how does wind angle instrument work?

skyflyer

skyflyer

Well-Known Member
Joined
26 Jan 2011
Messages
1,433
Location
Worcester, UK
Visit site
probably depends on the make, but this much I know about my Stowe unit and I think Raymarine ST50 work in a similar way

12v gets sent up to the masthead transducer where there are three potentiometers offset by 120 degrees. The voltage down each of the three lines (green, white, yellow) will vary between 0v and 12v (less line losses) depending on the angle of the wind vane.

The relationship of the three voltages allows you to mathematically calculate the wind angle -but i don't think that is how the instrument does it because....

when you measure the voltage down the green, yellow or white line, by disconnecting it from the instrument you get a 'sensible' reading, as above, between 12v and 0v.

But if you read the voltage when the wire is still connected to the instrument then the three voltages are very very close to being the same, e.g. 4.6, 4.9, 4.3v.

Can any electronic genius explain what is happening inside the instrument? I don't need to know, I'm just curious!
 
skyflyer said:
probably depends on the make, but this much I know about my Stowe unit and I think Raymarine ST50 work in a similar way

12v gets sent up to the masthead transducer where there are three potentiometers offset by 120 degrees. The voltage down each of the three lines (green, white, yellow) will vary between 0v and 12v (less line losses) depending on the angle of the wind vane.

The relationship of the three voltages allows you to mathematically calculate the wind angle -but i don't think that is how the instrument does it because....

when you measure the voltage down the green, yellow or white line, by disconnecting it from the instrument you get a 'sensible' reading, as above, between 12v and 0v.

But if you read the voltage when the wire is still connected to the instrument then the three voltages are very very close to being the same, e.g. 4.6, 4.9, 4.3v.

Can any electronic genius explain what is happening inside the instrument? I don't need to know, I'm just curious!
Click to expand...
By coincidence my B&G wind direction started playing up. Downloaded their instructions about how it works, same sytem as Stowe. When I phoned Tinsley the lady said that they could modify the Stowe to work with B&G, the B&G 496 MHU is no longer available although they can do some repairs. I put the multimeter on the terminals at the bottom of the mast and got readings off the three sectors. the bad sector however only read a steady 0.01v, the other two jittered between 1 and 6v which is correct apparently. so I suspect algorithms doing joined up thinking with a smoothing somethng to make it readable.
S
 
Given that these were originally analogue devices I wonder if they apply those voltages to coils in the display - creating a magnetic field that aligns a magnet mounted on the pointer.

Of course these days they do it all with electronics.
 
skyflyer said:
probably depends on the make, but this much I know about my Stowe unit and I think Raymarine ST50 work in a similar way

12v gets sent up to the masthead transducer where there are three potentiometers offset by 120 degrees. The voltage down each of the three lines (green, white, yellow) will vary between 0v and 12v (less line losses) depending on the angle of the wind vane.

The relationship of the three voltages allows you to mathematically calculate the wind angle -but i don't think that is how the instrument does it because....

when you measure the voltage down the green, yellow or white line, by disconnecting it from the instrument you get a 'sensible' reading, as above, between 12v and 0v.

But if you read the voltage when the wire is still connected to the instrument then the three voltages are very very close to being the same, e.g. 4.6, 4.9, 4.3v.

Can any electronic genius explain what is happening inside the instrument? I don't need to know, I'm just curious!
Click to expand...

That sounds like the signal down the line might be a current rather than a voltage.
Some industrial sensors are done like this, the voltage varies with line losses, the current does not.
Or it could be pulse width modulated, there are many was of doing the job!
The last B'n'G MHU I looked at was voltages though, there was a table in the manual, basically sinewaves offset by a voltage and 120 degrees apart.
 
skyflyer said:
probably depends on the make, but this much I know about my Stowe unit and I think Raymarine ST50 work in a similar way

12v gets sent up to the masthead transducer where there are three potentiometers offset by 120 degrees. The voltage down each of the three lines (green, white, yellow) will vary between 0v and 12v (less line losses) depending on the angle of the wind vane.

