convert power to voltage in a AO

[Yanissab]

Hi everyone!

I have spent my last 2 days on a problem that seems to me quiet easy but i can't find any sollution..

I have a Power detector which reads a power from the nanowatt to the Watt, this power is sent to an FPGA board in order to control it (PID). When i use it as a GPIB everything is so slow, so i decided to use the AO to transmit a continuous value.

BUT the power detector converts the power from Watts to Volts as follow: 10nW correspond to 1V, 15nW correspond to 1,5V etc... we could think that there is only a 10 factor to apply to get the right value.. the problem is when we have a grater value than 2 volts, the voltage returns to zero and the power scale changes and it becomes 1V for 10uW, 1,5V for 15uW and when it exeeds 2V again it goes back to zero and autoscale the power read etc etc. So the problem is that i try to control the power from the voltage so basicly i only see 0 -> 2, 0 -> 2, 0 -> 2 for the nW, uW and mW.

I hope i made my problem clear, sorry about my english...! i really don't know how to solve this..

thank you for your time.

Yanis

[NIquist]

You must be missing something.  There has to be an output (probably digital) on your device that indicates the range the analog voltage output is referring to.  Once you can find it and read it, you can make the proper corrections in your software to get the properly scaled power value.

[AnalogKid2DigitalMan]

My Newport 1835-C optical power meter does the same thing with the analog out when it autoscales. There is no auxiliary output that indicates what range it is in.

How slow is slow? Is the meter set up for filtering/smoothing? What make and model is it?

-AK2DM

[Yanissab]

Thank you all for your answer! but i jut found the solution! the VI is huge and you can find it attached. it does work.

the maximum speed with GPIB was 0.700 sec per cycle, now with the fpga and the AO it reaches 0.020s per cycle.

the power meter i use is an HP8153A.

In the loop on the left, I use the GPIB only to give the range of the power value (which doesn't need to be fast) 

In the bottom loop, i re-scale the value according to the range

And on the top loop, i control the power.

B->N is a subvi which deals with the conversion Binary to analog, N-B is the opposit conversion. P stands for Power and V for Volts.

TP is the Transmitted Power in volts (between 0V to 2V)

PV is the process variable.

All this to control a tunable optical filter which drifts.

I think this solution could be more "elegant" but it works

[NIquist]

---

@AnalogKid2DigitalMan wrote:

My Newport 1835-C optical power meter does the same thing with the analog out when it autoscales. There is no auxiliary output that indicates what range it is in.

---

Really???  I guess the manufacturers feel that if your readings seem off by 3 orders of magnitude you would know it jumped ranges.  Seems kind of dangerous though.  "Oh yeah, that beam is safe, only 70 micro-watts... No wait, I mean milli-watts!"  Too late, you're blind. 

[AnalogKid2DigitalMan]

NIquist:

Probably a trade-off between resolution and dynamic range.

The meter span is from uW to 100's-1000's of Watts.

-AK2DM