2001’s AE-35 Unit
The fictional AE-35 unit was a small but pivotal plot device in the film 2001: a Space Odyssey.
It was said to be a critical component of the Discovery One spacecraft’s antenna steering system; essential for keeping the communications array aligned to the distant Earth.
A publicity still from the film, featuring actors Gary Lockwood and Keir Dullea.
In real life the AE-35 unit was an off-the-shelf aviation component – a Sperry gyro magnetic compass – with a number of cosmetic alterations added by the filmmakers. And here is everything you ever wanted to know about it.
The gyro compass
The specific product used in the film was the Sperry G.M. Compass Mark 4F gyro unit type B, made in England by the Sperry Gyroscope Co. Ltd. Sperry was an American manufacturer of gyroscopic compasses, which had a British subsidiary based in Brentford and Bracknell, England.
The device should not be confused with a true gyro compass. These are electromechanical contraptions which use a spinning gyroscopic wheel to determine true north – they don’t rely on the Earth’s magnetic field to determine magnetic north. Such gyro compasses are quite large, and were commonly used in the last century for large ship navigation. The “G.M. Compass” used in the movie wasn’t one of these, and was probably named the way it was to sound cooler.
A Sperry device with its cover off, revealing the complex and precise gyro mechanism seen in the film.
It seems that the device was actually what we now call a direction indicator (DI in the UK) or heading indicator (HI in the US). These are aircraft navigational instruments containing gyroscopes.
Historically, magnetic compasses were the primary means of aircraft navigation prior to the rise of GPS. However, a magnetic compass can be thrown off by certain conditions, ranging from the Earth’s magnetic field at a given location to the movement of the aircraft itself. To compensate, planes carry gyro-backed direction/heading indicators to tell the pilot the correct course. Such indicators have to be reset to magnetic north regularly during flight unless they have external detectors for magnetic flux, the way the Sperry device did.
X-ray Delta One, this is Mission Control.
Roger your two-zero-one-three.
Gyros in space
More advanced gyroscopic devices are widely used aboard satellites and other spacecraft. Such machines use inertial navigation systems for control and navigation, and these include gyroscopic devices amongst other things. For example, the Apollo spacecraft used inertial measurement units (IMUs), containing three gyroscopes, to take astronauts to the Moon.
The Apollo IMU. Sperry manufactured the accelerometers in this device.
Therefore the inclusion of a gyro instrument aboard 2001’s fictional Discovery One makes a certain sort of sense. A gyroscopic device would be a key part of the communication control system, correctly maintaining the ship antenna’s azimuth and elevation (hence AE). The slightest drift or deviation would mean loss of communication with Earth; an obviously catastrophic situation.
Of course, a 1940s-era magnetic gyro compass wouldn’t have any value at all on a spaceship. For one thing it relies on the Earth’s magnetic field, it contains a single gyroscope when really you’d want at least three but ideally four (to avoid Apollo’s dreaded gyro lock), and it does nothing whatsoever to tell you your velocity and position. But it’s a nice prop approximation that looks awesome – especially in the X-ray images.
Nowadays even the bread loaf-sized single-gyro device seen in the film has been vastly eclipsed technologically. Modern smartphones contain non-rotating gyroscopic sensors capable of detecting the phone’s movement in space. They use MEMS (micro-electro-mechanical systems) solid state sensors to perform their magic. It’s kind of crazy that video games that use sensors that users can wave through the air rely upon these incredibly complex silicon sensors that people just take for granted!
The screen-seen device
We have a few views of the AE-35 prop in the film, most notably the scene when astronaut Dave Bowman tests the device using some sort of advanced diagnostics tool. From these shots, and publicity photos taken that day, it has been possible to confirm the exact model used in the film.
HAL’s eye of the test. This cropped screenshot is a bit blurred and distorted because it was shot using a fisheye lens.
However, there are a number of unlabelled sub-variants of this model. They’re all pretty similar, but the gyro compass had an internal body frame built of two separate metal castings, both painted orange. The specific model seen in the film has:
three low flat round posts at the top of each triangular gusset on the side,
two horizontal bars with slightly rounded edges at the outer edge, and flat where a metal plate is located (the plate was removed for the film)
Some Sperry gyro compasses have the triangular gussets but lack the low posts, and some have flat horizontal bars on the rear casting. (ie: they have the screen-seen casting on one side, and a different one on the other) There doesn’t seem to be any obvious way to tell which unit you have by looking at the markings on the outside of the unit. You have to open it up and examine it directly.
