watercart

Thanks Steve, I am one of the luddites that does not have a facebook account. I would like to keep in touch, but will PM alternative contact details, all the best, Damien

Steve, this is the closest I've ever come to having a Zeiss made No.7. Many WW1 captured and souvenired RblF were converted by Esdaile Instrument makers in Ultimo Sydney. They were intended to fill the gap in optical munitions until Australian manufacturers could gear up, which they did. The Australian version of the No.7 is the No.107 made by JW Handley in Victoria. It needed a new number as it was not very interchangeable with the No.7. The converted German sights were reverted for drill and training.

Many thanks Steve. I'll put a bid on! I have never seen the bottom side of one and did not realise that they have a tapered dovetail to attach to the tripod. It reinforces my understanding that the elevation arc is part of the tripod rather than the main part of the director. I agree that the No.1 and No.2 Directors are rare, and I have never seen either. A No.3 will do me for an early 18Pr gun, so I will not be competition for the No.1 or 2. If I see any, I'll let you know. There are dial sights that are very similar to directors that mounted on a bracket on top of the shields of 18 Pounder Mk.I carriages - up to the introduction of the No.7 dial sight. I did not know that No.8 dial sights existed until one turned up on ebay. The No.10 dial sight looks remarkably similar to the No.8. I was reading something on the WW1 British optics industry, and I was surprised to see that there were No.7 Mk.I dial sights made by a London branch of Zeiss. Finding an unmolested No.7 Mk.I would be brilliant, but one made by Zeiss London would be even better. cheers, Damien

Very nice! Ever seen the tripod? Here is a photo of a No.3 being used in the pre-Gallopoli training Giza. The Australians fielded 36 x 18 Pounders. Looks like I missed the No.3 on ebay. I have not been looking too hard for a while, but it's always the way...

I have been searching for one of these for a long while, but I think you have the most complete one I have seen. If I find any early Directors or parts that are surplus to requirements, I'll let you know. Are you after any other director, sighting or fire control gear? cheers, D.

Steve, a magnificent piece of research. I hope someone out there can provide some answers on where this dial sight carrier fits into the scheme of things. There must be a consolidated list of the various No. / Mk. designations of optical equipment somewhere, if for no other reason than to prevent gaps and overlaps in the nomenclature. Finding a gun carriage with the corresponding mounting interface would be good... The List of Changes is the only information source I can think of, but not sure how conscientious they were about it from the 1930s onwards. My area of interest is more WW1, and even then the LoC was a bit hit-and-miss as the war went on. Some pics of the No.8 dial sight for your info.

The range dials of these things often have the manufacturer and date. It is very well designed and made, so a lot of effort has gone into it. If I made it, I would be proud enough to put my name on it! Hard to believe there were not a few made, but this is the first one I have ever seen. Your theory of it being a prototype makes sense if it were made in the tool room of a Government ordnance factory. If a production item, it would have had IPL / catalogue part numbers on it, an official designation and maybe a production serial number. When was the No.2 field clinometer introduced? That may give an indication of date? There are some odd things in obscure Brit artillery optics. I found a circa 1915 made No.8 dial sight that fits numerically between the standard issue Goerz based No.7 (18Pr, 4,5in How, earlier 25Pr, etc) and the No.9 for guns post 1941 you have photographed, but seems to fit in an entirely different mounting bracket. Its closest relative is the No.10 dial sight made in WW2, although the No.10 is designed to fit the same mounting socket as the No.7 and No.9. Although undated, the No.8 had a broad arrow and makers name (defunct a long time ago) and used a bit of die casting. The only reason I could see why the No.8 was made was for mass production dirt cheap or to dodge the Goerz patents maybe. I have never seen another one, nor any reference to it despite it being in the main number sequence for dial sights.

