ARU-11/A Attitude Indicator - Part 2: Repair and Demo
[Marc] hello and welcome back if you follow the channel you know tHello and welcome back. If you follow the channel, you know that we love all things Apollo. We restored an Apollo Guidance Computer, and then got busy restoring the entire Apollo communication system. Recently, we restored a DSKY, the iconic computer interface that was sitting in the middle of the spacecraft’s control panel.
But there is another iconic instrument that we have been itching to play with: the Flight Director Attitude Indicator, or FDAI, aptly nicknamed the 8-ball indicator. Although it looks similar to a conventional aircraft attitude indicator, it is quite different. Traditional artificial horizon indicators rotate around two axes, pitch and roll. Which works well enough on earth. But in space, you need all three axes, hence the mesmerizing ball mechanism, which can rotate to any angle. We haven’t gotten our hands on a real Apollo FDAI yet, but we got the next best thing instead, some ARU-11/A fighter plane attitude indicators. They are made by Lear Siegler the same company that made the Apollo FDAIs, so they should be fairly similar.
Our blue-grey unit, is a scrap bin rescue and is missing some parts. In episode one, we opened it up, and discovered it had suffered some kind of catastrophic damage that melted part of the main slip ring. A slip ring is a rotating electrical joint that funnels dozens of signals to the rotating ball. That ended up causing pseudo shorts that released the magic smoke. However, after a lot of detective work by Ken and Eric, we managed to find our way around the electrical damage, and got the ball to rotate around all three axes, as recounted in the previous episode. That did feel like a victory already.
But our blue indicator is missing stuff. In particular, it did not come with the servo control electronics, which are essential to moving the ball in a controlled manner in response to gyro signals from an inertial platform. Fortunately, Patron R.J. Gritter came to the rescue and sent us not just one, but three ARU-11's, two of them having their screw-on electronic package still intact. [Marc] And also, I forgot to say but, Ken, you had figured out that we were missing the electronic amplifiers for the servo control? [Ken] Yeah. There's a separate amplifier module that plugs into the back, that converts the aircraft signal into the servo controls.
[Marc] Right. And the collector, it turns out, he said: oh I have others that have the this module, and he send us two of them. [Ken] So this is the amplifier in the back here. [Marc] Okay, can you tilt a little bit more away? Right. So that plugs in to the back. And they are different ones, they're not all the same plugs either. [Eric] You can see the plug here. [Ken] Yeah. So there's two plugs, one plug plugs
into the aircraft and then the other plug plugs into the amplifier unit. [Marc] Right, so that's the one that does the servo loop. [Ken] Yeah. [Marc] And then, you discovered that they were different configurations of that servo loop plug. And the collector gave us two of these, that are not the same. [Ken] Yeah, they all have similar connections, but keyed differently, and weirdly incompatible. So we need to figure that out.
[Marc] Oh, this one has an out plug too. [Eric] J1, yeah. [Ken] That's a puzzle! [Marc] All right, it's more reverse engineering to do. Moments later... [Marc] We got one amplifier off, and are discovering the innards. Doesn't seem ultra complicated. [Ken] Yeah, I was noticing there's the three amplifiers here, for roll, pitch, yaw.
There's a transformer that takes 115 volts and gives you four 2.33 volt supplies, an 18 volt and a 32 volt to power everything. [Marc] It's nicely documented on it.
[Ken] Now it looks like there's been some spray painting going on here. [Marc] Yeah, this has been eBay-ified. [Ken] Oh, it looks like conformal coating, my nemesis. So this plugs into here. So, aircraft signals go in here. This box sends signals and power through here to the amplifier, which then sends signals back to drive the motors.
It has a nice cover that that goes on it, to make it look beautiful. [Marc] Should we go further and take those modules out? I bet you they don't work. [Ken] You know the saying: if it ain't broke, break it! [Marc] Okay, twice the same thing indeed. All right. [Ken] Are the resistors all the same? [Marc] Uh, they might be tuned differently, that's possible. While I'm at it, I'll just take them all off. Okay, what is that one? Oh, that one, there's some there's some stuff that's...
