Century IIB autopilot troubleshooting

M

Mark Nestler

Pre-Flight
Joined
Dec 6, 2020
Messages
36
Display Name
Display name:
NESTIAM
I have a Century IIB autopilot which has not worked since I owned the plane. I've sent the Control Module out to be rebuilt (and reinstalled it). I've also replace the harness connection at the back of the Control Module. I can hear and feel the Servo snap when I place the autopilot into the ON position. However, during a ground check there is no response to turning the Bank either left or right and no response in the heading mode (either with engine running or without). Is there any simple steps I can perform to try and find the failure. I can use basic functions on a Multimeter. Visibility is not great trying to check Horizon connections or any other components. Avionics shops in central Florida are not interested in working on Century IIs, and a digital replacement are generally out of budget. Thanks.
 
Have you contacted Bob Weber (https://webairconsulting.com/)? I spoke to him regarding issues I had with my IIB, and he is very knowledgeable. If you're comfortable with a multimeter, he can do troubleshooting over the phone with you.
 
No, I have tried several local shops for recommendations and no one could offer any advice.
 
His site does look like it could be helpful I'll try to contact him. Thanks.
 
Replace the connectors!
Very simple and should be easily done by your IA.

Your symptoms are very common, and in my opinion, should not lead you to abandon this very capable autopilot. Another aid in diagnosis that helped me was to fly with your heading function on and bug within 30 degrees to either side. Then stick your hand behind the control and into the wires without pulling any out. You may find that the aircraft will feel like it engages and turns to the heading bug. If not the connectors, it might also be a faulty DG or AI, but I would consider the connectors first.
 
On the CD47 roll servo connector there are 4 pins A(solenoid power), B(ground), C(motor), and D(motor/common). If you feel the solenoid engage pins A and B are good, the ones in question are C and D. You can take a multi meter and check the drive voltage (DC) from the amplifier by using pins C and D. Be sure to to have one lead in C and the other in D. Do not use air-frame ground as a reference. Depending on the voltage of the aircraft you should get close to bus power on the drive pins to the motor when using the roll turn knob or heading error if the heading rocker is closed. In a perfect scenario AI level and using just the roll turn knob, under center there should be close to 0VDC moving left or right you should see a increasing voltage. One way will be positive and the other negative.

As others have mentioned the female split pins are notorious for being intermittent, you can try and use a small pick to push the pins back together but sometimes the connector just needs to be replaced. If you see the drive voltage from the amplifier and it is variable depending on the input error the problem might be in the drive motor assembly on the roll servo. You can also bias the drive motor directly if you have a portable variable DC power supply but that is a little more in depth.
 
Thanks for the advice on all entries. When checking the voltage at the Servo as described above: If voltage is present on pins C/D (When not in the center position), I suspect a bad Servo. If not, do I move back up to the Control Module pin block and check for voltage at a specific pin? I see on the schematic the C branches off to the E pin (Motor), and the D (I'm not certain) looks to go to the D Motor/Common pin with interconnect to the DG and Horizon.
 
Correct, if you see the voltage at the servo connector CD 47 and you can control it with the roll turn knob or heading mode the wiring from the amplifier to the servo is good and the amplifier is functioning properly. The problem would then be in the roll servo, possibly a issue with the carbon brushes in the brush cap. If you don't see the voltage you would have to back up to the amplifier connector and check it there.
 
Thanks to the OP for bringing up this topic, as I’m sure there are a lot of pilots with dysfunctional Century autopilots. I was among that group with my Piper Autocontrol 3B (Century 2(B?).
Mark, please let us know how it turns out.

I also feel indebted to Jacob Kinsey for giving such a technical answer to troubleshoot.
While I don’t need it now, I’m cutting and pasting this information for future reference to help track down specific gremlins.

With the lack of easy support in the field for those 40+ year old autopilots, it is even more important for each of us Century owners to help our local mechanics locate info that may not be easy to find.
 
I will provide feedback. Most likely Friday which is my next opportunity to get to the airfield. My plane is not in a hangar and I'm somewhat dependent on weather.
Yes, the technical advice is invaluable. Knowing where to tap is 99% of the solution.

If the Servo needs brushes, I hope someone still manufactures them. The plane is 1975 vintage (5700 hours on airframe) and I assume the Servo motor may have a number of hours on it.
 
Brushes and bearings are still available. If there is excessive wear on the commutator the drive motor assembly would have to go back to Globe for overhaul. Currently there is a part issue with some motor assemblies that I currently have at Globe and we are waiting on parts to finish the units that are there for overhaul.
 
I checked the voltages as described above on the C/D pins at the Servo motor. However, the results were not as specified. With the Multimeter on the C and D pins I was getting a voltage of 10.5V full left, 11V at the top center and 10.9V full right. The voltage never went negative with the rotation of the bank from left to right (unless I reverse the red and black leads against the C/D terminals then they were always negative). The plane was not running while performing these checks. Any ideas?
 
