The Dreadnought Project Forum Index The Dreadnought Project
Naval History in the years 1890-1920
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   Reg1sterReg1ster 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Dreadnought Battlecruiser Doctrine
Goto page Previous  1, 2, 3, 4, 5  Next
 
Post new topic   This topic is locked: you cannot edit posts or make replies.    The Dreadnought Project Forum Index -> Discussion
View previous topic :: View next topic  
Author Message
tone
Site Admin


Joined: 29 Oct 2004
Posts: 478
Location: Boston

PostPosted: Tue Apr 18, 2006 9:33 am    Post subject: Reply with quote

jwduquette1 wrote:
The 28cm spgr image is from: “Notes on German Shells, 2nd Edition, 1918”.


Is there more detail on this reference? I take it it is in German. Also, what does spgr mean? I have not studied German shells in much detail.


Quote:

The exterior ballistics problem is a bit of a head scratcher. I am getting two different BCs. One for the 13.5-degree launch angle, and a different one for the 16-degrees launch angle.

To get the trajectory to match the level line at 19790-yrds I need to use a BC = 8.284 -- so i=0.6476.

However if I use the same BC with a 16.5-degree launch angle I get the trajectory crossing the level line at about 21410-yrds. This is a bit further than the NavWeapon web sites quote figure of 20890-yrds.

I'm using an Ingals Drag function, and US Army Standard Atmospherics. Are you aware of what standard conditions were employed by the Germans during this period?


My entire simulation model revolves around fairly different bases than those you suggest (BC and form factor, etc, even though these were commonly the language of the day). Rather, I defer to the expertise of others in some regards:

1. I rely on a Java implementation of the McDrag program (google it) to create a custom drag-to-mach relationship for a given shell based upon its overall length, CRH, and mass

2. I take this general shape as the basis for the relationship, but employ a series of test firings to discover a scaling factor that will make the raw curve actually suffice to permit the simulated shell to exactly duplicate a single extreme range/elevation pair for the historical weapon. Typically, the scaling factor is in the neighborhood of 0.9 to 1.15 -- reflecting a fair vote of confidence in the general validity of McDrag.

From there, I do a lot of test firings and create a table through the spline-interpolated results. In many ways, I think my range tables are apt to be more accurate than historical ones (at least in the WW-I cases, as I have done some real studying into the manner in which these were compiled). Where they are not, however, improvements are best sought by examining the atmosheric model. For these WW1 shells, however, I am inclined to doubt anyone can produce data that would permit much review of the accuracy of my tables -- but I'd love to see such!

As to your questions about atmospherics... I have to punt. I recall my own are based loosely on the STP model, but I looked into this in excessive detail for awhile and got bored with it. I dedicated most of my time to honing the other elements of my code, as if I wanted to see this used in an interactive simulation, the most important aspect is that I can create splines that relate range-to-elevation and range-to-time-of-flight that faithfully describe performance of the simulated (not historical) shell. I am content if the grossest errors between simulated and historical performance are within (say) 1 degree in angle of impact, half a degree in angle of elevation, 1 second in time of flight, and within 2% in impact velocity (roughly). If a battle were to be won vs lost on the basis of such discrepancies, I'd eat my hat.

tone
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
jwduquette1



Joined: 20 Dec 2005
Posts: 90

PostPosted: Tue Apr 18, 2006 11:19 am    Post subject: Reply with quote

Hi Tony:

Send me an email at the address I used for registering here. I have the manual in a scanned form. I'll email you back a copy. It’s British.

Spgr is spreng-granate – or some such thing. It is sometimes used to refer to high explosive shell, but in this case it resembles British CPC. Thickened nose and a large bursting charge – but a base fuse rather than a nose fuse that one might commonly associate with high explosive shell.

I am familiar with McCoy’s work, and played with his various programs a fair bit. I have long suspected that many of the exterior ballistic programs floating about out there have his original Fortran program under the hood.

Just to be clear I am not being critical of your range table calcs. Most of this is pretty obscure. However, when I run across someone else that has been monkeying with this sort of think I am always interested in picking their brain.

As to my lingo – it is fairly common to the science. It is difficult to bounce ideas off of people if there is not some sort of common basis in language or jargon.

