Welcome to our resource for MG Car Information.



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 Midget and Sprite Technical - Testing a PCV valve

Some of you will remember my various threads about the PCV valve and erratic idling, eg these:

More recently I find that the problem is much worse when the engine is hot. So I took the PCV valve off and blanked the manifold connection, after a 30 mile run in warm weather. Idling was back to normal. Ran the engine for a few minutes to make sure, then put the PCV valve back on. Idling all to pot again. Took PCV valve off again, idling back to normal. So it's the PCV valve. I have replaced the diaphragm, top plate and metering valve. On testing by blowing or sucking I find the metering valve does not totally shut. Well it can't, there is no seal, so it will always leak. These devices are supposed to do 3 things:

1. Provide suction during normal running.
2. Stop sucking at idle.
3. Shut off if there is a blow-back from the manifold, for safety.

Mine does not do 2 properly, but it does do 3 - if I blow in reverse direction it shuts completely, so the check valve below the metering valve is working. I think there should be rubber seal on the metering valve, but the one I bought doesn't have one. It isn't shown as a separate part on the exploded diagram.

I think I will experiment by gluing some rubber to the metering valve. What do you think?

To save you from reading the other threads, my setup is 2 breather hoses from the front tappet chest and old fuel pump point on the block, via oil separators and a Y connection, and I have the correct filler cap that has been opened and cleaned.

I should mention that I have tried connecting the breathers to the carb upstream of the throttle, but this didn't control the oil leak as well as the PCV valve.
Les Rose

My guess is, that the valve is being defeated by the additional hose from the petrol pump blank.

The valve was the 1st version, originally connected from the tappet cover, and later from the timing chain cover.

The lower constant depression from the su carbs take off, allowed the valve to be elliminated, and isn't 'strong' enough to operate the valve.

When new, spring not interfered with, and diaphragm in good soft condition with no splits, it works as well as can be expected.

Pcv valves work on other cars too. But the later constant depression connection, with no valve, on our a series engines is at least as good, and probably better. Ironically, the proof is that it's strong enough to suck up oil -- on some engines, nuff said 😁.

If you can't stop the leak from your rear scroll, it's either the scroll, or too much crankcase pressure (blowby). But the scroll NEVER did stop all leaking anyway, that's why lip seals are better, but there isn't an easy/fully working lip conversion for Spridgets as far as I know.

Exactly how much oil is leaking?

I already have the Morris Minor Centre lipseal conversion, with a machined sleeve on the crank flange. I agree it's not perfect. The drip is not serious, ie the clutch stays dry, but there is always a little patch on the floor. Compressions are around 200 psi and even. Almost no oil mist from rocker cover.
Les Rose

I think you have the answers then.

The lip seal conversion isn't perfect. Or to put it less kindly, it doesn't work very well.

200psi is good. That's above the piston. Blowby is inevitable, and unless you have a better pcv system, and a better rear oil seal, there will be a leak.

Have you done Guy's rubber glove test? I seem to think you have.

Anyway, what it shows really well is that above idle, eventually suction INTO the engine via the oil filler cap, is overcome by the blowby past the pistons, and fills the glove. That pressure will find a way out. And since the first direct paths are in the crankcase, the front and rear of the crankshaft oil seals, that's where it will go. The scroll and the conversions can't keep it all in.

How about a separate vaccum pump, vented to a collection can? Better still install a scavenge valve in the exhaust manifold.

But if all you have is a small leak, I'd live with it.

Guy's castrol can conversion recommends itself I reckon. 🙂


Thanks for your thoughts Anam, but the question remains: The main issue is the idling. The PCV valve should not cause this wild variation. When hot it goes over 2000, then can suddenly drop to 1300 or so. As the valve controls the leak better than anything else, I need to make it work. I'm not happy with the on-board drip tray. When the leak was worse it fouled the clutch. I can live with the leak as it is, but not if I ditch the PCV valve. I've read that the exhaust-driven scavenge valve only works for race cars where throttle is either wide open or closed.

