Thanks Dave, to be honest it's hard to take it all in:) I did originally try a Maplin multicore. I learnt by trial and error that:
1. Firstly cable length was a factor,
2. Proximity of cables in the loom was also a factor - with a tiny extra gap solving the problem; hence interest in the dialectic cross you mentioned.
3. Not only was I getting circuits being made without switch buttons being pressed, I was creating an array of new, unintended circuits between all 12 wires involved.
The main learning point was that the sheath of the multicore was keeping all my cables very tightly compacted; just what I needed to avoid.
I've now made up a loose loom of 4 pairs of twisted pairs (16 wires) not sure if each leg going through both wires in the pair helps but it does work:)
I'm hoping to send this up the binnacle tomorrow afternoon, weather permitting, to see if it still works when surrounded by all that other wiring. Obviously need to test it with my OBM, GPS, VHF running.
Have been getting on well with other aspects of the AP install. I'll try to install the tiller extension - including modified AP ram end receptacle. Will post pics in due course.
Will also be measuring up for cutting into the engine well liner to accept the chestnut shelf I'm fabricating on which to anchor the AP unit.
Certainly learning a lot about sourcing all the relevant DIY bits cheaply. Eg was amazed to get a 200x200x3mm sheet of aluminum through the post in a couple of days for £6 including p&p. I'll be able to make two separate components out of that.
We have this joke at home now that "everything costs £7":) I'm certainly keeping the postman busy but it saves a lot of time and deisel.
—
MacGregor 26M 2009 - Sky's the Limit - Suzuki DF50
Thanks Rick, yes, the buttons connect directly to the microprocessor controller chip.
I'm actually just waiting for the tide so I can spend the afternoon on the boat just tinkering. Will try to test my "cat5 loose loom from buttons to binnacle" set up if I get time.
I caught my rudders on a mooring the other day and need to sort them out first; careless, I was trying to avoid a SUPer and didn't notice it.
—
MacGregor 26M 2009 - Sky's the Limit - Suzuki DF50
Still thinking of my relay idea, you can get these on eBay: http://www.ebay.co.uk/itm/192142396572. That's 5 x 12V relays with pre-wired holders. And just over £7 for the five!
Not sure what they're like, made in China, but there's quite a large selection of similar things. Should be easy to wire up.
Just a thought.
—
Rick Jones (Treasurer), former 26X & 19 owner, Isle of Wight
Thanks Rick, if my current plan fails I'll definitely give it a try, specially now I've opted for a 4 button approach.
Another issue I'm working on, as the unit will often be out of sight, is the motor burning out when the ram reaches the end of it's travel - there's nothing to tell the motor to stop.
These are some ideas:
1. Since the problem is mainly precipitated by undue lee/weather helm - ie full deflection fails to hold the course dialed in - better rig balancing should help.
2. Using a fuse that will blow when the load rises above normal but before any damage is done. A 12A fuse is recommended. I tried a 10A and this blew almost immediately the ram hit the end of it's travel thus saving the motor from damage.
3. The best idea, but trickiest, is to add micro-switches at each end of the ram's travel. These would be wired in series with the motor's supply to cut off the current just before the ram gets to the end.
In conjunction with this, diodes are wired across the switch terminals in such a way that they allow the current to flow in the reverse direction, when conditions prompt, at which stage the switch restores normal current path.
Problem with this lies in the fixing of the switches as they need to cope with the actuator running on for a while after power-off and not get caught up with any moving parts etc.
—
MacGregor 26M 2009 - Sky's the Limit - Suzuki DF50
From the way increasing the separation of your signal wire appears to help I suspect the EMC problem is being capacitively coupled. The capacitance of a reasonably short cable length (a few metres) will only be hundreds of picofarads. For this small capacitance to be having much effect the impedance of the circuit needs to be high (very likely with push button inputs that were only ever intended to be connected to local controls) and the transition edges of the signal need to be fairly fast. This does point to the pushbuttons being polled (or sampled) as Rick suggests. I would be wary of using relays to provide remote controls, there is a good chance of introducing new EMC problems from the substantial magnetic and voltage interference they will produce, given that we know the circuit is vulnerable. It would be better to use Opto isolators or CMOS switches to provide the remote controls. These both have the advantage of not introducing a new source of EMC problems and Opto isolators will give complete voltage and current separation from the existing circuits. If you are still having problems I am happy to do you a circuit diagram and provide the appropriate part numbers.
