Frequency Sampling Problem

[ccdubs]

Hi all,

I have been measuring wind speed at a location with an annenometer which has a rotating disc (like a cog) and a photodiode. The diode sends pulses to the data logger as the disc rotates, which are then converted to a wind speed value.

It has turned out that the wind speed value is incorrect and is infact higher, which is highly unusual as this means there are more pulses (an error in code has been ruled out). As it turns out, if the supply voltage is increased to 12Vdc from 5Vdc the windspeed values are correct.

We have been running 2 meters now at the same site both at 5V and the original is still incorrect and the new one is correct. I am baffled as to how the data logger can be reading more pulses. It should be mentioned that the wind speed error is for all intents and purposes constant.

Any ideas? Also there is a large AM broadcasting tower 120m away (675kHz), could this have any effect. It still wouldn't explain how one meter works and the other doesn't. The power supply is directly from a battery and nothing is earthed, if this is the case is it possible for there to be induced signal from the radio mast.

All help appreciated.

 

[cbarn24050]

Hi, it could well be interferance from the mast.Try some shielding,filtering and earthing, which of course you should have done in the first place.

 

[rodar]

Check the digital input threshold value for the
data logger. If the photodiode on or off output
voltage is close to the transition threshold of
the input it could cause unwanted readings.

 

It's probably 18-volt CMOS logic, whose noise margins increase with supply voltage. THerefore accurate measurement at 12V.

Just check the interface signals and logic family and do it right.

 

[ccdubs]

Thanks for the suggestions, what has been said makes sense. In my defence, we used a contractor who specialises in environmental measurement. Before this post, I asked about sheilded cable and they confirmed, I asked about grounding and they confirmed. Yesterday I went up and had a look at it myself and yes it looks like the shielded cable goes to the earth terminal on the logger but I could not see any external connection from the logger to earth. I will check this with them today.