How to overcome slow response time of photo-diode?

[Dushyant]

Hi,

The problem that I am encountering is related to slow response time of photo diode that makes the scan time quite large.

What I am trying to accomplish is:

(1) I am scanning the sample (in Y and Z direction) using two stepper motors  ; scanning dimension 3 mm X  3mm. The velocity of motor is 2 mm/sec and it moves in step size of 5-10 microns. The motors that I am using was bought from Newport.com and their model is UTM25PP.1.

(2) For each sample position, I am reading ampere value using photo-diode. The amplifier uses RS-232. I can read either watt or ampere and I am reading ampere.

The problem that I am encountering:

If in "Basic Serial Write and Read.vi", I change the parameter "delay before read (ms)" to be less than 1000 msec, the photodiode fails to update for successive sample positions. It only works if "delay before read (ms)" is 1000 msec or more. But, this makes my scanning extremely slow which is undesirable (If I follow 1000 msec, the time taken would be more than 300 X 300 sec = 25 hours).

The main vi is “Photo_Diode_N_Motors_Integrated” and other are subvis.

Please let me know if you have any suggestion.Thanks,

DK

Message Edited by Dushyant on 09-12-2005 06:19 PM

[LV_Pro]

Hi,
    If you can save your program using Save with options, Development distibution. There are some vi's missing. I'm not sure what the problem with photo-diode is, but an update of 1000mS seems very long. How is the photodiode connected, what is the circuitry between the diode and the rest of the system (you speak of RS232 interface)?

Thanks,
P.M.

Message Edited by LV_Pro on 09-12-2005 10:32 PM

[johnsold]

I think the problem is not the photodiode itself but the signal conditioning and RS-232 communication which follow. Even the slowest large area photodiodes have time constants on the order of a few milliseconds not a second. If the signal conditioning system does not have a mode for faster reporting of data, I doubt you will be able to do much. Can you monitor the photodiode output or an analog output from the photodiode preamp directly via an A/D converter? That could be much faster.

Lynn

[falkpl]

I agree it is probably the rs232 which is a very slow port and is might even be older than you.  I don't think the photodiode responses is slow, but you can try to use a simple photodiode setup with an analog-to-digital card which can be read at rates much faster than 1KHZ.  This will allow for an enormous increase in readout time.  The communication bandwidth should be your problem.  Its like using the latest dual-Pentium processor to connect to the we with dial-up,  the computer is not slow but is sure will act slow.  The cons of this new approach is that you will need to get an a-to-d card and a photodiode with analog output (not too expensive but still not free).  Serial is slow and easy to use.

Paul

[AnalogKid2DigitalMan]

What is the manufacturer and model # of the photodiode unit?

[Dushyant]

Hello guys,

Thanks a lot for your suggestions. I will try to impliment your suggestion.

I am sorry that some of subvis were missing. Now,I have attached all the subvis and I checked myself. So, if anybody have any suggestion, please let me know.

I am using labview 7.1.

Message Edited by Dushyant on 09-13-2005 12:37 PM

Message Edited by Dushyant on 09-13-2005 12:37 PM

[Dushyant]

Hi,

I bought photo-diode and controller form edmundoptics.com and here is the detail:

Silicon Photo Diode Detector System:

Details: • SILICON DETECTOR 100MM2 Biased: Normal	NT53-373	$98.50
PHOTODIODE AMPLIFIER	• NT57-601	$995.00

[AnalogKid2DigitalMan]

The photodiode is plenty fast (43nsec risetime when biased). Was hoping to find a manual for the amp, but none on edmunds site. I would suggest using a DAQ card with an analog input to monitor the +/-2V FS output of the amp instead of using it's serial port. Probably much faster. 3db Response of the amp at 20nA is 2KHz, it gets faster with more current (>20uA it is 40KHz).

Regardless of serial or DAQ approach, suggest that you bias the photodiode if needed. Be careful for stray EMI/RFI pickup, you may have to enclosed the photodiode in a copper or mu-metal enclosoure if you see any evidence of noise. An oscope is a good tool to use to verify signal integrity. Another issue you may encounter is stray light pickup (60/120Hz if US, sunlight will cause a DC offset too) that can even swamp out your signal of interest. Photodiode response is a broad 400-1100nm, use optical filters if needed.

[johnsold]

A quick look at the Edmund catalog confirms what I said above. The photodiode itself has a rise time of 43 ns. The amplifier has a bandwidth of 2 kHz to 40 kHz depending on current range. To use the full bandwidth, however, it appears that you need an external high speed measuring device such as an oscilloscope or A/D converter. The second paragraph of the Edmund catalog description for the amplifier essentially says that. It appears that the RS-232 connection is primarily for use as a power meter.

Lynn

[Dushyant]

I believe that the photo-diode amplifier that I am using only uses RS-232. I am not able to think of any way so that I can use DAQ card or anything.

Also, one of my friend suggested me that this may be because of using too many stacked sequence (which according to him not a very good practice). Do you guys think that it may be the reason?

Thanks,

DK

Message Edited by Dushyant on 09-13-2005 12:43 PM

Message Edited by Dushyant on 09-13-2005 12:43 PM