On form factors, McCoy did a number of correlations to nose length and form factors before his death. Much of this has been condensed into a series of figures presented in his work on exterior ballistics. For a simple crh ~4 head, a form factor of about 0.64 to 0.7 is in tune with his findings. So I felt like my back calculated value of about 0.65 was in the right range.

You’ll notice on the RN range tables that the form factor is indicated on the first page of each set of tables. In the case of heavy 13.5” CPC the “ko” (kappa sigma I think) factor of “0.7” is the form factor. That’s what British employed for determining the final BC for 13.5” CPC.

BC = w/(i x d^2)

W in lbs
D in inches

I typically bow to range tables or firing tables if I have them. I have all of the RN 1918 tables. I don’t disagree that there may be some subtle tweaks that would make the historic tables a bit more in line with current understanding of exterior ballistics. But in theory the range tables would have had actual firing trial data in which to assist in developing a ballistic coefficient\coefficients for a specific projectile. I’m not so concerned with the trajectory, terminal velocity, times etc as I am with the quoted 50% Zone figures. However this perhaps requires a different discussion.

Anyway that all aside – if you are employing a modified version of McTraj, can I ask how you are back calculating your descent angles? I’ve always used the last two range data outputs before the trajectory crosses the level line. Change in Height over change in range for the last two reads – take the ArcTan. Is this what you are doing?

Best Regards
JD
_________________
SHUFTI CUSH
Back to top
View user's profile Send private message
tone
Site Admin


Joined: 29 Oct 2004
Posts: 478
Location: Boston

PostPosted: Tue Apr 18, 2006 12:06 pm    Post subject: Reply with quote

jwduquette1 wrote:


On form factors, McCoy did a number of correlations to nose length and form factors before his death. Much of this has been condensed into a series of figures presented in his work on exterior ballistics. For a simple crh ~4 head, a form factor of about 0.64 to 0.7 is in tune with his findings. So I felt like my back calculated value of about 0.65 was in the right range.

You’ll notice on the RN range tables that the form factor is indicated on the first page of each set of tables. In the case of heavy 13.5” CPC the “ko” (kappa sigma I think) factor of “0.7” is the form factor. That’s what British employed for determining the final BC for 13.5” CPC.

BC = w/(i x d^2)

W in lbs
D in inches

I typically bow to range tables or firing tables if I have them. I have all of the RN 1918 tables. I don’t disagree that there may be some subtle tweaks that would make the historic tables a bit more in line with current understanding of exterior ballistics. But in theory the range tables would have had actual firing trial data in which to assist in developing a ballistic coefficient\coefficients for a specific projectile. I’m not so concerned with the trajectory, terminal velocity, times etc as I am with the quoted 50% Zone figures. However this perhaps requires a different discussion.


Thanks for the information that the ko is the form factor. It is good to talk to someone familiar with the style of expression common in the time and still (to some degree) in all the sniper wanna-bes out there.

The fact that such treatments are still used boggles me, as it is woefully approximate when mach varies significantly over the course of a projectile. I guess it matters little over simple rifle fire trajectories, but it is just an unsound basis for anything more.

In 1918, however, the excuse for this sort of "one BC for the weapon throughout its envelope" is understandable because they lacked the ability to create and run a physics loop. But I wonder why they did not take cleaned-up and splined together results of a good test firing series, as such a method would intrinsically embody the non-constant drag of the real world. I think that would have resulted in more accurate tables than the arcane method they employed.

You can find an extract that details the method used to compile the range tables HERE (3 MB PDF). I think you'll agree that the result would leave much to be desired at the level at which we are presently discussing it. Indeed, I feel downright unapologetic that I fudge the output of an app like McDrag to produce my mach-to-drag curves, as the RN's range tables seemingly reference a generic ballistic profile flavored by a constant ballistic coefficient and form factor. That some tables say "at elevations above N degrees, use this other BC" further demonstrates the deficiency of the simple model they employed.

Quote:

Anyway that all aside – if you are employing a modified version of McTraj, can I ask how you are back calculating your descent angles? I’ve always used the last two range data outputs before the trajectory crosses the level line. Change in Height over change in range for the last two reads – take the ArcTan. Is this what you are doing?


yes... simply that. I don't do any fancy integration at all -- I assume a simple step-wise integration is good enough. I was unaware of the existence of McTraj -- my outer loop and formulae are my own, and the trajectory is calculated using 3D coordinate spaces and orientations -- something far beyond the static vision of other ballistic apps. This makes my code amenable to use in 3D simulation.

