The Vokes air filter controversy

Having entered service in 1938, the Spitfire was showing its design age by 1942, overtaken in the dizzy pace of wartime technological change. Although generally a fairly fast aircraft by contemporary standards, the middle-war period of mid-1941 to mid-1943 found it at its lowest point, relative both to newer enemy machines and to the new American types entering service in that period – such as the P-38 and P-47. The comparatively sparkling performance of the earlier 1940-era Mark 1 had suffered from the additional of extra war equipment but without the compensation of great increases in power. Although the 1941 Mark V was indeed faster than the Mark 1, in the meantime new German fighters with new power-plants had become a lot faster again, establishing a speed advantage of about 20 knots. Such relative mediocrity on the part of the Spitfire would only be cured by the large increase in power obtainable by the Merlin 60 series engines with two stage superchargers. These engines were applied to Spitfire production in the second half of 1942, giving the type a new lease on life through 1943-44 in its Mark IX and Mark VIII variants.

While the development of this new variant was underway, squadrons in North Africa and Malta flew the worst Spitfire variant of all - the Mark V (Tropical), burdened with yet more weight-gaining equipment additions, including the Vokes carburettor filter. This was fitted in order to filter out the dust on the rough dirt airfields of the North African theatre, which would otherwise have resulted in rapid engine wear, reduced power output and sharply curtailed engine lives. The addition of the bulky Vokes filter to the Spitfire’s nose cowling not only marred the type’s emblematic good looks, but also focussed attention upon itself as the conspicuous cause of all the Spitfire V’s performance shortcomings relative to the opposition. In actuality, the reasons for this mid-war performance gap were a lot more complex than that (mostly they had to do with supercharging technology), but this somewhat simplistic way of attributing blame nonetheless travelled swiftly to Australia, landing there with the RAAF’s early batches of Mark VC aircraft.

Alarmed by the prevailing horror stories about the Spitfire VCT’s performance, the RAAF’s Eastern Area HQ in Sydney was soon pushing for an abandonment of the Vokes carburettor air filter in order to restore the alleged 20 knot performance loss. While the wing was still headquartered at Richmond in late 1942, prior to deployment to Darwin, work was begun to produce a locally-manufactured temperate air filter at 2 Air Depot, in pursuit of an ambitious intention on the part of HQ Eastern Area to convert the RAAF’s Spitfire fleet back to non-tropicalised configuration.

In order to support its claims, Eastern Area HQ from the comfort of its Sydney premises made the highly questionable claim that airfields in NEA and NWA had ‘in the majority of cases, now been sealed’, and therefore ‘it is not probable that an excessive amount of dust’ would be encountered, leading to the conclusion that it was ‘undesirable’ for dust filters to be permanently fitted.[1] Besides the complacent assumptions and inaccurate data that underlay this assertion, the technical ignorance behind it is breathtaking: Removing the Vokes filter in the Spitfire meant dispensing with the whole under-engine cowling assembly, requiring the manufacture of a replacement cowling from scratch! Given the distance to manufacturing centres in the UK, this was easier said than done.

In the absence of manufacturer’s drawings of the factory-standard cowling, the engineers at 2 Aircraft Depot fabricated their own version in impressively rapid time, but the experienced Spitfire pilot Flight Lieutenant Robin Norwood, Flight commander with 54 Squadron RAF, could not escape the suspicion that the 2 AD cowling did not deliver the same performance as the stock original. Indeed, comparative tests throughout 1943 continued to produce engine performance irregularities which strongly suggest that the locally-engineered intake was not production-ready.

Even had 2 AD’s intake been perfectly engineered, HQ Eastern Area’s position on the Vokes Filter shows little if any awareness of practical production issues: in a context where the RAAF disposed of only marginal reserves of Spitfire aircraft, and in the impending context of a defensive campaign over Darwin where losses were bound to be significant, it was scarcely credible to suppose that the Spitfire fleet could be changed over to the local cowling without major disruptions to the supply of aircraft to the frontline squadrons.