Maybe you are right about the extra breather hose. It is widely recommended but is it going to work if both air flows are going through the same restriction? However as the air should only enter through the tiny hole in the filler cap perhaps it's more a matter of pressure than flow?

I looked again at the metering valve, and it clearly isn't meant to seal as it sits proud of the lip under it. I've yet to find a clear explanation of how these things are supposed to work! I wonder about buying an aftermarket one.
Les Rose

The exhaust scavenge will work on any car as far as I know, and the effect increases with revs. At idle there is obviously less blowby, so less vacuum isn't an issue then.

Yep the leak and blowby are one thing, the idle is a different but clearly related issue.

The aditional vent was only meant to feed one vacuum source. If I understand your connections, they are fighting eachother. But that doesn't explain the fast idle when hot.

Can you post a picture of your setup?

I've got two vents. One on the timing chain cover, and one at the petrol pump blank. They are both connected to the su's via the y piece. It was done to reduce suction on the timing chain cover to prevent oil sucking, whilst still providing for sufficient crankcase ventilation.

I wonder if you have a heat related, random/intermittent air leak. Possibly the valve in the mushroom is sticking sometimes.


Here is a pic of my engine. As I only have one carb, there is only one vacuum source. I suppose I could drill the other branch of the manifold, but then I would need 2 valves and I'm not sure there would be any more vacuum. These valves are used on 1000s of cars and obviously do work for most people. The blowback valve is a possible source of sticking, but it actually works in the opposite direction so I doubt it.

Les Rose

The pcv valve is designed for use where the extraction is connected to the manifold. On over-run the manifold suction is very high and the pcv is designed to close off to prevent too much suction being applied. Its sort of counter intuitive as most of the time the pcv is open and provides a direct extraction of fumes (and crankcase pressure) with the engine running.

Later cars used the carb connection which not only produces a much lower and more appropriate suction, but also matches it to when the engine is pulling under power, which is when blowby is at a maximum. The other critical thing, at least on the latter design,is that inflow of air is limited to the vent on the oil filler cap. Additional orifices and leaks allow too much air in and upset the delicate designed balance of inlet and suction.

Have you tried omiting the PCV and venting directly to the vent provided on the carb - is it an HIF44/HIF6?
If so it should have a vent on the side which I guess you have blanked ?

Thats how mine vents on an HIF6 direct from the timing cover.

richard b

Les, you can buy a small bullet shaped pcv valve, as used on Fords. See pictures. They used to use the mushroom type, but changed to the bullet type on engines after circa 1964. It does the same thing, and works the same way. The early version can be stripped, the later is sealed.

I have both, I've fitted the later one to my Capri. I converted it from a draught tube system, to a fully sealed system.

The valve is NOT meant to fully seal at idle, because there is still crankcase pressure, and blowby gasses at idle, which need to be extracted.

I misunderstood how you had connected the valve. Ok, yep, in theory there shouldn't be a problem. But as Guy says, the valve is finely balanced, and counterintuitive.

The spring keeps the valve closed, against the manifold suction trying to open it.

Too much 'suck' and it stays open, less, and it's closed.

But then you have to add in the crankcase pressure, also opposing the spring, and acting to open the valve.

At higher revs, when the throttle is 'open', then the air draw down through the small hole in the oil filler cap, doesn't upset the mixture too much. But if the valve is open too much at idle, it's effectively an air leak. Rev's increase.

When hot, the crankcase gasses expand, more pressure; even at idle. That pressure is working against the spring in the valve, but aiding the manifold suction, to open the valve.

How weak/strong is your spring?

Do you have excessive piston blowby, and thus crankcase pressure? A compression test won't tell you.

But you CAN measure it. A 'simple' indication is Guy's glove. An accurate measurement like this. ---

"Not only can you measure crankcase pressure with a vacuum gauge or manometer, you can also use an accurate pressure transducer such as a Pico WPS500 to measure crankcase pressure with a scope. A scope and pressure transducer may also be able to show pressure pulses inside the crankcase that can be caused by excessive cylinder wall to piston compression leakage that escapes into the crankcase."