The relationship of the three voltages allows you to mathematically calculate the wind angle -but i don't think that is how the instrument does it because....

when you measure the voltage down the green, yellow or white line, by disconnecting it from the instrument you get a 'sensible' reading, as above, between 12v and 0v.

But if you read the voltage when the wire is still connected to the instrument then the three voltages are very very close to being the same, e.g. 4.6, 4.9, 4.3v.

Can any electronic genius explain what is happening inside the instrument? I don't need to know, I'm just curious!
Click to expand...

Or, in other words, how do you measure the position of a pot?
One problem you get with long cable runs is noise, an easy way to reduce this is to up the current. A low impedance input is one way to do this.
Your next question might then be, so why is it hovering round 4.5V and not 0?
One style of input buffer uses an op-amp, a very, very high gain device. These are configured as inverting amplifiers with the output fed back to the input. The effect of this is to generate a virtual ground at the input pin. This ground need not necessarily be 0V (as supplied) but is more commonly half way between the supply rails.
Next question, why, if it's half way is it 4.5V and not 6V?
My first guess would be that the gubbins has an internal 9V regulated supply.

All a bit glib, have a search and read about op-amps.
 
have a search and read about op-amps.
Click to expand...

I'm much better informed but none the wiser - well beyond my understanding, I'm afraid.

The mathematical calculation of wind angle does not require precise voltages because it uses the ratio of one line voltage to another. It is thus quite simple to use (as I have) an Arduino to read those voltages and calculate the wind angle and then output it in NMEA readable form so other devices onboard can have a wind input.

But the instrument is not NMEA it is analogue and designed as a 'stand-alone' so no doubt uses analog methods and I notice that op-amps are very prevalent in these!

What I don't understand is how/why a device that requires differential voltages to work out the angle, then changes those same voltages by feeding back into the circuit, and yet still works.

But i suspect I never will!
 
skyflyer said:
I'm much better informed but none the wiser - well beyond my understanding, I'm afraid.

The mathematical calculation of wind angle does not require precise voltages because it uses the ratio of one line voltage to another. It is thus quite simple to use (as I have) an Arduino to read those voltages and calculate the wind angle and then output it in NMEA readable form so other devices onboard can have a wind input.

But the instrument is not NMEA it is analogue and designed as a 'stand-alone' so no doubt uses analog methods and I notice that op-amps are very prevalent in these!

What I don't understand is how/why a device that requires differential voltages to work out the angle, then changes those same voltages by feeding back into the circuit, and yet still works.

But i suspect I never will!
Click to expand...

Nigel M has posted about this in the past and understands how it works.

I don't .... which is not very helpful I'm afraid.

PM Nigel to alert him about your thread.

Richard
 
I have always assumed sender and receiver are connected by synchronous motors, such that movement at the transmitter is reflected by a commensurate movement of the repeater. Both motors having three sets of windings (or a single winding tapped at three points) arranged at 120° to each other and current induced in the windings will differ in magnitude and direction depending on the position of the rotor shaft. But that's all supposition from ancient history and perhaps there are more up-to-date technologies to do the same job at much cheaper cost - synchronous (or de-synchronous) motors were always pricey back when I had anything to do with them.
 
I appreciate folks are trying to help but you're all telling me what happens at the masthead transducer whereas what I'm trying to work out is how the instrument itself resolves that info

I know what is happening at the transducer. I know what voltage is supplied to it and I know what voltages it sends to the instrument for any given wind angle

What I can't figure is why those voltages become hugely attenuated once the wires are connected to the instrument unless it's all to do with op-amps as suggested above

But by the look of it I'm not going to understand the explanation anywY. Thanks all for trying, anyway!
 
skyflyer said:
I appreciate folks are trying to help but you're all telling me what happens at the masthead transducer whereas what I'm trying to work out is how the instrument itself resolves that info

I know what is happening at the transducer. I know what voltage is supplied to it and I know what voltages it sends to the instrument for any given wind angle

What I can't figure is why those voltages become hugely attenuated once the wires are connected to the instrument unless it's all to do with op-amps as suggested above

But by the look of it I'm not going to understand the explanation anywY. Thanks all for trying, anyway!
Click to expand...

