Video walk-around of the HeS 3B replica cut-away at the Smithsonian Air and Space Museum. (I'm not sure it was there back in 2007 when I visited, but I don't remember it)
It wasn't until seeing this video a few years back that I properly understood the diffuser and combustion chamber arrangements of the HeS 3. It uses an annular combustion chamber of pretty simple and open design, with a splitter at the exit of the radial diffusor section, directing some air downstream into the chamber and most of it forward into the folded, reverse-flow or mixed-flow diffuser section, where it's split up further by a series of sheet metal diffusion/guide vanes, with a small portion tapped off and fed to the burners (combustors). The burners are a series of square flame tubes or square flame jets with vaporized fuel injection nozzles, with vaporizer fuel lines forming a grate-like structure at the exit mouth of the flame tubes. (I assume the square shape was chosen for simplicity of manufacture, as these engines were all made using Heinkel's airframe manufacturing facilities, with an emphasis on cut and formed sheet metal construction over machined parts: also a design feature that made for interesting mass-production and war-time resource conservation potential, but in any event, all those engines were hand-crafted by Heinkel craftsmen) The flame jets would presumably extend well into the very wide/open combustion chamber and would then combine with the air split-off from the compressor earlier on, mixing on its way to the radial inflow turbine guide vanes and the turbine itself. (the use of the radial inflow turbines is one of the more unusual and interesting elements of the engines and probably one of the reasons they worked as well as they did using solid Krupp stainless steel alloy blades: radial inflow turbines are subjected to far lesser stress than axial ones and would tolerate a good deal more abuse as well as less optimized designs or overall conditions)
Those vaporizer type burners were a solution to poor performance with attempted atomized fuel injection systems, inspired by the vaporizer for the flame jet of a gasoline or kerosene blowlamp. Due to the need for warm-up time to reach proper vaporizing conditions, these engines also had to be started with a gasceous fuel, typically hydrogen (at least in early testing), though I suspect propane, methane, dimethyl ether, carbon monoxide (or the carbon monoxide, hydrogen, nitrogen blend of coal-gas or town gas used as domestic heating gas in many regions at the time) would all have worked well enough. Once operating temperature was reached, the engine could be switched over to kerosene or gasoline fuel. Hypothetically, they should have been able to use a volatile, smokeless liquid fuel for start-up, like methanol, diethyl ether, a blend of the two, or ether-alcohol blends (ether-alcohol being a common/popular solvent form of diethyl ether, reducing volatility and largely preventing the formation of explosive peroxides during longer storage periods), but I've seen no reference to such starting fluids being used, or for such being used as primary fuels earlier on when combustion issues were even more serious. (pure Hydrogen gas was used instead to ease design of early prototypes and proofs of concept in the HeS 1 and HeS 2, a choice that likely also would have greatly sped up Frank Whittle's early developments, as would a number of other 'friendlier' fuel choices with higher flame speeds, more volatility, and lower flame temperatures: Whittle was even worse off, as he started with Diesel fuel and later decided on Kerosene)
Origins of German jet power
https://jqmgrdyk.home.xs4all.nl/jetpower/HeS-6-400.jpg
This might be the only image I've seen depicting the HeS 6, or at least the only one I recall off hand.
There seems to be relatively little information on the HeS 6 developed from the HeS3 but apparently not able to be flight tested in the He 178 due to size/weight issues. (the diameter appears to have been similar to the 930 mm of the HeS 3B, but it was quite a bit longer, with broader/taller compressor impeller, larger reverse-flow diffuser section, longer combustion chamber, and broader/longer turbine, plus weighed 420 kg to the 3B's 360) There's mixed information on whether fuel consumption was a major improvement or largely unchanged, but it appears to be approximately 1.6 lb/lbf (or kg/kgf) in any case, similar to the HeS 8 (or at least some early developments of that type), while the HeS 3 may have been as poor as 2.16 lb/lbf.
Thrust of the HeS 6 appears to have been 550 to 590 kgf, with the former being quoted more often, but the latter (or 1,300 lbf explicitly) being cited in some scientific journals or extracts on Jet Engine development.
One in particular: "Pioneering Turbojet Developments of Dr. Hans von Ohain — From the HeS 1 to the HeS 011" comes to mind.