I'd start though by increasing the spring pressure in the mushroom. I think that fine balance, is 'fluttering', when the engine is hot enough to make crankcase pressure at idle, sufficient to upset the balance.


Here's a picture of the early Ford mushroom. Top picture.

Haynes had a good picture showing the mushroom on our A series, and you can see how it works, tied in with my explanation in my previous post.

"The spring keeps the valve closed, against the manifold suction trying to open it. Too much 'suck' and it stays open, less, and it's closed. But then you have to add in the crankcase pressure, also opposing the spring, and acting to open the valve."

HOWEVER. looking at that haynes picture, the bmc mushroom, works the opposite way to Ford's bullet pcv valves.

In the bmc mushroom, the spring tries to open the valve, and the manifold suction tries to close it.

But still, too much crankcase pressure will tend to open it.

So WEAKEN the spring, to assist the manifold suction at idle.


Les, I wonder if, by adding that additional pressure relief at the petrol pump blank, it's upset the pressure differential of the pcv valve too much.

Try blanking off that 2nd feed to the pcv valve. Block the petrol pump opening, and block the 2nd opening into the pcv valve, put it back to original.

Yes, the oil leak at the rear worsens, but just to isolate the fast idle problem.

See what happens to idle then.

The carb is an HIF44. I previously tried the direct connection to the vacuum port provided on it, but got more leaking. It seems to me though that there is no point in having 2 breathers going to one vacuum source. Other posters here have recommended that, but I think it's more appropriate for breathing to atmosphere. What do you think?

I will have another go at the direct connection to the carb. I just set that up, warmed up the engine, and did the blowby test. Both the beer mat and the rubber glove were sucked well down at idle and when blipping the throttle. The revs rise when removing the filler cap. I have reset the mixture and idle speed so we'll see what happens with a longer run.

Oh, I also put an O-ring on the dipstick and fitted a spring to hold it down. There is already a bit of fuel hose glued to it but there was not going to be a good enough seal with hard rubber and only gravity.
Les Rose

I suspect your fast erratic idle could be caused by pooling of oil being ingested by the PCV. So oil mist collects in the horizontal bit of hose and when enough has collected the high idle vacuum gulps it in which increases the revs. I had something similar and ended up taking the hoses into a sealed catch tank and running another out to the PCV. This managed the oil mist issue but still allowed a sealed system.
If you have quite a high lift cam the alternative is simply delete the PCV and run the hoses to atmosphere. Idle can't be effected by going above atmospheric and ingesting a third party. But I think the PCV only works on a basically standard engine with factory tolerances. Once you change enough parameters it will tend to become compromised.
f pollock

Les, I use the double vent, NOT to reduce sump pressure more than a single vent, when going into a single 'vacuum' source. It may not.

I use it to reduce the suction at the timing chain cover, by splitting the suction. I did it ONLY to combat oil sucking, which it has very successfully.

Yes, a single port connection to a single hif44, will result in less suction, but at a constant depression. It would be greater if you had two hif44s, or even two original 1.1/4" SU's.

No, breathing to atmosphere, is no better, and possibly worse than a draught tube. At least with a draught tube, you get SOME suction when moving at over circa 20mph.

If you want open vent, then you need MORE vents to fully overcome the sump pressure.

That you have oil leaking from the rear oil seal, is either the seal is useless, or you still have sump pressure blowing the oil past the seal.

The glove/beer test only tells you that air is being sucked in. It doesn't really tell you if you have sufficient suction. Some race engines run at full sump vacuum pressure. It improves performance in some way. Better than open venting.

AND, a pcv system is not there only to reduce pressure. The V is for ventilation. By pulling air in through the top of the engine, and all the way through it, you 'Positively' carry out all the burnt gasses and condensed moisture, that otherwise degrades the oil. Your engine lasts longer with good oil.

As I asked before. How much oil are you losing at the rear oil seal?

It seems to me that you are trying to solve a problem that can't be/wasn't solved on a series engines. Because, as yet, nobody has come up with a rear scroll to lip seal conversion, that works 100%.

The best you can do is compromise.