[Ken] Is that anything important? [Marc] Uh yeah, something has corrosion on it. So this board is probably not in good shape. That looks like corrosion actually. Okay. [Ken] Yeah, that board must be rectifiers for the power supply. [Marc] Master, do I do an official handover? You get it. [Ken] Okay, I like the way they stacked one component just on top of the other here.
[Marc] Which one? [Ken] Um, here. [Marc] Oh yeah. [Ken] Is that a good angle? [Marc] Yeah, sort of. If I could focus on it, it would be good, but... [Ken] It's like some people just put parts on PCBs, and other people make it into a 3D assembly. [Marc] There you go. Yeah, I have it.
[Ken] Then there's another capacitor here, that's just sitting on top of the transistor. Looks like they just ran out of space by a little bit. [Marc] Yeah. But nothing that will resist you, I'm pretty sure. [Ken] Well, the resistors...
[Marc] The resistor will resist you! [Ken] And then there's a power supply rectifier board with some diodes, and some very nasty capacitors here. [Marc] Yeah, I bet you we will have to do some recap. [Ken] Yeah, these have seen better days. [Marc] Yeah the power supply board... [Ken] There is like corrosion. [Marc] ...is corroded, falling part. The usual capacitor mess that we see on those.
[Ken] We're not going to get good power out of that. [Marc] Tantalums that are not that reliable. Okay. [Ken] So yeah, I guess I'm going to be reverse engineering these little amplifier boards. [Marc] Alright, we have reached the unfortunate time when we need to fire up the elevator music, for a little refresher on synchro control loops.
So, as we had seen in the previous episode, each axis of our ball has a servo motor that spins the ball around, so, three motors total. The reference winding of each motor is connected to 115 Volts 400 Hz AC. But just doing that isn't enough to make them move. To make the servo motor turn, you need to also provide a smaller AC voltage to the other winding, the control winding. And that's exactly what we did in the previous episode. We provided a max of 28V AC on the control winding. That made the motor, and in turn the ball, spin continuously. And, in some cases,
it made it emit the magic smoke, but we won't talk about that anymore. Or, um, maybe we will. But we don't want the ball to just spin continuously. We want to move it to a specific orientation, under the control of the plane's gyroscopic inertial platform. For that, we need a more sophisticated control scheme.
First and foremost, we need a device to measure the angle of the ball. That device is the control transformer. Once again, there is one for each axis. It's the second cylindrical thingy next to each motor. The control transformer is mechanically connected to the ball, which is in turn linked to the motor via a gear reduction. So,
when the motor turns, so does the ball, and so does the control transformer. The control transformer is the electrical control input of our system. It receives the angle information from the gyro inertial platform. More exactly, it receives three AC signals, coming from the synchro for a gyro axis, depicted in green here. The three signals encode the angular position of the gyro. It is that angular position that we want to show on the ball. In order to do that, we watch the signal coming at the output of the control transformer. If the
gyro synchro and the control transformer are not perfectly aligned, as is the case here, an AC error signal will be generated at the output of the control transformer. The larger the misalignment, the greater the error signal. And that's where our electronics come in. They feed that error signal to the control winding of the motor, in such a way that the motor rotates until the ball is aligned with the gyro. At which point the AC error signal goes down to zero,
and the motor will stop. Which is the situation depicted here. You may have noted that alignment means, that the control transformer rotor winding is at 90 degrees to the synchro rotor winding. That's normal, that's how these things roll. The orientation of the devices is indicated by the smaller arrow, and you see that the two are now perfectly aligned. Our ball shows the position of the gyro platform. By now, you might have guessed that the servo electronics are trivially simple. If you ignore the dangling wires for a moment, they consist of a simple AC power
amplifier. And that's the whole point. We used clever electro-magnetics, which scare us now, but were in wide use then, so the electronics implementation is rudimentary. Originally, a servo amp consisted of just one vacuum tube, or a mag amplifier. However, our ARU-11 has the latest technology. It uses newfangled transistors, which are so small that the 3 amplifiers and their power supply can fit at the back of the instrument. It's the miracle of modern technology, really. There is one extra complication, which is needed to stabilize the loop. Our servo motor contains,
in the same package, a tachometer generator. It provides another AC signal, proportional to the speed of the motor. That signal is then subtracted from the error signal. That's what our dangling wires were for. This makes the servo loop dynamically stable, and prevents the ball from over shooting its final position. And as you can see, this essential refinement doesn't even complicate our primitive amplifier a bit. So, knowing all that, we have a good idea of what to expect after Ken reverse-engineers the electronics. It should be a relatively simple AC power amplifier, with a few transistors.