Kubrick, Lockwood, and Dullea taking a break during the filming of the AE-35 test sequence. Note the kind of hood behind the test bench “computer” monitor. This is part of the assembly used to project 16mm film loops up onto the screen, simulating a high-resolution flat-panel video display. You can also see the simple black outer case from the AE-35 prop, to the right of the gyroscope itself.
However these differences between sub-versions are extremely minor and difficult to notice unless you sit down and carefully compare the actual gyrocompass to a high-resolution screenshot of the film.
Sorry you fellas are having a bit of trouble.
We are reviewing telemetric information in
our mission simulator and will advise.
Discovering the compass
How was this thing discovered? Like many prop replica processes, it was multi-stage, involving different people.
At some point in the early 2000s it appears that American collector Dennis Gilliam had correctly identified the basic gyro compass. He posted a web page summarizing his work, and included a low-resolution photograph of the Sperry gyro compass that he had acquired. However, he did not provide any information to make it possible for anyone else to track down and buy the item, other than it was a “gyro”. Note that Gilliam’s found unit differs very slightly from the one in the film.
In 2020, Replica Prop Forum member “Abitofcredit” of the UK posted a few articles describing his findings and research, identifying the gyro as a Sperry unit. He used screenshots from the film, including the-often ignored X-ray images and wireframe shots of the purported AE-35 unit that are seen in the actual film. The assumption was these X-rays and wireframes were based around the real gyro device, which turned out to be correct.
Following this post, user “joberg” of Quebec, Canada, posted photos of his scratch-built replica AE-35 unit, pointing out that the wireframe images included elements added to the prop that did not exist on the actual gyro unit.
At the end of 2020, RPF user “Birdie” posted high-resolution screenshots of the AE-35 type prop that was painted red and used in the TV show Space: 1999. Close examination of these photos reveals that the TV prop was almost certainly the original 2001 prop, repainted. This is because it contains certain details that were added to the prop and which don’t exist on the actual gyro devices.
The probable AE-35 box seen in Space: 1999.
On 29 April 2020, 2001 effects man Douglas Trumbull was interviewed by author Michael Benson for the MOMI museum in New York. This virtual video interview included a slideshow in which Trumbull showed a photograph of an original gyro compass. The slide has the title “Sperry Gyro Compass”, and was clearly taken after production of the film was over since there’s a computer keyboard visible in the background. Note that the unit in the photo has flat side bars, and thus differs very slightly from the one seen in the film. Also the flat grey plate with the chamfered corners seen in this shot was removed for the film version.
In May 2021 “macropod80” of Australia posted some details of his work. He located a UK seller of used aerospace components, and found that the seller had numerous gyro units of various types available for sale. macropod80 also highlighted the differences between subvariants of the device and, working with the seller, eventually found one that was identical to the screen-used version. He has since embarked on a project to replicate the custom modifications made to the gyro unit.
Modifications
The AE-35 unit seen in the film had a number of modifications made to it. Essentially the stock outer casing was removed, and a set of custom-made and machined pieces were added to it. These included front, back, and bottom plates, mounting rails, truncated cone rings around the rear panel sockets, and a custom-designed orange-painted top plate printed with white text and legends for the testing scene.
The 4K release of the film has been invaluable for gleaning details of the AE-35 unit prop.
So to turn one of these gyros into an AE-35 you need to:
Remove and discard the outer metal case. This is the sheet metal case that’s painted black and has the memorable admonition DO NOT JAR HANDLE LIKE EGGS on it.
Unscrew and remove the pale grey side metal plate and its associated components.
Add extra square front and back metal plates. The front one will cover up the compass display.
Paint these plates semi-gloss black. I don't think they had wrinkle paint, like the housing did.
Add a rectangular bottom plate in thick aluminium. This has angled dovetailed aluminium bars screwed to it, that would match the base of the test bench in the pod bay set. Building a replica test bench is an exercise left to the reader.
Add two internal aluminium bars with machined-out screw notches for the screws that keep the new casing in place.
Replace the large socket on the back plate with a small one, resulting in two small Plessey mark 4 sockets.
Add conical aluminium rings around each socket. Remove threaded ring around each plug first.