I've tried timeslices of 1/20th and 1/40th of a second. Differences between the two are comfortably minute. I think that over-attention to these issues and related ones such as fancy integration methods is a common fault of such programs. That said, some custom interpolation is desireable in some cases, such as interpolating time-of-flight so the granularity of the timeslicing is massaged away to a large extent. Keep it simple, but keep an eye open for where simple != sufficient.

Re: 50% zones, etc

I support these in a sufficient but impure manner. If I have 50% zone data, my app allows me to input the points available and, upon firing a shell in simulation (that is, NOT in a simulation shot being used to compile a range table, but one where 50% errors are desireable to deliver the truest simulation), my code fudges it this way:

1. upon firing, look at MV and angle of departure and calculate a rough range for the shot
2. pick an interpolated depth and breadth for the 50% error likely to result at that range
3. roll some dice (so to say) to come up with an error THIS shot will have in range and lateral deflection.
4. fudge in a constant acceleration that will deliver exactly this error over the calculated time-of-flight

Pure? No. Statistically conformant to the best researched data available? PERFECTLY SO.

My drift simulation is likewise fudged in, and is not a result of a true physics model: an impure method which delivers a perfect result.

tone
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
jwduquette1



Joined: 20 Dec 2005
Posts: 90

PostPosted: Tue Apr 18, 2006 1:18 pm    Post subject: Reply with quote

tone wrote:
In 1918, however, the excuse for this sort of "one BC for the weapon throughout its envelope" is understandable because they lacked the ability to create and run a physics loop. But I wonder why they did not take cleaned-up and splined together results of a good test firing series, as such a method would intrinsically embody the non-constant drag of the real world. I think that would have resulted in more accurate tables than the arcane method they employed.


Maybe I have misinterpreted what you have said. If so I apologize for the following bit.

Even back in these fledgling days of exterior ballistics they didn’t assume drag was a constant. It varies as a function of the MACH number. The purpose of the common drag functions was to simplify the arduous calculations entailed for even one launch angle. The ballistic coefficient normalizes the common drag function to the foibles of a particular projectile and its unique form. The drag coefficient varies considerably. So while the ballistic coefficient was a constant (probably incorrectly so as you have pointed out for the higher angle shots), the drag coefficient is not.

Much of my interest in this area stems from trying to determe armor performance. To that end, I am of course most interested in descent angles and terminal velocities for given range settings and the like. Consistency is of course key. Moreover if the exterior ballistics model I am employing for projectiles that I do not possess firing tables is resulting in differences relative to projectiles for which I do possess firing tables, than I am not really comparing the two projectiles on an even basis – or fair playing field. Course the only way to tell is testing the model vs. actual tables. Hopefully that made since.

Is it your intent to at some point develop a computer game/simulation and sell it? Or is this a research tool you are creating?

Best Regards
JD

P.S. Thanks for the link to the RN Gunnery Manual.
_________________
SHUFTI CUSH
Back to top
View user's profile Send private message
tone
Site Admin


Joined: 29 Oct 2004
Posts: 478
Location: Boston

PostPosted: Wed Apr 19, 2006 1:14 pm    Post subject: Reply with quote

I'd love to make a commercial game.

I recently spent a week, however, isolating the portion of my old Java prototype that is needed to make the HTML range tables I linked to. I could consider making that available to others (in source code or compiled form). I have yet to create a nice GUI for it and would rather not, but if someone knew how to do this blithely it would clearly make a great web applet.

My Range Tables page explains in fair detail my methodology.

The bottom line is that I need these inputs to do a good job at calculating pretty darn accurate range tables for a weapon system:

Essential Inputs:
1. Shell mass
2. Shell diameter
3. Muzzle velocity
4. a reliable range/elevation pair at the highest possible elevation

Optional input:
1. CRH for the shell (describes its pointyness, as you know)

Luxury inputs (needed to simulate an actual gun, not merely calculate a range table):
1. 1 or more range/lateral drift pairs
2. 1 or more effective-full-charges to reduction in MV pairs
3. 1 or more 50% zone figures

tone
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
jwduquette1



Joined: 20 Dec 2005
Posts: 90

PostPosted: Thu Apr 20, 2006 6:59 pm    Post subject: Reply with quote

Hey Tony:

If your web site’s video and graphical ship models are any indication of what such a game would look like, I'd be first in line to plunk down my cash for a copy. But I doubt I am representative of the popular computer game market. I still reminisce about Avalon Hills old board game "Jutland".

