Issues with 1-wire Anemometer

[mrgadget]

Hi all,

I have a 1-Wire Wind Anemometer. I was under the impression that everything was working fine with it (seemed to read wind speed and direction - was using owwnogui under linux), but recently I have started to develop my own application to read my 1-wire network and it's highlighted an issue it has.

All the reed switches work fine - the main 8 points of the compass work without issue N, NE,E,SE etc). The only points that fail are three of the intermediate points at SSW, SSE & ESE (points 8, 10 and 12 on the following image). These are the points where two reed switches are supposed to be activated. The voltages I'm seeing are 0, 0.333 & 2.5. The three points that don't work are when I'm meant to see 2.5V on two channels at the same time - but instead I see nothing.

http://container.watje.org/index.php?title=Image:WSV3GDE.gif

When at these points, the reed switches on both sides are meant to be on, causing two A2D channels to read a voltage - instead there is no voltage on any channel.

Using point 8 as an example. Moving from point 7 to 9, through 8, should result in voltage only on CHANNEL_D while at point 7 (This is OK), then at 8 you should see voltage on CHANNEL_D & CHANNEL_C (I see 0v on all 4 channels), then at point 9 there should only be voltage on CHANNEL_C (This is also ok).

I wondered if this was due to voltage drop - but both with the instrument plugged directly into my 1-wire serial adapter, and also when supplemented with the optional 5V from a power injector, the problem remains.

I'm at a loss to explain this either mechanically or electrically - please assist!

Thanks in advace,

Ben.

 

[Lagerhead]

Can you disconnect the vane interface and verify that the two reed switches actually are closing at the intermediate points? On the face of it, it appears that there is no reed switch actuation at the in-between points.

 

[mrgadget]

Can you disconnect the vane interface and verify that the two reed switches actually are closing at the intermediate points? On the face of it, it appears that there is no reed switch actuation at the in-between points.

Hi, thanks for the response. This could indeed be the issue - how do you suggest I test this? I've tried moving the magnet bar up and down the shaft and it doesn't make any difference. I've disconnected it from the network and tried to use a multimeter to detect continuity but can't get anything - maybe due to the protective coating over everything? I know the reed switches are functioning, just not sure if they're on at this point.

 

[Lagerhead]

. . .how do you suggest I test this?

Well I thought there was a way to physically disconnect the vane from the electronics -- a cable or something (I don't know, I don't have one of these).

Looking again at the diagram, the relays on the three points you are having trouble with should pull the associated pins to ground, isn't that what you are seeing? Or, it may be significant that the three points are associated with only the ground-referenced relays.

 

[mrgadget]

Well I thought there was a way to physically disconnect the vane from the electronics

There's not, no. All the electronics, along with the reed switches are on one PCB.

It looks like I might have it solved - Eric has kindly pointed me in the right direction by indicating that the voltages I'm seeing aren't correct, and I've probably not correctly initialised the A2D converter (DS2450) correctly.

Apparently if you just convert and read the voltages - it's maxes out at 2.5V - causing issues as it's supposed to read 5V on some points. The key is apparently to set the range and resolution of the A2D when initialising it.

I'll report back here once I've found the solution so the answer is here, should anyone else have the same issue.

Thanks,
Ben.

 

[mrgadget]

As promised - reporting back. Thanks again to Eric for the help - and Lagerhead for the ideas.

The issue wasn't with the device at all - it was with how I was initialising it (as described in my previous post).

Cheers,
Ben.