View Full Version : Whatever happened to the 3d party He-100?

04-06-2005, 01:12 AM
I know 99,9% of you guys couldn't care less about this plane but have a special interest in the Heinkel (and it's story). The only chance I'll ever have of flying it is in simulations like these so... I have seen some great screenshots a while back but I can't find them anymore. I don't think the plane is on any "probable add-ons" list either?

04-06-2005, 01:12 AM
I know 99,9% of you guys couldn't care less about this plane but have a special interest in the Heinkel (and it's story). The only chance I'll ever have of flying it is in simulations like these so... I have seen some great screenshots a while back but I can't find them anymore. I don't think the plane is on any "probable add-ons" list either?

04-06-2005, 02:46 AM
I also like the plane very much. It could have been a better alternative for the Luftwaffe than the 109.

I very much doubt it will be released for IL-2 though... http://forums.ubi.com/groupee_common/emoticons/icon_frown.gif

04-06-2005, 03:01 AM
like 2 minutes before i read this thread i was talking about the Me109 & Me209 & He-100 to a old squadmate on why it wasnt accepted by the RLM & how it was a good thing it wasnt

BTW , if the model was ever accepted , it would have been AI , zero cockpit references . . . . . . apparently


04-06-2005, 03:06 AM

04-06-2005, 03:11 AM

nocockpit references, they didnt look hard enough.

I think its the other way around 99.9% of us would like to see this plane flyable and in this game.

More he100s seen service then the ta152 and i185 combined.

04-06-2005, 05:01 AM
I believe that's a replica isn't it? Not so sure if that cockpit is 100% correct. Still, it would be interesting to take a peek!

04-06-2005, 05:28 AM
Indeed it is a replica built to exacting detail by one skillful fella with the unlikely name of George Lucas. He has built a 109G and a 163 replica as well and I believe he donates them all to the Chino Planes of Fame Museum. The museum, in turn, helps him with his hobby. (They donated a genuine set of main landing gear from a 109 for his He100 build) He used to have a website which is gone now. Amazing projects.... http://forums.ubi.com/images/smilies/25.gif

