159: Finding U-Boats


The threat that the U-Boats posed was well known after the First World War, so how did the British plan to protect their trade from them?



Hello everyone and welcome to History of the Second World War Episode 159 - The Early War at Seat Pt. 9 - Finding U-Boats. This week a big thank you goes out to David, Tyler, Jan, Moritz, Trevor, and Alejandro for choosing to support the podcast by becoming members, you can find out more over at historyofthesecondworldwar.com/members. Over the last few episodes we have discussed the actions of the Admiral Graf Spee, which along with the Deustchland occupied many Royal Navy ships as they tried to protect British trade from the German surface raiders. There was of course another threat to British trade that the Royal Navy was also prepared for: U-boats. During the First World War the impact of German U-boats on British shipping had been a shock and it took time before the proper changes were made to combat the German submarines. For example convoys were not introduced until relatively late in the war. Then during the interwar years time would be spent by the Royal Navy preparing for the possibility of a future war in which submarines were a major factor. A core part of the plans that were developed was around the asdic technology, which we would refer to today as sonar. This was a very powerful tool that allowed surface ships to detect and firmly locate submarines, but there were key challenges when it came to making it a powerful tool in naval combat. It had short range, was susceptible to disruption, and provided low fidelity information. All of these problems would be solved in time, even into the modern day, but in 1939 they were serious challenges to making asdic an effective part of an anti-submarine campaign. Unfortunately for British seamen, the challenges were not well understood and so it was used ineffectively for a long time. I really like this quote from Britain’s Anti-Submarine Capability 1919-1939 by George Franklin to describe this problem: “The fundamental British weakness came not from a failure to develop A/S methods and equipment, but from a widespread misunderstanding of what had and had not been achieved. The mistaken belief held in certain quarters that asdic enabled a dived sub to be hunted like a surface ship led to an overreliance on patrol as a tactic.” Fortunately for British trade, the Germans were having problems of their own, particularly around just getting enough U-boats into the trade lanes on a consistent basis. They simply did not have enough U-boats, and the attrition enacted on them by even marginally successful Royal Navy tactics was enough to cut that number even further. This was true even though they had started their naval build up way back in 1935 with the signing of the Anglo-German Naval Agreement, an agreement that the British felt was a necessary evil. The next four episodes will be dedicated to the U-boat campaigns of 1939 and early 1940, but this episode will focus on the efforts of Britain to develop the technology and tactics to counteract the threat from submarines, and the German efforts to keep their communications secret via the Enigma encryption process, which would be a critical part of the naval war until it was defeated later in the war.

Britain’s preparations before the war for combatting the threat posed by German U-boats were a mix of good decisions, mistakes, and unexpected challenges. In the good decisions category the first place to start is simply that there was a good amount of time and effort spent on preparing the technology and manpower to detect and combat submarines. On the technology there was asdic, a known way to detect submarines so that they could be attacked and either damaged, sunk, or simply driven away from convoys. The goal was to eventually equip all destroyers with asdic, and then to build up a fleet of smaller trawlers and other dedicated anti-submarine ships that were simple to produce but could still provide results. These anti-submarine ships, either destroyers or smaller, would then be used to guard convoys of merchant ships as they moved to and from British and French ports. Manpower was slightly more difficult to handle. There were three major things working against building up enough qualified asdic operators and officers well versed in anti-submarine tactics: it was not a desired specialization, it was not part of mandatory training for officers, and it was also a new specialty. Starting with the first one: anti-submarine work was simply not a sexy branch of the Royal Navy, there were big battleships, new and interesting carriers, cruisers that roamed the world’s oceans, fleet destroyers prepared for their torpedo attacks….and then there was a small fleet of old, slow, unsexy ships dedicated to anti-submarine efforts. The second problem was around mandatory training. Within the Royal Navy the training regimen of young officers involved courses and instruction in many of the major areas of naval warfare, as it was felt that it was good for officers to have at least a basic understanding of many aspects of the ships they were serving on, even if they were later asked to specialize. Anti-submarine work was not one of those mandatory training topics, which limited the broad knowledge but also made it more difficult to convince officers to commit to it as a specialty when the time came. This training problem was eventually changed in the late 1920s, but the impacts of the early 1920s could still be felt when the war started. This was due to the fact that anti-submarine work was a relatively new specialty branch of the Royal Navy, which severely limited the number of retired sailors that could be reactivated in case of war. These retired sailors were a critical part of many navy’s plans to mobilize at the beginning of a conflict, as they would bring with them experience and skills that were often hard to find. Very few of these existed for the anti-submarie service, because it was simply too recently created. The final area of preparations that deserves some attention is around unexpected challenges. The first of these was mentioned earlier, asdic just was not as effective as hoped and would experience challenges as it was rolled out on the scope that was required for the war. This was exacerbated by the demands placed on the Royal Navy in the years immediately before the war. Other episodes discussed the remarkable build up of Royal Navy strength in the years immediately before the war, but money was still limited and when it came to discussions of increasing funding for anti-submarine efforts or ensuring that projects like the King George V battleships were completed, the larger projects often got priority, a tendency that would drastically change as soon as the war started. All of the challenges were then made worse by the events of the war, especially after the fall of France. The usage of French Atlantic ports completely changed the scope of the submarine war in the Atlantic, suddenly the U-boats could range further, spend longer on station, and could more rapidly deploy to intercept convoys. This forced not just more protection of convoys, but also forced that protection to be extended further out into the Atlantic which greatly strained the available resources of the Royal Navy.

