious how electronics would turn a laser pulse into even and odd RC decayed laser pulses.

Assuming the electronics charge a capacitor during the incoming laser pulses, and then alternately discharge the cap into a pair of LEDs or something, you could probably duplicate the circuit and then trigger the discharge of one of them with a one-shot timer circuit. The one-shot would be configured to 0.1ms and triggered by the charging of the input capacitor. The output of the one shot could fire a MOSFET that prevented discharge of the cap until it is asserted.
 

#12

Joined Nov 30, 2010
18,224
Here's an idea: Wait until after the photodiode produces an electrical pulse and delay that pulse.
 

Thread Starter

hwoarang

Joined Dec 28, 2015
18
I don't understand your problem. :rolleyes:
You introduce a deliberate 0.1ms shift between pulses and then say you don't want that shift.
If you don't have the 0.1ms shift and make the time delay zero then both pulses will occur at the same time. Is that what you want?

Sorry, I made you confused. Here, 'some electronics' is not important point. It can be whatever, optics or mechanics... (and this part is already done). For a detail explanation, the odd pulses contain meaningful signals (from samples) + meaningless background signals and the even pulses contain only meaningless background signals. This is the reason why I want to separate the laser beams into odd and even. And eventually, I'd like to get only meaningful signals which can be obtained by subtracting even pulses from odd pulses. For this, I think that the pulse timing between odd and even should be same. Do you have a interesting idea ?
Have a good day!
 

ronv

Joined Nov 12, 2008
3,770
Here is my read on his plan.
He has two channels. On contains information plus some background interference. The second only contains the interference - no data.
I think what he wants to do is subtract the interference from the combined signal leaving only "good stuff."
What would be nice to know:
What is different between the 2 signals? Amplitude? width, or ??? In other words how do you know the difference between the two.
Why are they offset in time? Why not turn them both on at the same time. Mirror or something?
 

GopherT

Joined Nov 23, 2012
8,009
Sorry, I made you confused. Here, 'some electronics' is not important point. It can be whatever, optics or mechanics... (and this part is already done). For a detail explanation, the odd pulses contain meaningful signals (from samples) + meaningless background signals and the even pulses contain only meaningless background signals. This is the reason why I want to separate the laser beams into odd and even. And eventually, I'd like to get only meaningful signals which can be obtained by subtracting even pulses from odd pulses. For this, I think that the pulse timing between odd and even should be same. Do you have a interesting idea ?
Have a good day!

Is this a spectroscopic application or some other?

If a spectroscopic application, why split pulses even and odd to get your background and signal data. Why not split each pulse with a 50% mirror and comment data and signal of each pulse?
 

WBahn

Joined Mar 31, 2012
29,978
Sorry, I made you confused. Here, 'some electronics' is not important point. It can be whatever, optics or mechanics... (and this part is already done). For a detail explanation, the odd pulses contain meaningful signals (from samples) + meaningless background signals and the even pulses contain only meaningless background signals. This is the reason why I want to separate the laser beams into odd and even. And eventually, I'd like to get only meaningful signals which can be obtained by subtracting even pulses from odd pulses. For this, I think that the pulse timing between odd and even should be same. Do you have a interesting idea ?
Have a good day!
You might consider using a lock-in amplifier. Capture the signal from the first pulse train and store the signal (on a capacitor or digitize it) and then do the same with the second signal and subtract the second from the first. I've designed imagers that stared into a pump-probe laser system with a signal-to-background ratio of 10^-7 and each pixel had an analog lock-in amplier in it, so it definitely can be done.

What is the signal to background ratio that you are dealing with?
 

Thread Starter

hwoarang

Joined Dec 28, 2015
18
You might consider using a lock-in amplifier. Capture the signal from the first pulse train and store the signal (on a capacitor or digitize it) and then do the same with the second signal and subtract the second from the first. I've designed imagers that stared into a pump-probe laser system with a signal-to-background ratio of 10^-7 and each pixel had an analog lock-in amplier in it, so it definitely can be done.

What is the signal to background ratio that you are dealing with?
Thank all of you for all above kind replies. And for your comment, WBahn, you are exactly right. I am using the pump-probe technique using femtosecond laser system and I am also using the lock-in amplifier. In a usual pump-probe technique, there needs two photodiode responses. One is the response which contains (real) signals + (dummy) background and the other one is the response which contains only background as I mentioned before. But these two responses (in this case, odd pulses and even pulses) should be in the same temporal position. Then I can remove dummy background response efficiently to get real signals. In principle, the lock-in amplifier can extract the real signals from just only one response, that is, odd pulses without subtracting even pulses. However, in terms of a signal to noise ratio (S/N), to apply lock-in amplifier after removing the background response is the best way, really (S/N is increased by 10 times). So that's reason why I need to subtract even pulses from odd pulses.
In the picture below, I put a little bit more details. The sample is excited by the periodic electric field. Only odd pulses contain the information about how the sample is modified by the electric field. This is what I want to extract. Odd and even pulses should be in the same time to subtract each other, but now 0.1 millisecond time difference between odd and even is the main problem which I want to solve now.

And for WBahn's question, I am not that sure about S/N in current experiment as this is a little bit different from typical pump-probe experiment measuring Kerr/Faraday signals or reflectivity/transmissivity. Just my guess is S/N : 1 ~ 5, which is not easy without 'odd pulses - even pulses'.
If you don't understand this clear, it is due to my inability, please don't hesitate to ask me further.

I am so sorry for my poor explanation making you distracted.
At the same time, I thank you very much for trying to help.

upload_2016-2-27_13-47-59.png
 

Attachments

Last edited:

dannyf

Joined Sep 13, 2015
2,197
If you want to have zero delay, why don't you just use one photo sensor so they are always perfectly aligned.