As for the He100.....fascinating history

On the 30th of March 1939 a prototype of the new Heinkel He 100 fighter design streaked into the record books at 746.6km/h €" the new world absolute speed record. Surprisingly it took the record away from a plane with well over twice the horsepower, and beat it by over 40km/h. Lessons learned from earlier Heinkel projects had been put to good use and its advanced aerodynamics resulted in a plane that was the best performing fighter in the air, even in its less slippery production line model.
Little information on the plane is available, and what there is often contradictory. All we know for sure is that Heinkel built the world's fastest plane, and it was suitable for use as a fighter (unlike many racing planes). We also know that after being built and proving itself in testing, the production line only built twelve planes before shutting down. The rest of the He 100 story is clouded in mystery, which makes it all the more interesting.
Basic specifications
Company: Ernst Heinkel Flugzeugwerke GmbH
Designer: Walter and Siegfried Günther
Year: 1938
Type: Single seat day fighter
Description: Low wing monoplane fighter with conventional control surface layout.
Fuselage: Egg shaped cross section with flattened sides, particularly around the engine. The front of the plane is largely flat horizontally, sloping down sharply behind the spinner. The rear of the plane slopes down toward the rear, starting at a high point at eye level behind the cockpit and ending up roughly level with the bottom of the canopy at the tail.
Wings: The wings are largely rectangular, with rounded tips. The inner portion of the wing is flat on the bottom and then bends up about 1/3rd along the span, but due to the thinning of the wing it appears to have a slight reverse gull-wing bend. The portion inboard of the bend is thicker with parallel leading and trailing edges, outside of the bend the leading edge tapers back slightly, and the trailing edge forward more strongly. Flaps span the area inside of the bend, and ailerons start at the half way point and run to the tips.
Other details: The canopy is similar to the Malcolm hood from later Spitfires, in which a rounded center section slides to the rear over smaller windows set into the fuselage. The windscreen at the front is well rounded and the flat plate in front of the gunsight is small and well faired. Fully retractable tail€"dragger landing gear were used, with the main gear retracting inward towards the fuselage, the wheels laying in the thick inner portion of the wing.
The Heinkel He 100 story starts in 1933 with the Reichsluftsfahrtministerium (Reich Air Ministry, or RLM) competition to produce the first modern fighter for the re-forming Luftwaffe. Four designs were submitted; Arado's Ar 80, Focke€"Wulf's Fw 159, Heinkel's He 112 and the Messerschmitt Bf 109. All four planes were tested competitively in early 1936 with interim engines, and the Ar 80 and Fw 159 were quickly eliminated. Both the 112 and 109 were considered worthy of further testing, and orders were sent out for 15 additional aircraft from both companies.
Although Heinkel was considered the favorite to win the contract, the more modern and better performing 109 won over the Flight Acceptance Commission. By late March of 1936 the 109 was considered the favorite. At that point Heinkel was allowed to redesign the 112, which resulted in the largely all€"new 112B. The 112B was considerably improved and was as good or better than the 109, but the 109 won anyway.
The 112 had a few problems that lost it the competition. The first was that the airframe was rather complex; it included a large number of compound curves and its elliptical wing was labor intensive. The RLM was looking to produced hundreds of planes, so cost in both dollars and manhours was a factor. The prototypes also suffered from a series of accidents, even if they weren't related to problems with the plane they still left a bad taste in the mouth.
But the biggest problem for the 112 was that after learning that Supermarine had started series production of the Spitfire, the Luftwaffe was desperate to get a modern fighter into squadron hands. Heinkel might have won the competition had the B model been available in early 1936, but by the time they were ready in the second half of the year the 109 was already in series production.
Nevertheless some small scale contracts for the plane were finally secured with a variety of air forces in Europe and Japan. Thirty were bought by Japan, but twelve of these were used briefly by the Luftwaffe during the Sudetenland Crisis. Another nineteen were then sold to Spain where they served long careers. Thirty were sold to Romania, they served in combat in 1941 but were quickly worn out. Finally three more B's were sold to Hungary as the vanguard of a license production series that never took place.
By 1939 production of the He 112 ended, and it appeared that Heinkel was out of the fighter business.
Development history
Even by early 1936 the RLM became interested in a new fighter that would leap beyond the performance of the Bf 109 as much as the 109 had over the biplanes it replaced. There was never an official project on the part of the RLM, but Roluf Lucht felt that new designs were important enough to ask both Focke€"Wulf and Heinkel to provide "super€"pursuit" designs for evaluation.
