Brake failure thread on the V8BB
Brake failure thread on the V8BB
As a contribution to the thread on the V8BB which started on 12th November 2008, Bob Owen provided some useful information together with an image demonstrating that the standard DOT4 brake fluid does not mix with DOT5 silicone fluid. (15.11.08)

I feel I should throw my two penn'orth into this thread which seems to have morphed into a brake fluid debate. My basis for contributing is that I run a V8 with standard DOT4 brake fluid and a TC with DOT5 (silicone) fluid, the latter having been converted by the previous owner some 4 or 5 years ago, and, whilst not a brake expert, I have a reasonably good understanding of physics due to my electronic engineering training.

It is true that standard brake fluid is hygroscopic and so the boiling point is lowered as time goes on and it needs changing every two years, or more frequently, whereas silicone fluids do not absorb water and so can be left in a system. It is also true that silicone fluid can affect rubber seals - with silicone fluid my TC master cylinder seal expanded leading to binding brakes. I turned down the master cylinder piston to shorten its length by 0.7mm to cure this because the expanded rubber seal was not exposing the fluid return drilling when the pdela was released. Pundits say that silicone should only be used after a complete refit - it is often reported that the seal problem seems worse if they have previously been used with standard fluids. And contrary to Robin's assertion, the two types of fluid do not mix. Try putting some of each in a jar - the violet coloured silicone floats on top of the straw coloured standard fluid.

I often find that what appear to be conflicting statements can both be true if examined closely. AP say silicone fluids have high compressibility whereas Robin says "fluids can't be compressed". The truth is that all things can be compressed - albeit solids and liquids can only be compressed by a very small amount. For example, if water could not be compressed then sound wouldn't travel through it and we wouldn't have war films with submarine ASDICs pinging. At 400 atmospheres pressure, water reduces volume by approx 1.8%, alcohol by 3.6% and steel by 0.025%. In general, less dense liquids have higher compressibility. Silicone fluids are less dense than standard brake fluids and so have a higher compressibility. But this is still very small. All liquids absorb gases (if water didn't absorb oxygen we'd have no fish) and generally lower density fluids absorb more. So silicone fluid

What happens if you "mix silicone and standard brake fluids? (Bob Owen)

absorbs more air than standard fluid and this further increases compressibility. This is still small and generally the expansion of brake pipes and flexible hoses under pressure would be a bigger factor. However, in motor racing circles, with high performance brakes and hoses, silicone brake fluid is not favoured because of the "spongy" pedal. However, racing types think nothing of stripping an engine between races, so frequent changing of their hygroscopic standard brake fluid is a mere soupcon.

It must be pointed out that the non hygroscopic nature of silicone can also be a disadvantage. For example, if any water does enter the braking system it will gravitate to the lowest point rather than being absorbed in a distributed way. If this is in a brake pipe the water may cause local corrosion: it will be there a long time as the reason for using silicone is that it doesn't need changing. Perhaps worse, it may freeze . . . . the consequences of a frozen section in a brake pipe could be very interesting . . . and difficult to diagnose when later examined in a warm workshop.

Slight differences in viscosity are not normally an issue on classic car braking systems, although with modern ABS brakes it may be a factor.

I have not had sticking brake cylinders on my V8 but it is a common problem on T types and is cured by using silicone fluid. Use of silicone fluid is widespread amongst older classics, eg T types, where it gives far fewer problems than standard fluid, but is not favoured by racers, as I mentioned above.

So, in conclusion, each type has its virtues - there is a large degree of "horses for courses" here . . . .
What were the postings in the brake failure thread on the V8 Bulletin Board that raised the brake fluid issues?

Matt Jones launched the thread with a query on brake failure on his RV8 saying "I recently suffered a long brake pedal in my RV8. After checking the master cyclinder reservoir and finding it empty I bled the whole system and concluded that due to the almost black nature of the sludge masquerading as brake fluid that it hadn't been changed for many years!
The pedal feel was dramatically improved but after a weekend of driving where I'd been keeping a careful eye on the fluid levels it was apparent that fluid was disappearing again from the master cylinder. I checked the cylinder itself and found no sign of leaks, there were also no leaks from any of the bleed niples on the calipers or drums.