Having said that, it is easy to do what you want to achieve. The simplest is to run the pulse through a RC network.
 

Thread Starter

hwoarang

Joined Dec 28, 2015
18
Sorry, I still don't understand what you mean by "same".
Do you mean the two pulses should be coincident?
Yes, odd pulses and even pulses should be in the same time by shifting 0.1 ms. Then I can effectly eliminate (dummy) background response in odd pulses. And I uploaded some further explanations and picture based on WBahn's comments.
Thank you, crutschow.
 

Thread Starter

hwoarang

Joined Dec 28, 2015
18
Is this a spectroscopic application or some other?

If a spectroscopic application, why split pulses even and odd to get your background and signal data. Why not split each pulse with a 50% mirror and comment data and signal of each pulse?
Thanks for the comment.
If I use the laser beam splitter like the picture below, two split pulse lines contain same amount of real signals and background as well. Then, after subtracting these two (odd and even), always null signal. So, that's the reason why I want to separate odd and even pulses, which contain real signal + background and only background, respectively.

upload_2016-2-27_14-27-22.png
 

Attachments

Thread Starter

hwoarang

Joined Dec 28, 2015
18
If you want to have zero delay, why don't you just use one photo sensor so they are always perfectly aligned.

Having said that, it is easy to do what you want to achieve. The simplest is to run the pulse through a RC network.
Thank you for the comment.
I am not sure if I catched your point. I hope that your comment is like the below picture.
A part of laser pulse (maybe 50%) goes through the sample and PD A and the other part of laser pulse goes to PD B directly without experiencing sample acting as a 'reference' response. Then, two photodiodes have same timing always and A-B has only real signals which I want to get. But the problem is here in that PD B does not go through the sample. In this case, BGs (background response) between PD A and PD B can be different, then the subtraction will not be complete. For example, when the sample is rotate by some angles or tilted, BG in only PD A can vary with maintaining BG in PD B. Then, real signals from A-B cannot be obtained efficiently because of remaining BG response in there. So I want to make both photodiodes go through the sample.
I am still thinking that the temporal shift between PD A and PD B is not bad idea as mentioned in the first thread, if feasible. Do you have some comments for this ?
Thank you, dannyf.

upload_2016-2-27_14-48-17.png
 

Attachments

GopherT

Joined Nov 23, 2012
8,009
Thanks for the comment.
If I use the laser beam splitter like the picture below, two split pulse lines contain same amount of real signals and background as well. Then, after subtracting these two (odd and even), always null signal. So, that's the reason why I want to separate odd and even pulses, which contain real signal + background and only background, respectively.

View attachment 101423
Please explain A) where is your sample in this drawing and (B) what is the important part of your real and reference signals? (Peak, decay rate or other?)
 

GopherT

Joined Nov 23, 2012
8,009
Thank you for the comment.
I am not sure if I catched your point. I hope that your comment is like the below picture.
A part of laser pulse (maybe 50%) goes through the sample and PD A and the other part of laser pulse goes to PD B directly without experiencing sample acting as a 'reference' response. Then, two photodiodes have same timing always and A-B has only real signals which I want to get. But the problem is here in that PD B does not go through the sample. In this case, BGs (background response) between PD A and PD B can be different, then the subtraction will not be complete. For example, when the sample is rotate by some angles or tilted, BG in only PD A can vary with maintaining BG in PD B. Then, real signals from A-B cannot be obtained efficiently because of remaining BG response in there. So I want to make both photodiodes go through the sample.
I am still thinking that the temporal shift between PD A and PD B is not bad idea as mentioned in the first thread, if feasible. Do you have some comments for this ?
Thank you, dannyf.

View attachment 101426
If you have data stored in software, the reference and signal are all collected by PD-A, why do you want to add additional noise and uncertainty by introducing a second Photodetector that has to be calibrated vs first detector? Also, the path and cell that contain the sample are not the same. YOur single-path setup is perfect. Now, you just need the software to subtract a reference signal mm editable before it.

If your sample is causing a time delay (fluorescence or phosphorescence), then you can still use a single beam but you need to collect data with a precision time signal as a second stream of data - using that second stream of data to trigger your laser pulses.
 

Thread Starter

hwoarang

Joined Dec 28, 2015
18
Please explain A) where is your sample in this drawing and (B) what is the important part of your real and reference signals? (Peak, decay rate or other?)
Sorry, I think that the below picture is more convenient to see. This is the case when the beam splitter is used.

upload_2016-2-27_15-15-28.png
About the important part of the signals, if I go back to the previous experimental scheme (odd and even pulses scheme with 0.1 ms temporal difference), the timing between odd and even is most important thing to achieve. But it is not that necessary to be perfect. The rising, decay rate and peak in photodiodes are optimized already.
Any reply will be appreciated.
Thank you, GopherT.
 

Attachments

dannyf

Joined Sep 13, 2015
2,197
That's a great description of your setup.

I think device differences and tiny delays you are trying to measure could be a challenge.

Two possible approaches.

First, have one beam and one sensor. Measure the phase difference between the signal driving the laser and the sensor signal.

Do the measurement twice, one without sample and one without sample. Delta the two.

You will need to control laser brightness. If you don't want to do that, use two beams plus two sensors and Delta the signal from the two sensors. And then swap the light path.

Alternatively, you can use the signal driving the laser to trigger a counter, and use the signal from the sensor to stop the counter.

A two laser two sensor version of this approach is also possible.

You probably need a high speed counter for this.
 

GopherT

Joined Nov 23, 2012
8,009
How are you collecting and storing your data from the detector(s)? An off-line manipulation of your data is the normal process. Identify which is a background signal and which is a "sample" signal. Subtract and move on.
 
Top