THE GREAT DH6 DEBATE
Mick Davis repeats the Grahame White assertion that the ‘swamp spruce’ supplied for building the DH6 was inadequate. Notes are made about the mods (including removing a forward section of the wing to reduce the extreme chamber) and the point is made that many of these aircraft never received all the mods. Comments are made about accidents – though these seem few in number – and, if anything, Mick stresses that the DH6 was a lot better than popular history records as it has (unfairly) been in the shadow of the Avro 504. Generally, it seems, the DH6 was used as a replacement for the Shorthorn and Longhorn and mostly used by units training Corps pilots (RE8 etc). The Avro went to scout pilot training units as a priority, presumably because of its better performance.
I am forced to enquire whether the aircraft were supposedly burnt after they had been accepted by the RFC or did they fail the acceptance test. If they failed the acceptance test then I would think that this would make G-W liable to replace them and I would have thought that there would be a record of litigation about this. As the materials to be used were specified as part of the contract (or because G-W was ‘Government Controlled’ ?) then G-W could not have been liable. Furthermore, the components would have been passed by the AID prior to assembly; again negating any G-W liability. In any case, the AID inspection system existed to identify substandard components BEFORE they became part of the aircraft. If the aircraft were accepted then they would have been allocated serial numbers in line with the allocation of the contract. Presumably the serial numbers of the first batch are known, or can be found easily from the Crowood or ‘British Military Aircraft Serial’ books ? If this is the case then all that needs to be done is to find photographic and/or documentary evidence of first batch aircraft in service. I rather think that this is part of G-W’s ‘rage against the machine’ or, to be kinder, a lapse of memory combining issues relating to the poor quality of materials supplied to G-W and the post-Armistice destruction of surplus aircraft.
The Ministry of Munitions responded to the limited supplies of seasoned wood by setting up factories to kiln dry the wood prior to use.
The Air-Britain RNAS serials book confirms that many of these were retained
for anti-sub patrols :
A9565 – 252 Sqn by 8.1918
A9569 – 260 Sqn by 5.9.1918
A9570 – Cramlington by 16.4.1918
A9571 – 507 Flt by 5.1918
Of the second batch, C2016, C2021, C2067, C2078, C2098, C2119, C2149 etc were all retained for these duties too. C1972, C2101 & C2136 survived to gain postwar civil registrations. At a time when a surplus Avro was £50 the buyer could afford to be choosy and so it is fairly clear that there was nothing wrong with these aircraft.
The other thing than concerns me about G-W’s tale is the assertion that the shortage of spruce was due to the Americans entering the war in April 1917 and the ‘commandeering’ of American spruce for the US aviation industry.
To what extent is that true ? I rather gained the impression that the
Americans spent a great deal of 1917 deciding what they were going to build (ie Hispanos, Liberties, Bristol Fighters, DH4s) and reorganizing the infrastructure to do it. As the AEF was equipped with European built aircraft does it make sense to risk the loss of these supplies by withholding timber ?
Perhaps the more likely explanation is that the U-boats were sinking too many timber carrying ships from the US and so alternative supplies had to be found?
Does anyone know where cypress comes from ?
Also, work on constructing government subsidized kilns for drying timber doesn’t seem to have started until June 1918. This seems to be in response to the increased demand for timber brought about by the creation of the National Aircraft Factories which started production a few weeks before this. Bearing in mind that it took a matter of weeks to set up these kilns (two were in operation in July 1918), one wonders why they were not built in 1917 if the problem was identified then ?
The origin of the claim about the use of unsuitable cyprus timber for the construction of Grahame-White built DH6 trainers comes from Graham Wallace’s 1960 biography. There is, for instance, no mention in the 1957 ‘British Aeroplanes 1914-1918’ by Jack Bruce – understandably established as the definitive source in 1960.
One can cast doubt upon the accuracy of Wallace’s reporting : the dates are not clear and the sources, other than hearsay via G-W, are not quoted.