So, with red lead on CD47 C or E at the amp and black lead on CD47 D or D at the amp with the roll knob full right I had 12.2VDC and full left was -12.2VDC. Test was ran on the 66D1000 Century bench test set with a Century 2 amplifier (1C338). My guess would be that there is a fault in the amplifier possibly bad drive transistors. It is common when they fail to produce a constant drive voltage only one way. Just a question but what was the orientation of the AI when you were doing the test? Ideally it should be as close to level as possible if not it will introduce an error into the system and it is possible that with a large enough error the roll turn knob would have no effect. The error would need to be greater than 30 deg of indicated bank.
 
My horizon was sitting about 3 to 5 degrees low and not more than 5 degrees of bank. The Control Module was rebuilt and certified by a shop in Oklahoma a few months back. Is that the same as the amplifier?
 
Yes the 1C338 and 1C385 are the main amplifier/control head. With the AI only having a couple of deg roll error it should have been controllable with the roll turn knob or heading error if the heading rocker was closed. You could check the resistance of the drive motor, a 14v motor should be around 3-4ohms and a 28v motor should be around 8-12 ohms. You should not have a dead short when checking resistance of the motor windings. If you had a dead short I could easily see that taking out a set of drive transistors.
 
To check the resistance of the drive motor would I insert the OHM meter black/red terminals into the C/D slots on the Servo side or must I have access to the motor windings?
 
If the C/D terminals are producing a voltage of approximately 11V no matter what the position of the Bank control. Should not the motor responds to the voltage in some way. Such as full deflection of ailerons in one direction?
 
Mark Nestler said:
To check the resistance of the drive motor would I insert the OHM meter black/red terminals into the C/D slots on the Servo side or must I have access to the motor windings?
Click to expand...
Yes, the roll servo connector pins C and D with meter on resistance.

Mark Nestler said:
If the C/D terminals are producing a voltage of approximately 11V no matter what the position of the Bank control. Should not the motor responds to the voltage in some way. Such as full deflection of ailerons in one direction?
Click to expand...
Ideally yes, the motor should drive with that type of voltage reading. However, if one of the brushes is not properly seated or the windings are open/shorted the motor will not work.
 
Jacob Kinsey said:
Yes, the roll servo connector pins C and D with meter on resistance.


Ideally yes, the motor should drive with that type of voltage reading. However, if one of the brushes is not properly seated or the windings are open/shorted the motor will not work.
Click to expand...
If one was just starting the diagnostics, it would seem that the servo/motor needs to be checked for dead short and resistance and function first before connecting to a new or serviced amplifier/control module to avoid the possibility of blowing the transistors. After the multimeter readings, the use of a 12 volt power source directly connected to the C and D would seem to do the trick, and then reversing the wire connections to demonstrate reverse motor operation. Is this correct?

This is probably stupid of me(not an A&P), but where is the servo/motor located and how easy is it to get to replace it or just the brushes.

Also, you mentioned the need to keep the AI relatively level. What if the AI is functioning, but the pickup giving attitude info to the autopilot is defective. Could this cause the control module voltage readings to stay positive throughout the rotation of the wing leveler knob, rather than the transistors being blown? And further, if the connectors were not making good contact, might that further confuse the situation?
 
MajorTurbulence said:
If one was just starting the diagnostics, it would seem that the servo/motor needs to be checked for dead short and resistance and function first before connecting to a new or serviced amplifier/control module to avoid the possibility of blowing the transistors. After the multimeter readings, the use of a 12 volt power source directly connected to the C and D would seem to do the trick, and then reversing the wire connections to demonstrate reverse motor operation. Is this correct?
Click to expand...
Yes you could use a 12VDC supply and bias the motor directly to see if it would run, I would suggest to have a current meter so you can see what the current pull is. Even if the motor was faulty and was pulling more current than normal it might run but due to the current pull it would cause the transistors to fail eventually. A dead short would almost instantly kill the drive transistors.

MajorTurbulence said:
This is probably stupid of me(not an A&P), but where is the servo/motor located and how easy is it to get to replace it or just the brushes. Probably needs to be done in conjunction/or by an IA.
Click to expand...
Depends on the aircraft, a V35 bonanza for example is in the left wing wheel well, kind of a pain to get to easily. I would suggest to have a A&P remove the servo and send it out for repair if the brushes were faulty.