Regardless of any challenges, the faith of the Royal Navy in asdic was absolute, with a 1936 memo from the Admiralty stating that due to the presence of asdic on British ships “the submarine should never again be able to present us with the problem we were faced in 1917.” The origin of the word asdic seems to be disputed, but it was a relatively simple device which had a transducer that protruded from the underwater hull of the ship and which was protected by a dome. The control officer could then point the transducer in a direction, send electrical current through the device which would send a sound pulse out from the ship. The reflections of that sound pulse was then analyzed as a way of determining if there was a submarine that it had hit. The challenge with the prewar asdic setups is that they generally good a good amount of time to do sweeps around the ship. This was primarily due to the fact that the detection arc was small, only around 5 degrees, and they required that before moving to the next 5 degree section all of the sound returns had to occur. This meant that it could take several minutes to do a full sweep around the ship. Another challenge was around detection range, the outer edge of the range of an asdic system was often just a few thousand meters, although it would improve during the war. And even this range was a major improvement over what it had been during the early interwar years. But the challenges were not just around limited range, but also a minimum range. This dead zone was only around 100 meters around the ship, but even at that distance the submarine could still react to the presence and attack of a destroyers or trawler. This meant that it was important to try and not be detected by the submarines hydrophone when lining up for an attack, and then to launch that attack quickly. Then once the attack was launched, it generally took time for asdic to become functional again because a system that relies on sound waves loses most of its effectiveness when there are underwater explosions. Even with all of these challenges though, it is important to state that asdic completely changed the dynamic between submarines and surface ships. Before its introduction submarines often felt invincible, and after its introduction and widespread use it was critical for U-boat captains to properly plan and react to it. A major way that this was done was through surfaced night attacks, because asdic was incapable of detecting surfaced U-boats, and radar was in its infancy at the start of the war, although again this would rapidly change. In September 1939 radar was not powerful enough, or precise enough to really be able to find and pinpoint the location of small surface contacts, there was just too much noise that could not be properly filtered. This did not prevent the Admiralty from believing that radar, either based on ships or on aircraft could be a powerful anti-submarine tool, which they were absolutely correct on, it would just take time for radar to live up to its potential. One area of technology that had not seen much evolution between the wars was around anti-submarine weapons, and in 1939 the Royal Navy would enter the war primarily relying on the simple depth charge. It was an unsophisticated weapon, a bunch of explosives timed to go off at a certain depth, and it did function, just not spectacularly.