You won't get a fuse to reliably protect the motor and survive the end of travel overload. The motor current will rise typically 3x to 10x its normal running current when the motor stalls at end of travel. Anyway I guess you don't want to have to keep changing fuses! Far better to either prevent the overtravel with limit switches as you suggested or detect the overcurrent and shut it off fast enough so a fuse doesn't have time to blow. A simple but effective method is to use a reed relay in the motor circuit with the correct number of turns of wire around the reed it will not switch on under normal motor load current but will as soon as the motor current rises. The number of wire turns needed can be calculated from the reed sensitivity and the target current e.g if it is a 20 Ampturn reed and you want it to switch at 4A you wind 5 turns round it, 40 turns and it will switch at 0.5A. You then need to use it to switch a latching relay so that it doesn't chatter the motor on and off repeatedly. Use a T rated antisurge thermal fuse (not an F rating fast fuse) and it will survive the very short overload currents.
PTC self resetting Polyfuses are a possible option for the motor, at least then they self reset once the motor is switched off.
Dave.
Rick, do you want the issues of ram end travel limit on a separate thread?
Looking at the other (much more expensive) autopilots, they all seem to have "ram travel limiting" covered in some way, e.g. Rudder position sensor etc.
There is now a new thread on the whole issue of limiting ram travel - see here.
—
MacGregor 26M 2009 - Sky's the Limit - Suzuki DF50
Do you think I can use this to effect a wireless remote for my autopilot.
As far as I understand it (which is not very far) the unit receives commands via Bluetooth and effects one of 5 modes of switching, separately in 4 circuits; each of which can carry up to 100v DC.
Thanks Dave, to be honest it's hard to take it all in:) I did originally try a Maplin multicore. I learnt by trial and error that:
1. Firstly cable length was a factor,
2. Proximity of cables in the loom was also a factor - with a tiny extra gap solving the problem; hence interest in the dialectic cross you mentioned.
3. Not only was I getting circuits being made without switch buttons being pressed, I was creating an array of new, unintended circuits between all 12 wires involved.
The main learning point was that the sheath of the multicore was keeping all my cables very tightly compacted; just what I needed to avoid.
I've now made up a loose loom of 4 pairs of twisted pairs (16 wires) not sure if each leg going through both wires in the pair helps but it does work:)
I'm hoping to send this up the binnacle tomorrow afternoon, weather permitting, to see if it still works when surrounded by all that other wiring. Obviously need to test it with my OBM, GPS, VHF running.
Have been getting on well with other aspects of the AP install. I'll try to install the tiller extension - including modified AP ram end receptacle. Will post pics in due course.
Will also be measuring up for cutting into the engine well liner to accept the chestnut shelf I'm fabricating on which to anchor the AP unit.
Certainly learning a lot about sourcing all the relevant DIY bits cheaply. Eg was amazed to get a 200x200x3mm sheet of aluminum through the post in a couple of days for £6 including p&p. I'll be able to make two separate components out of that.
We have this joke at home now that "everything costs £7":) I'm certainly keeping the postman busy but it saves a lot of time and deisel.
MacGregor 26M 2009 - Sky's the Limit - Suzuki DF50
Thanks Rick, yes, the buttons connect directly to the microprocessor controller chip.
I'm actually just waiting for the tide so I can spend the afternoon on the boat just tinkering. Will try to test my "cat5 loose loom from buttons to binnacle" set up if I get time.
I caught my rudders on a mooring the other day and need to sort them out first; careless, I was trying to avoid a SUPer and didn't notice it.
MacGregor 26M 2009 - Sky's the Limit - Suzuki DF50
Still thinking of my relay idea, you can get these on eBay: http://www.ebay.co.uk/itm/192142396572. That's 5 x 12V relays with pre-wired holders. And just over £7 for the five!
Not sure what they're like, made in China, but there's quite a large selection of similar things. Should be easy to wire up.
Just a thought.
Rick Jones (Treasurer), former 26X & 19 owner, Isle of Wight
Thanks Rick, if my current plan fails I'll definitely give it a try, specially now I've opted for a 4 button approach.
Another issue I'm working on, as the unit will often be out of sight, is the motor burning out when the ram reaches the end of it's travel - there's nothing to tell the motor to stop.