Since the super€"pursuit type was not an official recommendation, it was possible that Heinkel would be told to stop work on the project. Thus the work was kept secret, in a company Memo No.3657 on January 31st this was made clear; "The mockup is to be completed by us... as of the beginning of May... and be ready to present to the RLM... and prior to that no one at the RLMis to know of the existence of the mockup."
Walter Günter €"one of Heinkel's most talented designers€" looked at the 112 and decided that nothing more could be done with it. He started over with a completely new design known as "Projekt 1035". Learning from past mistakes on the 112 project, the design was to be as easy to build as possible while still offering good performance. That good performance was set at an astounding 700km/h (435mph). Keep in mind that fighters with this sort of performance didn't appear on the battlefield until 1944.
To ease production the new design had considerably fewer parts than the 112, and those that remained contained considerably few compound curves. In part count the 100 was made of 969 unique parts and was held together with 11543 rivets, in comparison the 112 had 2885 parts and 26864 rivets. The new straight-edged wing was a source of much of the savings, after building the first wings Otto Butter reported that the reduction in complexity and rivet count (along with the Butter brothers's own explosive rivet system) saved an astonishing 1150 man hours per wing.
In order to get the promised performance out of the plane, the design included a number of drag reducing features. On the simple end was a well€"faired cockpit, the absence of struts and other draggy supports on the tail, and fully retractable gear (including the tailwheel) which were completely enclosed in flight. These and similar changes applied to the 109 for the F model would boost performance of that plane 50km/h. The engine was mounted directly to a strong forward fuselage as opposed to internal struts, so the cowling was very tight fitting and as a result the plane has something of a slab sided appearance. The design used a shorter wing than the 109, trading altitude and turn performance for speed.
In order to provide as much power as possible from the DB 601 engine, the 100 used exhaust ejectors for a small amount of additional thrust. In addition the supercharger inlet was moved from the normal position on the side of the cowling to a location in the leading edge of the left wing, where the clean airflow improved the ram-air effect and increased boost.
For the rest of the designed performance increase, Walter turned to the risky method of cooling the engine via surface evaporation. Inside the engine the fluid is kept under pressure which stops it from boiling even though it's allowed to heat above its normal boiling point, the fluid is then run to cavity with lower pressure where it quickly starts to boil and releases steam. Since steam contains considerably more energy than the same temperature water, if you can remove the steam you can remove a lot of heat. The stream can be cooled by allowing it to condense in a series of pipes inside the plane. With no external openings at all, it's basically a zero-drag cooling system.
On the down side the system is complex and hard to maintain. Worse, it greatly increases the chance of killing the engine in combat due to a "radiator hit" on the now much larger cooling system. Other designs would attempt to use the same sort of design, but invariably returned to conventional radiators due to the complexity. A number of people had already tried the system and given up on it, but Heinkel had good experiences with it on their He 119 high speed bomber project and decided to press ahead.
In the Heinkel system €"designed by Jahn and Jahnke€" the engine was run at 110 Celsius and the superheated fluid was then sprayed into the interior of a centrifugal compressor, allowing the pressure to drop and steam to form. The water, being heavier, was forced to the outside of the pump by centrifugal force and returned to the engine. The weight of the water forced the steam into the only available space, the inside of the pump, where it was removed. The steam was then allowed to flow into a series of tubes running on the inside surface of the leading edges of the wings, where it would condense back into water and be pumped back to the engine. A number of pumping systems were tried, and eventually a system of no less than 22 small electric pumps (all with their own failure indicator lamp in the cockpit) was settled on.
Unlike the cooling fluid, oil cannot be allowed to boil. This presents a particular problem with the DB 601 series of engines, because of a particular design technique that results in a considerable amount of heat being transfered to the oil as opposed to the coolant. To cool the oil a small semi-retractible radiator was fitted under the wing.
This radiator was later replaced on some of the prototypes with a system in which the oil was sent to a heat exchanger where it boiled methyl alcohol to carry away the heat. The alcohol was then cooled in a similar fashion to the engine fluid, by running it to tubes on the top surface of the rear fuselage and leading edge of the vertical stabilizer.
