Photos of Kearfott Ring Laser Gyro-Based Inertial Reference Units

Note: Most links open in a single new tab or window depending on your browser's settings.

These are photos of Kearfott navigational Inertial Reference Units (IRUs) or parts of them that use HeNe laser-based Ring Laser Gyroscopes (RLGs). Only two samples of one model at present.

The terms Inertial Reference Unit (IRU) and Inertial Measurement Unit (IMU) may be used interchangeably. The Kearfott Website uses IMU; the label on the actual device has IRU. Go figure. ;-)

Kearfott Inertial Reference Unit K600A374-01

This is a complete system, probably intended for "Sea Navigation" - e.g. in submarines, larger UAVs, and the like that have degrees of freedom similar to those of an aircraft. But for ships, 2 of the 3 axes would be largely wasted, except perhaps in tsunami. ;-) Exactly what the difference is compared to "Land Nav" systems is probably more in the interfaces and other capabilities than the specific hardware and functions. The primary physical diffference appears to be of the paint color in the Kearfott brochures: Land Navs are black; Sea Navs are biege. I suppose that black doesn't show dust and dirt like a light color would. ;-)

           

Kearfott IRU: Exterior (left), Interior Labeled (center-left), Inertial Sensor Assembly Platform (ISAP) and its Major Components (MRLG PCB, MRLG Block, and Accelerometers, center-right), MRLG Block Larger than Life (right)

The rectangular portion of the IRU case is 10-15/16" (L) x 6-7/8" (W) x 7" (H). Including the extensions at the ends with the mounting holes and connectors, L increases to 12-1/2 inches. Its weight is around 14 pounds. For reference relating to what follows, the dimensions of the MRLG block itself - the 'heart' - are less than 2-1/2" across x 2-1/2" tall and it weighs under 8 ounces.

Note that most of the photos here and in the Web Albums below have the MRLG Block inverted compared to how it is mounted within the cylindrical ISAP housing in the IRU. It's more photogenic that way and the photons don't care. But for this reason, some of the terminology like the designation for the "Base" of the ISAP Housing (which is really at the top) may be a bit confusing. Sorry. ;-)

For much more information on this IMU, go to Kearfott K600A374-01 IMU with Monolithic Block Triple RLG.

Kearfott Inertial Reference Unit K600A374-01 #1

(The photos in the first Web album are courtesy of Ismael Tremblay.)

See Kearfott Inertial Measurement Unit Assembly K600A374-01 #1 (Web Album, 25 photos).

As noted above, the RLUs in this IRU are interesting because all three axes are fabricated in a very compact single block of special low expansion glass or similar material. The accelerometers are totally separate. At present not much else is known about the overall IRU other than that the RLGs do power up, at least briefly.

The last pic shows an ideal application for this IRU. ;-)

Kearfott Inertial Reference Unit K600A374-01 #2

But first, here are some short videos of what happens with the RLG Block when the IRU is powered. Click on the small image to see the corresponding video.

                   

Overall Setup for Powered ISAP RLG Block Photos (left) and Closeups

These videos show the naked RLG Block from IRU #1 tethered to the power supply and electronics from IRU #2. Minor rearrangement and camera angle variations similar to the next three are used for all the powered photos in the Web Album, below. The last one has the RLG Block oriented in approximate agreement with the schematic derived from the Kearfott patent. However, even with the thing in-hand and powered, it's virtually impossible to figure out the discharge and beam paths due to the contoured exterior of the block.

The clicks near the beginning and end of the videos are from the outlet strip power switch - on and off. Around the mid-point of the video, the RLG Block glows brightly three times in succession accompanied by a continuous audible whine (likely from the dither PZT). This behavior is 100 percent repeatable and IRUs #1 and #2 do exactly the same thing, so it's very likely a feature, not a bug. The assumption is that there needs to be some handshaking with whatever the IRU is connected to for it to remain active. The indicator LED does stay lit but that is the only remaining sign of life. So these videos get rather boring after almost no time at all.....

