Suspension, brakes and steering
Brake and clutch reservoirs, and why replacements leak
Reproduction reservoir bottles leak often enough that owners replace them repeatedly. The cause is usually one of two things, and both are checkable before you buy.
Three white plastic bottles sit in a bracket on the bulkhead of a six-cylinder E-Type. Two feed the brake circuits and one feeds the clutch. You can tell them apart without moving anything, because the brake reservoirs carry a level sender in the cap with wires running from it and the clutch reservoir has a plain cap and no wires. Early 3.8 cars used round bottles before the change to the square ones, and the Series 3 is arranged differently again, with the rear brake reservoir mounted separately under the heater box rather than in the row on the bulkhead.
Owners replace these bottles more often than any other single item of their kind, and the usual explanation offered is that modern reproductions are simply poor. That may be true, but it is not a finding, and it is not something a buyer can act on. There are two specific causes underneath it, and both can be checked before spending anything.
The first cause: the outlet spigot, and the tube inside it
The failure reported most often is not a seam splitting or a bottle cracking in the middle. It is the outlet at the bottom of the bottle, where the hose pushes on. It cracks around its base, or in the worst accounts it snaps off entirely with the hose still attached to it, and the circuit empties.
There is a documented physical reason. Original bottles were moulded with a metal tube and mesh filter set into that lower outlet, which reinforces the spigot from the inside and stops it being crushed when a hose clip is tightened onto it. At least some reproductions have been supplied without it. Owners on two separate forums have identified the same missing part as the reason their outlet cracked at exactly that point, which is the kind of agreement between independent sources that this archive treats as worth reporting.
That turns an unactionable complaint into a purchasing decision. Martin Robey's product description for their brake reservoir assembly states in plain terms that it includes the strengthener tube with plastic filter. Other suppliers' descriptions do not mention it either way. If the failure mode is the spigot, buy from a supplier who will tell you the reinforcement is there, and ask the ones who do not say.
The related mistake is at the other end of the same joint. An unreinforced spigot is crushed by an over-tightened worm-drive clip. Owners who have cured leaks here report doing it partly by changing to lighter fuel-line clips rather than jubilee clips, which spread their load rather than concentrating it.
The second cause: the hose, and why fuel hose destroys it
A great many leaks blamed on the bottle are the hose. This matters because the wrong hose is easy to fit, looks correct, and fails in a way that mimics a leaking reservoir.
Glycol brake fluid is not a general-purpose liquid that rubber either survives or does not. It is defined by the rubbers it is designed to live with. The American federal standard FMVSS 116 defines brake fluid as a liquid for use in a system where it will contact components made of styrene butadiene rubber, ethylene propylene rubber, polychloroprene brake hose inner tube stock or natural rubber. SAE J1703 names the same family: natural rubber, SBR, or EPDM. Everything else is outside the design set.
Ordinary fuel hose is nitrile, also called NBR or Buna-N, and it is optimised for precisely the opposite chemistry. Nitrile carries polar groups that glycol ether is attracted to, so the fluid diffuses into the rubber and swells and softens it. Separately, the fluid extracts the plasticisers, which can leave the hose hard and shrunken instead. Either way the push-fit joint at the spigot stops sealing. Parker's own elastomer data rates nitrile as not compatible with glycol based brake fluid, and rates EPDM as compatible up to 149 degrees Celsius. Owners report fuel hose failing in this application inside twelve to eighteen months.
There is a subtlety worth getting right, because the usual advice overstates it. Neoprene is often described flatly as incompatible with brake fluid. The regulation says otherwise in a narrow sense: FMVSS 116 specifically names polychloroprene brake hose inner tube stock as a material the fluid is designed to contact. Neoprene is not a seal material for a brake system, but a particular neoprene hose liner compound is a recognised one. Precision matters here more than the usual shorthand allows.
One further failure belongs to the hose alone and is easy to misread. Some modern hoses sweat brake fluid through the wall rather than leaking at a joint. It shows as sticky beads on the outside of the hose and fluid appearing below with no visible drip at either connection. That is permeation, not a bad fit, and no amount of tightening will fix it.