Hi Mike, I have gone through the rest in some detail now. Comments are: 1. An excellent technical description of the gun and its production variants. Well laid out and well illustrated. 2. A detailed description of the production history of the gun, trailer and ammo, including the contractors and sub contractors. I found this especially interesting, as I am keen on industrial history amongst other things. My own research has not drilled down to the same levels, so I was pleased to see references to the subcontractors' trades and products. It certainly filled in a few gaps in my knowledge 3. Good summary of operational history and salient examples without trying to be a full regimental history. Generally tells us what the regimental issues were where the guns were used. 4. Great narrative stringing all of the production, design, trials and operational info together , as well as the personalities involved. Makes a potentially dry list of facts come alive into an interesting read. (BAMs interaction with Kirby a case in point. Kirby was rather preoccupied with aircraft engine production, so he was drinking through the proverbial fire hose for other war effort projects. Still, his taking on the 25 Pr recuperator was something he should have avoided with hindsight). 5. In 239 pages, info is densely packed and covers every conceivable angle of 25Pr from artillery to tank use. Good value for money IMHO. 6. Comprehensive referencing and end notes. A readable book for anyone interested in the subject, with academic rigor. I like the publications list 😉 7. On arcane points, I was uncertain as to why Ruwolts marked the saddle data plate No.11 Mark. 1, for what should have been marked No.2 Mk.I, (was there a No.11 saddle I didn't know about?) but I was pleasantly surprised that you covered that point. As you say, the standard way of marking is to have model Number (No.) in Hindu-Arabic, and Mark (Mk.) number in Roman numerals. This was done by GMH and everyone else in the British arsenal system. I agree that Ruwolts have intended to use Roman numerals (II) where they should have used a '2', but on short 25 Reg. A8, Ruwolts they have been cheeky and used a '1' instead of a 'I' stamp. Maybe it allowed Charles Ruwolt to economise on stamps! Glad I've cleared that up now.... 8. One of my own projects, now I am in semi-retirement, is to republish my book on Australian manufacturers' codes. That gives the actual makers code marks stamped on the components / ordnance for most of the government factories and contractors you mention. Overall Mike, thanks for writing this book for us artillery aficionados, it is a cracker. I cannot see anyone being bothered to write another book on the subject, as there would be little extra to write about unless they went to the same level of detail on UK and Canadian production. That said, I think your book would cover off on that to the satisfaction of the vast majority of the readership. All the best, and good luck with it, I am sure it will be very successful. Damien

A way of valuing a good buying price may be to work out what a restored on is worth (GBP 12,500?) minus the cost of the restoration. What is missing, and how likely are you going to find these missing bits? How much will they cost to buy if they turn up? One of the killers of 25Pr value is the steel wheel rims and tyres. Left outside for a long time, the rims rust out along the tyre beading. This writes the rims off, but the tyre can be salvageable if very carefully removed. A time consuming process that requires patience from my experience. If the rims are stuffed, where do you get spares? PS. Use all normal precautions in inflating the tyres on split rims in a cage. Avoid using corroded and damaged rims. If there is any sudden and unexpected way to die in gun and vehicle restoration, it is via a wheel's locking rim suddenly letting go.

Mike, just received the book in the mail. Congratulations on a magnificent effort! This is the first book on Commonwealth field artillery I have seen in years with lots of new information that is not a rehash of other work done by Ian Hogg and Terry Gander, etc. In a nutshell, I would call it original and close to definitive - definitely a worthwhile addition to the library. I have naturally perused the 25Pr short section first. It is definitely a wealth of information, and I can't see how someone would read something more comprehensive without having all of the referenced archive files and gun manuals on hand. I see what you mean about the production numbers - they seem pretty tight. You seem to know all of the guns by name and where they live, so to speak. How the third Mk.II carriage (production number B3) ended up with Rego No. A224 / 225 remains a mystery, but that may be revealed in time, who knows. You mention discrepancies in production numbers and how they may arise, so that may give me some clues for ongoing pondering.... there was a bit of chopping and changing (literally) that may have contributed to that, and also the possibility of guns plucked from the factory for war bonds work. The Rego Number is Army allocated, so that is where the discrepancy may lie.... I have only had the chance to skim through the rest of the book so far, but I can see that there is a lot of interesting info that is not available published works to date. It is a valuable one-stop reference to anyone interested in the 25 Pounder generally, not just those in Australian service. I will be putting other books aside for a few days until I have a good read through the rest. ATB, Damien.