MajorTurbulence said:
Also, you mentioned the need to keep the AI relatively level. What if the AI is functioning, but the pickup giving attitude info to the autopilot is defective. Could this cause the control module voltage readings to stay positive throughout the rotation of the wing leveler knob, rather than the transistors being blown? And further, if the connectors were not making good contact, might that further confuse the situation?
Click to expand...
It is possible if the roll pick off was faulty to produce a large enough error that would prevent the amp from functioning properly. If the AI was in question one could simply unplug the AI and the amplifier would think it was getting a null. Maximum command of the roll turn knob is around 30 deg of bank. For instance on the 66D1000 test set I added in a 40 deg roll attitude error and when looking at the drive coming out of the amplifier it would not change due the excessive roll attitude error. You could turn the roll turn knob to end of travel either way and it never changed the output drive.
 
In the Grumman AA5 the Servo Motor is located in the center console. Just to the right of the pilot's foot. Took about half an hour to remove console panels for access. For the rest of the tests I'm going to pull the Servo out for better access and see if there is a brush inspection panel. Better than standing on my head juggling a Multimeter.
 
With the Multimeter set on 3 KOHMs I got a reading of .01. What is the interpretation of this reading?

upload_2022-1-8_11-13-49.png
 
Can the brushes be viewed in this motor without major disassembly?

upload_2022-1-8_11-34-43.png

upload_2022-1-8_11-35-14.png
 
The problem is the you can't get bearings, and you can't fix the clutch.

The way to deal with this is to get you and the aircraft to Edo-Air Mitchell in Mineral Wells. They provide complete diagnosis and repair in ONE SESSION. I do that....
 
Servo is reassembled. Cleaned out any remaining brush residue dust. Tested on 12V car battery. Motor rotates with power supplied to the C/D terminals (reversed black and red leads between the C/D terminals to confirm both directions function). Solenoid engages/disengages when power supplied to the A/B terminals. Can the resistors be checked while still on the circuit board? What is the procedure if that is possible.
 
Based on the picture of the multi-meter I would say that it looks like you had 10ohms of resistance. Brush wear looks to be ok on the commutator no bad threading or grooving that I can see. How was the back bearing, was it stiff or packed with carbon? Also, what resistors are you referencing for checking?
 
After I reinstalled the brushes I ran the motor for about a minute and then rechecked the resistance on the C/D terminals. Using the same settings in the Multimeter I got a reading of .004. Which fell right in line with your earlier post of 3 to 4 Ohms for a 14V motor. The bearing above the commutator was clean, but not stiff at all, felt easy to rotate. I did not disassemble further to look at the other bearing. You mentioned a bad resistor may be the cause of the Voltage not reversing it's polarity (it was always positive on the C/D terminals in the aircraft). I looked at what they repaired on the control Module when they tested it a couple months back. They replaced the Potentiometer, a Tantalum capacitor, and a 200MF capacitor (I hoped they tested all the other components). Possibly they welded the wrong leads into the Potentiometer? Below is a picture. The red lead on the left terminal, orange on right and the white/orange coming off the HDG on/off rocker on the middle terminal of the Potentiometer. Does that seem correct? There is a multitude of resistors on the circuit board. I do not want to have to un-soldier each to check resistance. Is there a way to check them on the board?

upload_2022-1-10_10-26-10.png
 
The 2 small can transistors on the top right of your pic and the 4 transistors on the bottom of the unit are the most common parts to fail. If you know how to check transistors that is where I would start or possibly send the unit back for evaluation. Sounds like the servo is ok, just need to figure out why you are not getting the correct drive out of the console amplifier. Here is a pic to compare the wiring. There is a MM online at CSOBeech, just search for 1C338 in a google search and it should show up.
p
1C338_3__58326.1518205036.1280.1280.jpg
 
It is interesting that 10 ohms resistance was found the first time checking but after running it a bit and after cleaning up the brushes and dust, the hoped for 3-4 ohms was obtained. But the initial 10 ohms resistance would initially make one think there could be an eventual problem that would lead to transistor failure had you not done the cleanup and visual inspection. I guess this step should not be omitted to avoid coming to the wrong conclusion about resistance.

So far, obviously not a direct short since motor was functioning, and thus a hypothesis of blowing the transistors not an issue after you reconnected the certified control unit. And hopefully the changing resistance is now stabilized to the 3-4 ohms and not a concern going forward to reduce the life of the control panel transistors.
 
For anyone thinking of checking the brushes it is not an easy process. Evidently, they have a special tool to hold the spring loaded brushes in place during reinstallation. The way I got around it is to remove all 4 screws on the top releasing the spring tension. Place the brushes back in place making sure the spring is just touching the back side of the brush (brushes have a slight grove in back for the spring). Make sure the breaded copper wire attached to the brush is not impinged in any way. Move the brushes back in the cap to clear the way for the bearing and commutator. I laid that flat on the table so I could tip the motor upside down and place the shaft in the cap. Then fully seat the cap and turn right side up. Now turn the springs (one at a time) counter clockwise to put tension back on the brush and screw down the top screws with the terminals attached. Make sure the breaded wire is still free to move and you didn't pull it off the soldier connection. Double check spring is forcing brush in towards commutator. You can now run it on a 12V car battery if it is not the 24V motor. The motor rotates freely as long as you do not engage the solenoid with the A/B terminals. The other headache is the cable attached to the front of the motor. Do everything possible to keep that in place on the pully. Mark the location on the Aileron cable with tape to keep it roughly in the same place on the Aileron cable during reinstallation. Then recheck resistance if you like. Still have to tackle the resistors part. Thanks for the coaching.
 