Regarding what might be included under the trunk, if your discussions on exterior ballistics and gunnery are any indication of where you are headed, it would certainly be the first time that I can think of that a computer game designer went into this level of detail and accuracy. Most computer naval game designers would skirt over this sort of thing in favor of graphics. Don't get me wrong -- eye candy is what sucks you into a game -- initially. But computer game designers typically gloss over ballistics and gunnery in favor of presenting something that they think "feels" about right. Than can mean anything.

I'd love to look at your base code for firing table development. I haven’t had any issues getting reasonable match ups with your descent angles with my own various runs. However I haven’t been able to match up terminal velocities with your output for German 11" APC. For intermediate ranges of say 12 to 17 thousand yards my terminal velocities have been running anywhere from 10% to 25% higher. I'm perfectly willing to concede that the error is on my end, but I would be interested in seeing how you are modeling drag.

Best Regards
JD
_________________
SHUFTI CUSH
Back to top
View user's profile Send private message
jwduquette1



Joined: 20 Dec 2005
Posts: 90

PostPosted: Mon Apr 24, 2006 8:52 am    Post subject: Reply with quote

Hit Percentiles from Jutland. From John Campbell's "Jutland, An Analysis of the Fighting", Conway Maritime Press, 1986. Pgs 354 - 355.

Campbell's numbers indicate that Beatty's Battle Cruiser's over all hit percentage was lower at Jutland than at Dogger Bank.




_________________
SHUFTI CUSH
Back to top
View user's profile Send private message
Adrian Dobb



Joined: 13 Nov 2005
Posts: 41
Location: Devon, UK

PostPosted: Tue May 02, 2006 3:20 pm    Post subject: Reply with quote

Well if the figures are accurate that is by comparison quite poor.

Has anyone read an article in Journal of Military History (62, Jan 1998) by Nicholas Lambert entitled 'Our Bloody Ships or Our Bloody Sytem? Jutland and the loss of the Battlecruisers' I mention it as I remembered it has a number of things to say about the propellant. Having re-read the article, what it says is a little different to what I remember but worth mentioning all the same. Briefly and amongst a number of other points, t appears that pre-war the ammunition allocation per gun was increased. This was ok for shells as the shell room had the capacity but the magazine was short of space so propellant cases were stowed on the floor in magazine gangways. This coupled with the BCF ethic of rate of fire over accuracy resulted in a dangerously overprovisioned supply chain between magazine and gun. There is much more besides this in the article including Jellicoe suppressing a report by rear Admiral Tudor (3rd sea Lord) essentially to the effect that this was the cause of the losses. It was particulary contoversial as it implied that therefore the ships crews own lax cordite safety procedures where a part of there own demise. Tudor's hand however was also behind the original increase in projectile allocation. Anyway well worth a look I think if you haven't already seen it. JoMH is on JSTOR if you have a university web address.

NB Lambert cites New Zealand firing 422 shells and with at best 3 hits she had a 0.7 percent hit rate.

Adrian
Back to top
View user's profile Send private message
desmond



Joined: 04 May 2006
Posts: 2
Location: Austin, Texas

PostPosted: Thu May 04, 2006 9:53 pm    Post subject: british vs german battlecruisers Reply with quote

British naval vessels from the pre WW 1 and WW 1 era were designed and required to serve anywhere on the globe. Thus they had to be minimally liveable in everything from arctic to tropical seas and the crews lived on board. German naval vessels from the same period were designed to fight the Royal Navy's Grand Fleet and defeat it in the North Sea. The crews lived onshore in barracks and were on board for a few hours up to a few days at a time. Thus less consideration was given to livability than in British vessels.