Ok... Maybe a bit more bamboozlement or maybe an explanation...
When you configure an op-amp as I suspect is done on your piece of kit, what happens at the input is what is called a virtual earth. In order to maintain a fixed voltage at that point the op-amp must fight the signal source, it's output will oppose the input signal. So the following circuit will "see" a buffered, possibly amplified version of the transducer output. The purpose of this is to, firstly give the transducer a fixed impedance load while generating a voltage source signal for what follows.
After the input buffering stage, who knows, I suspect some A to D conversion (one of the reasons for an input buffer), followed by some signal normalisation, source selection, and ending up with a look up table for one sextant of the possible range of inputs.
 
nigelmercier said:
Or bad connections, perhaps in the power feeder.
Click to expand...

Don't thinks o Nigel as I later tried to simulate the feed using some pots to produce three specific line voltages as per the table produced by Stowe. Everything spot on, i.e. if i set the voltages then connect the instrument it reads the correct angle. But once the wires are connected the voltages are nowhere near what they were originally set at!

This is all of interest as I am splitting (paralleling) the voltage lines and sending them to an Arduino that calculates the wind angle (and speed) using the aforementioned ratios and then outputs a proper NMEA0183 sentence so the a/p can sail to a constant wind angle.

It seems to work, so the attenuated voltages must presumably still be in the correct ratios to one another.

But Dougal's reply goes a long way to explaining why this might be.
 
skyflyer said:
Don't thinks o Nigel as I later tried to simulate the feed using some pots to produce three specific line voltages as per the table produced by Stowe. Everything spot on, i.e. if i set the voltages then connect the instrument it reads the correct angle. But once the wires are connected the voltages are nowhere near what they were originally set at!

This is all of interest as I am splitting (paralleling) the voltage lines and sending them to an Arduino that calculates the wind angle (and speed) using the aforementioned ratios and then outputs a proper NMEA0183 sentence so the a/p can sail to a constant wind angle.

It seems to work, so the attenuated voltages must presumably still be in the correct ratios to one another.

But Dougal's reply goes a long way to explaining why this might be.
Click to expand...

Eek, in that case your poor Arduino will not be getting an optimal signal. You might want to think about buffering the signal, then splitting them, use resistors to feed the direction indicator.
 
Dougal Tolan said:
Eek, in that case your poor Arduino will not be getting an optimal signal. You might want to think about buffering the signal, then splitting them, use resistors to feed the direction indicator.
Click to expand...

Yep, you've found my problem in one. But Arduino limited to 3.3v max input so even these low voltages (after going through a voltage divider) are well resolved as Arduino can resolve 1024 discreet values between 0 and 3,3v as I understand it?

But i'd love a better idea of how to do what you suggest - buffer and output. There are limited analog ports on the arduino so the only way i can think of replicating the output is with pulse width modulation, but when i tried that there was an audible whistling sound (pulses?) and i gave up before i fried something.

The weird thing is that it does seem to work ok at the moment, but I'm not happy that its is very accurate.

Of course for the a/pilot to steer to a constant wind angle it doesn't really need to know an accurate value, it just needs to accurately resolve any changes!
 
skyflyer said:
Yep, you've found my problem in one. But Arduino limited to 3.3v max input so even these low voltages (after going through a voltage divider) are well resolved as Arduino can resolve 1024 discreet values between 0 and 3,3v as I understand it?
Click to expand...
Yes, believe that's right. The arduino ADC is 10 bit, so 0 - 1023. And to make matters worse the internal reference can drift a bit depending on what voltage is powering the board. For about a tenner ebay has 4 channel 16bit ADC's , very accurate and communicate over I2C.
http://www.ti.com/product/ADS1115
https://www.adafruit.com/product/1085
 
You must log in or register to reply here.

Share this page

Top