Except nothing is ever straightforward in these mesmerizing military contraptions. [Eric] All right! [Ken] So the amplifier has some dodgy capacitors. [Marc] Oh, okay. I can work on that, I bought a whole bunch. Can you can you show the amplifier? It's been all reverse-engineered now? [Ken] Yeah, mostly all. [Marc] And it has the... So, those are wet tantalums, or something like that?
[Ken] So the the ones in the power supply... [Marc] Oh, yeah. [Ken] ...are the most dodgy. They're like covered in fungus.
[Marc] Yeah, those are gone. [Ken] There's three amplifiers for roll, pitch, yaw, to control the servos and this is the power supply [Marc] Is there any way to test them out of circuit? [Ken] I guess so. I mean, we can stick signals in and get signals out. [Marc] Mhm, okay. We'll recap, and we'll test those. So while Eric recaps the amplifiers, let's see what Master Ken found out. As usual, he first traces the circuit, mostly as it is laid out on the board. He then moves the components around, to make a comprehensible schematic out of it. And we got what we expected, or almost. It's
an AC amplifier alright, with 3 stages. Q5 and Q6 form the two first stages. But the output stage appears duplicated. There are two out of phase outputs, driven by Q7 and Q8. However, we sort of expected that. Ken had found out earlier, that our motors had split control windings, with
two inputs and a center tap. So the amplifiers match our servo motors, which makes sense. Here is a more complete schematic, showing the extra power supply circuit at the bottom, where Ken has circled all the suspect caps. We ended up replacing only two, the ones in the power supply, with other military spec tantalum caps. [Marc] And, I see you have the safety squints for some reason? [Eric] Got the safety squints, just in case! [Marc] And so, you have hooked up the amplifier for the first time, basically, the controller. And then, you had to recap a few...
[Eric] Three amplifiers, and that's the power supply. We had to replace these two tantalums. One of them leaked, and made kind of a mess. But that one's gone, we've got a new one in there. [Marc] Okay. And then we have three synchros, one for each axis.
[Eric] One, two, three. [Marc] And so, that should work, maybe. And then we have 400 Hz supplies. So, if I go on the other side, I should see the thing rotate, maybe. Okay, so let's try it again, see if that gives anything. So,
I haven't remounted the ball or anything. [Eric] Okay. Are you ready? [Marc] Yep. Ooh! [Eric] Oh, it's servo-ing. [Marc] Yeah. Somebody turns the...
[Eric] Grab one of the axis? Oh, oh, it's moving! [Ken] Yeah. It's the roll axis. [Marc] Well, you're flying a little fast, Ken. [Ken] Yeah, I think that's the pitch axis. [Marc] Okay. [Ken] And then... [Marc] ...the third axis. Oh, yeah, we see it moving!
[Ken] You can see the motor, but without the ball attached, it's not as interesting. [Marc] Ah, so we might have our FDAI ersatz working finally, after all the smoking. [Eric] It's a little bit warm, but not too bad. About 27 to 30°. So maybe 5° above ambient, it's not that hot.
[Marc] All right. And these are the components you replaced, on the side over there? [Ken] Just the two of them. [Eric] These guys right here. So this one which lost its wires,
and that one, which is pretty crusty looking. [Marc] Okay, so just two bad components, a shorted - for a while - indicator, but it seems to be working. [Eric] Did we see if the off flag is moving at all? [Ken] So, I think we have to hook up another wire. [Eric] Ah, okay. [Marc] Should I put the ball back together? Or are we are we that far off yet? [Ken] I think we're at that point now. [Marc] Okay.