Thanks Mike, I am hoping to do A8 the justice it deserves. Although the main sections were sandblasted and primed before I obtained them, there is a spare shield with lots of original paint that I can sample to ensure an authentic paint scheme. The ordnance number is hard to read, as it is pitted from lying on the ground for years, but it has the standard muzzle cone. That will have to do for the moment, but I am hunting an early barrel with the early type cone. That said, there is a Whitelaw photo that shows a prototype with the later type cone, which could have been A8 for all I know. You may already be aware, I completed an 18Pr Mk.I in 2016 as shown in posts in HMVF, but mainly Gunboards in the Commonwealth Weapons section. It required extensive research into all aspects of the 18 Pounder, so with similar research into the Short 25, my I hope to get A8 to that standard of condition and completeness. Very lucky for me, Mr Belfield traded me a sight cone for very early short gun (possibly prototype) that was manually graduated, and a later production type made by the Union Can Company (UCC/V), so beyond my wildest expectations, those fine details will be correct. There was a report of a bulk lot of the barrels lying in a row somewhere in Victoria, according to a colorful Victorian collecting identity we both know (I have not spoken to him since 2003), so if that were ever to be found by others, there may be some interesting things to come out of that. Best of luck with your book, I am sure it well do very well. ATB, Damien

Well, I'm about to pack away my Short 25 Pounder files again, until my recently ordered copy of Mike's book arrives. I have followed the link Mike provided and it has a preview of several pages and the table of contents. Looks very good, so I wait with anticipation! Attached are a couple of photos from my own files that I hope the viewership will find interesting. This is the first prototype. Salient differences to the standard gun are: 1. The underslung axle as per the standard 25Pr gun 2. The light truck wheels and hub caps akin to 2 Pounder. 3. Short barrel with no muzzle cone / blast deflector, nor attachment screw thread tor attachment of same. 4. Short cradle and recoil block. No brass data plates on the cradle. 5. No hump in the trail, as no traversing platform was intended to be fitted (at this stage) 6. The four securing pins and lugs holding the front and rear of the trail together. These appear to be very similar to 2 inch tow balls with a longer shank and a cross bar handle. This is the carriage of gun Registered No.8, complete with shield as I found it about 20 years ago. It has some residual features of the standard trail that were not expected, such as the redundant attachment faces for the shield support arm brackets. The new light weight shield is attached to the saddle as per the remaining bracket visible in the photo above. The data plates on the saddle are also standard Ruwolt 25Pr types, as being a prototype there is no proper acknowledgement of the new trail pattern. The Reg No. is stamped A8 dated 1943, and the saddle data plate is marked No.11 Mk.I. Both data plates had "A8" marked on their reverse sides in black paint. I have almost all of the bits required to reassemble A8 into a complete gun, so am looking forward to starting the rebuilding process. ATB, D.

For the Australian membership, a 25Pr was auctioned in NSW in the last fortnight or so. It was deactivated and looked OK (needing cleaning up and a new paint job), and had been under cover for a long time, which is better than most. Completeness of the gun itself was acceptable, but no CES. Hammer price was $20,000 plus 20% commission, so pretty much aligns with the GBP 12,000 suggested above.