Why not try those suggestions I made earlier? 😉


Anam, I'm not sure which suggestions you mean, you made quite a few! Anything to do with the PCV valve I have put on the back burner. The actual volume of oil lost is hard to measure but it's not enough to empty the sump. After my 200 mile run recently the level was unchanged. In the past, when the sleeve on the crank fell off and the seal could not work, the clutch was soaked and unpleasant to use. I am not getting that now, just a 5-10cm patch on the garage floor. The leak is much better with the seal kit than without, and I suppose you are right in that I am trying for perfection. However the primary reason for this thread may have been solved, as the idle is better. Not perfect, but it is a 270 degree cam. I actually wonder if the sealing of the dipstick might be a game changer, but we'll see.

...and yes, the engine is far from standard, so I should expect problems with the ordinary PCV valve.
Les Rose


1. Try blanking off that 2nd feed to the pcv valve. Block the petrol pump opening, and block the 2nd opening into the pcv valve, put it back to original. What happens to idle? Is it stable when hot?

2. Try WEAKENING the spring on the valve, to assist the manifold suction at idle. What happens to idle? Is it stable when hot?

The oil is very likely NOT leaking out of the rear at IDLE. It's happening at rev's, when the suction is mostly always lower than at idle.

Suction at idle doesn't tell you what suction is at normal road going rev's.

Ok, you have a performance cam. Read the following, suggetions for different 'strength' pcv valves.


"PCV Valve Suitable for Performance Camshaft with Lower Vacuum Signal

I have tested various PCV valves to determine which perform better with the lower vacuum signal of a performance engine, using a camshaft with more duration and overlap. PCV valves are not all calibrated the same, and perform differently depending upon the amount of vacuum an engine produces. A performance camshaft will typically produce a lower vacuum signal than the stock engine/camshaft, and the stock PCV valve may not close enough at idle, which can cause an irregular and/or high idle, or otherwise not function efficiently throughout the operating range of the engine. An improperly functioning PCV valve can also cause the premature break-down of engine oil, sludge build-up, leaking gaskets and seals, water vapor build-up, as well as allow excessive oil to enter the intake manifold. A PCV valve that closes too easily and completely, can stay closed too long above idle, and can also interfere with the proper evacuation of blow-by gases.

The following two PCV valves function better at a lower vacuum signal, but still open once the engine vacuum changes during all-around operation. They both begin to close at a lower vacuum signal of ~1-2” hg. One is standard flow and the other is high flow. The high flow unit increases flow by 40% over the standard flow unit. It works well with a stroker motor, or shorter stroke, high rpm race applications."


Read the rest here.


The difference is pcv valves, is going to be largely governed by the internal spring strength.

Hence my suggestion to start fiddling with the spring.

The spring in the pcv keePs the valve open in its normal position. As suction increases the cone (Ford) or diaphragm ( BMC) is pulled with the ait flow against the spring. Eventually as suction increases the spring is overcome and the valve closes.

Weakening the spring will cause the valve to close more readily cutting off the flow of crankcase gasses at lower revs, though not at tickover when both manifold suction and crankcase pressure are at a minimum.
Weakening the spring will not assist crankcase extraction.

Thanks Anam, I'll put all that into my back pocket while I evaluate the direct connection option. Watch this space!
Les Rose

Direct connection to manifold without a pcv, will suck up oil, as well as cause a massive air leak at idle, if you could get it to idle.

Direct to su, with no pcv is fine, but reduced suction, whilst also constant, will resolve idke speed issue, but won't reduce oil leak.

I agree Guy. Weakening the spring won't improve ventilation. The issue seems more to be about increased, uncontrolled idle speed.

Actually, I WAS right the first time. I justed tested my spare ford valve. The Ford and bmc valves work in the same way. The drawing confused me. lol.

On BOTH, the spring is trying to keep the valve open. Suction is trying to close it.

A pcv valve is a controlled air leak.

To reduce the 'air leak' at idle, if the spring is weakened, the suction will overcome the spring, and idle speed will be more controlled.

The issue as Fergus pounted out, is your crime in souping up your engine. 😁 Now you have to alter the spec's of the pcv sysyem to match it.