Given the various technology and the expected threats there were many plans for how to counter the threat posed by German U-boats. On the completely defensive side of preparations there were mines, mines were the hidden and silent killers during the war, and they could be very effective against submarines, and they had in fact sunk more U-boats during the First World War than any other weapon. They were also not sufficient, because of course mines are good at protecting areas but they are completely defensive weapons. On the complete opposite side of the spectrum were the planned hunter-killer groups that would be setup in wartime where an aircraft carrier and then a collection of destroyers would be sent on active patrols in expected areas of U-boat concentration. The theory was that the combination of airborne assets and asdic equipped destroyers would be a potent way of finding and then sinking submarines, but instead they would turn into a debacle with there being multiple instances where U-boats would be able to attack, damage, and even sink the carriers. The aircraft carrier based hunting groups would be disbanded within the first year of the war. Another tool that was to be used was land based aircraft of coastal command, which were solid observation platforms, but there was little training provided to the pilots and observers and just the difficulty of spotting submarines from the air was discounted. The Admiralty also planned to dedicate most of the British submarine fleet to anti-submarine operations. Finally, the most well known anti-submarine tactic, and one that had been used late in the First World War was one of the most simple, the convoy system. The official plans were always to enact the process of convoys, plans that were solidified and practiced during exercises during the 1920s and 1930s, but that did not mean that the convoy system was not without critics. The common complaint about convoys, at least from naval leaders, is that they forced the navy to dedicate a large number of resources to a purely defensive function. When ships were grouped into convoys they had to be protected by destroyers and other anti-submarine vessels, but all of these ships were tied to the merchant ships and were simply reacting to German actions. This was deeply unsatisfactory for some naval leaders who wanted naval resources spent on more proactive offensive operations. But these critics were generally not able to change the mind of convoy advocates, and really the people who supported convoys at any cost were almost certainly correct. This was true even though at the start of the war there were not enough anti-submarine platforms to provide the needed protection for all of the convoys that were running. At the very least it forced the German U-boats to funnel into the areas that were protected, which was even more important when there were not enough escort vessels to go around. When it came to protecting convoys the best setup was for ships to work in pairs to detect and attack submarines because it allowed one ship to attack while the other was dedicated just to keeping contact with the submarine, which was always a challenge due to the asdic minimum range and the interference of depth charges. The escorts would be situated several thousand meters from the convoy in an attempt to provide a perimeter of defense, with the perimeter biased towards the forward movement of the convoy, especially for fast convoys which were faster than the submarines that were hunting them. If a U-boat was detected in front of the convoy the merchant ships might simply turn away, and if this was successful the escorts might not even be detached to attack it, as it was often felt that it was more important to keep the escorts with the merchant ships to prevent another U-boat from being able to take advantage. If an attack was ordered, ideally two escort vessels would move in to attack with one of the vessels dropping depth charges in a centered square pattern. This was not always possible due to a lack of escort vessels though, especially in the opening months of the war. During this time the entire escort for a convoy might just be two escorts, and so a proper attack against the U-boat would completely strip the entire convoy of protection. This was balanced against the struggles of the German navy to get enough U-boats, and other complicating factors like a lack of experience, the very restrictive orders that U-boats operated under early in the war, and then the efforts made by British intelligence against German naval radio transmissions.


A critical part of the war at sea, in all theaters, would be around signals intelligence. There would be many instances where the ability of one side or the other to intercept and decypher the other’s communications would have a drastic impact on naval actions, with the most famous probably being the Battle of Midway. But even without the ability to read the contents of radio messages, there were other things that could be learned just from the existence of those radio communications. One of these was the location of ships, with radio direction finding being an important part of naval intelligence gathering dating back to the First World War. But of course both sides knew that the other side wanted to try and read their radio transmissions, and so there was an arms race to keep ahead of the ability of the enemy to crack codes. Enter the most famous encryption device of the Second World War, the Enigma machine, honestly it might be the most famous encryption mechanism in all of history. It its core the Enigma machine was built around a set of initially 3 rotors, each with 26 contact points. Then these rotors would spin as a message was entered, with the first rotor spinning 1/26th of the way around, and it would spin 26 times, then on the 27th both the first and second rotor would spin 1 time, then after the first rotor moved 26 times, the second rotor would spin 1/26th again, then after the second rotor had spun all the way around, the third rotor would spin one time. Every single time that the rotors moved, the electrical connections between them changed. This would cause lights on the machine to light up, with one light for each letter so that when the operator hit say an X the light that lit up with an F, and F would be the letter used when communicating the message via radio. Then on the receiving in, when the person input F and X would light up on their machine. The key on both sides was making sure that the machines were set up exactly the same because as the rotors spun the letters were constantly translating into different letters, so maybe the first time the letting F represented X, but then the next time X was used it might be represented by a B. And it could be made even more complicated through the use of a reflector, which essentially double encoded each letter, which effectively turned 3 rotors into 6. As with any encryption that is built around character translation, the most important piece of the puzzle when it came to the security of the encryption was around the number of permutations. The math gets very large very quickly here, but essentially there were 3 rotors, there would be more later, but at the beginning we start with 26 x 26 x 26 is 17576 possible combinations for the three rotors. Then each of the rotors could be moved to any position in the three, and since each rotor had different electrical connections the order was important, with three rotors that resulted in 6 possible combinations which means we have to take 17576 and multiply it by 6, 105,456. 