One can question G-W’s motives : the DH6 was chosen over the Hedderwick
designed G-W trainer; G-W had persistent problems with ‘red tape’,
particularly with AID inspections which he thought too harsh (he wasn’t
alone in this); G-W was clearly still embittered over loss of Hendon. Added to this is the fact that Wallace’s interviews with G-W were 40+ years after the event.
However, some supporting information does comes from Penrose’s “British
Aviation : Great War & Armistice”. Penrose asserts that Lt Colonel ACH
Maclean, CFS, submitted an urgent report to ‘HQ RFC’ (in England ?) noting that the wood used to built DH6s ‘splintered on every heavy touch down’. Penrose goes on to note that it was General Alexander, head of the Air Board Contracts Department, who was contacted by G-W.
In 1918 Lt Colonel Mervyn O’Gorman (late of Farnborough) was appointed as chairman of the Accidents Committee to investigate why the DH6 was a
‘dangerous trainer’. The inference here is that this led to the recommended modifications rather than large scale scrapping.
Returning to Wallace’s book, there is mention of a ‘high number of crashes at CFS’ (ref Maclean’s report above) and the order canceling the use of cyprus timber 3 months later.
Without study of the individual aircraft record cards it is difficult to assess the true number of crashes due to structural weakness. However, if fatal crashes are taken as an indicator then seven aircraft were lost in 1917 and a further three in January 1918 (when I stopped looking). These were :
11.7.1917 A9579 16TS ; 12.8.1917 A9619 66TS ; 1.11.1917 A9595 37TS ;
12.12.1917 A9744 31TS ; 22.12.1917 C1953 17TS ; 23.12.1917 B2631 1TDS* ; 23.12.1917 B2656 1TDS* ; 7.1.1918 A9637 20TS* ; 7.1.1918 A9593 20TS* ; 26.1.1918 A9669 53TS
* indicates aircraft lost in mid-air collisions which can be discounted. Of the others, 5 aircraft were from the G-W first batch and one from the second. For a WW1 primary trainer was this an unusually high level of loss ?
The ‘high number of crashes at CFS’ (ref Maclean’s report above) led to the destruction of 100 G-W built DH6s, together with the parts for another 75.
G-W received orders for 750 DH6s, though this seems to have been in four batches : A9563-A9762 (200); C1951-C2150 (200); C7601-C7900 (300);
D951-D1000 (50). The first batch order for 200 aircraft was placed on 13 January 1913.
37 G-W DH6s had been accepted by the RFC by June 1917 (presumably all from the A9563-A9762 batch). The C1951-C2150 batch was ordered prior to the end of August 1917 (by which time the C series had been allocated) but I have yet to discover when production started. Given that G-W built only 37 aircraft in the first 6 months of 1917, I think it fair to assume that few of the C series aircraft were built that year (though note that C1953 was lost at 17TS on 22.12.1917). Note also that the G-W factory was supposedly idle for some weeks due to a lack of materials.
So, according to Wallace the G-W factory in less than 6 months built at
least 163 aircraft (plus at least three from the second order) plus another 100 aircraft which were scrapped plus components for another 75 aircraft. All this in spite of the fact that only 37 aircraft were built in the 6 months previous and the factory was idle for some weeks due to lack of materials.
What of the aircraft which had been accepted ? If they were of inferior
quality why weren’t they scrapped also, particularly as there was need to investigate their flying characteristics and they were deemed ‘surplus’ later (thus making them available for anti-sub work).
One can identify enough G-W first batch aircraft to confirm that they were both delivered and remained in service, viz :
A9564 68TS; A9564 21 TDS; A9598 207 TDS; A9599 207 TDS;A9659 260 Sqn; A9612 98 Sqn; A9611 121 Sqn; A9576 13TS; A9579 16TS; A9580 16TS; A9582 17TS; A9588 20TS; A9590 20TS; A9591 20TS; A9593 20TS; A9611 25TS; A9590 50TS; A9607 53 TS; A9619 66TS;
According to Wallace the first aircraft were accepted but suffered structural failure in service as a consequence of which all components in production, and complete but unaccepted aeroplanes were ordered to be burnt. I am currently looking for evidence of the crashes. The instruction for the burning I have been unable to find although I haven’t been able to access whatever G-W papers they may have at Hendon.