These are some ideas:
1. Since the problem is mainly precipitated by undue lee/weather helm - ie full deflection fails to hold the course dialed in - better rig balancing should help.
2. Using a fuse that will blow when the load rises above normal but before any damage is done. A 12A fuse is recommended. I tried a 10A and this blew almost immediately the ram hit the end of it's travel thus saving the motor from damage.
3. The best idea, but trickiest, is to add micro-switches at each end of the ram's travel. These would be wired in series with the motor's supply to cut off the current just before the ram gets to the end.
In conjunction with this, diodes are wired across the switch terminals in such a way that they allow the current to flow in the reverse direction, when conditions prompt, at which stage the switch restores normal current path.
Problem with this lies in the fixing of the switches as they need to cope with the actuator running on for a while after power-off and not get caught up with any moving parts etc.
MacGregor 26M 2009 - Sky's the Limit - Suzuki DF50
From the way increasing the separation of your signal wire appears to help I suspect the EMC problem is being capacitively coupled. The capacitance of a reasonably short cable length (a few metres) will only be hundreds of picofarads. For this small capacitance to be having much effect the impedance of the circuit needs to be high (very likely with push button inputs that were only ever intended to be connected to local controls) and the transition edges of the signal need to be fairly fast. This does point to the pushbuttons being polled (or sampled) as Rick suggests. I would be wary of using relays to provide remote controls, there is a good chance of introducing new EMC problems from the substantial magnetic and voltage interference they will produce, given that we know the circuit is vulnerable. It would be better to use Opto isolators or CMOS switches to provide the remote controls. These both have the advantage of not introducing a new source of EMC problems and Opto isolators will give complete voltage and current separation from the existing circuits. If you are still having problems I am happy to do you a circuit diagram and provide the appropriate part numbers.
Dave
Dave Newton Sailbadthesinner
Many thanks Dave, will see how it goes.
MacGregor 26M 2009 - Sky's the Limit - Suzuki DF50
Mike,
You won't get a fuse to reliably protect the motor and survive the end of travel overload. The motor current will rise typically 3x to 10x its normal running current when the motor stalls at end of travel. Anyway I guess you don't want to have to keep changing fuses! Far better to either prevent the overtravel with limit switches as you suggested or detect the overcurrent and shut it off fast enough so a fuse doesn't have time to blow. A simple but effective method is to use a reed relay in the motor circuit with the correct number of turns of wire around the reed it will not switch on under normal motor load current but will as soon as the motor current rises. The number of wire turns needed can be calculated from the reed sensitivity and the target current e.g if it is a 20 Ampturn reed and you want it to switch at 4A you wind 5 turns round it, 40 turns and it will switch at 0.5A. You then need to use it to switch a latching relay so that it doesn't chatter the motor on and off repeatedly. Use a T rated antisurge thermal fuse (not an F rating fast fuse) and it will survive the very short overload currents.
PTC self resetting Polyfuses are a possible option for the motor, at least then they self reset once the motor is switched off.
Dave.
Rick, do you want the issues of ram end travel limit on a separate thread?
Dave Newton Sailbadthesinner
Looking at the other (much more expensive) autopilots, they all seem to have "ram travel limiting" covered in some way, e.g. Rudder position sensor etc.
There is now a new thread on the whole issue of limiting ram travel - see here.
MacGregor 26M 2009 - Sky's the Limit - Suzuki DF50
Managed to squeeze the 16 wires for the remote up the binnacle tube this morning (no wind:() and the good news is that they work fine.
Just waiting for a few new tools from eBay and I can get on with the other bits.
MacGregor 26M 2009 - Sky's the Limit - Suzuki DF50
Ok, just did a bit of Googling and found this - http://www.ebay.co.uk/itm/5V-12V-24V-4-Channel-Bluetooth-Relay-Remote-Sw...
Do you think I can use this to effect a wireless remote for my autopilot.
As far as I understand it (which is not very far) the unit receives commands via Bluetooth and effects one of 5 modes of switching, separately in 4 circuits; each of which can carry up to 100v DC.
Sounds ok to me but am I missing something?
ps just seen the same thing for £4.98 - http://www.ebay.co.uk/itm/12V-4CH-Channel-200M-Wireless-Remote-Control-R...
MacGregor 26M 2009 - Sky's the Limit - Suzuki DF50
Pages