Does this sound familiar to people? I read some notes on seal failure in the Servo, could it be this and if so does anyone know where to get a new Servo or is it possible to just replace the failed seal? Any help appreciated as always". (12.11.08 @ 13.32)

Ralph Coulson responded saying "I guess you will be getting a lot of 'press' on this one. Black fluid invariably means perished seals, or damage caused by rusty cylinder bores & pistons. Did you fluid have a 'tinge' of brown? If there was no trace of fluid leaks from cylinders or hydraulic lines, including flexi hoses, then almost certainly your servo is drinking it (see note below) and passing it to the engine to be burnt off. This would go un-noticed because the volume of fluid is small.

Brake fluid, except the silicone based type, is hydroscopic and will over time absorb water from the atmosphere. This will present two problems.
First, rusty cylinder bores and pistons, including the servo.
Second, water 'boiling-off' in the brake fluid at high temperatures, causing temporary brake failure (when you need it most). The system will then return to normal when the hydraulics cool down.

It is recommended that hydroscopic brake fluid be changed every two years minimum, regardless of mileage because it's the exposure time that counts. Many MG and classic car owners have fitted Mr Owen's very excellent 'Brake Fluid Alarm Kit', see V8 Register Home Page for details. I have fitted one to my car and I consider it to be a life saver.

The conclusion is, new cylinders / seals all round, new servo, inspect/replace hydraulic lines and flexi's, fresh fluid and the cream on the cake, Mr Owen's alarm kit. The reward will be, your brakes will never have felt better and the car will be safe to drive again. Sorry to be the bearer of a costly news, but as we all know, brakes are at the top of the critical list.

Footnote: the RV8 servo is in tandem with the brake master cylinder. I am assuming the thirst of a failing servo is the same as for the remote type as fitted to earlier MGBs. (12.11.08 @ 22.21)

Al Barnett added a brief comment - "Exchange servos are available from Clive Wheatley. I had one fitted last year and am entirely happy with the result. The symptom in my case was a slight hissing noise which Steve quickly diagnosed as a failing servo; Good Luck". (13.11.08 @ 14.43)

Matt Jones responded saying "Thanks guys, I've spoken to Clive Wheatley and will be sending him my Servo on exchange". (13.11.08 @ 21.06)

Geoff King provided some useful guidance with "the RV8 (and the late model MGB) has a direct acting servo; a servo failure alone cannot result in the symptoms you describe. For fluid to leak into the servo it has to be a master cylinder seal leak. There may also be a servo problem but your description of the fault doesn’t suggest this at all and the servo simply cannot suck fluid out of the master cylinder.

I don’t know what your skill level is but some checks are fairly simple with a few basic tools. Before you change the servo remove the vacuum hose from the plenum and check if there is evidence of brake fluid – I doubt if there will be. If the hose is clean remove the pedal box cover to see if there are leaks from the master cylinder, however, the master cylinder is directly connected to the servo and fluid could leak without any obvious signs and you’ll need to remove the master cylinder and servo assembly and separate them.

Have you removed the rear brake drums to check for fluid leaks?(14.11.08 @ 08.43)

A couple of hours later Geoff King added a footnote - " apologize; my comments concerning the removal of the pedal box cover to see if there are leaks is misleading - the master cylinder is attached to the servo on the opposite side to the pedals and leaks may be visible at the joint face between the servo and the master cylinder".

Robin Gell then added that he "would recommend refilling the system with SBF, Silicone Brake Fluid, available from specialists and on eBay. It is expensive at around £20 a litre, but is not hygroscopic and does not damage paint. In fact it puts a nice shine on things when wiped over! It is DOT 5, has a higher boiling point than other fluids, and retains this due to the lack of any moisture absorption.