Mike, I am not saying that there were necessarily 225 guns made, but am asking the question as to why the carriage registration numbers go to at least 225. The registration number is not applied by the factory, but by Army on acceptance of the item. So unless there is a deliberate gap in the registration numbers as applied by Army, I have no answers. Rego A224 was one of the earliest Mk.2 carriages with the recoil block rego number 225 installed. This recoil block is (factory test?) dated 3 March 1944, which is almost 6 months after rego A103 and over 6 months before Regos A 107 and A 140. Rego A 183 is dated 8 Nov 1944. So a bit all over the place. However, I think it is clear that B3 / Rego A 224/225 sat on the factory floor for a long time and was one of the last carriages offered up for acceptance and registration. The other thing not covered so far is the ordnance situation. Generally, the breech rings were taken from standard production and I am unaware of any that were inscribed with the carriage Rego number. However, the barrel was a special component for the Short 25 Pr and the jacket was marked with an "A" number similar to the carriages. While one may imagine that a complete gun left the factory with carriage Rego and barrel numbers matching, I have never seen a complete gun with matching barrel and carriage numbers. We are all aware that the ordnance system managed barrels and carriages as separate items, so if the survivors are a representative sample, there must have been a lot of barrel changes throughout the fleet if they were all matching on Army acceptance. That said, most guns would have had such a small amount of work that a barrel replacement between manufacture in 43/44 and retirement in 1946 would seem unlikely on the basis of wear and tear. The breech rings and barrel jackets I have seen have all been made by either the ordnance factory at Maribyrnong or by Ruwolts, and while early barrels are MO, there is a mix of MO and CR barrels as production proceeds. This may be reflected in the contracts you have researched. So amongst other things, I am wondering when the barrel jackets were allocated their A number? Were the jackets numbered on production at Maribyrnong / Ruwolts and the carriage given the same Rego number on acceptance? Or perhaps they are completely separate numbering systems and streams of management? So to your question regarding whether I have accounted for spares...I have accessed a lot of archive material on the Short 25Pr, but not the production contracts, so the answer is "no". Given the fact that MO provided ordnance that was mounted onto Ruwolt made carriages at various stages during the project, were the ordnance and carriage contracts separate? MO made ordnance would not appear in the Ruwolts contracts unless it was in the Schedule / List of GFE (Government Furnished Equipment), whereas the balance of ordnance would be contracted to Ruwolts. So the obvious questions are: 1. how many carriages were contracted to Ruwolts? 2. how many breech/barrel assemblies were provided by MO as GFE? 3. how many breech/barrel assemblies were contracted to CR? 4. how many spare carriages and breech barrel assemblies were in the Ruwolt's contracts? 5. Given that the carriage production numbers and rego numbers are quite separate, did Army bother stamping unassembled spares with a Rego number? and 6. were there sufficient spares for army to build up more complete guns after final deliveries under the contracts, to which they applies a Rego number? I'm looking forward to reading your new book and thanks for making the effort to publish. I was going to do a monograph my self one day, but something published is worth much more than a manuscript that may get chucked in the paper recycling when they clear out your estate. ATB, D.

Mike, although I have a mountain of stuff unpacked from a recent house move, the boxes with the Short 25 material were miraculously near the top, so I can answer your questions and clarify a couple of my observations made above. Firstly, the shield against the wall is indeed for a 17 Pr. Consulting the list of observed examples, there was a consistent dislocation of three between Registered number and Production number in the Mk.I carriages, eg, Rego A26 has production number A29 and Rego A 104 has a production number of A107. As for the "improved prototypes", the Rego and production numbers were the same, at least up to A8. It is pretty clear that the Rego and production numbering of the officially accepted Mk.I carriage continued directly from the "improved Prototypes", so presumably the first rego number for the Mk.I carriage would have been A13 by that reckoning. At what point the dislocation between Rego and production numbers happened I don't know, but Rego A15 had production number A19, and Rego A26 has production number A29. My reference to the discrepancy in the improved prototypes was poor recollection abetted by one of the data plates on Rego A8 bearing the inscription "No.11". This was the saddle designation number rather than a production number. On that basis, it looks like there may have been some kerfuffle around the last of the improved prototypes and the first of the production model Mk.Is. The simple explanation is that three guns produced by the factory were not accepted into service and therefore not registered. What guns could these have been? Well, I don't know, but I would have a couple of suggestions, being the high saddle model and maybe the one using Beaufort aircraft wheels. As far as production chronology goes, Rego 22's recoil block is dated 20 March 1943 and Rego A 104 is definitely a Mk.I carriage with its recoil block is dated 20 Sep 1944. The earliest Mk.2 carriage I have listed is Rego A139 with a production number of B23. If Mk.2 production kicked off at Rego A113 and production number B1, then that is sort of in the ballpark. However, the Rego vs production number correlation, such as it is, becomes really scrappy and out of sequence. I have confirmed that the Mk.II carriage Rego 224 has the recoil block for gun Rego A225. as this would appear to be the number applied by the Army / MGO(?) then it would seem that there was at least 225 Short 25s accepted. Going back to the transition period between the improved prototypes and the accepted production Mk.I, there was a lot going on in terms of converging the prototype design to the production design. More later...

