Home Models Docs Forums Plans Sources Contact
A Barr & Stroud FT 27 Coincidence Rangefinder, c1915

Rangetaking with a Barr and Stroud FT 27

rangefindersI am lucky to own several short baselength rangefinders from the World War I era. From top to bottom: a Barr and Stroud FT 27 in fine working order, a Barr and Stroud FT 27 (inverted upper image) with muddled optics, a Bausch and Lomb 1 meter rangefinder (non-working)

All these devices were suitable for naval use, though the Barr and Strouds were also apparently employed in infantry roles. The Barr and Strouds are fitted with different receivers which appear to be for tripod mounting (probably for land use) and feature fold-out handles for unmounted operation, though this is rather clumsy. Barr and Stroud is a Scottish company which provided the bulk of the rangefinders for the Royal Navy.

The Bausch and Lomb seems to have received extensive naval use, being heavily painted-over in naval gray. Brass rings on this device seem to be for mounting in a heavier fixed shipboard mounting. The painted-over tube would make cracking this open to attempt to restore its utility problematic. I'm open to ideas!

These are short base length instruments, and while this fact makes them relatively handy, it restricts their accurate use to ranges in the 250 to 2000 yard envelope. In land service, this would have proven helpful in setting a sniper's rifle sights or in placing a mortar's first round more accurately.  In naval use, I would think these could only have proven useful for station-keeping or to support torpedo aiming calculations, but not for gunnery at ranges where hitting was apt to be truly problematic.

Calibrating the Rangefinder

I did all my testing with my best rangefinder:  the FT 27 with the clearer optics and with the upright upper image.  To set up an FT 27 for use, one must first calibrate it.  
The first step in calibration is to set the "halving adjustment".  This is the means by which the pitch of the prisms is adjusted to ensure that no part of the scene is visible in both top and bottom halves of the combined image, and that no part of the scene is left out of both halves.  That is, the two halves must perfectly cut the scene in two.  If some part of the scene is visible in both halves, this is called an error of duplication.  If there is a gap between the halves in which part of the scene is lost, this is termed an error of deficiency.  A small thumb knob usually hidden under a sliding panel is used to adjust the halving.

Once the halving is correct, the second step is to set the device for infinite convergence.  This adjustment ensures that the range scale reads "infinity" when the device has perfectly halved an object that can be considered to lie at infinite distance.  By the book, this is done by ranging on the moon.  I decided to approximate this by ranging on a flagpole I knew to be at 875 yards from my vantage point and to adjust not for infinite distance but 875 yards.  The nature of the instrument makes these two steps essentially equivalent.

Testing the Rangefinder

To make this more fun, I came up with the concept of the "Rangetaker Challenge".  Visitors to Cambridge are routinely hauled out to experiment with the RF by taking range cuts on various landmarks at known distances.  The goal for each rangetaker, of course, is to obtain ranges that are nearest to the known ranges.

The FT 27 can converge on targets at ranges from 250 yards out to infinity.  To really appreciate the capabilities of the device, one needs a vantage point with a relatively clear view to readily identifiable landmarks at least 250 yards away.  Ideally, the vantage point should oversee a vista offering a variety of landmarks at ranges extending well beyond 250 yards.  For me, living in Cambridge, Massachusetts, a convenient choice is to sit on the bridge over the Charles River running between Harvard Square and Harvard Stadium.   From this perch, the view along the river is unobstructed, and Boston offers a wealth of distant landmarks well over a mile away.

I have learned that it is easiest, for purposes of carriage and not alarming the public, to transport the rangefinder in a soft guitar gig bag.  The rangefinder bears a strong resemblance to a bazooka or RPG, and so it is always with some trepidation that I wield it openly.  The best tack is to ensure that your expression and body language help convey that you're not out to blow anything up, and to answer any questions asked by passersby.  I suppose that a police officer could panic and shoot at us.  However, they use handguns -- hardly the sort of serious firepower that rangetakers usually have distracting them from their duties.

I selected several prominent features for use as ranging targets and used Google Earth's ruler tool to obtain a fairly accurate range to each from the vantage point for use as a point of comparison by which our results might be judged.  This method was not completely accurate, but must be trusted to have been pretty darn good.  Factors that render its accuracy a little moot are that we are not stringent about standing in a given location on our bridge, and tend to roam a bit to obtain the clearest view.

The experimental method is fairly simple.  The guest rangetaker is directed to range on each target in turn.  I'll check his or her "cut" for the first one or two, to ensure that they understand what was being asked of them.  And then, just as in the Royal Navy, their results are duly recorded in a special Happy Alligator Spiral Notepad.

The closest target is the nearest point of Harvard Stadium.  Google Earth indicates the range is 345 yards.  The stadium is the ideal first target due to its prominent vertical, which makes it easy to align the half images.

