Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Simulate an auto pilot's rudder feedback unit

Status
Not open for further replies.

R_C

Member
I am trying to bench test various parts for a Robertson AP300 boat auto pilot. I can't get past a "rudder feedback failure" error with an RF300 rudder feedback unit that is likely bad. The RF300 has two wires that connect to the main junction box running at 12VDC. Its specs show a frequency of 3400 Hz indicates the rudder is at center position and each +/- 20 Hz indicates one degree of rudder change to port or starboard.

The RF300 is mechanically connected to the rudder control arm so its arm is probably rotating a potentiometer. If I put an ohmmeter on the two wires its readings fluctuate wildly while the unit is stationary. Measuring frequency with my DMM shows around 4600 Hz at center position, which is way off the 3400 Hz spec so I think this unit is probably bad. How can I create my own circuit to simulate the RF300? I would love to breadboard a simple device that I can use but I don't know where to begin? For example, how is resistance related to frequency? Thanks in advance for any suggestions.
 
A simple 555 timer chip running as an astable, where a pot is used to control the frequency?
 

Attachments

  • DF53a.jpg
    DF53a.jpg
    291.1 KB · Views: 364
Last edited:
Try measuring at the endpoints. What you may find is that the values are stable at or near the end points. If it appears resistive, then substitute a potentiometer.

The always operating "straight ahead" probably causes wear.
 
Mike's solution looks good. As for repairing the RF300, is it easy to open the unit? A worn pot could be replaced, or perhaps a LED plus photocell could be substituted for it.
 
If the circuit does discriminate to 1 degree then it will be a fancy potentiometer and expensive.
Have you verified the coupling to the rudder shaft is good?
A can of servisol might improve things, dust and crud can get around the sliding contact on pots.
 
Unfortunately, the RF300 is a completely sealed unit so there is no access to its internals. Only the shaft protrudes from the top and the underside looks like all the electronics were potted. This unit is a spare I picked up on ebay so I could bench test all the other components of the autopilot: junction unit, compass, control heads, etc.

Mike's circuit looks like just what I need to fool the main junction unit into thinking an RF300 is connected and I can turn the pot to simulate rudder movement. I haven't used a 555 timer chip before but I'll read up and give Mike's circuit a try. Thanks everyone.
 
Measuring frequency with my DMM shows around 4600 Hz at center position, which is way off the 3400 Hz spec
Is your DMM expecting a sinusoidal input? The RF300 output may be far from sinusoidal and confuse the meter if so. Can you confirm the frequency using a 'scope?
Is the RF300 frequency affected by the supply voltage? What is the peak-to-peak or average/RMS voltage of that frequency signal?
I see from the RF300 spec that its only connections are via a screened twisted pair cable, so presumably it relies on a phantom power supply. Not sure how that will affect the simulation using Mike's circuit.
 
Last edited:
...
Mike's circuit looks like just what I need to fool the main junction unit into thinking an RF300 is connected and I can turn the pot to simulate rudder movement. I haven't used a 555 timer chip before but I'll read up and give Mike's circuit a try. Thanks everyone.

The 555 pin 3 output will have a near square-wave, amplitude a bit less than the Vcc of 12V, capable of sourcing/sinking ~200mA. Make sure that is what it takes to drive the other box?
 
Is your DMM expecting a sinusoidal input? The RF300 output may be far from sinusoidal and confuse the meter if so. Can you confirm the frequency using a 'scope?
Is the RF300 frequency affected by the supply voltage? What is the peak-to-peak or average/RMS voltage of that frequency signal?
I see from the RF300 spec that its only connections are via a screened twisted pair cable, so presumably it relies on a phantom power supply. Not sure how that will affect the simulation using Mike's circuit.

I don't have a scope so I can't confirm the frequency. I do recall testing the frequency of the working RF300 on our boat with the same DMM and it was close to the expected 3400 Hz at center rudder. I positioned the arm on this RF300 so it read close to 3400 Hz then cycled the power on the junction unit but I still get the rudder feedback failure alarm.

The RF300 connects to two pins on the junction unit and it is not polarized according to the manual. I am powering the junction unit with a bench power supply. I measure about 12.7 VDC at the RF300 pins when I vary the junction units supply from 11 to 14 VDC. If I slowly spin the RF300's arm the pin voltage varies from 12.6 to 12.8 VDC. I just tested the ohm meter with a 75K pot to see if it would remain stable then disconnected the RF300 and attached the ohm meter and today it remains stable at about .5Mohm as I spin its arm. Yesterday, I couldn't get a stable reading.
 
The 555 pin 3 output will have a near square-wave, amplitude a bit less than the Vcc of 12V, capable of sourcing/sinking ~200mA. Make sure that is what it takes to drive the other box?

I get only 22mA at 12.7V to this RF300. The specs don't show any power requirements. Of course, this unit could be bad but I think it's worth trying your circuit. I don't have any nanofarad capacitors around so it may take me a little while.
 
I want to thank everyone for your help and let you know I have things working. I read up on the 555 then created an astable with a fixed approximately 2900 Hz frequency and hooked it up to the junction unit with no luck. Next, I did a master reset of the junction unit's memories then connected the RF300 and went through rudder feedback calibration in the setup menus and it worked. Apparently, the previously saved calibration settings were sufficiently different from this RF300 unit that it was not recognized. I marked my own center at 3400 Hz and 30 degrees to port and starboard at about 4000 Hz and 2800 Hz. Next, I connected the 555 timer circuit and it was successfully detected at a fixed starboard rudder angle.

Now that I have both the spare RF300 and the 555 timer circuit working, I am confident I could use Mike's circuit to create my own rudder feedback simulator and I may just do that for fun. Thanks again for your help.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top