And now for the good stuff. ;-)

See Kearfott Inertial Measurement Unit Assembly K600A374-01 #2 (Web Album, at least 148 photos so far).

Description of the Web Album photos for #2:

• Overall views (Front1, Right1, Front_Right1, Right1, Back_Right1, Back1, Back_Left1, Left1, Front_Left1): These are virtually identical in appaerance to the sample on the Kearfott Website in the year 2025 under "Sea Nav IMUs".

• Connectors1 and Connector1-5: These are the panel with the five military-style circular connectors and closeups:
  • J1 (66 pins): Control / Signal / Information.
  • J2 (15 pins): Power input. 28 VDC nominal, 30 W max. 24 VDC OK. +Vin: Pins A,B; -Vin: Pins M,N. Case GND: Pin P. No other pins used.
  • J3 (13 pins): ????
  • J4 (22 pins): ????
  • J5 (13 pins): ????

  (J3 and J5 have the same pin configuration but are keyed differently.)

• Label1: This has both the numbers as described above. This unit is SER 0009

• Indicator1, Lit1. amd Indicator2: This is also visible in the lower right corner of the Label photo. Originally I thought it was some sort of fancy optical sensor because it does NOT look like a photodiode or LED. But the mystery has been mostly solved after powering the unit multiple times for the photo shoots and actually paying attention to what it does, which would appear to simply be a high priced mil-spec LED in a hermetic package no doubt costing the U.S. Taxpayer BIG BUCKS. It turns on bright green a couple seconds after power is applied to the IRU. Whether it is simply a fancy Power LED or something more sophisticated is not known.

  And in the "well that's interesting department", SER 0003 - which is identical in every other respect to this one (SER 0009) including all the labeling - has what looks like a green dome over the pricey hermetic LED as shown in the 3rd pic from SER 0003. Perhaps it fell off of SER 0009. Or it was a cost saving of 0.00001 cent. ;-)

• Caution_Label1: Like we didn't know. ;-)

• Inside_Top1: This is what greets you after removing the cover. It's a very clean layout with only 3 or 4 major subassemblies (depending on how they are counted): Main PCB Assembly (top of photo, mostly hidden), "Inertial Sensor Assembly, Platform" (ISAP, the round can on elastomer mounts, lower left) - the heart with the ring laser gyro block and accelerometers; and the power input, conditioning, and interface modules (lower right).

• Main_PCB_in_Situ1: What's visible is the top PCB; a similar size board is mounted on the other side of the aluminum frame.

• Connector_PCB_in_Situ1: At least for some of the connectors.

• Top_Main_PCB1: Some LSI parts but mostly smaller ICs. The black polka-dot rectangular affair at the right is the high voltage power supply for the RLGs and possibly the dither PZT.

• Bottom_Main_PCB1: More chips including a Texas Instruments TMS320 DSP. I would guess that the black painted-over 32 pin packages contain the firmware. ;-)

• Power_Connector_Module_Inside1: This is what is behind J2. It shows the very simple but almost impossible to follow wiring due to the use of all white insulation. ;-)

• Power_Filters1: These appear to be RFI-type filters for the input power. (Look up their Martek part numbers!) Also visible is the Power Connector Module.

• DC Power_Converter1: This is another readily identifiable Martek module. It is a DC-DC converter that takes 28 VDC (nominal) and provides +5 VDC and ±15 VDC for all the electronics. The spec is actually 14 to 40 V in, so fortunately it will run happily off 24 VDC, which is a lot more common for powering on the bench with a fixed supply. ;-)

• Interface_PCB1: Function of this PCB is unknown but it has a bunch of opto-isolators on it so it must have something to do with interacing in a nasty electrical environment.

• RLG_HV_Connector1: This is really just a white wire with high voltage insulation and a female contact secured with a black plastic cap for the HeNe lasers inside the RLG block and possibly the dither PZT. It has approximately -600 VDC on it with respect to the case starting a few seconds after the IRU is powered. There is enough high quality capacitance somewhere such that sufficient charge may still be present hours after powering down to result in a mildly shocking experience. ;-)

• ISAP_in_Situ1: Another view.