The Short 25Pr is one of the more interesting guns in my opinion, although a bit of an evolutionary dead end. From what I can tell, a lot of what has been written on them (including Mike's good work). The source info appears to be the same, so published production figures are fairly consistent at 212 guns. The development of the gun is not a subject I see written about in any detail, but I managed to have a long chat with the nephew of the gun's manufacturer. The guns were developed and made at Charles Ruwolt's factory in Richmond - a suburb in Melbourne Victoria. Charles was an outstanding production engineer, and was constantly prowling the factory floor applying new design modifications on the spot. As a result, very few of the surviving guns are identical. This would may seem to be a nightmare from the configuration control perspective, but all assemblies remained interchangeable over the two year production life of the gun (1943-44). About 15 years ago I conducted a survey of all known surviving guns and parts bearing numbers, and was surprised to find that they were inconsistent with the accepted history (for want of any challenge to the original source material). The gist is that there appears to have been at least 225 guns produced that achieved an official Registration number, so the question of "what does the figure of 212 refer to??" is a good one. In my research, I had the good fortune of talking to the elderly MAJGEN John Whitelaw, who was a junior officer in the RAA during WW2. It so happened that his father was the MAJGEN RAA involved in the Short 25Pr project. MAJGEN Whitelaw (Jr) was a keen artillery historian and provided me with several photos of the Short 25 Pr gun development from his father's collection. One photo was of the original prototype constructed at Ruwolt's showing a sawn-off 25Pr with 2 Pr wheels. It was clearly cobbled together, as you would expect, from available 25Pr and 2Pr parts with some cutting here and there, but trail length being the same as the standard gun. The next two prototypes constructed were tested at Fort Gellibrand, and there are several photos of this pair available in the AWM photo database. The next prototype batch of 12 guns were closer to the production guns, and featured a shield and muzzle cone, but had the full length trail joined by horizontal cross pins of the earlier prototypes. One of these guns is pictured next to a standard gun in talltom's 2019 post above. Note also a number of unfitted shields leaning against the wall. There was a very wide variety of experimentation done with these guns, from troop trials in New Guinea (lots of photos of this in AWM's photo database), to extended height saddles to get a mortar-like trajectory, and aircraft balloon tyres, and generally lots of small design refinements. The other interesting thing about the third prototype run was that they were assigned production numbers and were accepted by the army with Registration numbers, however, these numbers are out of sync by three, indicating that all prototypes were assigned a production number, but that the first three prototypes were not accepted by Army. This numbering sequence continues into the first of the production guns, being the officially accepted 25Pr, Short, Mk.I, so the first production gun is production number A16, Registered number A13 (lucky?). The production and Registered numbers go in lockstep initially, but it becomes more ragged as the production numbers rise. The second run of guns (Carriage 25Pr Short, Mk.I/1) have the production number commence at 'B1', but registration numbers continue the sequence from the first run. With Rick's example, his gun is the second pattern with machined rather than riveted cradle. It is the 75th gun of the second run (hence B75), but the allocation of the Registration number by Army is A207 rather than the expected(?) A199. My example is Production number B3 with Registration number A224, with cradle A225. My gun was one of the first of the second production run, but was one of the last (if not the last) to be delivered to Army for registration. It obviously sat on the factory floor until the production run had finished, so contributed to the discrepancies between the production and registration numbers from early in the second run, The production number is stamped on both sides of the trail joint, and the Registration number is obviously on the brass Registration fixed to the left side of the saddle. The Registration number is also stamped on the recoil cradle as well. So, where did the number 212 come from? 112 first model(Mk.I) and 100 second model (Mk.I/1)? Maybe this was the number of production guns rather than prototypes, but it ignores the fact that the prototypes were accepted and Registered and used by the Army. The list I have compiled indicates that the survival rate for the gun type, rare as it is, is quite good, so I wa able to achieve a good sample size. Using actual serial and production numbers suggests that the 212 number is not quite as simple to explain as that. Anyway, probably enough for now... ATB, D.