The correct hose, and a number that is widely misread
The hose owners name for this job is Cohline type 2337, which is a low pressure reservoir feed hose from a German manufacturer certified to IATF 16949. Cohline's own catalogue gives its construction as EPDM inner, aramid braid reinforcement, EPDM outer, and states resistance to polyglycol fluids to DOT 3, DOT 4 and DOT 5.1. Its compatibility with DOT 5 silicone is not stated by the manufacturer and this archive has not found it stated anywhere else.
The part number causes confusion that is worth clearing up. Cohline 2337.0613 is frequently quoted as though 0613 were a size, and it is then repeated as six millimetres. It is not a bore. It encodes a nominal DN6 with a thirteen millimetre outside diameter, and the actual inner diameter of that hose is seven millimetres. The larger 2337.0915 is nine millimetres inside and fifteen outside. Anyone who has been told six millimetres and anyone who has been told 0613 is being told about the same hose, and neither figure is its bore.
The other figure in circulation is three eighths of an inch, which is 9.53 millimetres. That looks reasonable until the joint is measured. One owner measured his reservoir outlets at about 8.5 millimetres and the hard pipe from the master cylinder at six. On those figures a seven millimetre hose stretches onto the spigot with roughly a millimetre and a half of interference, which is what makes it seal, and clamps down onto the six millimetre pipe at the other end. A three eighths hose over an 8.5 millimetre spigot is a clearance fit that seals only when the clip crushes the spigot, which is the exact place these bottles break. The archive reports this as reasoning from a single owner's measurement rather than as a factory dimension, because no supplier publishes the bore of Jaguar's own hose, part number C18717.
Cohline also specify a minimum bend radius of forty millimetres for the smaller hose. In a crowded engine bay that is easy to violate, and a hose bent tighter than its rating pulls sideways on the very spigot that has just been identified as the weak point.
Why nothing will glue it
A common suggestion when an outlet starts to weep is to seal around it with an epoxy such as JB Weld. It does not work, and the reason is worth knowing rather than simply being told.
Adhesives have to wet a surface to bond to it, and whether they can is governed by surface energy. Both 3M and Permabond put the practical threshold for adhesive bonding at about 36 millinewtons per metre. Polyethylene and polypropylene, the two polymers that brake fluid reservoirs of this era are made from, sit at about 31. They are below the line by design, which is the same property that makes them useful for holding aggressive fluids in the first place. Bonding them at all requires flame, corona or plasma treatment, a polyolefin primer used with the one adhesive type it suits, or a purpose-formulated structural acrylic. Permabond also note that surface treatment decays, and can return close to its untreated state within a week, so even a successful treatment has to be bonded immediately.
The correct repair for a polyolefin is welding rather than gluing, because welding does not depend on surface energy at all. The archive has not found any account of that being done successfully on an E-Type reservoir, so it is offered as the principle rather than as a recommendation.
Whether it matters
It does, on three counts, and the paint is the least of them.
- The car has split brake circuits, so a reservoir that empties takes a circuit with it. The worst first hand account found describes the outlet parting company with the bottle and the car losing its brake fluid.
- The MOT inspection manual does not have a clause naming a leaking reservoir, but a reservoir leak reaches a serious finding by more than one route. Brake fluid not visible is a Dangerous defect. Fluid significantly below the minimum mark is Major. A leak at a hose or pipe connection is Dangerous. Fluid level checks apply only to transparent reservoirs or where an indicator is fitted, which is a point against opaque replacements.
- Castrol's own data sheet for DOT 4 warns that spilling the fluid on paintwork may have a damaging effect. The archive quotes the manufacturer's warning rather than explaining the mechanism, because it could not source that explanation to any named authority.
What to do about it
In order of how well supported each option is. Fit the correct hose first, since a proportion of reservoir leaks are not reservoir leaks. Buy a replacement bottle from a supplier who states that the outlet reinforcement is present. Use clips that clamp gently rather than worm-drive clips wound tight. Route the hose so it hangs without side load and within its bend radius.