Walter was killed in a car accident on May 25th, 1937, and the design work was taken over by his twin brother Siegfried, who finished the final draft of the design later that year. The wing started out flat and then bent upwards about 1/3rd along the span, and the portions inboard of the bend were thicker to hold the wheels. The gear retracted inward and thus were wide set when opened, resulting in a significant improvement in ground handling over the 109. The rear of the fuselage sloped down to the tail from a point at about eye level at the rear of the cockpit, so while it didn't have the visibility of the 112's bubble, it was still significantly better than the 109. A small retractable radiator was added for running on the ground where the surface cooling system wouldn't work. The plane was small, slightly smaller than the 112 that spawned it, and considerably lighter.
At the end of October the design was submitted to the RLM, complete with details on prototypes, delivery dates, and prices for three planes delivered to the Rechlin test center. At this point the plane was being referred to as the He 113, but the "13" in the name was apparently enough to prompt Ernst Heinkel to ask for it to be changed to the He 100 (even though it had previously been given to Feiseler).
In November Messerschmitt took the speed record for landplanes in a modified 109. In response Ernst Heinkel made plans to use the He 100 design as a record setting plane (less serious plans for this appear to have been in the works all along). Much of the fuselage was as smooth as it could get, so the modifications were limited to the canopy and a newer set of much shorter wings. The racing version would need another airframe, so a fourth prototype was added to the series.
In a December meeting at the Heinkel factory with Ernst Udet and Roluf Lucht the plans were changed slightly. V1 through V3 were to be used for testing and record attempts, V3 sporting the clipped wings. V4 was to a testbed for series production. The RLM went ahead with the plan, due in no small part to Udet's (Generalluftzeugmeister, Minister for Aircraft Production in the RLM) plans to fly the plane in a series of record attempts.
The first prototype He 100 V1 flew on January 22nd, 1938, only a week after it's promised delivery date. The plane proved to be outstandingly fast. However it continued to share a number of problems with the 112, notably a lack of directional stability. In addition the Luftwaffe test pilots disliked the high wing loading, which resulted in landing speeds so great that they often had to use breaks right up to the last 100m of the runway. The ground crews disliked the design too, complaining about the tight cowling which made servicing the engine difficult. But the big problem turned out to be the cooling system, largely to no one's surprise. After a series of test flights V1 was sent to Rechlin in March.
The second prototype addressed the stability problems by changing the vertical stabilizer from a triangular form to a larger and more rectangular form. The oil cooling system continued to be problematic so it was removed and replaced with a small semi-retractible radiator below the wing. It also received the still€"experimental DB 601M engine which the plane was originally designed for. The M version was modified to run on "C3" fuel at 96 octane, which would allow it to run at higher power ratings in the future.
V2 was completed in March, but instead of moving to Rechlin it was kept at the factory for an attempt on the 100km closed-circuit speed record. A course was marked out on the Baltic coast between Wustrow and Müritz, 50km apart, and the attempt was to be made at the plane's best altitude of 18000ft. After some time cleaning out the bugs the record attempt was set to be flown by Captain Herting, who had previously flown the plane serveral times. At this point Ernst Udet showed up and asked to fly V2, after pointing out he had flown the V1 at Rechlin. He took over from Hertingand flew the V2 to a new world 100km closed circuit record on the 5th of June, 1938, at 634.73km/h (394.6mph). Several of the cooling pumps failed on this flight as well, but Udet wasn't sure what the lights meant and simply ignored them.
The record was heavily publicized, but in the press the plane was referred to as the "He 112U". Apparently the "U" stands for "Udet". At the time the 112 was still in production and looking for customers, so this was one way to boost sales of the older design. V2 was then moved to Rechlin for continued testing. Later in October the plane was damaged on landing when the tail wheel didn't extend, and it's unclear if the damage was repaired.
The V3 prototype received the clipped racing wings, which reduced span and area from 30ft 10in and 155sq ft, to 24ft 11in and 118.4sq ft. The canopy was replaced with a much smaller and more rounded version, and all of the bumps and joints were puttied over and sanded down. The plane was equipped with the 601M and flown at the factory.
In August the DB 601R engine arrived from Daimler-Benz and was installed. This version increased the maximum RPM from 2200 to 3000, and added methyl alcohol to the fuel mixture to improve cooling in the supercharger and thus increase boost. As a result the output was boosted to 1776hp, although it required constant maintenance and the fuel had to be drained completely after every flight. The plane was then moved to Warnemünde for the record attempt in September.
On one of the pre-record test flights by the Heinkel chief pilot, Gerhard Nitschke, the main gear failed to extend and ended up stuck half open. Seeing as the plane could not be safely landed it was decided to have Nitschke bail out and let the plane crash in a safe spot on the airfield. Gerhard was injured when he hit the tail on the way out, and made no further record attempts.