The second and third targets are found to the east along the Charles:  the Weeks footbridge and a tower beyond it.  The Weeks is a beautiful brick bridge arcing over the rowers (finally... some shelling!) and is a tricky object, as it lacks any ready verticals, and is not perpendicular to our line of sight.  I opted to make the actual object the rightmost of 2 cement medallions on its footings, a range Google Earth indicates is 430 yards.  The tower, on the other hand, is deceptive.  It appears to be an ideal ranging target, with a clear vertical, but close examination of the geometry of our sightline reveals that we are looking almost directly down one of its sides, and the rangetaker runs the risk of ranging alternately on the near (700 yards) or the far (725 yards) corner of the side he is looking almost perfectly along. I just decided to call its range 715 yards, but we should be suspicious of this target. One the plus side, the eastward view to these objects (indeed, all our objects except the stadium) offers the rangetaker the luxury of steadying the rangefinder on the structure of our own bridge.

The fourth target is a dorm on the south side of the river with irregular tinted windows.  It is a large target in depth for its range, and so we select a particular feature of it as our ranging target -- the rightmost of 2 vents or stovepipes.  This creates some possible ambiguity for us, however, as it is not certain that this stovepipe's location on the roof can be identified in Google Earth.   A first look offered a guess of 615 yards, but its disagreement with all ranges taken prompted me to go back and analyse its apparent position from our perch as being just to the right of the near corner of the dorm, and so we judged that it must actually be near the far side of the roof, placing it perhaps as far as 625 yards away.   Making an executive decision, I decreed its range of record to be 620 yards.

The last target is the truly challenging one... the Prudential Tower in Boston.  Its massive antenna provided a nice vertical, but at such a distance it is very hard to clearly see something as subtle as this piece of metal, particularly against a cloudy gray sky.  At a range of 4485 yards, the results we obtain when ranging on it are abysmal.  Their contrast to the much higher accuracy obtained against nearer objects seems to provide good indication of a short base rangefinder's limited utility at anything but very short ranges.  Even though 4500 yards is not an extreme range for WW-I gunnery by any means, a photograph of my friend Albert ranging on the mighty "HMS Prudential" shows how it is lost in a cowlick.


To date, I have walked just a few visitors through the Rangetaker Challenge.

Date weather Guest Rangetaker bio notes
all Tony Lovell gadfly, fairly good vision in right eye
20070611 bright with light overcast Dr. Reed Riddle astronomer, wears glasses
20070620 gloomy dark clouds, light rain Albert Xthona high end computer graphics, very sharp eyesight
20070805 bright sun Doug, Kim, Garrett and Kelsie Hiscano a dangerous family, but all results recorded of this session were lost due to a failure to bring the Happy Alligator Spiral Notepad.
20070809 sunny, high clouds Chris Buja marketing executive
20070809 sunny, high clouds Spencer Buja menacing teen
20070921light hazeChris Pennytriple mocha
20081025cloudy, light rainKossuUber-Finn
20081025cloudy, light rainPossummusician
20090812overcastSimon HarleyTrouble from Leeds
20110327cold, clearTim LanzendorferDeutschlander!
20110327cold, clearByron AngelNaval historian
20110709partly cloudy, windNick JellicoeBorn 90 years lateRF was reading long

Harvard Stadium -- Google Earth Range 345 yards

Rangetaker Date Cut
reed 20070611 363 +18 +5.2
albie 20070620 352 +7 +2.0
buja 20070809 345 0 0
spencer 20070809 340 -5 -1.4
penny 20070921 349 +4 +1.1
simon 20090812 338 -7 -2.0
tim 20110327 348, 343 +3, -2 +0.1
byron 20110327 351, 352 +6, +7 +1.9
jellicoe 20110709 357 +12 +3.5

Weeks Footbridge -- Google Earth Range 430 yards

Rangetaker Date Cut
reed 20070611 460 +30 +7.0
albie 20070620 422 -8 -1.9
buja 20070809 441 +11 +2.6
spencer 20070809 419 -11 -2.6
penny 20070921 418 -12 -2.8
simon 20090812 418 -12 -2.8
tim 20110327 448 +18 +4.2
byron 20110327 433 +3 +0.7
jellicoe 20110709 441 +11 +2.6

Tower Beyond Weeks -- Google Earth Range 700 (or 700-7301) yards

Rangetaker Date Cut
reed 200706111 711 -4 -0.56
tone 200706111 715 0 0
albie 200706201 718 +3 +0.42
albie 200706201 732 +17 +2.4
buja 20070809 700 0 0
spencer 20070809 713 +13 +1.8
penny 20070921 665 -35 -5.0
simon 20090812 716 +16 +2.3
tim 20110327 685, 679 -15, -21 -2.6
byron 20110327 687, 692 -13, -8 -1.5
jellicoe 20110709 717 +17 +2.4

1.  Prior to the failed Hiscano rangetaking day, we aligned on the right edge of this tower.  However, I soon realized that the variability of the readings were due to the fact that we were looking almost directly down that edge of this structure, and some rangetakers might be measuring the near corner (700 yards) and others the far corner (730 yards).  We have since standardized our ranges by doing a cut on the central column of windows, which assures that all ranges should correspond to 700 yards.  As a consequence, I consider the true range for sessions prior to 20070809 to be 715 yards, and 700 yards thereafter.