In addition it must be remembered that British industry was in decline by the late 1890s and early 1900s and was no longer the technological leader it had been a few decades previously. Germany, in contrast, had become the leading technological nation in the world with the U.S. serving as its rival over the long period of British decline from the 1890s through to the Second World War. German ships of the period had much better levels of internal subdivision, and in general German steel was better quality, their optics were of superior quality, and consequently their fire control was better. They also produced better quality heavy guns although the British 12 inch and 13.5 inch guns were successful armaments. The British 15 inch gun of the day had a fault in its armor piercing shells which caused them to explode on impact rather than piercing the innards of German ships. To a lesser degree all British armor piercing shells (the 12 and 13.5 inch) were defective and were more likely to explode on impact than were German shells (this had something to do with a less sophisticated tip on the shells which often was cracked in the British models.

Add to this the emphasis that the British BattleCruiser squadron placed on rapid fire and the attendant sloppy shell handling practices that led to the magazine explosions that destroyed the three British BCs at Jutland. German fire control was better, used far better optics and was more effective at long range than British practice. A more or less equivalent system was developed in the UK but the Royal Navy declined to use it preferring an inferior "in house" system as I recall. Beatty drilled his squadron to fire as rapidly as possible and to increase their rate of fire the turret and barbette crews left all the flash proof doors open while firing to facilitate rapid shell and propellant movement. Even worse the British handled the cordite in silk bags which proved to be very flammable under the right circumstances.

At the Dogger Bank Battle the Seydlitz almost blew up from similar practices and was saved by heroic crew member who turned red-hot valves flooding the magazines and saving the ship. The Germans learned their lesson and observed scrupulously careful ammunition handling during battle from then on. The incident on the Lion did prompt the British to the same conclusion and they continued courting disaster in their shell and propellant handling.

Thus the inferior RN ships, with thinner armor and dangerous ammunition handling practices, and also inferior targeting technology engaged the HSF battle cruiser squadron at Jutland. Then to make the situation even worse the British fought much of the battle silhouetted with the sun behind them while the Germans were difficult to see with only gray sky behind them. Of course nobody knows exactly what happened on the British BCs that day but the thin armor is probably not the critical factor. The dangerous shell and propellant handling practices undoubtedly led to the destruction of Beatty's ships. It in notable that the Queen Elizabeths fought well and took numerous hits without the disasters that overtook the BCs. The QEs were innovative ships that, as other commentors here have noted, went far towards making the BC concept moot.
Back to top
View user's profile Send private message
jwduquette1



Joined: 20 Dec 2005
Posts: 90

PostPosted: Fri May 05, 2006 3:56 pm    Post subject: Reply with quote

The British examined the FCS on either Bayern or Baden after the war. They described the German fire control system as quite “primitive” compared with their own technology. They did however comment that the optics on German Range Finders were superior to their own material.
_________________
SHUFTI CUSH
Back to top
View user's profile Send private message
desmond



Joined: 04 May 2006
Posts: 2
Location: Austin, Texas

PostPosted: Fri May 26, 2006 2:09 am    Post subject: German fire accuracy Reply with quote

Thank you for catching my error about the High Seas Fleet fire control. I checked and according to "Battlecruisers" pp. 90 - 95 by John Roberts (Caxton Editions, 2003) British fire control was more advanced though indeed they had chosen a somewhat less technologically sophisticated system (developed within the RN) named the Dreyer Table over the more sophisticated "Argo Clock" system. The Argo Clock system would have had greater ability to contend with movement of the ship it was installed on while tracking the target ship. In fact the 12 inch RN Battlecruisers even used a simplified version of the Dreyer Table. Germany led the world in optics for many years including this period and actual optical equipment on German ships would always be top quality.

Nonetheless the German fire was more accurate than British fire. Germans more quckly found the range and began to score numerous hits. Again according to this secton in "Battlecruisers" the British used a system of firing single shells and waiting to see where they fell and correcting until they got the range. The Germans would fire three salvoes in quick succession, walking the range out over about 1000 yards, without waiting to see where the shells fell. They would then observe where the shells of these salvoes did fall and base their succeeding fire on that. This enabled them to find the range more quickly than the RN and begin hitting their targets.