105,000 is a lot, but it probably would have been manageable, and would have resulted in it being cracked quite quickly, most likely, but there was one more complication. Each of the 26 rotors could be started at any position, and each position that any of the rotors started resulted in a different output. This takes the number of combinations up above the normal numbers that we use in our normal lives, and where you start hearing phrases like million millions or thousand trillions. And there were also easy ways that these numbers could be increased even more, the easiest being always having more than 3 rotors, and then choosing 3, which multiplied the number of options by the number of available rotors because it was impossible to know which ones were being used if you were not a German radio operator. It was believed, and would be believed for a very long time during the war that the Enigma was unbreakable, in a time before computers the combinations were simply too vast. The Enigma system was honestly incredible, and it was also very very flexible, because the encryption could be completely altered simply by having a different rotor. This is how the different groups within Germany would handle encryption, and each group could simply have different rotors with different internal electrical connections and it completely changed the resulting values. This allowed each military branch, and the government, to all have different and mutually exclusive encryption, increasing the security of all. After building it up like this, the story of Enigma is that it was eventually compromised, and so how did that happen. Well it all started with message protocol. Rotor settings and starting positions were always set up with some kind of pre arranged system, some sort of menu, or timed based setups, those types of things. But then when a message needed to be sent it would always begin with a three letter sequence which was repeated two times, with it being repeated in case of bad radio conditions. The theory was that these three characters would always be random, but the humans actually using the machines quickly fell into patterns, just like modern people do with passwords, there were a lot of AAAs and ABCs or three characters in a row or in a column. It also became known that the first three characters repeated, so this meant that the 1st and 4th, 2nd and 5th, and 3rd and 6th characters were always the same, which is the exact type of information that starts people down the path of cracking these things. Another challenge that would always be present in military signals traffic is around the fact that there are many many similar phrases and words used in military communications, divisions, regiments, objectives, etc. when something was repeated often the possibility of it being decoded was much greater. These were problems, but they did not make enigma messages impossible to read all of the time. In fact before the war in January 1938 Polish cryptographers were actually able to decode most German enigma messages even with a relatively small team of people actually working on the project. Unfortunately there were two primary changes made to enigma protocol before the war that would take Enigma back to being unreadable. The first was the change to give every Enigma machine possible rotors, and then to change them out on some kind of pre-arranged schedule, this really just greatly increased the number of possibilities. The second change was around where the starting positions of the rotors were set, up to September 1938 the rotors on the machine were set via pre-arranged settings for the message, and instead they were told to simply pick the settings at random and then transmit them in the clear at the start of the message before beginning the actual encrypted transmission. This made it much much more difficult to decode messages because it changed the encryption every single time, instead of on a periodic basis. Even with these challenges processes were put in place that still allowed the Poles to reach the Enigma messages, although it became much more difficult, the first was an automated machine that used sets of 18 total rotors at one time, and a total of 108 total rotors, and basically just allowed the user to brute force the encryption. Another way that was developed at this time was through the use of large perforated sheets with horizontal and vertical columns of letters that could be laid on top of one another as a decryption tool. The problem was that there were 156 separate sheets of characters needed and each sheet had around 1,000 total holes in it, this allowed for a three rotor Enigma setup to be decrypted, but it required a lot of manual work and preparation, but the introduction of 5 total possible rotors defeated the sheet method, at least for the time, because it just increased the possibilities too much. And that is basically where things stood at the start of the war, the general theory of how to crack some of the earlier Enigma setups was sound in theory, but it was simply impossible to directly translate those methods into the more complex Enigma codes that were being used when the war started. Eventually the British would be able to adapt the sheet stacking method introduced by the Poles to begin to read the Luftwaffe transmissions of early 1940, but the naval enigma codes, the most important for these particular episodes continued to allude them. Part of this was due to the different nature of the naval signals traffic, and part of that was simply that there was far less of it than among the other arms of the German military. Eventually even the naval Enigma would be cracked, but not until August 1941 and after countless hours of work by British code breakers, a story for another day.