Stovepipe on Dorm -- Google Earth Range 620 yards

Rangetaker Date Cut
reed 20070611 619 -1 -0.16
tone 20070611 641 +21 +3.4
albie 20070620 630 +10 +1.6
tone 20070620 633 +13 +2.1
buja 20070809 630 +10 +1.6
spencer 20070809 639 +19 +3.2
penny 20070921 619 -1 -0.2
simon 20090812 618 -2 -0.3
tim 20110327 600, 616 -20, -4 -1.9
byron 20110327 630, 621 +10, +1 +0.9
jellicoe 20110709 632 +12 +1.9

Prudential Tower -- Google Earth Range 4485 yards

Rangetaker Date Cut
reed 20070611 3675 -810 -18
tone 20070611 5800 +1315 +29
albie 20070620 4740 +255 +5.7
tone 20070620 5600 +1115 +25
buja 20070809 5900 +1415 +32
spencer 20070809 9100 +4615 +100
penny 20070921 4525 +40 +0.9
simon 20090812 not seen
tim 20110327 4750, 4580 +265, +95 +7.9
byron 20110327 3975, 3700 -510, -785 -14.4
jellicoe 20110709 4800 +315 +7.0


The FT 27, in our hands, does a fair job at ranging on objects under 1000 yards.  At some point beyond that, however, things quickly go far astray.
As of the end of June 2007, with 16 observations taken, we see no errors larger than 7% on targets 1-4, but only a single observation on the Prudential Tower falls anywhere near this standard of accuracy.  

There seems to be a slight miscalibration of the device, in that the observations seem to tend to be higher than the ranges taken by Google Earth, though I have not done an analysis on this.

I am puzzled by the relatively large errors on the stadium, but as this objective is along the axis of the bridge and we did indeed move about along this axis, an undue amount of experimental error might be embodied in our actual vantage point versus that eyeballed on Google Earth.

I regret that I have no photos taken through the rangefinder's eyepiece.  I have not mastered the technique required to do that well.  However, I can offer the following subjective impression of the view, and on the factors of your stance and support that impact the process for better or worse.  The most vital element to stress is that vibration and rotation of the instrument degrade your perception of the clarity of the view quite a bit.  Those cuts where we took advantage of the bridge's wall as a support for the rangefinder were the easiest to take, by far.  It is also important to stress the fine angles involved in ranging on the distant Prudential Tower.  If you range on it twice, you may obtain ranges that are over a thousand yards different.  The reason for this is obvious if you think about the geometry, or just look at the range scale on the instrument.  The markings on the range scale are irregular and become increasingly fine as the ranges increase.  This makes sense, as infinity is one of the readings on the scale, and the mechanism can move the prisms through a sufficiently broad range of angles that divergence is actually possible.  Think for the moment of the angles that are involved.

The FT 27's baselength of 0.8m equates to about 0.875 yards.  If you are converging on a target that is 400 yards away, the proper angle of convergence which achieves this harmony between the half images is atan(0.875 / 400), or 0.1253 degrees (0 degrees would purely parallel vision -- as when ranging on the moon or other effectively infinite ranges).  If your target moved 10 yards further away to 410 yards, the required prism angle of convergence would alter to 0.1222 -- a change of just 1/300th of a degree.  

Now, if you imagine a target further off at 4500 yards, the prism angle would be 0.0111 degrees.  If the operator moved the range cut knob the same distance as in the 400->410 yard example above, the range scale would not go to 4510 yards or even to  4612.5 yards (that's 4500 * 410/400).... it would zoom up to 6267 yards!  The same adjustment that corresponded to a 2.5% change in range now would change the reading by 39%.  Things get more outrageous as the ranges approach those of WW-I gunnery:  a 10,000 yard range would increase to 25,230 yards.  

The fact is plain... the angles of adjustment of the prisms and the angles the operator's eyesight must permit him to align are very fine.  Magnification can help, but this diminishes the amount of light available, and a balance must be sought.  I can readily imagine how rangetaking efforts go right down the tube when
  • the target has smoke all around it
  • the target is 15,000 yards away
  • you are surrounded by sea spray throw up by wind and enemy shells landing nearby 
  • you are surrounded by smoke from your own vessel's gunfire and smokestacks
  • you are continually rattled by the vibration of your ship's engines
  • your ship is rolling, pitching and heaving on the waves
  • you are startled and jolted by the shock of your guns firing
  • you are frightened by the fact or the prospect of enemy shells hitting your ship
It is my own conviction that rangefinders were a fine theoretical instrument that rarely lived up to their promise in action in this period, and indeed, seldom thereafter.  A greater reliance had to placed upon spotting the fall of your own shells to obtain a working understanding of the enemies range and its derivatives -- and that art also had its very real challenges.