Add an engraved plate beneath the two sockets. I don’t think this is the original Sperry plate, though it's close in terms of size and shape. The one on-set photo we have is kind of low-rez, but suggests maybe four lines of text and no rivets. The Sperry plate has six lines of text in two typesizes, and is held on with two rivets.
Strip the black paint from the device’s front panel chamfered switch lever, and polish it up.
Replace the two black front panel knobs with machined aluminium custom ones (I can’t find any commercial equivalents still sold, so these would have to be custom-machined or else 3D printed). It’s possible the film-used knobs were the original unit’s knobs, painted silver. I don’t think so, however, because I think the gyro unit knobs have fewer ribs than the knobs in the film, and are parallel-sided, rather than slightly tapered.
Add a D-type stainless steel wire handle to the front plate.
Add a brownish-red plastic card to simulate a circuit board into the side of the device, as seen in the top-down sequence. Assuming your gyrocompass doesn't have one - some seem to have this card on the opposite side.
Add an orange-painted aluminium top plate with bent sides.
Drill suitable holes in the plate for the probe.
Print suitable white legends and logos onto the plate.
Optionally add a sheet metal housing that goes over the whole thing. Simple bent sheet metal painted wrinkle black.
Roger your plan to go EVA and replace
alpha-echo three five unit prior to failure.
The top panel testing plate
This is a tricky one. By examining the 4K scan of the movie we can determine the basic layout of access holes, white lines, and symbols. We can also make out some of the lettering, which were almost certainly dry transfers, laboriously hand-applied. Type is likely uppercase Futura, Letraset’s version. However, most of the alphanumeric text on the top plate is not fully legible. Since no photos are known to exist of the finished prop, we do not have any way of determining this lettering with certainty. We can just guess at the shapes and make up something that seems reasonable.
For example, two of the three words on the top line can’t be read. My version has the text “AE BORESIGHT AOS” since that lines up reasonably closely to what the blobs look like. Plus it kind of makes sense - the term boresight (suggested by RPF user Aahz) is used in reference to antenna alignment by analogy to weapon adjustment, and AOS is a NASA Apollo acronym meaning “acquisition of signal”. The blobs on the bottom block of text, however, are difficult to devise text for.
My version of the top plate text and diagrams.
The oscilloscope probe
The scene in which Dave Bowman tests the supposedly faulty AE-35 unit has him poking a grey-handled probe into a series of tiny holes on the top plate of the unit. Various diagnostic displays then flash up on the screen. In reality, of course, actor Keir Dullea coordinated his movements to a pre-filmed sequence of animations that were projected onto the surface of the table.
The device in his hand is a real object – it’s a Tektronix P600x passive oscilloscope probe. The exact model is not known, because Tektronix produced a number of probes with nearly identical external appearances. It’s most likely a P6006 or P6008 probe, since those models had BNC plugs on the end, and the test bench set in the film has BNC sockets on it. The P6007 or P6009 are less likely, since they have different plug types, but since you don’t see the plug in the film anyway, it really doesn’t matter which one you go for as long as it has a grey tip.
A later-model P6008 Tektronix oscilloscope probe.
Note that these probes were produced for many years. The one seen in the film has a subtle difference from the more commonly available revision of the device. The ribbed or fluted plastic handle (the “locking sleeve” in the Tektronix manual), as seen in the film, has slightly curved or cylindrical ribs. The more common variant seen on auction sites has flat ribs. However, this difference is really minor, and you can only tell that that’s the version used if you examine the 4K scan of the film really closely. Note that some versions of the probe lack the flexible shiny black strain relief seen in the film.
Also, the probe was made with a multitude of tip ends. The one in the film was the simple flat wire version, not one of the hook tips.
We concur with your plan to replace number one
unit to check fault prediction. We should advise
you, however, that our preliminary findings
indicate that your on-board niner-triple-zero
computer is in error predicting the fault.
I say again, in error predicting the fault.
Thanks
This project, like most prop replica expeditions, is not the work of one person. Thanks to “Abitofcredit” for the RPF post containing the valuable information about the gyro, to “macropod80” for finding the UK seller of gyros and his excellent work on the reproduced components, and Lee Stringer for other helpful comments.
And if you find this sort of thing interesting, have a look at my HAL 9000 faceplate research.
Contact
© 2021-23 NK Guy, contact@theageofplastic.info