So what breathing system did the last of the Classic Minis have? I haven't been able to find out.
Les Rose

One thing I think they had was a factory fitted lip rear oil seal.

What you need to deal with competing issues, is an engine management system. Obviously not possible.

The other suggestion I made was an electric vacuum pump, adjustable. That would completely isolate the air intake from the crankcase ventilation, and allow you to set a constant level of vacuum, perhaps sufficient just enough, to pull back any oil at the rear oil seal. If you found one with a feedback input, you could fit a sump pressure sensor, to dynamically increase/decrease vacuum as required.

Mini rear oil seal.

Yes I recall that the late Mini had a lipseal. But this makes me wonder, why was the upstream vacuum point fitted to the HIF carbs? How was it plumbed at the factory?

I have rummaged through my large stock of springs and there is nothing suitable - all are too stiff. I think I will see how the direct connection works for a week or so, and then try the single breather going into the PCV valve.
Les Rose

The crank lipseal on later A series engine is far better at controlling oil leakage,but does nothing for excessive crankcase pressure, which if not dealt with will blow out the front crank seal (on a mini). So cars fitted with the HIF carb, such as the Maestro still had a connector for the extraction system.

Yes Guy, but what I am curious about is what plumbing it used. Was there a PCV valve?
Les Rose

You don't need a pcv valve when the connection is a low suction point like at the carb. The pcv is only there to limit the very high manifold suction which is taken off downstream of the carb butterfly.

Les, all su carbs are constant depression.

" It is now well over fifty years since the Skinner brothers first introduced their constant depression carburetter. "

The pcv valve can be omitted because of that. All that's needed is a connection to the area where the constant low depression is. It does vary a bit, but not by much. Near perfect for a pcv system. But, not if you have excess sump pressure, and or an imperfect oil seal at the ends of the crankshaft.

Modern engines have a far more sophisticated system, with feedback and control over the intake.

I think I have known SU carbs are constant depression for at least 50 years! I just wondered if there was anything else involved in the late Mini plumbing.
Les Rose

What makes you think the A+ engines fitted to late minis and metros, didn't have problems?

My final suggestion is you do away with the mushroom, connect to the su, and live with the oil leak. Because unless you can perfect the lip seal conversion, you'll always have an oil leak from the rear seal.

Actually, here's another suggestion. Copy bl, dump the a series, and fit a k series.😉

BMC / BL attempted to refine and improve on the crankcase breather system throughout the life of the A series, from the primitive draught tube of the 948cc cars through the mushroom pcv on the manifold set up on 1098 and early 1275 to their 1275 system via the su. Non worked 100% but the latest su induction version was the best, especially considering the increased displacement and likely higher blow by problems.

So why not go with that? Given that the su connection is also a constant depression it is also likely to give a far more consistent "air leak" making mixture tuning much easier. The key to getting it working at its best is not to mess with extra breather holes to atmosphere, use the correct oil filler cap and seal the dipstick tube air leak - which you have done.

Yep I agree Guy, that would fix the fast idle issue completely. But Les wants to reduce/elliminate the rear oil seal leak too.

My argument is, because the rear seal is ineffective, and he has a tuned engine, it can't be done; even with the best pcv system for a series engines as you describe.

It may be considered that I was joking about a separate vacuum pump for pcv. I wasn't.

"Pulling vacuum with a dedicated pump has been the preferred method for well over two decades now and the technology has steadily advanced with important performance benefits that include:

Improved ring seal by relieving crankcase pressure below the rings
Reduced friction by allowing lighter tension rings
Reduced windage around the rotating assembly
Improved oil scavenging
Reduced parasitic pumping losses
Minimized cavitation (in wet sump applications)
Reduced lubricant contamination
Improved bay to bay breathing in Y-block configurations
Improved power via reduced mixture contamination
Reduced gas port contamination"

An external vacuum pump evacuates excessive crankcase pressure, resulting in less-than-atmospheric pressure in the sump. This pressure is readable with a vacuum gauge. The size and speed of the evacuation pump, along with the size of the vacuum line, control the amount of vacuum."