V4 was to have been the only "production" prototype and was referred to as the "100B" model (V1 through V3 being "A" models). It was completed in the summer and delivered to Rechlin, so it wasn't available for modification into racing trim when V3 crashed. Although the plane was unarmed it was otherwise a service model with the 601M, and in testing over the summer it proved to be considerably faster than the 109. At sea level the plane could reach 348mph, faster than the 109E's speed at its best altitude! At 6560ft it improved to 379mph, topping out at 416mph at 16400ft before falling again to 398mph at 26250ft. The plane had flown a number of times before its landing gear collapsed while standing on the pad on the 22nd of October. The plane was later rebuilt and flying by March of 1939.
Although V4 was to have been the last of the prototypes in the original plans, production was allowed to continue with a new series of six planes. One of the airframes was selected to replace V3, and as luck would have it V8 was at the "right point" in its construction and was completed out of turn. It first flew on the 1st of December, but this was with a standard DB 601Aa engine. The 601R was then put in the plane on the 8th of January 1939, and moved to a new course at Oranienberg. After several shakedown flights, Hans Dieterle flew to a new record on March 30th, 1939, at 746.6km/h (463.9mph). Once again the plane was referred to as the He 112U in the press. It's unclear when happened to V8 in the end, it may have been used for crash testing.
V5 was completed like V4, and first flew on November 16th. It was later used in a film about V8's record attempt, in order to protect the record breaking plane. At this point a number of changes were made to the design resulting in the "100C" model, and with the exception of V8 the rest of the prototypes were all delivered as the C standard.
V6 was first flown in February 1939, and after some test flights at the factory it was flown to Rechlin on the 25th of April. There it spent most of its time as an engine test-bed. On the 9th of June the gear failed in-flight, but the pilot managed to land the plane with little damage and it was returned to flying condition in six days.
V7 was completed on the 24th of May with a change to the oil cooling system. It was the first to be delivered with armament, consisting of two 20mm MG/FF in the wings and four 7.92mm MG17's arranged around the engine cowling. This made the 100 the most heavily armed fighter of its day. V7 was then flown to Rechlin where the armament was removed and the plane was used for a series of high speed test flights.
V9 was also completed and armed, but was used solely for crash testing and was "tested to destruction". V10 was originally to suffer a similar fate, but instead ended up being given the racing wings and canopy of the V8 and displayed in the German Museum in Munich as the record€"setting "He 112U". It was later destroyed in a bombing attack.
Overheating problems and general failures with the cooling system motors continued to be a problem. Throughout the testing period failures of the pumps ended flights early, although some of the test pilots simply starting ignoring them. In March Kleinemeyer wrote a memo to Ernst Heinkel about the continuing problems, stating that Schw¤rzler had asked to be put on the problem.
Another problem that was never cured during the prototype stage was a rash of landing gear problems. Although the wide-set gear should have eliminated the gear failures that plagued the 109, the 100's were built very thin and as a result they were no improvement. V2, 3, 4, and 6 were all damaged to various degrees due to various gear failures, a full half of the prototypes.
He 100D-0
Throughout the prototype period the various models were given series designations (as noted above), and presented to the RLM as the basis for series production. The Luftwaffe never took them up on the offer. Heinkel had decided to build a total of 25 of the planes one way or the other, so with 10 down there were another 15 of the latest model to go. In keeping with general practice, any series production is started with a limited run of "zero-series" machines, and this resulted in the He 100D-0.
The D-0 was similar to the earlier C models, with a few notable changes. Primary among these was a larger vertical tail in order to finally solve the stability issues. In addition the cockpit and canopy were slightly redesigned, with the pilot sitting high in a large canopy with excellent vision in all directions. The armament was reduced from the C model to one 20mm MG/FF-M in the engine V firing through the propeller spinner, and two 7.92mm MG17's in the wings close to the fuselage.
The three D-0 planes were completed by the summer of 1939 and stayed at the Heinkel Marienehe plant for testing.
He 100D-1
The final evolution of the short He 100 history is the D-1 model. As the name suggests the design was supposed to be very similar to the pre-production D-0's, the main planned change was to enlarge the horizontal stabilizer
But the big change was the eventual abandonment of the surface cooling system, which proved to be too complex and failure prone. Instead an even larger version of the retractable radiator was installed, and this appeared to completely cure the problems. The radiator was inserted in a "plug" below the cockpit, and as a result the wings were widened slightly.