Moments later... [Marc] So, in the meantime, after we got the indicator to work, I realized it wasn't working at full voltage, because it was not enough... [Ken] Current for... [Marc] ...current for the inverter. And I powered it up correctly,
and it emitted the magic smoke! We smoked a resistor out here. And we're all perplexed, whether it was still a short in the thing? [Ken] We didn't know if it's the amplifier, or the indicator, faults in both? [Marc] And it worked when it was powered by half the voltage. [Ken] It worked beautifully! [Marc] Yeah! And I wanted to make it work correctly and I had the smoke event again. So that's the second smoke event for that indicator. [Ken] Yeah, more voltage was not more spinning.
[Marc] Okay. So, yes, that's the yaw motor, right? That's the one that behaved bizarrely before. [Ken] Right. [Marc] Okay, so you want me to... So, we know we have a bad transistor,
and a bad resistor, but we're not sure what's the cause or the effect. [Ken] Yeah the transistor shorted, but that could be because it got too much current. [Marc] Yeah, because of something else. Okay.
One week later... [Marc] So, it's a day after, actually a week after, and we have repaired the amplifier. So, you can point to the new resistor? [Ken] We have a new resistor here! [Marc] And also a transistor from my collection. And that repaired it.
But before we did that, we made sure we understand why it blew up. And it appears that there was a short in the slip ring? Another one. [Ken] Yes, there was a short between the 115 volts that drives the motor, and the control line from the transistor to the motor. And so, that fed 115 volts into the resistor and transistor with dramatic effect.
[Marc] And you were able to find... actually we have the schematics now of the thing, right? [Ken] Yeah, Eric made an awesome schematic here. [Marc] So yikes, there was a second short in the slip ring, in addition to the one that had caused our first hot spot and smoke release. To track this one down, it was Eric's turn to go into super reverse-engineering mode, and after many hours, he produced this superb and complete schematic of the wiring of the ARU-11. The short was located,
and in retrospect, it explains some weirdness we had seen before. That's why we saw the yaw axis move all on its own before we even applied voltage to its control winding. Mystery explained. Better even, Eric and Ken found a way to rewire the slip ring to bypass the fault.
[Ken] It turned out there were two separate slip rings that provided 115 volts. So we could just bypass the bad one and use the good one for both 115 volt. [Marc] So we just disconnected the 115 that was going through the slip ring to that particular motor... [Ken] ...and then rewired the motor to go to the good slip ring. [Marc] Right. At which point, it started to kind of work. Except,
the pitch axis started to misbehave! So, we repaired - that was the yaw?... [Ken] Right. [Marc] ...that was burned out. And then the pitch axis started to just continuously spin. And then, we traced it to yet another short, in the slip ring. So that's our third slip ring short? A partial short, right? A 400 ohm short, not a full short. [Ken] But 400 ohms at 115 volts is plenty to vaporize a resistor.
[Marc] Right, but in this case, it didn't vaporize the resistor, it leaked onto the tachometer. [Ken] Yeah, this one was at 18 volts. So it was much lower voltage, just enough to mess up the tachometer signal and confuse the control, and send it off spinning in a random direction. [Marc] Right. So, we got it kind of working by disconnecting... So Eric
did that, he disconnected tachometer on the... [Ken] That's this loose wire here. Eric disconnected this wire here from the slip ring. [Marc] Which is the tach. [Ken] So, without the tachometer feedback, the motor will you run too fast rather than slowing down as it nears the zero point.
[Marc] Right, it got undamped, but it worked. The servo loop worked again. [Ken] Just not as well as you'd like. [Marc] Yeah. [Ken] So my next plan is to reconnect the tachometer, and disconnect the 18 volt side. Because 18 volts is only used for the pitch trim, which we don't have a knob for anyway. So it's not functional.
[Marc] Okay, so that's the last hurray on this one. But the good thing, is that the amplifier works, and is safe. So we can use it on one of the better units that we have. This was our try out unit, we knew it was bad. [Ken] And now we're ready to do an FDAI. [Marc] (laughter) Yeah and then we do the Apollo one.
We are almost ready to try the good one, but before that, I wanted to demo our rescued one, now mostly repaired and free of shorts. So, I put it all back together. [Marc] All right. I hope I didn't undo the wires, and then we stick the amplifier at the back. All right, and it should be ready for its final demo. All right, time to see if I can demo it. I need this, and... power! Uhhuh! So, this one is pitch.