My words.
This would even sort out the rear scroll I would think. And since this is about improving the rear oil sealing, rather than increasing engine power, it wouldn't need a pump capable of 12-to-14 inches of vacuum. Probably 5 inches would do it, so a smaller pump, electrically driven, rather than belt drive.

I wasn't commenting on your suggestion Anam
Your suggestion of about 5" suction for a pump extraction is probably about right. I recall that I measured suction at the SU connection as about 2". I know it was very low, surprisingly so, I thought.

I know you weren't Guy.

Yep looking at the specs, pcv valves seem to range around that 2 inch figure. Only a guess, but perhaps that's the most you can do with air intake as a source of vacuum, without upsetting carburation too much. Hence exhaust scavenge or an external pump.

What I found interesting reading from that link, that is now obvious to me but wasn't before, is how much piston ring sealing is affected by sump pressure.

Pressure from above the rings, pushes the rings out against the bores. Pressure from below the rings, caused by blowby, counters that and reduces ring sealing, causing more blowby.

Create a high enough vacuum in the sump, circa 14/15 inches, and ring sealing will mostly only be limited by the bores and condition of the rings etc. Better sealing, more power. Maybe or Probably not noticable on even a tuned a series, but significant on bigger bhp engines.

Might be worth playing with, just for a bit of fun. 😁

Here is the new setup. Fuel pump point blanked, oil separator on front side cover connected to carb upstream of throttle. It's only petrol hose but I don't think 2" vacuum will collapse it 😁. I will tell you how it performs.

Les Rose

Well whatever it does, it looks bloody nice.🙂

Progress report. Short run the other day and the idling is better, but still sometimes sticks at 1300. Hard to say re the oil leak. What is odd is that there is a bit of oil around the top of the separator. The new vacuum hose is tight, so the welded top must be leaking. It's a swine to get off as there is no room behind the exhaust manifold to get the side cover off, so the manifold has to come off as well. If the separator is leaking oil out, it could also be leaking air in. I can put a bead of weld around the edge - after flushing out the oil of course!

While I'm at it I will replace the manifold gasket, just to be sure there are no air leaks there.

I am still puzzled by the physics of all this. I know I am getting vacuum at idle, but with the connection upstream of the throttle I would be surprised if changes in suck caused significant speed variation. It should be almost entirely regulated by the throttle.
Les Rose

Les, read this for the explanation.

I thought you all might appreciate an update. The idling is much better, but still rises a bit when the engine gets hot. Actually this isn't too bad a thing as the radiator fan takes over 10 amps. With the dynamo this just ran the battery down a bit, but the alternator responds with more juice and slows the engine! I think a bit more fiddling can get the idle even better, as it settles after maybe 10 seconds so there may be something else going on. The oil leak is about the same as it was with the PCV valve, so I have lost nothing by ditching that. Considering the on-board drip tray! The leak is not enough to foul the clutch so I will just have to live with it.
Les Rose

I have done some longer trips at speed so the testing has been more rigorous. Basically the new vacuum tube to the carb does not work. The idling is better but the oil leak is much worse. Today I attended the big car show at Farnborough, over an hour up the M3. Allowing that amount of leaking would not be tolerable for that distance, so I changed back to the PCV valve yesterday. I carefully set the mixture and idle speed. Indeed the idle was more even if I backed off the mixture a quarter turn - about 900 rpm.

Towards the end of the outward trip the engine started to run hotter. By the time I arrived the idle was up to 1500 or more. It got even hotter on the return, with the cooling fan switching on even at 65 mph. Back home the plugs told me that it was running weak.

What seems to be happening is that the PCV valve is changing its setting when hot, allowing more air into the manifold, pushing up the idle and weakening the mixture, which then makes the engine run even hotter. This is confirmed by a noticeable drop in oil pressure, and no doubt the thinner oil leaks even more. Vicious circles.

What I need is a PCV valve that works. I am thinking about an aftermarket one, but not sure how to choose.
Les Rose

Les, you need to resolve the overheating issue before anymore changes.

I don't have an electric fan. I can drive at 80mph up a long incline in very hot weather, and not overheat.