G-W’s personal photo albums are with the R Ae S at Hamilton Place but I don’t know if any company records found their way there too.
Apart from details of those DH6s which survived for service after U-boats – which appear in the Air-Britain RNAS serials book – I guess that the only way to determine details of the non-fatal crashes will be to consult the aircraft record cards. I assume that these must still exist, and be largely intact, as they form the basis for the Air-Britain series.
The first G-W DH6 was accepted some time between 13 January (when the order was placed) and June 1917 (by which time 37 aircraft had been accepted). The first fatal DH6 crash is not until 11 July 1917, and this is A9579 the 16th production aircraft. All the G-W DH6s which crashed between 11 July 1917 and 26 January 1918 (inclusive) which came from the first G-W order batch were aircraft accepted prior to June 1917 (ie part of the 37 aircraft). Furthermore we know that aircraft accepted prior to June 1917 were still flying in August and September 1918. Therefore I must conclude that either the Wallace claim cannot be supported or first batch aircraft were not built of cypress.
However, if one considers the second batch aircraft (C1951-C2150) that is another matter. This order was placed before the end of August 1917 (when the serial allocation was complete) and the first of these aircraft must have been delivered before 22 December 1917 as we know that a 17TS aircraft from this batch crashed on that date.
The aircraft concerned, C1953 (3rd aircraft built), is the only 2nd batch aircraft to suffer a fatal crash prior to 26 January 1918. So was the second batch built with cypress ?
I will have to check Hobson for 1918 to look for further crashes with 2nd batch aircraft and also re-check the Air-Britain book for surviving aircraft which became ‘scarecrows’. From the notes to hand, I have no scarecrows with serials below C2016; though C1972, C2101 and C2136 all survived to gain civil registrations. Assuming that these were OK – using the ‘why buy a dodgy DH6 when you could have a good Avro for £50’ rule of thumb – we will have to be looking at the first 20 aircraft of the second batch as being possibly built from cypress.
I have to say that the creditability of the reason for using cypress being the shortage of spruce looks greater in the latter half of 1917. I can imagine Weir agreeing that lower quality materials could be used for second line aircraft, especially as the Ministry of Munitions priority was, by then, moving towards the increased production of aircraft which would have greatly increased the demand for spruce this side of the pond as well.
Wayne Biddle’s ‘Barons of the Sky’ – if what he says is true – I would not be in the least surprised to learn that there was large scale buying of spruce in the US in order to control the market and make a killing when Government contracts were issued.
Inevitably the scale of everything in the US was/is greater and G-W’s clams – if they are true – pale into insignificance compared with the 1600 Standard J1 s which were condemned in June 1918 after “tests revealed dangerous vibrations and fire hazard due to an engine mismatch”. This resulted in a write-off of $17.5 million !
This aircraft was unrelated to the DH4/DH9 series and was a 2 seat biplane trainer and attack aircraft with 36′ wingspan and a Royal Aircraft Factory air cooled 90hp V8 engine. They were initially made by Grahame White’s or Airco and these aircraft can be told easily by the rounded top to the rear fuselage. Later versions made by a collection of 7 contractors had angular tops. The DH6 had a max speed of around 65mph and a ceiling of just 6,500′. Endurance was around 2.5 hours and it could carry up to 100lb of bombs. The aircraft was used for antisubmarine patrols by the RNAS and RAF and served at Waddington with 48 Training Squadron in the latter stages of the war, by which time it was totally obsolete. C2018, C7363, C9381 & C7299 all served at Waddington, the latter ending its days in a crumpled mess impaling the roof of the WRAF accommodation block! One aircraft was made in Canada in case the Curtiss JN-3 was a failure as their standard trainer.