Secondhand originals are the other route, and owners who have gone back to them report no further trouble. They are identifiable by SOVY and MADE IN ENGLAND moulded into the bottle. The caution is that the steel reinforcing tube that makes them desirable is itself sixty years old, and at least one seller has described corrosion in it.
Aluminium replacements come up in discussion and the archive could not find one made for the E-Type. Generic billet reservoirs exist for custom builds, and the reasons they are not a straight swap follow from their own published specifications rather than from anyone's opinion: far smaller capacity, a threaded union rather than a push-on spigot, no level sender and therefore no dash warning, and an opaque body that removes the visual level check.
Part numbers
From supplier catalogues rather than from a factory parts book. Fitment varies by chassis number and should be confirmed against your own car.
- Square brake fluid reservoir with level sender: C36606.
- Square clutch reservoir and cap: C36607.
- Clutch reservoir cap, square bottle: C19515.
- Reservoir bracket set, right hand drive: C24128ASS.
- Round clutch reservoir, early 3.8: C18616. Round brake reservoir: C18617. Round brake reservoir with sender: C23638.
- Low pressure hose, pipe to reservoir: C18717.
- Original bottles carry SOVY and MADE IN ENGLAND moulded into them.
Common questions
Which of the three bottles is the clutch reservoir?
The one with a plain cap and no wires. The two brake reservoirs carry a level sender in the cap with wiring running from it. This is the quickest way to identify them without disturbing anything.
Can I use ordinary fuel hose between the reservoir and the master cylinder?
No. Fuel hose is nitrile, which glycol brake fluid swells and softens, and whose plasticisers the fluid extracts. Owners report it failing within twelve to eighteen months. The material required is EPDM, which is one of the rubbers the brake fluid standards are written around.
Is Cohline 2337-0613 a six millimetre hose?
No. The 0613 is a nominal DN6 with a thirteen millimetre outside diameter. Its actual bore is seven millimetres. The number is very widely misread as a bore size.
Will JB Weld or epoxy seal a leaking reservoir outlet?
No. The bottles are a polyolefin, with a surface energy of around 31 millinewtons per metre against the roughly 36 that adhesives need to wet a surface. Bonding them requires surface treatment or a specialised structural acrylic, and surface treatment decays within about a week. Welding, not gluing, is the correct repair for this class of plastic.
Why do reproduction reservoirs crack at the outlet when originals do not?
Original bottles have a metal tube and mesh filter moulded into the lower outlet, which reinforces the spigot against being crushed by a hose clip. At least some reproductions have been supplied without it, and owners on two independent forums have identified that omission as the cause of cracking at that point. At least one current supplier states that their part includes the reinforcement.
Sources
- FMVSS 116, 49 CFR 571.116, for the definition of brake fluid by the elastomers it is designed to contact, and for the compatibility test method.
- SAE J1703, for the corresponding statement that fluids are designed for systems using natural rubber, SBR or EPDM.
- Parker Hannifin O-Ring Handbook, for elastomer compatibility with glycol based brake fluid.
- 3M and Permabond published technical guidance, for surface energy figures for polyethylene and polypropylene, the bonding threshold, and the decay of surface treatment.
- COHLINE GmbH catalogue and Cohline UK, for the construction, dimensions, pressure ratings, bend radius and fluid compatibility of type 2337.
- Martin Robey, Moss Motors, Terry's Jaguar Parts, Welsh Enterprises and the SNG Barratt parts catalogue, for part numbers, and for the statement that a current brake reservoir assembly includes the strengthener tube.
- Castrol DOT 4 product data sheet, for the warning about paintwork.
- DVSA MOT inspection manual for private passenger and light commercial vehicles, section 1, for the defect categories quoted.
- Owner accounts on the E-Type Forum, Jag-lovers and Jaguar Experience, used as evidence of what owners consistently report rather than as statements of manufacturing fact, and labelled as such in the text.
Last reviewed .