While the plane didn't match it's design goal of 700km/h once it was loaded down with weapons, the larger canopy and the radiator, it was still capable of speeds in the 400mph range. A low drag airframe is good for both speed and range, and as a result the He 100 had a combat radius between 900 and 1000km compared to the 109's 600km. While not in the same league as the later escort fighters, this was at the time a superb range and may have offset the need for the 110 to some degree.
By this point the war was underway, and as the Luftwaffe would not purchase the plane in its current form, the production line was shut down.
Specifications for the He 100D-1c
Engine: 1,175hp (876kW) Daimler-Benz DB 601M liquid€"cooled inverted V12
Dimensions: span 9.42m (30ft 10 3/4in)length 8.20m (26ft 10 3/4in)height 3.60m (11ft 9 3/4 in)
Weights: empty 2070kg (4,563lb)max loaded 2500kg (5,512lb)
Wing Area: 14.5m2 (156ft2)
Wing Loading: 29.25lbs/ft2
Performance: maximum speed 668km/h at 6400m (416mph at 21,000ft)560km/h (348mph) at sea levelcruise speed unknownservice ceiling 11000m (36,090ft)range 900km (559miles)
Armament: one 20mm MG/FF-M firing through the propeller spinnertwo 7.92mm MG17 in the wings
He 100 in service
In 1940 the He 100's were publicized by Goebbels in a propaganda effort aimed at convincing people that a new fighter was entering service with the Luftwaffe. The plan involved taking pictures of the remaining D-1's at different air bases around Germany, each time sporting a new paint job for various fictional fighter groups. The pictures were then published in the press with the He 113 name, sometimes billed as night fighters (rather silly since you could see they didn't even have a landing light).
The plane also appeared in a series of "action shot" photographs in various magazines like Der Alder, including claims that it had proven itself in combat in Denmark and Norway. One source claims that the planes were on loan to the one Luftwaffe staffel in Norway for a time, but this might be a case of the same misinformation working many years later.
It's unclear even today exactly who this effort was intended to impress €"foreign air forces or Germany's public€" but it seems to have been a successful deception. British intelligence featured the plane in AIR 40/237, a report on the Luftwaffe that was completed in 1940. There the top speed was listed as 390mph (interesting that it also states the wing was 167 square feet) and it noted that the plane was in production. Reports of 113's encountered and shot down were listed throughout the early years of the war.
The remaining twelve He 100D-1c fighters were used to form Heinkel's Marienehe factory defense unit, flown by factory test pilots. They replaced the earlier He 112's that were used for the same purpose, and the 112's were later sold off. At this early stage in the war there were no bombers venturing that far into Germany, and it appears that the unit never saw action. The eventual fate of the D-1's remains unknown.
Foreign use
When the war opened in 1939 Heinkel was allowed to look for foreign licensees for the design. Japanese and Soviet delegations visited the Marienehe factory in late October, and were both impressed with what they saw. Thus it was in foreign hands that the 100 finally saw use, although only in terms of adopted design features.
The Soviets were particularly interested in the surface cooling system, and in order to gain experience with it they purchased the six surviving prototypes (V1, V2, V4, V5, V6 and V7). After arriving in the USSR they were passed onto the ZAGI institute for study, there they were analyzed and its features influenced a number of Soviet designs, notably the LaGG-3. Although the surface cooling system wasn't copied, the addition of larger Soviet engines made up for the difference and the LaGG-3 was a reasonably good performer. It's perhaps ironic that German planes would later be shot down by German inspired planes.
The Japanese were also looking for new designs, notably those using inline engines where they had little experience. They purchased the three D-0's for 1.2 million DM, as well as a license for production and a set of jigs for another 1.8 million DM. The three D-0's arrived in Japan in May 1940 and were re-assembled at Kasumigaura. They were then delivered to the Japanese Naval Air Force where they were re-named AXHei, for "Experimental Heinkel Fighter". When referring to the German design the plane is called both the He 100 and He 113, with at least one set of plans bearing the later name.
In tests the Navy was so impressed that they planned to put the plane into production as soon as possible as their land-based interceptor €" unlike every other forces in the world, the Army and Navy both fielded complete land-based air forces. Hitachi won the contract for the plane and started construction of a factory in Chiba for its production. With the war in full swing in Europe however, the jigs and plans never arrived. Why this wasn't sorted out is something of a mystery, and it appears there isn't enough information in the common sources to say for sure what happened.
The DB 601 engine design was far more advanced than any indigenous Japanese design, which tended to concentrate on air cooled radials. To get a jump into the inline field, Kawasaki had already purchased the license for the 601A from Daimler Benz in 1938. The adoption process went smoothly, they adapted it to Japanese tooling and had it in production by late 1940 as the Ha-40.
At the same time Kawasaki was working on two parallel fighter efforts, the Ki-60 heavy fighter and the Ki-61. The former was abandoned after poor test results (the test pilots disliked the high wing loading, as they always did) but work continued on the lightened Ki-61 with the Ha-40 engine. The Ki-61 was clearly influenced by the He 100.