I just noticed you’re looking at other autopilot conversions, so maybe my questions are no longer germane for you. So, if still appropriate, I’m not sure whether you will be addressing the control panel drive voltage anomaly, whether checking the transistors yourself or sending the unit back to the Oklahoma repair station. But your other point as to whether +- 10.5 volts should be enough to move the servo/motor was answered affirmatively earlier in the thread. Since you demonstrated normal function of the servo/motor after cleaning, there should be no reason that voltage should not be present at the servo unless the wiring/ connectors are preventing that. Did you verify that that 10.5 voltage was present at the servo, or at least continuity of the wiring?

Secondly, if you verify the transistors are working normally, or even before that, did you remove the connector from the AI per Jacob’s suggestion, to eliminate an AI error confounding the situation?
 
The problems associated with the Control console became insurmountable. I do not have any circuit board soldiering experience. To check the transistors they would need to be removed. I thought maybe just replacing all transistors may be the way to go. However, the Motorola 2N3116 are extremely scarce. I could only find an after market version (13 were available) at Quest electronics for $10 a piece. That was assuming the problem is in the transistors. There could be other components bad or ready to go bad. I'm already into the unit for $650 for the rebuild earlier this year. For the other part of your question, no I didn't remove the wire at the AI. I didn't want to break another wire like the wire in the blue connector behind the Console module (try finding those blue connectors, they are almost as hard to find as the transistors). Just a maze of tie wrapped wires behind the instruments. The 10.5 to 11V was read on my Multimeter directly at the terminal tied to the Servo motor. My servo had the blue connector flush on the motor, so the line was delivering voltage up to the connector on the motor at the very least. Without the Control console reversing polarity when turning the Bank control knob, it would only drive the motor in one direction. I tried calling Century Flight systems in Texas to see if they still worked on them. The last guy with experience retired 2 weeks before Christmas this year. It was worth a try to see if it was a simple solution, but time to cut my losses.
 
Totally understand. You have already done more than most would do to get it to work. It would be upsetting if the Oklahoma shop that worked on the control panel did not replace transistors that were faulty if they found them. I assume no warranty this remote from that repair date.

I don’t know the specific Oklahoma shop that did the work, but I just called “Autopilots Central” in Tulsa. They had been referenced by many over the years as great with our Century units. They are still working on them and the person I talked to was unaware of unobtainable parts for them, or alternative workarounds. In some old threads, they were able to completely rebuild entire systems at 1/3 to 1/2 the cost of something entirely new.
If I was having a big problem with my Century 2B, I would fly from CT to Tulsa to them.

At some point, support will be unobtainable in some respect. Jacob had mentioned that he was still awaiting parts for some of his units, and Dr Bruce Chien mentioned that bearings for his Century 3? are no longer made, so making a switch to a shiny new autopilot makes sense in the long run.
 
Last edited:
Bearing are still available and parts a like transistors are still available. If you have a warranty from the previous repair on the unit I would send it back in for a evaluation. If not I would be willing to take a look at it if you wanted. I work for Bevan Aviation in Wichita KS, as the head autopilot tech. If anyone has questions about Century autopilot systems feel free to ask me for help if needed.
 
Jacob, very happy to include you at Bevan in my contacts for autopilot problems. You are almost equidistant from me as Tulsa. Hope I never need you though.:fcross:
 
I am having similar problems with my Century IIb autopilot. With the plane running, AI flat/erect, and the unit on (left switch on, but not right switch, so in ROLL mode but NOT in heading mode), the unit commands a left turn when the dial is moved left, but no command to the servo when the dial is turned to the right (yoke stops moving when dial turned past midline to the right). I tested the 4 power transistors and they all read correctly as 'double diodes' (btw Mark, they don't have to be removed to test them, just slide the wire connectors off of them, takes 5 minutes). I don't see anything else obviously bad on the board. Also, when someone else's control amplifier unit is put in the plane, the autopilot works fine, so i know its the amplifier control unit itself. What would you guys look at next??? Thanks
 
I would go after the 2 small can transistors on the motherboard should be 2n3116 or 2n3700. They are used to bias the main drive transistors that are on the bottom of the heat sink.
 
You must log in or register to reply here.
Back
Top