• ISAP_with_Mounts1: Front, Right Side, Back, Left Side, Top, Top_Label, Bottom. The upper cable is for the RLG, the lower cable is for the accelerometers. The +HV socket for the RLG is also visible.

• ISAP_Label: And ISAP S/N 670. ;-)

• Inertial_Measurement_Unit_Labeled1: And an interior view with major assemblies annotated. ;-)

• ISAP_Housing_Mounts_Shroud1: Exposing the cylinder with the RLG block and accelerometers.

• ISAP_Shroud1: The purpose of the Shroud is not entirely obvious as it is not structural. But it is made of a thin soft ferrous material which is perhaps Mu-metal, and thus may be a serving as a magnetic shield. It is conceivable that even weak magnetic fields could affect the behavior of the lasers in the RLG at the margins of precision, even resulting in Zeeman splitting. If that isn't its function, its purpose must be to make disassembly more annoying. ;( ;-)

• ISAP_Shroud_Label1: The wording of the label on the shroud is strange (and probably unrelated to the shroud itself): "KEARFOTT CORPORATION IS THE INITIAL TRANSFERER OF THIS PRODUCT, WHICH CONTAINS THORIUM LICENSED UNDER 10 CFR 40.13(c)(i)(ii)". It's even stranger looking up the definition of 10 CFR 40.13. ;-) 10 CFR 40 is part of the "Code of Federal Regulations" dealing with nuclear energy. ;-) Part 13 states that the company is exempt from regulations because there is very little of the stuff present. Thorium is a weak Alpha emitter so a trace amount may have been placed inside the RLG block to assist starting of the HeNe lasers. Any Alpha particles would be totally blocked by the block (no pun intended intentionally) so no need for your Gieger counter. ;-)

• ISAP_RLG_PCB_Top1 and Close1: The RLG monolithic block is below this PCB. The wires in the 3 ribbon cables (5, 6, 7 wires each) and 3 sets of individual orange, yellow, and green wires need to be carefully unsoldered to remove the PCB. That's a total of 27 wires and they are a lot thinner and closer together than they appear in the photos. In fact everything in the "can" (which is around 8 cm or just over 3-1/8 inches in diameter) is actually much smaller than it appears in the photos. ;( ;-) Detaching (and re-attaching) the wires requires a steady hand, super fine tip soldering iron, and Mark II upgraded eyeballs (or a good magnifier or microscope).

• ISAP_Accelerometers_with_PCB1 and Close1: The three accelerometer units are attached to a metal frame and wired to their respective PCBs.

• ISAP_Accelerometer_Interface_ICs1 and Close1: A fancy part for each one. ;-) This may be custom as the only hits via a Web search return info on an older Kearfott IMU.

• ISAP_RLG_Block_in_Situ1: The visible portion of the RLG block exposed when the accelerometer assembly is removed. Three of the six mirrors are visible with their PZTs for cavity length control. The purpose of the what appears to be Home Depot Fiberglas foam is not known. It's very soft and fluffy and not likely to provide much damping or thermal isolation.

• ISAP_RLG_Block_Insulation1: After removal. I hope the specific locations of each clump were recorded. ;-)

• ISAP_RLG_Block_in_Situ2: And with the foam removed. When the RLG lasers are powered there is around -500 VDC on the central pillar as well as the small test-point at the 1'oclock position, which is actually connected directly to it. Go figure. ;-) However, this is NOT the common cathode for the 6 discharges, but separated from it by a significant gap within the block which probably serves as a HeNe gas reservoir. So there is a voltage difference of around 100 V (for a total of -600 V) on the actual cathode at the other end of the RLG block and a faint glow can be seen in the space between them. Assuming the material for both parts of the pillar is the same, either could really be used for the cathode connection.

  I wanted to take photos of RLG Block powered in situ but for some reason, neither the RLG discharges or PZT drive will come on with the cover in place even if not attached. There is just a brief faint "weep" and it then apparently aborts. ;( ;-)

• ISAP_RLG_Block_Cover1: Removing this is more than a matter of three screws. There are 27 teeny wires that need to be carefully unsoldered. And everything is must smaller and more closely spaced than it appears in the photos.