Of course the entire situation was complicated by weather, poor visibility, smoke, spray, and the impossibility of detecting hits by Armor Piercing shell that worked well (i.e. actually pierced the armor and penetrated into the innards of the target ship). Fire spotting positions were moved up out of armored protected areas into exposed mastheads and still obeservers sometimes never saw much of anything. The British did have the disadvantage of being somewhat silhoueted by the light while the High Seas fleet was generally less visible.

In both cases the crews were firing at tiny targets and under difficult conditions. It is a testament to scientific warfare that they managed to hit anything.
Back to top
View user's profile Send private message
Harley



Joined: 23 Oct 2005
Posts: 131
Location: Great Britain

PostPosted: Fri May 26, 2006 7:53 am    Post subject: Reply with quote

Out of interest, what are your sources for the judgement on British fire control? From my experience there are many different accounts regarding the effectiveness of British fire control at Jutland and elsewhere, and unfortunately there are a large number which rely purely on hearsay which is especially true of retrospective "popular" accounts of the Battle of Jutland.
Back to top
View user's profile Send private message AIM Address MSN Messenger
Horsa



Joined: 09 Dec 2005
Posts: 32
Location: England

PostPosted: Fri May 26, 2006 8:40 am    Post subject: Reply with quote

I believe Desmond's ammended post ( last one) is about as close an asessment of relative fire control methods as any you are likely to get.

I'd add two things :-

There is some arguemet about how much of an improvement the Argo system was over the Dreyer Table. Some would say it was very marginal, so it all depends on who you want to listen to. I really don't know.

At the Dreadnought Conference Andrew Lambert made the point that the German rangefinding system was indeed very effective but relied on carefully matching up some etch marks ... a process that was very dependent on a largely undistracted and/or alert operator. As Andrew Lambert put it so well himself " the German shooting was fine until they started taking a pasting " . In other words , once the operators got fatigued or shot at, German rangefinding deteriorated. The British system was less effective initially, but held up well under protracted battle conditions.
Back to top
View user's profile Send private message Send e-mail
jwduquette1



Joined: 20 Dec 2005
Posts: 90

PostPosted: Sun May 28, 2006 8:53 am    Post subject: Reply with quote

About two years back, I had been looking for information on intro dates for various improvements conducted by the RN during WWI on larger caliber shells – ballistic caps, penetration caps, and bursting charge improvements. I had been poking about the libraries of the nearby universities for such information. I was merrily photocopying pages from various books I was pulling from the shelves. I came across the following bit which was the source of my original comment. Unfortunately I had assumed that the title of book was included in the header or footer of the page I was photocopying. It wasn’t. So I don’t have an actual cite.

I assume this snippet is derived from conclusions put forth in a post war ADM report on the Baden. However, as yet I have not been able to track down the precise ADM report title that details the post war RN examination of the Baden’s FCS. Baden permeates a number of post war ADM reports – such as Progress in Gunnery Material in which firing trials were conducted against the Baden’s Armour plate. So I am certain the primary source of the FCS exam is floating about – perhaps at PRO, Kew. It would be nice to track this down as the amount of information – at least in English – detailing German WWI naval FCS seems to be fairly limited – or non-existent.

Best Regards
JD


_________________
SHUFTI CUSH
Back to top
View user's profile Send private message
tone
Site Admin


Joined: 29 Oct 2004
Posts: 478
Location: Boston

PostPosted: Mon May 29, 2006 12:03 pm    Post subject: Reply with quote

Quote:
The Argo Clock system would have had greater ability to contend with movement of the ship it was installed on while tracking the target ship


While you make some good points on rapidity of finding the range, this particular statement is not true.

The Dreyer table Mark III (the first regular production version) was fully helm free, with a gyrocompass continuously and automatically keeping own ship's heading current on the Dumaresq, requiring only that own ship's speed be manually set -- a tiny deficiency considering how infrequently and slowly speed changed. Indeed, an indifference to own ship's maneuvering was a core strength of the Dreyer table (I will have to check on the simpler Mark I tables retrofitted in the earliest dreadnoughts to see if they were also helm-free).

There are two schools of thought on the relative merits of Pollen and Argo, with the widely influential work of Prof Jon Sumida ("In Defence of Naval Supremacy" from 1989) having the greatest mindshare and a recent countervaliing conclusion from John Brooks ("Dreadnought Gunnery at the Battle of Jutland" from 2004) is only just starting to create a sense that there is something to discuss on the point.

