Wiring an up/down counter to magnetic sensor in Quick windlass

[BelleSerene]

I've been thinking of making a cheap chain counter for our Quick 1000 anchor windlass using something like these, connected to a magnetic sensor on the windlass:
http://www.ebay.co.uk/itm/DC-12V-4-...all-Proximity-Switch-Sensor-NPN-/221714536440
http://www.ebay.co.uk/itm/DC-12V-24...wn-Hall-Proximity-Switch-Sensor-/321697613873
http://www.ebay.co.uk/itm/0-56inch-...l-Counter-Totalizer-Meter-DC-12V/201466548390

Such gadgets will count up or down depending on a signal I can control from whether the windlass' down or up button is being pressed. They also retain their count while the power is off so I only need to have the counter drawing current while the windlass is being powered in either direction or when I want to view the count. If I trigger these devices on windlass revs, they'll give a count of X per metre of chain, but I am happy to handle the multiplication factor for the price. Then the counter will always hold the amount of chain that's currently out - and of course there's a reset line in case an error develops.

I just found that there already IS a magnetic sensor that's been sitting in the windlass all these years, unused, with two wires sticking out. I see that Quick's own (highly expensive) counter connects this between ground and a sensor input. Yet the magnetic hall effect sensors on these counters have three wires for +(12V), 0V and signal: so presumably there's an amplifying circuit in there - which would imply they won't trigger on the signal sent from the magnetic sensor in the windlass.

I would like to adapt the counter to take a signal from the sensor already built into the windlass. Could someone suggest how please? I'm able to take reasonably involved instruction - once upon a time I studied electronics.

Yes, my rubber markers in the anchor chain work fine in practice; it's really just a technical challenge, to combine a cockpit remote control with a cockpit-based count of chain length deployed.

 

[lpdsn]

I've been thinking of doing something like this. I'm partway through fitting a new quick windlass at the moment and a chain counter would be handy, but their accessories are pretty expensive.

I guess the first thing is to see exactly what comes out of the inbuilt sensor. I've got no further than briefly loooking at USB oscilloscopes. If it's not enough to drive a counter like yours then a basic amplifier circuit would be needed.

 

[William_H]

Magnetic sensors come in 2 types either a magnetic reed switch which simply makes a contact when magnetism is present or a Hall effect sensor. The reed switch is less sensitive but is characterised by low resistance as checked with a Digital multimeter on ohms range followed by an open circuit shortly after as the winch is rotated. So a resistor feeding a small current from 12v supply will give a 12v pulse for each step. That should be OK for the counter or may need a voltage divider if voltage is too high.
The hall effect sensor is a bit more complicated in that it is fed with 5v and the output varies from 2.5v no magnetic field to 2v or 3v depending on field direction. This change of voltage must be detected by a simple voltage comparator to give a pulse. The comparator (amplifier) might be in the sensor or in the counter. So a volt meter on the output might indicate what is happening as you slowly rotate winch. I hope this might give you some ideas. olewill

 

[vas]

hi,

If you have a Garmin GMI10 or GMI20s installed in your boat, I could send you my arduino code and instructions to built a blackbox that sends NMEA0183 commands to the GMI showing screen chain out, windlass speed, resetting rode out etc.
All part of the AutoAnchor AA601 black box simulation I've been working on for some time now and only last night seems like I've finished it... To be tested on board in an hour or so.
Yes, it will cost you a bit more than 12quid, but under 50 (if you're careful) as main cost would be the Due or teensy, a TTL->serial converter and a few resistors.
More info on this thread

ah, no matter what you do, MAKE SURE you don't route the reed switch signal back to the helm NEXT to the thick power cable going to the windlass. It picks up EMI like hell, took me some time to figure it out and sort out some hardware debouncing...

cheers

V.

 

[JohnGC]

It sounds as though the counters are designed to work with a Hall Effect sensor (3 terminals) but you have a reed switch in the windlass.

Before spending any money, I would connect a multi-meter to the 2 wires, slowly rotate the gypsy and check that the sensor operates. I'd expect the resistance to alternate between open circuit (>1M ohms) and closed circuit (<5 ohms).

If that's OK; You can connect the sensor to the counters 0V and input terminals. Then add a "pull-up" resistor between the input terminal and the counter's +12V terminal. Try something around 4k7.

Almost all mechanical switches suffer from contact bounce when they close. Reed switches are better than many but not perfect. I would expect the counter input to have a low pass filter which will remove or reduce the bounce before the signal reaches the active part of the counter. If it doesn't, you may have to add an external RC filter to prevent false counting. A "bounce" could be counted as several pulses rather than one; the number is likely to vary. (I once made an electronic dice, the random element of which was contact bounce, it was very random.)

 

[BelleSerene]

Thank you; this is very helpful. Certainly clear now that it's a magnetic reed switch and not as I had assumed a Hall effect sensor.

I just took the windlass gypsy apart, with the minor benefit of de-gunking and re-greasing it. The face of the sensor is under the gypsy, and nestled in an 8mm diameter recess in the underside of the gypsy is a small magnet. Or rather, was: it had rusted so badly that I wouldn't expect it to trigger the reed switch. I chiselled at it with a screwdriver and much of it has fallen away. I've bought an 8mm diameter neodymium magnet on eBay (well, actually ten for £2.29); I plan to chisel the rest of the old magnet out of the recess, drill out the recess if necessary and mount a new magnet in there with epoxy. I'll test it on the sensor with an ohm meter first, of course. I'll use a twisted-pair cable to bring the signal aft and if the counter multiple-counts signals I'll low-pass filter it.

Many thanks to all.

 

[vas]

I'll use a twisted-pair cable to bring the signal aft and if the counter multiple-counts signals I'll low-pass filter it.

get a shielded cable for that job!
I used a two pair green bus cable, but near the windlass motor I swapped it for a SFTP cable with 4pairs, shielded individually AND shielded all together. Don't forget to ground the shields...