These are the GW 2nd batch aircraft, which we know were ordered before
August 1917 and deliveries started before December 1917 :
22.12.1917- C1953 – 17TS
10.6.1918 – C1969 – 193TS
10.4.1918 – C1971 – 193TS
4.6.1918 – C2028 – 5TDS
17.9.1918 – C2039 – 193TDS
5.11.1918 – C2045 – 18TDS
5.6.1918 – C2046 – 21TS
18.9.1918 – C2093 – 35TDS
11.3.1918 – C2135 – 17TS
Note the absence of any aircraft in the C1972-C2027 range, but even if
these aircraft were accepted as part of the contract and then burnt prior to delivery that still falls short of the 100 aircraft noted by Wallace. The serial allocation for the 2nd GW batch was C1951-C2150. Again we know that C2135 was delivered prior to 11.3.1918, which doesn’t really fit with the idea of there being a 7 week delay between the stoppage of using cypress and the arrival of the replacement spruce.
These are the 3rd batch fatal crashes :
22.3.1918 – C7663 – 13TS
13.6.1918 – C7708 – 191TS
13.9.1918 – C7890 – 242 Sqn
This serial allocation was C7601-C7900. Note that at least 60 aircraft had been delivered by 22.3.1918, which again doesn’t fit with the idea of a 7 week production stoppage, nor does it fit with the “alarming number of crackups, some unfortunately fatal” supposedly reported by CFS.
I have some writing by Sir Ralph Sorley who claims that when stationed at Aldeburgh in 1919 he delighted, when the wind was strong enough, to fly up the High Street, throttle back at the top, and fly backwards down the street, He claims it would fly at speeds of less than 30 mph and had a top speed of about 75 mph. . He also commented that in his view,’ it was the answer to the instructor’s prayer because it was devoid of vices and you could do with it what you wanted. In fact it was too placid and did not do those things, which got you into trouble in most aeroplanes.’
Sorley flew B2659 at Mudros April-May 1918 and F3340 and F3389 at Yarmouth in January 1919.
This information seems to imply that the delay in G-W production of DH6 trainers was likely to have been during March/April/May 1918, with cypress being substituted for spruce in airframes built in November and December 1917 and January 1918. Therefore the ‘100 complete and 75 sets of spares’ claimed to have been destroyed would have been from late 1917/early 1918 orders.
Of the 3rd batch of G-W aircraft (C7601-C7900) only 163 can be accounted for, suggesting that the ‘complete’ aircraft could have come from this batch.
The 4th batch G-W aircraft (D951-D1000) are noted as having either ‘no record of delivery’ or were supposedly sent to Egypt as spares to assemble there on arrival. I have yet to find evidence that these aircraft were assembled. Therefore the ’75 sets of spares’ may have included this batch.
Other possibilities are two batches of AirCo built aircraft, for which G-W may have been a sub-contractor or which may have been included in the numbers given to the Air Board :
B9031-B9130 – 4 delivered, 96 cancelled.
D8581-D8780 – no record of delivery/cancelled.
C5451-C5750 Harland & Wolff – 100 built and 200 canceled.
Known builds from spares :
C3506 – built from spares at 2TDS (but prior to August 1917)
C994 – built from spares at 27ARD, Waddington
When war broke out in 1914 there were good supplies of aircraft grade spruce in stock in Britain. However, by October 1916, a shortage was forseen. Thus at this point it was agreed that the Admiralty would assume responsibility for buying all timber to be imported in to Britain, including inspection and shipping. This situation continued until October 1917.
When the US entered in the war in 1917 efforts were made by the US government to control and increase production of aircraft grade timber.