Fortunately, HQ Eastern Area’s overly-sanguine ideas were immediately challenged by the Air Board in Melbourne: The Air Member for Engineering and Maintenance reiterated the RAAF’s policy for dust filters to be fitted to all aircraft because of the ‘almost universal premature wear of aero engines’ that had already been demonstrated under war conditions on northern Australian airfields. Moreover, the AMEM emphasised the especially ‘bad’ position of the Spitfire’s carburettor intake - fitted low under the nose and thus close to the dirt – and emphasised the reality of the ‘dust menace’ on the undeveloped airfields of the NWA, which made any taxying a serious threat to engine life, especially in the absence of a dust filter.[2]

Meanwhile, at RAAF Richmond, Group Captain Walters, CO 1 Fighter Wing, sought the advice of Robin Norwood concerning this matter of Spitfire performance. Norwood had been flying Spitfires since 1940, and had 500 hours on a range of marks from the Mark.1 to all models of Mark V – all of which had had the temperate intake. Contrary to expectation, he refuted the VCT’s poor reputation:

'These aircraft at height, with the Vokes Filter, are just as good, and probably better, than the Mark Va & b’s we used to fly at home, and will, I think, give a good account of themselves…at height these are the best yet. I make great insistence on height but then these are designed for high speeds high up, not low down.'[3]

Norwood’s emphasis upon height relates to the fact that the RAAF’s Spitfires were fitted with the high altitude Merlin 46, which produced its maximum power output at 21 500 feet, rather than the 11 000 feet rated altitude of the Merlin 45 fitted to most Spitfire V aircraft in the UK. The Merlin 46 produced a modest 1115 hp at take-off, but thanks to its supercharger was still producing 1150 hp at its rated altitude of 19 000 feet, and with a maximum power output of 1210 hp at 21 500 feet (both outputs were achieved at 3000 rpm and plus 9 pounds boost).[4]

However, to settle the matter, conclusive data had to be obtained. In pursuit of this, HQ Eastern Area rapidly initiated comparative tests between standard aircraft fitted with the Vokes filter and aircraft modified with one of the temperate carburettor intakes fabricated by 2 Aircraft Depot, delivering the modified aircraft to Richmond in December 1942. Norwood conducted the trials, finding that although the standard VCT was ‘not particularly fast’ below 10 000 feet, and although its climb rate was 100-200 feet per minute less than the aircraft with the temperate cowling,[5] the speed difference was minor: about 3 knots under 10 000 feet, and no greater than 8 knots up to 20 000 feet.[6]

It is noteworthy that the standard VCT aircraft demonstrated a maximum speed of 316 knots at 22 500 feet in several separate tests - achieved once again at 3000 rpm and plus 9 pounds of boost. This was a much better performance than might have been expected, given the Vokes Filter’s bad press, for it was only a little less than the 321 knots achieved by temperate-intake aircraft in the UK powered by the same Merlin 46.[7]

In short, it appears that the adverse reputation of the Vokes Filter was greatly exaggerated! Under the pressure of further perceived performance shortfalls while chasing speedy Mitsubishi Ki.46 reconnaissance machines during the 1943 Darwin raids, the matter was revisited and comparative tests re-run, but these confirmed that the performance difference was less than 5 knots. A few of the re-engineered temperate air intake cowlings saw service with aircraft of 79 Squadron in New Guinea , but the bulk of the Mark VC fleet soldiered on to the end with its unsightly air filters doing the prosaic but necessary job for which they were designed.


[1] Letter, HQ Eastern Area to Air Board, 5.12.1942, in NAA A1196: 1/501/478.

[2] AMEM to HQ Eastern Area 16.12.1942, in NAA A1196: 1/501/478.

[3] Memo, 1 FW to HQ Eastern Area, dated 29.12.1942, in NAA A1196: 1/501/478.

[4] Signal, RAAF London to RAAF HQ, dated 12.5.1942; & Memo from 1AD dated 7.10.1943 in NAA A705: 9/41/74.

[5] The actual climb difference was at the upper end of this estimate, temperate-intake Mark.VC aircraft taking 7.4 minutes to reach 20 000 feet, while the Vokes Filter aircraft took 8.1 minutes – Signal RAAF London to Air Board 26.12.1942 in NAA A1196: 1/501/478; & test figures by RAAF Special Duties Flight dated 5.4.1943 in NAA A705: 9/41/74.

[6] F/L Norwood to OC 1 FW, 27.12.1942, in NAA A1196: 1/501/478.

[7] Signal, RAAF London to Air Board, 26.12.1942, in NAA A1196: 1/501/478.