  The overall dimensions of the cover cylinder are 3-1/4 inches in diameter (not including the small protrusions) and 5-3/4 inches tall. The RLG Block itself is around 2-1/2 inches at its maximum width. So this is really so much smaller than it appears in the photos and videos! ;-)

• ISAP_RLG_Block_Top1-2: Two views almost straight down. Note that when inside the IRU, the MRLG block is actually inverted from how it is shown here and in most of the subsequent photos. The base is at the top of the housing.

• ISAP_RLG_Block_Down1-5: Six Views at roughly a 45 degree angle.

• ISAP_RLG_Block_Side1-6: Six views at roughly a 15 degree angle.

• ISAP_RLG_Block_Close1-5: Five closeup views of the interior attempting to show details of the pillar area, but the block gets in the way distorting everything within it.

  Two of these shots show what appear to be coils of magnet wire near the bottom, specific purpose unknown, though a guess would be that they are part of the pulsed starter for the six HeNe laser discharges. With a supply voltage of 600 V and "tube" discharge voltage of 500 V, the current would be around 0.33 mA though the 300K ohm ballast resistor for each discharge path, which is consistent with measurements made on other similar RLGs. There is probably no actual current regulator, only a transistor or MOSFET switch to disable the discharges when the RLGs are not running since the 600 VDC supply remains on constantly.

• ISAP_RLG_Block_Not_Powered, Powered, and Dark_Powered: Several sets of photos from various orientations and distances showing the RLG Block in room lighting, with the discharge powered, and the same in the dark. However, being tethered to the HV and RLG cables severely limits options for creative photo composition. ;-)

  The first set of three shots shows the overall setup for taking these photos with the naked ISAP RLG Block next to its clothed buddy inside the IMU.

  Since this IRU is not connected to anything except power, it does not stay on for more than a few seconds. This is not thought to be a fault since unit #1 does the same thing. So it's a feature, not a bug. At least that is the assumption: Its microbrain expects some sort of hand-shaking or acknowledgement and quits when this does not occur. Further, the RLG Block is only powered three (3) times for a second or so and then shuts off until the next power cycle. So the photos of the RLG Block with the discharge glowing required applying power to the IMU with the camera already focused and its shutter mode set to "Repeat". ;-)

• ISAP_RLG_Block_Central_Pillar1: This is in two parts - The bottom section is the common cathode for the HeNe discharge paths with a gap between it and the top section. The bottom section receives approximately -600 V directly from the HV power supply but there is only around -500 V with respect to the IMU case on both the top of the top section of the pillar itself and the little test-point at the 1 o'clock position when the lasers are lit. (The test-point is electrically connected to the pillar even though it looks like it is separate, strange!) So, ~100 V is lost somewhere, likely due to the ionization voltage of the gap. What is not known is whether there is an additional positive voltage applied via the anode circuit for regulation. It isn't obvious where that goes on the RLG PCB. It may just be GND.

• ISAP_RLG_Block_Ballast1: The blue resistors standing up (each 100K ohms) are the main ballast for 2 of the 6 HeNe discharges. Their common point is connected to the other two pairs. The blue 10M ohm resistor laying down along with the a HV diode are part of the starter which pulses the metal ribbons visible in most views of the RLG block to initiate the discharges via the thin white wire. It is near the ballast resistors but has no coonnection to or association with them.

• ISAP_RLG_Block_Tip-off1: This is where the air gets sucked out and the HeNe gas mixture gets sucked in. ;-) Unlike some other RLG designs, it is NOT a cathode. The glob covering it is primarily to protect fragile humans as the metal is very sharp after it is pinched off with several tons of force forming a cold weld. The large disk ia probably frit-sealed to the RLG block to achieve permanent gas integrity.

• ISAP_RLG_Block_Mirror+PZT1-3: These are closeups of the three mirrors that are dithered for cavity length control. These shots are in no particular order.