However supply continued to lag behind production and crisis point was reached in October 1917. During October and November no spruce arrived in Britain from the USA, and only 10 standards were delivered in December. By 5 January 1918 only 150 standards of spruce had arrived in Britain since the end of September 1917, against contracts for some 40 000 standards. Furthermore the U-boat war was at its height in late 1917.
Therefore the British government, which already controlled the supply of materials to certain industries, had to introduce even stricter controls and seek alternatives. The latter included wood, which did not meet the previously agreed specification.
The situation worsened in early 1918 despite increases in production of usable spruce. This was because the transportation of troops was given priority in response to the manpower shortage and the German offensives.
Timber imports increased to modest levels by May 1918, largely due to reorganization in the industry. By this time though British industry had evolved a ‘built-up’ spar design and identified alternative sources of supply.
The extent to which the timber shortage affected aircraft manufacture was limited because airframes were being produced quicker than the aero-engines needed to power them and so stockpiles had been created.
Source : The War in the Air, Volume VI, P80-1
So it appears that, whilst there were some problems with deliveries of spruce, the kindest thing that can be said for Wallace’s story is that it inaccurate. But how much that is due to G-W’s poor memory of a, probably inflated, claim for compensation, and Wallace’s need of a good scandal to help sell the book, we can only guess.
I think that the jury must still be out on this as there are significant blocks of DH6s (including G-W ones) which I have not accounted for to my satisfaction. I would like to view the aircraft record cards (or talk to someone who has) before I make any conclusions. I would like to know more about the new ‘built up spar’
mentioned.
To US Navy : C7736, C7741, C7745, C7747, C7753, C7754, C7756, C7759, C7760
Civilian DH6 : C7620 – G-EALS; C7739 – G-EAPH; C7763 – G-EAUS; C7768 – G-EARD; C7797 – G-EAVR
“Materials were no easier to get than management and labour. In September 1917 Weir spoke of shortages in timber, in three-ply (supplies from North Russia were cut off), in “dope”, in alloy steels, in ball bearings (the best came from Sweden, and Weir suggested collecting them by warship), and in scientific instruments. Here the country’s backwardness before the war was still a hindrance. “This country,” Weir said, “was not laid out for scientific instrument making, necessitating delicate plant and highly skilled labour.”
“Architect of Air Power : The Life of the first Viscount Weir of Eastwood 1877-1959”, by WJ Reader (Collins 1968).
So efforts were being made to increase home production and decrease imports of wood generally. However the imported wood replaced by homegrown wood was largely for mining (ie pit props). One assumes that home grown alternatives to spruce were looked at – and one must not forget that ash is used in aircraft manufacture, particularly for longerons – but the only alternative used for spars etc which we know of is cypress (and this via Wallace/G-W).
Also cypress is an imported wood in any case. How likely is it that there would have been stockpiles of cypress in Britain in late 1917 ?
We know that the need to give priority to shipping men over material in early 1918 slowed the recovery of spruce imports, but it was the disruption in supply (from the USA) which caused the shortage.
“Wood, in short, was more than ever indispensable. But we could not spare the tonnage space to carry this bulky commodity here from distant countries. As early as 1915, the Government set up the Home-grown Timber Committee to promote the development of timber supplies from our own native resources, and early in 1916, when the first steps were taken to effect a reduction in our imports, it was decided to cut down the import of pit-props and seek to obtain them more largely from sources in this country.”
“The Final Report of the Foresty Sub-Committee of the Reconstruction Committee, issued in May 1917, estimated that the 1915 timber imports
represented 75% and those in 1916 62% of the average imports in the five years before the war.”
“The serious shipping situation… at the beginning of 1917 called… for further reductions of imports. Lord Curzon’s committee…. recommended that of a total restriction of 500 000 tons, 200 000 tons should be at the expense of timber imports. The War Cabinet examined the Committee’s Report on 16 February 1917, and in order to ensure that such a reduction should be carried out without leading to such a timber shortage as would hamper our military efforts, we decided to appoint a Director of Timber Supplies at the War Office….. On 19 February, we decided at a War Cabinet meeting that timber imports must be cut down by a further 100 000 tons a month – making a total reduction of 300 000 tons a month.”