Pioneering Turbojet Developments of Dr. Hans von Ohain — From the HeS 1 to the HeS 011 | Volume 1: Aircraft Engine; Marine; Turbomachinery; Microturbines and Small Turbomachinery | GT1999 | Proceedings | ASME DC
There's also a good summary of Ernst Udet's involvement and his support for Heinkel, including the agreement to allow Heinkel to absorb Hirth if the He 280 met its 1941 test flight deadline.
The combustion chamber layout (and possibly the axial diffuser design itself) of the HeS 8 also seemed to prove major problems to combustion stability, general performance, and development time/energy required to make a flight-worthy design and it's always puzzled me a bit as to why the HeS 6 wasn't reconsidered or at least returning to the more known and (apparently) workable diffuser/combustion chamber layout already proven. Being only about 6 inches larger in diameter, it may have even fit acceptably well in underwing mounts on the He 280, unmodified, though the ground clearance obviously would be reduced. Still, compared to the massively greater weight of the Jumo 004A engines (plus 1.5 inch larger diameter) eventually trialed as well, the difference in size seems fairly minor, especially if it had been considered from the start in the He 280's design phase. (for that matter, redesigned wings for mid-mounted engines in the style of the Gloster Meteor also would have worked, and given the many months the He 280 prototype was ready before the first flight-worthy HeS 8 engines, it seems like building a whole new, alternate wing would be quite realistic, too)
Besides all that, the HeS 8 was always intended as an interim substitute for the exceptionally appealing HeS 30, and from that perspective, it seems like falling back on the HeS 6 fit the job much better, especially once difficulties in the HeS 8 began to develop.
On the other hand, the HeS 8 likely did provide valuable experience in developing and refining the axial diffuser arrangement, which could have been quite valuable in a scaled-up second-generation design, but that didn't materialize either for whatever reason. (scaling up the HeS 8 1.26:1 should result in an engine approximately twice as powerful, twice the weight, but with a much better frontal area to thrust ratio and diameter of about 38.4 inches or 976 mm) Though combining the combustion chamber and turbine design of the HeS 30 with the simpler centrifugal compressor of the HeS 8 seems like it would've been a winning combination, including avoiding potential issues of the larger radial turbine consuming too many rare materials and possibly being more difficult to adapt to air cooling. (though the radial design does seem better suited to cooling slots/jets cut into the hub at the base of the blade roots rather than needing the more complicated drawing or folded+welded hollow blades to be effective ... and investigating air cooling methods for the HeS 6 may have been yet another better choice than dealing with the HeS 8's struggles)
That above document also mentions, contrary to some other claims (including wikipedia's) that the centrifugal compressor was built-up from a steel hub and sheet metal aluminum blades, which should have been an appealing route for mass production in war-time, avoiding the high speed steel consumed in precision machining. (designing an impeller suitable for casting would be the other route, and notable one used for the post-war American J44's arguably mixed-flow compressor) The axial inducer fan stage was made of forged aluminum blades, though I'm not sure of the distinction of forging and stamping in this case, but it was not simply formed/curved sheet metal of uniform thickness and was not machined. (also possibly noteworthy was the J44's use of inducer type blading on its compressor being a separate piece with individual blades that mated to the blades of the magnesium alloy casting, allowing simple, straight blades on the cast portion; Ohain's impellers were not completely straight though, but had a slight curve to them on the upper edges receiving intake air) The Whittle and De Havilland (as well as most war-time American and British supercharger impeller designs) had more pronounced curved, angled intake blade tips and overall shaping, plus tended to be made of large, monolithic aluminum forgings, quite complex ones in the case of double-sided impellers as Whittle used. (also notable in the Napier Saber supercharger)
The HeS 011's mixed-flow impeller appears to have been made of machined aluminum, but mass production likely would've switched to something else, possibly a casting or a two-piece casting with additional stamped-blade section. (I'm not sure the sheet metal blading of the earlier Ohain designs would adapt well to the more complex aerodynamic shape of the impeller) The HeS 011's diagonal flow compressor is also not the same type of mixed-flow compressor as the J44 uses (the latter is likely closer to the mixed-flow compressors used on a couple experimental V-1710 supercharger designs) with the J44 using either an axial or diagonal diffuser with fairly heavily curved/shaped centrifugal impeller blades with inducer blades angled sharply enough to arguably count as an axial-centrifugal hybrid stage. (so it goes from axial to centrifugal flow, vs the centrifugal to axial flow in the Heinkel engines: also something present in the HeS 8 with axial stages added
after the centrifugal one rather than having one or more axial compressor stages feeding a centrifugal compressor as the final stage as is pretty much the only arrangement used in commercial turbojet and turboshaft designs ... and slightly odd or ironic given the axial inducer stage Ohain had already included: simply developing that into a full axial compressor stage with stator blades flowing into the centrifugal stage would've seemsed straightforward enough, then potentially stacking more axial stages onto the intake of the engine to increase mass flow and compression without increasing the diameter significantly ... for that matter, taking the HeS 30's 5-stage compressor or omitting one or two stages of it, and adding a single centrifugal stage behind that should have made a quite good class-2 engine and been a lot simpler than the HeS 011, though required the precision machined reaction type compressor blading of the HeS 30 ... though I suppose switching to the simpler impulse blading and retaining all 5 stages, plus the centrifugal one also would've worked)
Fairchild J-44s: Minijets (for some reason it's half in french and half english) Some dramatic cost-saving measures there, like using off the shelf burners and stainless steel tubing for the combustion chamber casing.