Like the D's it lost the surface cooling system (although an early prototype may have included it), but is otherwise largely similar in design except for changes to the wing and vertical stabilizer. Since the Ki-61 was supposed to be lighter and offer better range than the Ki-60, the design had a longer and more tapered wing for better altitude performance. This also improved the handling to the delight of the test pilots, and the plane was put into production. The Hien would prove to be the first of the Japanese planes that was truly equal to the contemporary US fighters.
Further developments
In late 1944 the RLM went shopping for a new high altitude fighter with excellent performance. It's unclear exactly why this happened, as the Ta 152H version of the Fw 190 was currently in limited production for just this task. Nevertheless Heinkel was contracted to design such a plane, and Siegfried Günter was placed in charge of the new "Projekt 1076".
The resulting design was similar to the He 100, but many detail changes resulted in a plane that looked all-new. It sported a new and longer wing for high altitude work, which lost the gull-wing bend and was swept forward slightly at eight degrees. Flaps or ailerons spanned the entire trailing edge of the wing giving it a rather modern appearance. The cockpit was pressurized for high altitude flying, and covered with a small bubble canopy that was hinged to the side instead of sliding to the rear. Other changes that seem odd in retrospect is that the gear now retracted outward like the original 109, and he re-introduced the surface cooling system. Planned armament was one 30mm MK 103 cannon firing through the propeller hub, and two wing-mounted 30mm MK 108 cannons.
The use of one of three different engines was planned: the DB 603M with 1825hp, the DB 603N with 2750hp or the Jumo 213E with 1750hp (the 603M and 213E both supplied 2100hp using MW-50 water injection). Performance with the 603N was projected to be a shocking 880km/h (546mph), which would have stood as a record for many years even when faced with dedicated racing machines. Performance would still be excellent even with the far more likely 2000hp class engines, the 603M was projected to give it the equally amazing speed of 855km/h (532mph).
These figures are somewhat suspect though, and are likely just optimistic guesses that could not have been met €" something Heinkel was famous for. Propellers loose efficiency as they approach the speed of sound, and eventually they no longer provide an increase in thrust for an increase in engine power. Even the advanced counter-rotating VDM design is unlikely to have been able to effect this problem too much.
The design apparently received low priority, and it was not completed by the end of the war. Siegfried Günter later completed the detailed drawings and plans for the Americans in mid-1945.
In 1939 the He 100 was clearly the most advanced fighter in the world. It was even faster than the Fw 190, and wouldn't be bested until the introduction of the F4U in 1943. Nevertheless the plane was not ordered into production. The reason the He 100 wasn't put into service seems to vary depending on the person telling the story, and picking any one version results in a firestorm of protest.
Some say it was politics that killed the He 100. However this seems to stem primarily from Heinkel's own telling of the story, which in turn seems to be based on some general malaise over the He 112 debacle. The fact is that Heinkel was well respected within the establishment regardless of Messerschmitt's success with the 109 and 110, and this argument seems particularly weak.
Others blame the bizarre production line philosophy of the RLM, which valued huge numbers of single designs over a mix of different planes. This too seems somewhat suspect considering that the Fw 190 was purchased shortly after this story ends.
The RLM version of the story is accepted largely at face value; that the production problems with the DB series of engines was so acute that all other designs based on the engine were canceled. At the time the DB 601 engines were being used in both the 109 and 110 aircraft, and Daimler couldn't keep up with those demands alone. The RLM eventually forbade anyone but Messerschmitt to receive any DB 601's, leading to the shelving of many designs from a number of vendors. After all, the 109 and 110 were better than anything out there, so another plane that was even better didn't seem important at all.
The only option open to Heinkel was a switch to another engine, and the RLM expressed some interest in purchasing such a version. At the tim the only other useful inline was the (inferior) Junkers Jumo 211, and even that was in short supply. However the design of the He 100 made adaptation to the 211 difficult. Both the cooling system and the engine mounts were designed for the 601, and a switch to the 211 would have required a redesign. Heinkel felt it wasn't worth the effort considering the plane would end up with inferior performance, and so the He 100 production ends on that sour note.
For this reason more than any other the Fw 190 became the next great plane of the Luftwaffe, as it was based around the otherwise unused BMW 139 (and later BMW801) radial engine. Although production of the engines was only starting, the lines for the airframes and planes could be geared up in parallel without interrupting production of any existing design. And that's exactly what happened.