“Our timber imports in 1917 were … cut down very drastically. Whereas in 1913 they had amounted altogether to over 11.5 million tons, in 1917 they were reduced to 2 875 000.”
“War Memoirs of David Lloyd George, Volume 1″ (Odhams, nd)
I have obtained a list of the G-W papers held by the RAF museum at Hendon. These are mostly press cuttings and the like but the following might shed some small light on the DH6/ cypress discussion if anyone is nearby and can look them up;-
B779 Expenditure, balance sheets and accounts for G-W Aviation Co. Ltd 1914-1918.
B782 Agreement between G-W and ministry of Munitions relating to financing of government contracts (1-3-18)
Incidentally, the Admiralty was apparently dissatisfied with the quality of G-W built aircraft as early as 1915 ( see item B777).
The Royal Botanic Gardens at Kew has this to say about Swamp Cypress (Taxodium disticum);-
It is a deciduous conifer found in the lowlands of southeastern USA. It has straight grain, is non-resinous and of medium weight for a hardwood. It is said to contain much water but dries slowly and well. It is very durable and its strength compares well with European Redwood. So if G-W got a batch that was shortgrained he was very unlucky? I have checked its physical properties and although its specific gravity is slightly higher than that of spruce ( 0.422 against 0.396) it Modulus of Elasticity is equal to that of Spruce, its compressive strength, both parallel and cross grain, Modulus of Rupture, and bending stress are all superior. However its sheer, and tensive strengths are both inferior to those of spruce ( the latter being only 75%).
Ministry of Munitions
Department of Aircraft Production
Technical Department
TDI No 505 (incorporating amendments of TDI No 505a)
Handbook of Instructions
February 1918
Section I
Use of Standard Materials and Parts (p1)
1 All material for which Air Board material specification exists is to conform with that specification.
4 Complete sets of Air Board material specifications and standard
fitting specifications may be obtained on application to the
Technical Department.
5 Non-metallic material, such as dope, glue, varnish, 3-ply etc., is to be supplied only by sub-contractors approved by the Director of Inspection.
Section IV
Miscellaneous (p13)
214 The use of wooden tailing edges is discouraged on seaplanes, as they are liable to warp and distort.
215 all splices in longerons are to be the ordinary ‘chip-lapping’
scarf, well bound with fabric, or kite cord, which is to be glued
whilst wrapping on. The length of the splice is to be not less than 8 times the depth of the longeron and some positive grip is to be provided in addition to the glue.
216 Except in the case of small single seater machines, main spars may be spliced with an approved type of splice at an approved point in the spar.
217 As an alternative to splicing, spars may be of built-up
construction. Before use, approval is to be obtained for any suggested method of built up construction, from the controller, Technical Department.
218 Spruce for main spars, longerons and Interplane struts is always to be ‘Grade A’ spruce, excepting in the case of training machines.
219 Ribs may be strengthened in the way of the spar, by binding with
1/2″ glued tape, as follows :- Turn the end of the tape and tack to the upper flange with a brass tack, then wrap round the rib 1 1/2 times and finish off by turning the other end and tacking to the lower flange.”
In September 1917 supplies of spruce would have been getting very low, though this is before they stopped. As A9605 was a first batch aircraft this would suggest that either it was spruce built or the cypress was good enough ?
One thing we haven’t consider in all this is that the ‘acceptable standard’ for wood being passed may have been lowered as shortages became more acute. If this was so then one would expect that the lowest grade wood would have been used cDecember 1917 when stocks were at their lowest and before the gradual recovery in imports had begun.
DH6 A9605 was sent to Farnborough for structural testing in September 1917. C6350 (75 hp R-R Hawk) was at Martlesham Heath between 23 Feb 1918 and cMarch 1918 for testing.