This one is roll. And this one is yaw, or direction. Tumbling capsule! I don't think that thing is well balanced either, so I think we certainly have an issue there.
Of course, you can do two at the same time, and your spacecraft is in distress. Ooh, upside down... Right side up...
So, yay! Our left for dead unit is back to life, and our servo electronics are repaired. We now felt confident enough to try the repaired amplifier on the good ARU-11 unit that had been loaned to us by channel Patron R.J. Gritter. Thanks R.J.! Several days later... [Marc] So, why I was not looking, the boys have been doing some good work.
[Eric] Oh yeah! [Marc] And both of you had repaired the bad one, right, that was our trial one. [Eric] Yeah, that was terribly painful to work on, but it had short circuits in the slip rings. And it was just... [Ken] Missing parts! [Eric] ...horrible to work on. [Marc] Yeah. But eventually, we got it to work. And then we could repair the amplifier.
[Eric] And we got the servo amplifier working, which was a very important goal! [Marc] All right. So, therefore, we now put it on the good unit... [Ken] And it just works! [Marc] ...with certainty. Oh it's too easy! Let's move the other one out of the way.
So, this one, you just hooked it up? And since you have solved all the problem, now it works? [Ken] Yeah, so watch the off flag there. [Marc] Okay. Oh, wow! And this one has the needles. [Eric] Yes. [Ken] So we wired up two of the many needles and flags. [Marc] Mhm. [Ken] And we have our roll axis.
[Eric] Oh sorry, I moved one of the other axis. [Marc] That's okay, it proves it works! [Ken] Azimuth, and our pitch. [Marc] Yeah! Why is it that we always work on stuff that's broken? It's so much easier when the stuff actually works! [Ken] And the pitch trim knob gives you a little bit of pitch adjustment here. [Marc] Oh, explain the pitch trim knob! This gave us all kind of problems.
[Ken] So the idea is, when you're flying your plane level, it may not be pointed exactly level. So you can adjust this knob, so it's level when you're flying level. And then, the fancy feature of this indicator is, if you turn your jet to go straight up... [Eric] Pretty far up. [Ken] ...it then eliminates the effect of the pitch trim. [Marc] So, the pitch trim is only put in gradually... [Ken] ...in level flight. And so when you're going up at 90°, you know it's really 90°, not 90-ish.
[Eric] They have a special four terminal potentiometer inside there just for that pitch trim mixing. [Ken] And a patent. [Eric] And patent! [Marc] And that's the one actually, I think they had taken that potentiometer meter out of our the gray device, right? [Ken] Yeah.
[Marc] Which caused us all kind of problems. [Ken] So when you're going straight up... [Marc] ...there's nothing. Yeah, put it put it back on the... [Eric] Let me try adjusting it from this angle. See? [Marc] Yeah, nothing. And then, put it back horizontal.
And now, you adjust it, so even though you have a few degrees nose up, so it flies, it shows zero on the horizon. [Ken] Exactly. [Eric] Flying straight, you're not gaining or losing altitude. [Marc] Excellent! Yeah, those are not used often. Most of the time it's just two dimensional,
and then the direction indication is on another instrument. [Ken] Yeah, they got a compass. This way you've got everything you need in one place. [Marc] This is way cooler. [Eric] I agree! [Marc] This is way cooler. Guys, make it spin several axes at the same time. Slowly, like we're flying somewhere.
We need pitch, we need to we need to make a loop. It's such a cool instrument. And on the Apollo one, they had like, I think, four more needles. They had the rates, and they had some kind of navigation error thingy. [Ken] And as soon as someone sends us an Apollo FDAI, we'll wire it up and watch it go.
[Marc] Yeah, that's the idea. That we would get familiar, well, step one, the grey instrument that's broken. Step two, a good instrument, that's not an Apollo one. And now that we feel more comfortable with it, I think we're ready to do the Apollo FDAI.
Very very cool instrument though. I should have one in my car! Thank you guys. Awesome.
2025-03-29 16:58