Clive, I recall that they are called "hydromotors" in the main gun manual. These systems are usually a constant speed AC motor driving a hydraulic system (back to back pump and motor) using a swashplate arrangement (I think). It allows an infinitely variable speed output to the elevation and traverse drivetrains. I would love to get hold of the British printed Illustrated Parts List for the predictor, as it would confirm (or not) the above theory and tell me what I am missing. I do not think what I am missing out of the predictor would be hard to make, if only I knew what was missing. D.

This one is set up for the No.3 Kerrison Predicter, with electric hydromotors for traverse and elevation. Very original. I have the Predictor and the hydromotors, but yet to fit them to a gun. There is a lot of wiring and electrical stuff out of sight in these mountings, so it is not a simple restoration. There is also a diesel generator that comes with it, and the motors inside are marked 50 Volts, 50Hz 3 Phase. The Kerrison was literally a barn find in its original crate, but the previous owner had repurposed the aluminium gables and the telescope mounts / linkages. So I'm still looking for those bits...D.

Hi Coppo, I would be interested. Can you PM me with the details? cheers, watercart (Damien)

I have been doing a bit of research on another Brotherhood item, being a Brotherhood-Crocker car c.1904. It is a 4 cylinder T-head engine, but interesting to note that the Brotherhood logo was based on their 3 cylinder radial design that could be adapted to either being a pump / compressor or a motor. The early 3 cylinder torpedo motors are straight Brotherhood designs. A photo of the car engine is included for interest sake. However, no evidence that it was used in a military application...yet. Brotherhood did design some very innovative features for the car engine. Most of the car and engineering journals at the beginning of 1905 have large articles on the engineering side that can be seen using Graces Guide if you are curious. A lead on any parts for a Brotherhood car engine and gearbox would be much appreciated. But back to the main subject, - out of the blue, I have also been offered a Japanese torpedo engine for the museum. It happens to be an 8 cylinder radial unit with an 18 inch diameter. This was a bit of a surprise, as I have not seen any references to it. The Japanese definitely produced the Whitehead long stroke 2 cylinder designs for their 18, 21 and 24 inch torpedoes, so where an 8 cylinder radial for an 18inch fits in, I do not know (other than very tightly).The present owner was a salvage diver after WW2 and recovered the torpedo. He has had the motor running on compressed air, so he says. I will be following it up. ATB, DNA

Simon, the best indicator of the torpedo diameter is measuring the maximum diameter of the engine. If it just squeezes inside a 21inch circle, then it will certainly be a 21inch weapon. If the weapon was 24 inches, the designers would have used up every bit of that additional diameter in their quest for maximum power. If 21 inches, then your observation that higher pressure (400 bar) or higher percentage O2 was used will be the only alternatives. The 8 cylinder stats quoted with respect to a similar range and double the air volume tend to make me think it was not high purity O2, but you never know. In general reading, I was under the impression that several countries including UK experimented with varying percentages above 20% purity. As we know, the Japanese seem to have been the only ones to get higher purity O2 wrangled properly for their various torpedoes, which gave an old Whitehead long stroke 2 cylinder design a new lease on life. D