• ISAP_RLG_Block_Mirror+Detector1-3: These are closeups of the three mirrors which have the interferometers combining the clockwise and counter-clockwise beams and detectors for each of the three axes. They are in no particular order.

• ISAP_RLG_Block_Window1: This shows the clear more-or-less planar port hole showing the glow in the space between the cathode and discharge paths.

• ISAP_RLG_Block_Anodes1: This shows two of the 6 anodes. Note how the snode pin extends almost into the discharge path. Like the tip-off, these are also probably frit-sealed.

• ISAP_RLG_Block_Schematic1: This diagram is derived from the Kearfott patent and shows the arrangement of the mirrors and beam paths inside the RLG Block. The leg length of the discharge paths in the RLG Block is between 1.5 and 1.75 inches - it's difficult to measure precisely or even if the geometry is a precise square. With 4 legs, the result is a total ring cavity length of between 6 and 7 inches for an enclosed area of between 2.25 and 3.0625 square inches. Therefore, the sensitivity is higher than that of the Honeywell GG1320, which has a two inch leg length but only 3 legs for a 6 inch ring cavity length with an enclosed area of 1.73 square inches..

• ISAP_RLG_Block_Oriented_Similar_to_Schematic1: The photo was rotated so that the locations of the mirrors are very approximately like those in the schematic. More or less. ;-)

• ISAP_RLG_Block_Ring_Paths_Marked1: Each ring cavity is color-coded to more clearly show how it is embedded and oriented in the RLG block. Each of them is a tilted square that surrounds the central pillar. It is very difficult to visualize the relationships in the actual device in part due to the distortion caused by the block, perspective when viewing the actual block in any photo of it, and the (incorrect) assumption that the ring paths must be close to the outer surfaces. This rendering may have errors and it has nothing to with AI! ;-)

The next series of photos is of an orphan ISAP with RLG. Apparently, it had be forceably removed from its IRU and mounts against its will, thus the cut wires and mangled edge of the RLG PCB.

• ISAP_Housing_Hatchet_Job1: This is how it appeared just after removing the Mu-Metal shroud and the remains of some cable clamps. The wires cut, or more accurately hacked off. Poor thing.

• ISAP_Housing_Unclothed1: Another similar view but standing up and hiding the poor cables. ;-)

• ISAP_Housing_Top1: and ISAP_Housing_Top2: are pretty much identical to the unit above. And the SNs of the IRUs differ by only 3 or 4.

• ISAP_Housing_Bottom1: For some reason the label on this one differs from the other. It should be the same internally.

• ISAP_RLG_PCB_Top_Close2: The is virtually identical to ISAP_RLG_PCB_Top_Close1, above, except possibly for some wire colors.

• ISAP_RLG_Block_in_Situ3: Can you spot the difference between this one and ISAP_RLG_Block_in_Situ2, above? ;-)

• ISAP_RLG_PCB_Removed_Top1: The RLG PCB had the three ribbon cables and 3 sets of discrete wires un-soldered so it can be completely removed.

• ISAP_RLG_PCB_Removed_Bottom1: The 2 black cylinders are transformers. One is a pulse transformer for starting the HeNe discharges but its exact specifications are not known as it appears to be a special part with the Kearfott 88818 CAGE code. The other is a standard commercial part - a PICO S21670 power transformer spec'd at 115 V input, 12.8 VCT output, and a frequency of 400 Hz. Its 6.4 V output windings are connected in parallel here. It is used in reverse for driving the PZTs.

• ISAP_RLG_Block_PCB1: This is the cylinder base with the 3 dither PZTs attached to the struts. The green, blue, and violet wires are for the detectors; the detached discrete colored wires are for the PZTs; and the HV cable goes directly to the center post, which is attached to the common cathode for all 6 discharge paths.

• ISAP_Housing_Top_Cover_Top1-2, Bottom1-2: The housing covers including, apparently, their part numbers, perhaps.

• ISAP_Empty_Housing Top1, Top2, Bottom1, and Bottom2: These show the empty shell. ;-)

• ISAP_RLG_Block_in_Housing_Base1:: This is similar to ISAP_RLG_Block_Down3, above.

