Simrad Robertson RF300 Testing

[R_C]

I am bench testing Robertson AP300 auto pilot parts and I'm getting a rudder feedback error with a used spare RF300. At its center position the RF300 tests at 4460 Hz when the specs say it should be 3400 Hz. At 20 Hz per degree that's over 50 degrees off. Could that be the cause of the rudder feedback error? I tried running the rudder feedback calibration in setup's dockside settings but it always shows a zero angle as I move the RF300's arm. The frequency does change above and below 4460 Hz as I move the arm. Could it be too far out of range to be recognized?

[tayana55]

Can you rotate the rudder arm to get 3400Hz ?
If you can, rotate it further to port and starboard +/- 50 degrees approx and check that the frequency is in spec.
Some RF300 units were assembled incorrectly at the factory and the pot zero was misaligned

[R_C]

Many times I positioned the rudder arm at 3400 Hz then turned on the AP300 but always got the rudder feedback failure. I assumed the problem was the RF300. Next, I breadboarded a simple 555 timer as an astable at about 2900 Hz and it too was not recognized. I was planning to create my own RF300 simulator with a 555 and a potentiometer. Finally, I did a master reset of memories on the J300 then went through the dockside settings and rudder feedback calibration using my marked 3400 Hz center and it worked. It now recognizes both the RF300 and my 555 timer circuit. I can only guess that the J300's stored rudder calibration data was either bad or way out of spec.

I still get the rudder feedback failure alarm at power on but after several minutes it recognizes the RF300 and all is well. I don't understand why the delay because the frequency measures fine at power up. Any ideas?

[tayana55]

Originally Posted by R_C

I still get the rudder feedback failure alarm at power on but after several minutes it recognizes the RF300 and all is well. I don't understand why the delay because the frequency measures fine at power up. Any ideas?

If you power cycle the J300 after it has been on a while, do you still get rudder feedback failure alarm ?

[R_C]

Originally Posted by tayana55

If you power cycle the J300 after it has been on a while, do you still get rudder feedback failure alarm ?

No. It will turn right back on without the alarm and correct rudder angle if it has been running a while. Have you seen this before?

[tayana55]

Originally Posted by R_C

No. It will turn right back on without the alarm and correct rudder angle if it has been running a while. Have you seen this before?

In that case it sounds like a capacitor on the feedback circuit, that's the most likely component that would introduce a time delay.

If the RF300 is giving the correct output, then it is likely the J300

[R_C]

Originally Posted by tayana55

In that case it sounds like a capacitor on the feedback circuit, that's the most likely component that would introduce a time delay.

If the RF300 is giving the correct output, then it is likely the J300

This morning it took almost half an hour at power up before it recognized the RF300. Turning it off then right back on it continues to work without the alarm. A bad capacitor makes sense. I understand the only proper way to test electrolytic capacitors is out of the circuit with a dedicated meter. In that case I'll probably just order up a bunch and replace them.

[tayana55]

Originally Posted by R_C

This morning it took almost half an hour at power up before it recognized the RF300. Turning it off then right back on it continues to work without the alarm. A bad capacitor makes sense. I understand the only proper way to test electrolytic capacitors is out of the circuit with a dedicated meter. In that case I'll probably just order up a bunch and replace them.

I would check all of the power rails for stability first.
The caps can be checked in circuit with an ESR meter, high impedance is a good indicator of a worn out cap. Any other suspects, I usually remove and do an out of circuit test.

[R_C]

Originally Posted by tayana55

I would check all of the power rails for stability first.
The caps can be checked in circuit with an ESR meter, high impedance is a good indicator of a worn out cap. Any other suspects, I usually remove and do an out of circuit test.

Where are the power rails and what are the expected voltages. I see a row of what look like voltage regulators in TO-220 packages. Any idea what those are?

[tayana55]

Originally Posted by R_C

Where are the power rails and what are the expected voltages. I see a row of what look like voltage regulators in TO-220 packages. Any idea what those are?

The T0-220 packages are mosfets, diodes and voltage regulators.

The Mosfets form a PWM bridge driver that controls the motor speed.
The voltage regulator supplies the system voltage, around 26V if I remember correctly and the diode is for reverse polarity protection.

There are various other voltages generated internally, 15V and the 5V logic supply.

I haven't got round to reversing the complete board, only areas that I've needed to effect repairs.

[R_C]

Originally Posted by tayana55

I would check all of the power rails for stability first.
The caps can be checked in circuit with an ESR meter, high impedance is a good indicator of a worn out cap. Any other suspects, I usually remove and do an out of circuit test.

I bought an ESR meter and checked all the capacitors. Five of the six SMD caps on the board tested either very high or wouldn't register at all. I replaced all six and the system now turns on every time without the rudder feedback error and with a valid rudder angle. I do get a low supply voltage alarm at power on but that's because the bench power supply I'm using maxes out at 5A. Apparently, charging all those capacitors takes a brief burst of current.

[tayana55]

Originally Posted by R_C

I bought an ESR meter and checked all the capacitors. Five of the six SMD caps on the board tested either very high or wouldn't register at all. I replaced all six and the system now turns on every time without the rudder feedback error and with a valid rudder angle. I do get a low supply voltage alarm at power on but that's because the bench power supply I'm using maxes out at 5A. Apparently, charging all those capacitors takes a brief burst of current.

It sounds like you've had some success, that's great!
Were the PSU caps OK or did you have to replace these too ?

[R_C]

Originally Posted by tayana55

It sounds like you've had some success, that's great!
Were the PSU caps OK or did you have to replace these too ?

All the electrolytic through-hole caps tested fine with the ESR meter. Only the surface mount caps were bad. Which ones are the PSU caps, the 2200uF ones? The two caps I could not test are the green (ceramic?) caps labeled 151K and 221K. Are these 150pF and 220pF? If so, what chemistry are these and why so large? Current 150pF and 220pF ceramic caps are tiny.

[tayana55]

Originally Posted by R_C

All the electrolytic through-hole caps tested fine with the ESR meter. Only the surface mount caps were bad. Which ones are the PSU caps, the 2200uF ones? The two caps I could not test are the green (ceramic?) caps labeled 151K and 221K. Are these 150pF and 220pF? If so, what chemistry are these and why so large? Current 150pF and 220pF ceramic caps are tiny.

Can you send a photo or attach to this thread ?

[tayana55]

Originally Posted by R_C

All the electrolytic through-hole caps tested fine with the ESR meter. Only the surface mount caps were bad. Which ones are the PSU caps, the 2200uF ones? The two caps I could not test are the green (ceramic?) caps labeled 151K and 221K. Are these 150pF and 220pF? If so, what chemistry are these and why so large? Current 150pF and 220pF ceramic caps are tiny.

Measure the DC resistance, I suspect that they are inductors
15 microH and 22 microH. See here MCNL453232-150K - MULTICOMP - INDUCTOR, 15UH, 200MA, 10% | Newark