It does not need to be changed at all and therefore is effectively a fit and forget operation; especially useful on cars which do not see much use, as this avoids rusting and sticking cylinders etc due to absorbed moisture and lack of regular use. Rubbers are supposed to last longer too. Filling is easy as systems don't need to be flushed before filling, other than with the new fluid. It mixes and works with ordinary brake fluid, although naturally removing as much of this as possible is to advantage. When bleeding through, the SBF can be seen coming out from the bleed tube as it is purple and looks like meths. Thus you know it is all flushed out with new fluid. As disconnection of the master cylinder means a full fluid change, this is the time to do it. (14.11.08 @ 18.41)

Mike Howlett then referred fellow members to a warning note from AP Lockheed reproduced in V8NOTE228 released in June 2001. He concluded by saying "AP Lockheed glycol based fluids do not contain the adverse properties described above. The recently introduced Supreme DOT 5.1, which exceeds the performance criteria of DOT5, is suitable for all conditions likely to be encountered in modern driving conditions". (14.11.08 @ 18.41)

Linked to that V8NOTE is V8NOTE228A with contributions from fellow members discussing the DOT4 and DOT5 brake fluid issues.

Dave Wellings then provided a note on his experiences with an MGBGTV8 over many years by saying "This old chestnut has been around ever since silicon fluid hit the shelves. I would only consider using it in a completely rebuilt system, because it can't flush the old fluid properly from around the seals, so at the very least a full set of new seals and blow through with compressed air would be recommended. The thing is, I had this decision tomake in 1991 when my V8 braking system was completely new. I agonised over this choice for weeks, and listened to all the evidence, plenty of which was anecdotal. Some well respected concours entrants swore by silicon and had had no problems over long periods, but with less than average mileage. In the end, the issues highlighted in Mike's post led me to conclude that mineral fluid - changed regularly would do the job. And so it turned out.

One other point about servos - the internals are well protected from temperature/ atmospheric changes. I stripped my remote servo when it was 18 years old - it worked properly and there was no trace of corrosion in the bore. I fitted the full repair kit and it lasted another 18 years before sucking the fluid from the master cylinder - fortunately on the drive at home. (14.11.08 @ 20.31)

Robin Gell returned with another contribution - "It's good to bring these things into the open, so thanks Mike for posting this. I for one however, am very sceptical over what AP say here, but at the same time however take it seriously. The issues for me are:

. Fluids don't compress. Basic physics? Even water will work in brakes by itself when cold.

Entrapped air leading to spongyness and slow filling. There shouldn't be any air in the system if it has been bled properly?

Similarly moisture globules. What moisture? This should be bled out! There is no difference between the fluids in this respect. What free unmixed water? Glycol fluid absorbs water leading to lowering of boiling point. Where would free water come from otherwise? Why would this lead to greater lowering of boiling point than absorbed water? I don't buy it at the moment. I don't buy any of it without very much more technical and scientific substantiation for them as none of them seem logical to me at all.

Twice the viscosity? Don't think so somehow, but will check.

Lower lubricity? Sceptical about this claim too, but again, will check.

I know that I have been using it for a number of years with none of these problems, and so have many others. I will send this note to who market SBF and ask for their view on the subject, and of course, they are bound to stand their corner, as both they and AP have their own interests to defend. AP have their recommendations and insurance liabilities to consider and so are not going to support something outside their box. It is like many other instances of specification of particular consumables. "Only use xyz sort of stuff" appears regularly all over the place, but as we well know, there are usually many other choices, just as good, (Comma oils for example), but which are not "recommended" and so would produce no come back to the manufacturer.