Simon, looking back at the stats quoted earlier, I am thinking about the power you will have to dissipate when running it up, if you decide to go for full power that is. If the 8 cyl vs. 4 cyl Mk.VIII / IX engine stats apply to this engine, you are consuming almost twice the air in about 3/4 of the time that a Mk.VIII engine would. On a dynamometer, we were getting about 550HP out of a Mk.VIII engine, which is significantly higher than the handbook minimum. Are we therefore contemplating in the ballpark of 1200 HP for this engine? Much as I would like to see this thing go full power, I am not sure how you would set this up for a run without the risk of things getting out of hand! We have been concentrating on the engine somewhat, but the actual torpedo that contained this 8 cylinder beast would need to be somewhat different to a Mk.VIII / IX. For a start, doubling the air capacity would need to double the reservoir length or the storage pressure. I had a quick look through the previous 5 pages buy did not see reference to the diameter of the engine / torpedo (sorry if I missed it). As a 21 inch torpedo is a more or less standard length to fit in in-service launching tubes, this would require the same size reservoir and doubling of the reservoir pressure. Otherwise, to keep pressures under 250bar, are we looking at a 24inch diameter weapon? Getting ignition would also be interesting, as I cannot tell whether there is a pre-combustion igniter on the engine. In Mk.VIIIs, this as a device with three hammers that are cocked and released to fire three blank cartridges as the first blast of compressed air and fuel go to the engine. I assume that the fuel admitted to the cylinders ignites by the usual diesel principle, as the pre-ignited mix is now warm as opposed to actually burning (??). If I were in your situation, I may be thinking about finding a tame racing engine workshop with a dyno that could take 1200HP and a fairly sizable test rig containing fuel and compressed air. And a method of directing the exhaust outside... Maybe just sticking to turning it over with compressed air for a few seconds would prove functionality as well as avoiding accumulation of corrosive combustion products internally, which may cause seizure in future. After all, Mk.VIIIs where fuel ignition failed were still capable of about 22 knots over a few hundred yards. Cheers, Damien

Simon, the standard approach for conservators in most situations is to only do what is reversible, so your comment about bites of the cherry is prescient. Overall, the objective is to get the motor looking decent, rather than being a frosty looking ball of corrosion, so I would go with the steel wool and oil approach to begin with. At least that will have the least risk of doing irreversible damage. If it still looks a bit rubbish, you can ponder the alternatives then. Some of my professional work deals with the effects of salt water corrosion, so that may be another avenue of consideration if the motor is a combination of steel, aluminium and brass that has been sitting in salt water for a long while. The bad news is that chloride ions penetrate deep into metals such as steel over a period of time and can play havoc with the metal once it is out of the water, eg, anchors that have been sitting on the bottom of the bay for 200 years OK, and explode into giant puff balls of rust when within months of being landed. Lots of work has been done in this area with Mary Rose and USS Monitor, etc. Bottom line is that regardless of your treatment, chloride contamination may make life interesting. You will know you have it if you get aggressive rust forming under paint / preservative on what you thought was a clean and well prepared surface. For steel tanks and ship hulls that have not had long term seawater exposure onto bare steel, ultra high pressure water blasting (approx 30,000psi) can be effective, as it can remove salt embedded in the surface of the object. However, this can be very harsh on aluminium, and is not too effective on a long term salt water immersion. I do not have the specific info on the preservation products, but I saw a can of it at a local automotive paint supplier. It is a silicone based product used by the oily rag / barn find restorers where original paint and patina are the priority.

I took a measurement of the internal diameter of the air line from the Mk.VII torpedo air reservoir. It is 3/4 inch, so capable of discharging a lot of air quickly. Given that it is an historic artifact, keeping a bit of patina on the steel and brass would be standard museum practice. Fine or medium steel wool and oil would remove the soft rust scale but leave a stable patina and paint. If you want to go down that path (as opposed to attempting to make it look brand new), you can get products now that apply over the steel components' patina that makes them look good and protects against further rusting. From the photos, it is not clear whether there is much brass as the earlier engines, but it would be the crank case material and some pipework if anything.