• ISAP_Base_Top1-2, Bottom1, and Side1: The HV connector with its cable attaches using a stud and nut to the center post and cathode of all 6 discharge paths. The 3 PZT actuators are glued to the 3 struts via the 3 sets of orange, yellow, and green wires.

• ISAP_Base_PZT_Close1: Shows details of the actual PZT with its 3 electrodes. The large one is probably for drive with the others for feedback, or something. ;-)

• ISAP_RLG_Block_Removed_Top1: View from above without housing base.

• ISAP_RLG_Block_Removed_Side1-6: These views of the naked ;-) RLG block are similar to those in ISAP_RLG_Block_Side1-6 except for the lack of the housing base.

• ISAP_RLG_Block_Removed_Bottom1-4: The 6 thick bare wire leads are the 6 100K ohm ballast resistors, which are all shorted together on this PCB. The 3 sets of green, blue, and violet wires go to the 3 detectors. The poor abused ribbon cables ;-) include the positive anode voltage (which may simply be ground) and PZT signals for the mirrors to fine tune or dither cavity length, among others as yet to be identified.

• ISAP_RLG_Block_First_Light1: After removing the base, this is the first time any of the discharge paths have been lit from an external HeNe laser power supply. Electrically, all 6 discharge paths are in parallel, only isolated by 100K ohm ballast resistors. So if driven by a constant current power supply like one for a normal (single) HeNe laser, the first discharge to strike will pull down the voltage on all the others preventing them from starting. And indeed only a single discharge path is lit in the photo with a current of around 0.4 mA. In the IRU, the controller applies a pulse via a HV diode in parallel with a 10M ohm resistor to the metallic strips visible in most of the RLG photos. So it's like a xenon flashlamp trigger.

• ISAP_RLG_PCB_PZT_Transformer1: This is a standard power transformer made by PICO with specs of 115 V input to dual 6.4 V outputs at 400 Hz. It is used in reverse to provide for the high frequency drive to the PZTs. The yellow and green wires to all 3 PZTs are connected directly across its primary (now secondary) winding. The orange wires are connected together but go somewhere else, probably some type of sense signal.

• ISAP_RLG_PCB_Start_Transformer1, ISAP_RLG_PCB_Start_Post1, and ISAP_RLG_Block_Removed_Trigger1: These provide the starting pulse to initiate all 6 discharge paths. The black cylinder is the trigger transformer mounted on the underside of the RLG PCB (see photo above). The 7 conductor ribbon cable including trigger wire passes through the slot on its way to the RLG Block PCB. The 10M ohm resistor and HV diode laying down connect between the trigger wire soldered to the post on the RLG PCB and the white wire which runs to the copper strips (trigger / start electrodes) glued to the RLG block in proximity to the 6 discharge paths.

• Kearfott-K600A374-01_ISAP_RLG_Block+External_Power_Supply1: This shows the RLG block on an acrylic stand running off a custom stand-alone power supply constructed on a solderless breadboard. The input is 12 VDC which is boosted to 600 to 1,200 V by the inverter transformer (center of circuit). The micro-switch in the lower left discharges a capacitor into the trigger transformer (right side of circuit).

• Kearfott-K600A374-01_ISAP_RLG_Block_External_Power1-6: Closeups of the RLG block on external power.

• Kearfott-K600A374-01_ISAP_RLG_Block+External_Power_Supply2: The power supply has been transferred to a Perf. board to be more permanent. I was considering designing a PCB for the power supply but the market would be quite small unless Kearfott was interested in using the design for testing. Then it would be a total of 3. ;-)

• Kearfott-K600A374-01_ISAP_RLG_Block_Size_Comparison1-2: To appreciate how small the entire RLG block really is, these show a bog standard 2 inch Acrylic cube and bog standard regulation size baseball beside it. In fact, the entire block itself would fit comfortably inside the baseball with room to spare, though the tips of the "attachments" might poke through the cover. ;-)