It will be interesting to see how Automec defend themselves. Who does one believe? I suspect that the reported problems here are more to do with incorrect setting up than the stuff itself. I stick my neck out however, but am loathed to take this at face value! Currently I have to admit that I take AP's warning seriously, but with scepticism, as SBF is apparently used as standard by the US Military in all their vehicles, and also conforms to the highest acknowledged standards, the US Department of Transportation (DOT)5 standard as well as the Federal Motor Vehicle Safety Standard (FMVSS)116 and US Military Standard MIL B46176. I will see what Automec have to say on the subject and report again. (14.11.08 @ 20.53)

Victor Smith then added - "I feel that before any suggestion that the advice issued by the Technical Service Department at AP Lockheed is influenced by "their own interests", I do think the technical matters over which you are sceptical should be thoroughly researched first. Surely the technical people at AP Lockheed are engineers with a special knowledge of auto hydraulics and in particular how the various available hydraulic fluids perform in both classic car and modern car braking and clutch systems. Why not telephone the technical department at AP Lockheed and discuss with them first hand and in detail each of the concerns you have raised, get clear information as to the technical basis for the concerns expressed in their warning note (which we reproduced in our V8NOTES series some time ago when this matter was discussed amongst V8 enthusiasts) and write up a note or article for the V8 website which would then be a substantial and authoritative contribution to clarifying the matter? I fear that at present all we have is renewed confusion." (14.11.08 @ 22.19)

Robin Gell responded saying "Victor, I agree. I am not leaving it there and have already emailed Automec to ascertain their view on the points. It is only fair that they have the chance to reply to these points before taking it farther. It seems as though up to now, they have not had the opportunity. Then perhaps, Lockheed will be approached for further comment.

I will take it from there, but at the end of the day, I suspect there will always be some polarity between the two. All I am doing is to try to stimulate discussion on the matter in making suggestions, not accusations, and by questioning what to me, as an engineer myself, seem on the face of it to be some strange statements. Sorry if it is thought that this has generated confusion. I will report factually, anything which I am able to discover. (15.11.08 @ 00.56)

Bob Owen then posted his informative piece featured at the head of this webpage and sent the V8 Webmaster the image for publication. (15.11.08 @ 11.46)

Rob Collier then noted Bob's piece was "a very informative explanation of the pros and cons of DOT4 and DOT5 brake fluids. We learn sommething new everday and following your explanation I now understand why DOT5 silicone has not totally replaced DOT4 fluid in all applications." (15.11.08 @ 12.59)

Victor Smith then added "On re-reading V8NOTE228, which was issued back in June 2001 along with the comments on the brake fluids topic from fellow members in V8NOTE228A, there are two matters where reliable and authoritative information is necessary to help clarify some of the issues over the use of silicone brake fluid in the brake and clutch systems fitted to the MGBGTV8 model and the later RV8. In V8NOTE228 an experienced motor and aero engineer says "says silicone fluid attacks rubber seals and causes swelling". The informative posting from Bob Owen seems to support that comment where he says "silicone fluid can affect rubber seals" and that he had to "turn down the master cylinder piston on his MG TC" to overcome that effect where the hydraulic system on that car was using silicone fluid. Does any member have any authoritative or factual information on the effects of silicone fluid on the rubber seals fitted as original equipment and current replacements and in particular the ways the fluid attacks the rubber seals?" (15.11.08 @ 15.06)

Robin Gell responded saying "Bob's is a very authoritive and believable explanation. Thank you for the illumination Bob.
In my mind it does go to show that many of the cited problems, while having bases in fact, are actually very small in proportion to the whole picture, and are probably insignificant. (400 atm. only 1.8% reduction in volume of water against 3.6% for alcohol. 400 atm is collosal pressure, and if it takes this pressure to make alcohol compress twice as much as water, it puts it into perspective. This is really confirming what I was thinking. I'm not sure that the analogy with sound travelling through liquids is a good one though, because these are travelling vibrative pressure waves rather than volumes of liquid compressing to smaller volumes.

Talking about steel, as an opposite to compression, railway engineers stretch rails by maybe more than a foot in length (depending on ambient temperature), with hydraulic pullers(over maybe a quarter of a mile free length) to gain a length equivalent to a temperature of 27 degrees, and then weld them up. Thus they are in considerable tension and so compressive forces in hot rails, up to 50+ degrees can be resisted in the fastening system. I digress, although it does show how significantly the volume of steel can be changed in this way.

It seems that there is no ideal answer to the problem of which fluid is best, and indeed, as has been said, it is often a case of horses for courses.

Indeed I know that the two don't mix in a homogenious nature, having used it myself, as they are slightly different in density and form layers when in the same container. What I meant is that cross contamination, if it happens, say by bleeding new fluid through a previously glycol filled one, if one wants to do this, is not a problem as the two can work together, even if this is not exactly ideal. (15.11.08 @ 16.01)

Rob Collier then posted some useful information on the effects of brake fluid on rubber seals quoting regarding rubbers seals. "
Silicone brake fluids are not hygroscopic, and tend to retain their dry boiling points for very long periods of time. For this reason, silicones are favored by owners and restorers of classic and antique cars, as there is minimal danger that seldom-used and possibly irreplaceable brake components will be lost to corrosion.

Silicone will cause natural rubber to swell, even when it's compounded with synthetics. The seals in modern brake systems are no longer 100% natural rubber, but blends of natural rubber and synthetics like nitrile. Glycol fluids will also tend to swell blended rubber seals, but to a much smaller degree then silicone. Swollen seals may leak, or cause caliper pistons to bind, resulting in brake drag.

Silicone has several other properties that make it less then desirable for street or track use. When forced thru small orifices under high pressure, like the solenoid valves in an antilock brake system, it tends to foam, generating bubbles. Bubbles in brake fluid make for spongy brakes. Silicone also tends to become slightly compressible at temperatures near its boiling point, which makes it generally inappropriate for racing.

To get the maximum benefit from silicone, the entire brake system MUST be flushed of old glycol fluid. A brake system cannot be completely flushed using the bleeder fittings, as they are purposely at spots in the system to allow air to be bled, you simply can't get all the old fluid out by bleeding. The best way to completely flush a brake system is to dismantle and overhaul it, cleaning everything with alcohol, and then coating all the parts with the new fluid as they are re-assembled. Going to this much hassle just doesn't justify changing to silicone, IMHO." (15.11.08 @ 18.57)

Victor Smith reported that "from a call to Brovex Nelson at Camelford in Cornwall this afternoon, I learned from their technical manager that they make a variety of parts for classic car brake and clutch systems including rubber seals and rubber flexible front brake hoses. They supply most MG specialists and parts suppliers in the UK and elsewhere. He confirmed that their rubber compounds work well with either glycol or silicone hydraulic fluids and their seals do not swell when in contact with either fluid. In response to the question "do rubber seals, that have been fitted to classic cars using glycol fluid and then later have been in contact with silicone fluid following a change from glycol to silicone, swell and increase in size" he said “yes that is a very good question!”. But he then added that Brovex Nelson rubber seals performed reliably as they use their own rubber compounds and they manufacture their seals. He then mentioned that many existing seals fitted to classic cars and some replacement seals in the market are supplied by other manufacturers and he was not able to say how they might perform in a switch from glycol to silicone fluid.

I then enquired how you could identify a Brovex Nelson rubber seal. It was a slight surprise to learn that their seals do not bear their name or a brand mark, only the text “made in England”. However there are three or four small letters or numbers on the seal which enable them to identify their own seals. So if a V8 enthusiast is buying replacement seals from a parts supplier, mere inspection of the seals will not reveal whether you are getting a Brovex Nelson seal or one from another manufacturer. Your only option is to ask the parts supplier which manufacturer has supplied them." (17.11.08 @ 16.54)

See useful brake fluid articles. More

Review of the brake fluid debate
Bob Owen is preparing a paper on the glycol-slicone brake fluid debate with the aim of presenting the facts and the reported observations and views of experienced motor engineering specialists together with a dispationate analysis. More

Which brake fluid?
An article providing the information you need to make your own choice has been prepared by Bob Owen. (23.1.09) More

V8 Register - MG Car Club - the leading group for MG V8 enthusiasts at