Packard XJ41 - Wikipedia (one of the few other mixed-flow turbojets of that period, and like the J44, designed for expendable drone or missile use, and the two might even be related, but I haven't seen any reference to such) Also slightly odd the J41 was marginalized, as the thrust was competitive with the J33 and J35 of the period, fuel consumption similar or better, and weight much lower, while diameter was between the two at 48 inches (not listed on wiki). It seems closer in size, weight, and performance to the Derwent V, though possibly simpler and cheaper (especially if it was much like the slightly later J44) but somewhat larger in diameter. (the Derwent V was capable of 4,000 lbf maximum thrust, but was down-rated for operational use in Meteors, though I believe it was used at max RPM for the speed record Meteor flights)
Also, the HeS 6 becomes all the more appealing for use in testing of other jet aircraft developments, not just Heinkel's, and was producing more power in 1939 than the BMW P.3302 (003) was producing a couple years later (and possibly still at the time of being fitted to the Me 262 V1, where those engines failed) and almost as powerful as the 004 up to its mid 1941 endurance tests, though the 004A developed fairly rapidly later that year. The diameter and certainly weight seem to have been small enough to be substitute engines for the Me 262, and potentially low enough in weight to not lead to the wing sweep modification the heavier-than-expected BMW engines caused. (wing loading and stall speed should also have been reduced a good deal on the Me 262, though thrust to weight ratio would be poorer than the 004A, particularly if overreved to 10,000 RPM, but the take-off run might still be reduced due to the lower wing loading, and landing speed would be greatly reduced, plus dead-stick glide landings would have more margin for error, and of course, roll rate would be greatly improved with all that mass on the wings removed) As a possible operational/production design, even with approximate 10-hour turbine life, 1.6 lb/lbf specific fuel consumption and the moderately greater diameter, it seems reasonably appealing with the better thrust to weight ratio and relative ease of producing flight worthy examples. (plus, with the lower weight and nominal thrust, actual fuel consumption or fuel economy in terms of range might have improved over the much heavier Jumo 004 in an aircraft of the Me 262's size)
Also odd is the lack of interest in bomber or recon aircraft with Heinkel. 4 or 6 HeS 6 engines should have worked well for high-speed light or medium bombers and been very appealing for getting RLM attention (or enticing Hitler himself). The Arado 234 also likely would've fared well with 4 HeS 6 engines, being much less underpowered than the 004 installation for similar weight and much lower weight than the 4 003s later tried.
HeS 6 engines also would've made sense on conventionally powered bombers and transport aircraft as booster engines. (small and light enough to make sense in that role, and the relatively short turbine life would be extended if only used for take-off assist and potentially for emergency thrust) And running on diesel, or aviation kerosene compatible with diesel engines, they could also serve alongside Junkers diesel engines and potentially offset the poor ability to use maximum power for combat/emergency conditions with those engines while having a common fuel system using the cheaper, more available diesel or kerosene (or a blend of such). Jet-boosted, turbocharged 2-stroke diesel powered aircraft would be quite interesting. (and apparently turbo-diesels were easier to engineer on the high-temperature turbine materials end due to the lower exhaust temperatures than gasoline counterparts, though the turbine blades of Jumo 207 engines do appear to be hollow for air cooling as well)
File:Jumo 207 im Technikmuseum Hugo Junkers Dessau 2010-08-06 01.jpg - Wikipedia