MG-Cars.net

Welcome to our resource for MG Car Information.

Recommendations

Parts

MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG MGA - Master Cylinder Contaminaion

A cautionary tale.

About 8 months ago I put new seals in the master cylinder as part of an investigation to cure a hard to find weep. Then, about 3 months ago I replaced the front pads and rear shoes. Since then, I have suffered brake squeal which I have been unable to cure. During the last month or so I have also noticed that when pulling to a stop on slight inclines or slopes I have been able to take my foot off the brake pedal and the car has not moved.

I got the car on jacks this after noon, applied the brakes and confirmed that they were not releasing. I slackened one brake bleeder and all wheels then turned okay. The problem was therefore in the master cylinder.

Stripped the MC down. I could barely move the brake piston. There was sticky contamination on the brake cylinder wall, piston rear seal and on the rear contact surface of the piston. (see photo).

So what caused it? When I was chasing the weeping problem all those months ago, I put a bit of Wellseal jointing compound on the rear plate. It would seem that some of the excess has been absorbed into the fluid and caused the piston to stick.

Moral of the story? Either don't use non setting sealing compounds, such as Wellseal, in these areas or be very cautious in the amount used.

Steve


Steve Gyles

The piston seems to have taken on the gold colour of the Wellseal Steve, is that what it looks like in real life? What type of brake fluid are you using?
Lindsay Sampford

Lindsay

It's more of a yukky dark brown (Wellseal colour). It's definitely got the stickiness of the product. Surprisingly the clutch side is pristine.

I use Silicon - no, let's not start that debate!!! I put it down entirely to my over enthusiastic use of the compound; albeit uneven use, in that the clutch side is ok.

Steve
Steve Gyles

If you spread Wellseal on in a thin film it is gold, maybe thats where you went wrong Steve!
Lindsay Sampford

Steve, when you say rear plate do you mean the cover plate where the two pushrods go through? My picture shows a gasket there.
Lindsay Sampford

Lindsay

I will concede the point on colour. Eyesight not all it used to be. Must be all this open top driving - See Barney's question on your other thread!!

Maybe this problem is the root cause of why my mpg has not improved in the way I expected after much general tuning and longer commutes. I wondered why my front wheels have been so dirty with brake dust!

Steve

Steve Gyles

Lindsay

Yes. The plate is as you have identified.

The fluid in my MC kept slowly dropping. I could not see a leak. I eventually discovered the cloth wiring loom that passes through the MC bracket was sopping wet. I applied the Wellseal as an additional measure to cure the unidentified leak. It worked. I will now check the mating surfaces and flatten as necessary.

Steve
Steve Gyles

No wonder my front plate was weeping before I put the seelant on the gasket. I have just put a straight edge on the machined face and can get a 15 thou feeler gauge in the gap. See photo.

Steve


Steve Gyles

Steve-
There should be NO, ZERO, NOT ANY fluid behind the MC piston to get to that rear cover plate. Not EVER! If there is, then the MC seal (rear one toward the pedal pushrod) is leaking.
(I know the terminology is backward, most MC being mounted pedal end rearwards)
I have never been able to understand why they (Lockheed) put a gasket there, it just screws up the at-rest location of the piston, and the cover plate gets bent from people over tightening it in a futile effort to stop leaks. If you remove the gasket, it is sometimes possible that the piston at-rest is far enough forward to block the compensating fluid port in the cylinder = no workee.
No other MC uses such a gasket; most have a circlip as the piston stop, and a dust boot with a vent hole to keep crap out of the bore. MC for servo systems have an extra seal there, but the purpose is to seal vacuum for the servo, not to keep leaking fluid in the MC.
Aside from general age and use, I have noticed that the Magnettes with the same type cylinder have a serious pushrod alignment problem that eventually wears the bore badly on the bottom, in the at rest position. That means the seal will always leak at rest since the piston is no longer centered in the bore, even if the cylinder otherwise appears good = no pitting. Haven't had a chance to check a MGA since I figured that out, but it's possible.
Go buy a MC, or get yours correctly rebored & sleeved, and don't use any sealant!