04-06-2005, 06:33 AM
Nice summery of the history of the He-100. In a way, we already have a He-100 in the game as it's obvious that the Ki-61 was based on the Heinkel. Ofcourse the Japanese preferred a much lower wingloading and so gave the wings more span.

Still, it would be great to test (in the virtual world) if the Heinkel would have made a difference. It would no doubt have made a good B&Z fighter with it's high speed and good endurance (BoB??) but I'm not sure if it could have evolved as well during the war as the 109 did. Also, landings seemed to be difficult due to the higher landing speeds, although the landing gear was better designed than the 109's.

04-06-2005, 11:41 PM

i knew they recieved Emil Bf-109's , but didnt realise they had such close contact withe He-100 design

stupid german RLM , its lucky for the rest of the world that they dis-regarded the He-100

04-07-2005, 12:06 AM
<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Badsight.:
its lucky for the rest of the world that they dis-regarded the He-100 <HR></BLOCKQUOTE>Lucky? I dont know.. I mean the He-100 may have extended the war a few months.. But it surly wouldnt have changed the final outcome. If anything Germany is lucky, because had the war lasted a few more months Berlin would have been nuked too.

04-07-2005, 12:44 AM
erm, he100 in BOB = maybe german win....

africa is now WIDE open too axis attack, and india asia, russia is much weaker too...

04-07-2005, 01:19 AM
Here's a skin -HH-Beebop made for me for an as-yet unmade co-op. Looks pretty good though. The vertical stab. and lack of kink in the wing being the most noticeable difference. Oh and the big ventral radiator intake. But for AI He100s...pretty good. http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif


04-07-2005, 02:05 AM
<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by p1ngu666:
erm, he100 in BOB = maybe german win....

africa is now WIDE open too axis attack, and india asia, russia is much weaker too... <HR></BLOCKQUOTE>

Doubt it

Although tactically the plane was very fast, the cooling system was also very vulnerable to battle-damage. So there would be that many fewer Jagerfliegen limping back to base, and that many more crash-landing in Kent to be captured by the Home Guard along with their wonderfully advanced aircraft...

And of course all the strategic advantages would still be on the side of the RAF

Interstingly, though, in 'his' book Angels 22, on several occasions during the BoB George Barclay reports seeing formations of He 113 at high altitude and moving very fast. These were almost certainly misidentified 109s

Regardless, it was a highly beautiful aircraft

04-07-2005, 03:27 AM
<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by p1ngu666:
erm, he100 in BOB = maybe german win....

africa is now WIDE open too axis attack, and india asia, russia is much weaker too... <HR></BLOCKQUOTE>thats the whole point , the HE-100 was a much better A/C for what the SCW veterans were trying to do

they saw early that BnZ & slash & run attacks were better for survivability & in 1939 , the HE-100 totally was the better plane for this than the Emil

would be interesting to see how they compared in manouverability . the plane the HE-100 developed from performed worse in the 1935 tests than the original Messerschmitt 109 for "manouverability" , but both A/C changed a lot by the time of the polish invasion

unlike the Me-209 , the HE-100 that set that speed record was just that , a HE-100

with a superior plane at the start of WW2 , it would have been that much harder for the RAF

04-07-2005, 04:59 AM
kinda OT for a bit but:





04-07-2005, 05:06 AM
<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Vanderstok:
Nice summery of the history of the He-100. In a way, we already have a He-100 in the game as it's obvious that the Ki-61 was based on the Heinkel. Ofcourse the Japanese preferred a much lower wingloading and so gave the wings more span.

........................... <HR></BLOCKQUOTE>

thats much to far !
they have a similar shape, like other planes too , but they were totaly different.
inner konstruktion, wingfoil, wingshape and and and.... .

btw ,the japanese Navy purchased some He112 as A7He1 http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif