Mind boggling dual circuit braking system

For almost a year now I am faced with the fact that the brake fluid reservoir that sits on the master cylinder and that serves the rear brakes (and the clutch as it has two outlets) empties itself in the reservoir that sits on the slave cylinder. This reservoir serves the front brakes. Emptying happens when the car is parked, and then takes about a week. It also happens under braking, and it takes about a 50 km run to have its contents fully dumped.

This all started after I refurbished the rear brakecalipers, as these were leaking badly. After numerous bleeding attempts, which I think all were successful, I suspected the slave cylinder, as this is the only part I could imagine where a transfer of fluid between the two reservoirs could somehow take place. So I refurbished that one as well, to no avail unfortunately.

And now I am left stumped, having no idea how to cure this problem. I sincerely hope one of you may have a clue as I really wish to have my Rover back on the road again.


Well-Known Member
Hi, Brake fluid reservoirs that serve brakes and clutch have in the bottom a dam between the two outlets so that if the clutch hydraulics leak it doesn't starve the brakes of fluid. So check which side is causing the loss of fluid. It could be the clutch side. Can you put a picture up of the arrangement?

Thank you for your responses.
This is a picture from the Haynes manual of the dual braking system arrangement:
haynes dual braking system rover p6 (2).jpg
So, it is the reservoir E that sits on the brake master M that dumps its contents in the reservoir B that sits on the brake slave N. As a result, reservoir B overflows. In my case, reservoir E has two outlets, one to the brake master and one to the clutch master cylinder. There is no visible sign of any leak in the clutch circuit, and the clutch operates as it should. Combined with the fact that reservoir B overflows, I do not suspect anything caused by the clutch.
What puzzles me as well is that the transfer of brake fluid from reservoir E to B not only happens under braking but also during standstill.
BTW, my car is a 1973 3500S with a LT77 manual box.
In addition, the above picture contains an error: the two outlets of the brake slave cylinder are reversed. In reality, and in line with a description of the brake slave cylinder a few pages later in the same Haynes manual, the left most outlet of the slave cylinder goes to the front brakes (that's L in the picture) instead of the rear brakes and the other outlet (of course) goes to the rear brakes.
I hope this all clarifies things a bit.


Well-Known Member
I would suggest to refurbish the master cylinder AND the air valve that is attached to it.
From your description, it seems that the stored vacuum in the servo opens the air valve of the master cylinder and pushes the fluid from the first reservoir to the second. You can confirm this theory by disconnecting the vacuum hose from the inlet manifold to the servo just after parking the car and stopping the engine. Unless there is height difference between the 2 reservoirs, there should be no fluid transfer after that.
It is quite usual for master cylinders to develop faults after a bleeding session, because the seals tend to work in areas of the bore that are not reached during normal use, and may well be deteriorated.


Active Member
I have a similar car, 1972 3500s with lt 77 gearbox and dual circuit brakes. My brakes are connected exactly like your haynes manual drawing with the left most outlet going to the rear brakes. That is also how it is stated in my original workshop manual. But I have seen other drawings from an american guy where the lines were switched the opposite way. I cant see why this should not work either. I think you have to refurbish the slave cylinder, because if that is in good order with no pitting and fresh seals, there should be no possibilty for the fluids from the separate reservoirs to mix.
Thanks again for your help.
I'll start with the master cylinder then and refurbish that. Although it is rather fresh, all the bleeding I did may have been too much for it. The slave cylinder has had its turn a couple of weeks ago so it should be OK but I may have done something wrong with it.

Ah, the 3500S system. Not at lot of experience with those. There are only two places where the two hydraulic circuits are just separated by a rubber seal, an intermediate seal inside the slave cylinder and the brake warning switch. Both should be absolute seals and not allow bypass, so I think they should be the focus. If the slave was just rebuilt, perhaps a backwards seal?

I have to agree with Vern. I'd expect if one of the seals in the brake warning switch were not working the other one would halt the flow so i'd be redoing that booster/slave cylinder with particular attention to the free piston seals and the trap valve in the end outlet (probably good to check its opposite number in the master cylinder as both contain 5 -7 parts and you can do them in the car) the trap valves are there to stop the fluid flowing back when you take your foot off the brake, so the pads don't retreat way back into the housing.