FRM
FR Millmore

Steve

I concur with FRM, there shouldn't be any fluid there anyway. I also concur on his observation on uneven wear on the cylinder and piston, which is something I observed on my recent seal replacement (fortunately confined to plating wear on the piston at the moment, but I strongly suspect that the next time the clutch MC seals go it will be time to replace the whole MC as a result. It may also be the reason that the seals have worn out twice in less than ten years. I can see that the pushrod is off the centre line but as everything else seems to be straight I don't really want to start bending things to try and get it centred.

dominic clancy

FRM is right. No fluid should ever get to that plate. Any fluid in that area means the piston seal is not sealing. Those two rubber pieces on the front are dust seals only.
R J Brown

Thanks FRM and all for the education. At the time of the weep I generally tightened up all the unions, the top extension bolts and the front cover. I assumed that the front cover was the problem; maybe it was not. However, the contamination caused by the sealing compound did not do the seals any favour - replaced them this evening. The MC seems to be in very good condition with no measurable wear that I have determined. It is 10 years old and only ever held silicon fluid. I will monitor it with interest during these coming weeks.

I flattened the front plate anyway on a glass and am now doing the same with the MC extension which is a newly manufactured item and had only been coarsely flattened. I noticed previously that when I used Eezibleed I could detect air escaping through the extension joint. That said, I was using quite a bit of pressure - more than specified - 25 PSI!

Steve
Steve Gyles

Lindsay mentioned in another thread the importance of closing off a problem with the solution so that it is meaningful to others when trawling the archives.

Happy to report that the MC is back in operation with not a sniff of sealant in sight. I replaced all the seals as a precaution, even though they looked okay after cleaning (only 7 months use). Brakes working A1 again and no leaks from the MC.

Steve
Steve Gyles

When I was using silicone brake fluid I ground off about a sixteenth (or more) of an inch from the piston so that it comes to rest that much further towards the front of the car (sorry for being a bit wordy--it is awkward trying to explain this!). I did this on both of my MGA 1600's. The reason was that the seal swelled a little due to the silicone brake fluid and did not clear the return port, causing the front brakes (since they have no positive return arrangement such as a spring) to bind. There is no detrimental affect on the brakes (I went back to conventional brake fluid many years ago and the "mod" remains).
Barry Bahnisch

Barry

That was a modification recommended to me by Bob West, regardless of type of fluid used. I recall taking about 10 thou off the piston face.

Steve
Steve Gyles

This problem is likely not due to swelling of the seals, especially not from silicone fluid which is particularly inert. Some years ago (back in the 80's maybe) some seal cups were made too long and would cover the fluid hole in the reservoir at initial assembly. The interference (overlap) was small and could usually be cured by cutting the hole in the paper end gasket large enough for the piston to move more forward to touch the end plate. If that wasn't enough, then double layering the paper gaskets would be enough to make it work. I have never heard of seal swelling subsequently making the condition reoccur after this small adjustment, which is why I'm pretty well convinced that it never was a problem with swelling seals.
Barney Gaylord

Glad you said that Barney and not I. I agree with you but I was not wanting to regenerate the silicon debate. I have never had a problem with it; just my own inadequate engineering.

Steve
Steve Gyles

Alignment of the pushrods is pretty poor, Fletcher. I recently put a new set of seals in, and the signs were there, including a very eccentrically-shaped working seal on the clutch side, where travel is full. No discernable cylinder wear up there, luckily.
Nick

Nick-
Lateral misalignment can be dealt with by spacing the pedals correctly on the shaft, or the pushrods. The vertical is built into the geometry of the linkage.
On the Magnette, with the pedal at rest, the pushrod is pointed noticeably to the bottom, and that's where the wear shows up. With some travel, it lines up pretty well on center. So, your reward for keeping your brakes well bled and adjusted is to receive a worn out MC! Of course it also means that applying a lot of force with your good brakes wears out the cylinder faster, but most of the added effort goes to side loading and worse wear. Guess people who don't bleed and adjust really need good panic stop capability.

FRM
FR Millmore

I too have no desire to re-open the silicone brake fluid topic but I suspect that "modern" seals (that look more like shiny plastic than rubber) may not be as susceptible to swelling as the older ones were. There is no question that the older type used to swell, I have had personal experience and have replaced many of them (in a variety of MG models) over many years.
Barry Bahnisch

This thread was discussed between 09/10/2009 and 22/10/2009

MG MGA index

This thread is from the archive. The Live MG MGA BBS is active now.