While I know much more about the Dreyer tables than the Argo system, my own opinion on this is that the Dreyer was the better system for the RN, and any failures that were experienced with it would have been realized sooner and more confoundingly in ships that instead employed the Argo gear in the form envisioned by Pollen.

The Pollen system encompassed some brilliant design elements, but when delivered in the complete form Arthur Pollen intended was hamstrung by a fascinating but useless plotting device that traced from the input of a single rangefinder/bearingtaker a true course plot of the battle -- a sketch of great value as a souvenir of the battle but of little direct utility in hitting your enemy.

The Dreyer table plotted something much more useful -- the ranges over time -- and this sample set was directly useful in determining the range rate, a vital discovery indeed. Moreover, the Dreyer was amenable to having a number of rangefinders supply data to inform the solution whereas the Argo was cyclopic. The Germans, in a slightly different manner, employed a similarly robust system that reflected the reality of the first problem confronting a fire control computation: that accurate range cuts will not arrive in sufficient number.

And yet, everyone keeps on about Argo. Why? I think part of the answer is that the machines were beautifully made and more reliable in their variable speed drives (they were less prone to slippage than Dreyers' were when their rate was altered). Additionally, Dreyer clocks modified their rates in discrete steps of 25 yards per minute whereas the Argo could do it in a purely continuous manner. That is a nice hallmark indeed. But how nice?

"Dreadnought Gunnery at the Battle of Jutland" makes it pretty clear that the frailties of the Dreyer gear in this regard was not a limitation in practice. Taking a fairly cartoon-like example that would stress the Dreyer-Elphinstone range clock to the utmost (two 30 knot ships on opposing courses, passing beam to beam at 7000 yards -- a scenario that would require the clock to compute a range whose derivative was changing rapidly and continuously), the Dreyer clock never wandered off a perfect modeling of the range by an amount greater than the salvo pattern. This, I feel, is a resounding illustration that the quantization is not a true deficiency in the Dreyer equipment despite any claims to the contrary.

But what of issues of slippage? There are remarks in Dreyer handbooks to the effect that some clocks had slipped in practice, but these failures are only ever noted to have occurred in combination with unauthorized (and unspecified) adaptations to individual Dreyer tables perpetrated by clever officers who wound of placing additional load on this rather weak variable drive device, or in cases where oil had slipped into the friction interface between a spinning plate and a wheel. Argo's clock was superior in many fine points, and the non-slip performance of the variable speed drive was one, but I think it unproven that the inferior design of the Dreyer in this regard was ever demonstrated to the misfortune of its users. I think it is possible that many people who are impressed by the elegant design of the Argo clock and by its ability to reliably function under a continuously changing speed have simply never realized how small the errors are which accrue when this optimal performance is compared to the Dreyer/Elphinstone range clock operating under a step-wise, discontinuous change in speeds. Was the Argo clock better in this regard? Yes. Did this superiority amount to a measurable advantage over the Dreyer counterpart? No.

That said, I should repeat that I do not know as much of Argo gear as I do of Dreyer equipment, and I don't pay any mind to other points of contention that surround the two devices such as issues of plagiarism and political influence, as they are not relevant to a comparison of the tactical merits of the two fire control systems.

One thing I am sure of, however, is that the output of a ship's fire control solution whose crew is relying on a Dreyer table would be CLOSER in character to a German ship's efforts than would an Argo table-equipped crew's shooting for the subtle reason that the German fire control, though differing in many ways from the RN's, also centered its data collection on the ability to average the range cuts being supplied by a considerable number of rangefinders. I would not want to be shackled to a fire control process that incorporated the Argo plotter when there were alternatives that allowed a quorum of rangefinders to inform the fire control solution. I would, on the other hand, feel quite at ease using either a purely Dreyer system or a Dreyer table Mark II which replaced the dumaresq and Dreyer/Elphinstone clock of the other Dreyer tables with the Argo clock and his well-crafted range clock.

tone
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
Display posts from previous:   
Post new topic   This topic is locked: you cannot edit posts or make replies.    The Dreadnought Project Forum Index -> Discussion All times are GMT - 5 Hours
Goto page Previous  1, 2, 3, 4, 5  Next
Page 4 of 5

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group