When was your rad last renewed? Standard core?

I changed to the later crossflow rad on my '66 sprite, and keep a 180 stat in all year.

If you don't want to change, I think you can have the vertical flow rebuilt with a bigger core.

The standard vertical rad has been recored twice in the last 15 years. It is cleaned and flushed regularly. After enriching the mixture the engine ran cooler today. After last year's dyno session there was no overheating. I am sure it is related to mixture.
Les Rose

Les, if you are using a PCV system, still getting oil leaks and think it is causing the mixture to weaken too much those could all be symptoms for the same thing. For the crankcase ventilation to work as a negative pressure system the amount of air intake needs to be correctly restricted. Too much air flow into the crankcase will mean the extraction cannot keep up and you don't ever attain the negative pressure in there so it continues to leak oil. At the same time, it draws too much air through and you get excessive weakening of the fuel mixture.

You need to have only the correct oil filler cap with the restricted air slot. And all other possible air entry routes need to be as near sit tight as you can make them. Adding any secondary routes for the crankcase to breathe is self defeating.

I guess you know all this! 😁 But I thought I would say it anyway.

There are many things I don't know Guy, but this at least I think I do know! I have the correct filler cap and it is clean. The dipstick has an O-ring and is held down with a spring. I even checked the oil separator for leaks in a bucket of water. I am obviously not getting enough suck. I tested the valve by blowing into the intake and it a hard blow opens it. I thought the idea was to stay open at moderate vacuum (cruise) and close at high (idle). That doesn't seem to be happening, but I have replaced everything in the valve that I can.
Les Rose

My 1275 Sprite had a lip seal conversion and no Pcv. The leak from the Bell housing was horrendous. I bought two original PCVs on eBay, lifted the engine, ditched the lip seal and refitted the original scroll top cover, fitted the PCV with a new diaphragm and success no oil drips at all. I now have a spare PCV which may need servicing and will post it to anyone if they will put a couple of pounds in a cancer charity box

Jan T
J Targosz

It's worth a try Jan. Maybe the body of mine is too worn for the spring loaded valve to work properly. Please email me and I'll send you my street address. Very happy to pay the postage and support a charity.

lesrose1959 at
Les Rose

Sorry to keep boring you with this, but I have been trying to test the valve and I'm still confused. The parameter is the pressure difference across inlet and outlet, so blowing the inlet is the same as sucking the outlet. If I blow on the inlet there is initial resistance and as pressure increases the valve opens and allows more flow. I am not sure this is right. I thought the idea was to limit flow at high vacuum, ie at idle, and allow more at cruise.

In addition there is the temperature issue. When hot the idle goes up. Obviously I can't test this off the car, and I can't visualise what is causing it.

I have searched for days and can't find any reliable guidance on how the Smiths valve should work. I am amazed at the amount of frankly erroneous advice out there, with people even thinking it is an oil separator.

I could go back to the carb port vacuum connection, which gives a stable idle but the oil leak is worse. I originally vented to atmosphere, but went for the PCV because of the persistent oil leak. That was better for a while, until the idling problem appeared about 2 years ago.
Les Rose

Barney Gaylord, who drives his MGA hundreds of thousands of miles, is usually very good so how about this if you've not already seen it. -
Nigel Atkins

Thanks Nigel, no I had not seen that, despite extensive searches. I have not been testing it correctly. It is probably working. I checked the port in the filler cap and it is 3/16". Is that correct?

I still can't explain the variation of idling speed with temperature.
Les Rose

Is the filler cap the plastic vented (horizontal slots at top) and filtered type as per photo?

Nigel Atkins

A shot in the dark Les; re temperature and idle.

What distributor, what kind of ignition timing advance?

Yes Nigel, it's exactly that type of cap.

Anam, the dissy is A Plus 998 Mini, Powerspark electronic pack. I changed back to points recently and no improvement.

This has got to the point where last night I started looking at the parts of the spare engine. Interesting to see that the surface of the rear main bearing cap that abuts the scroll has wear marks.
Les Rose

Les, I wasn't wondering about points vs electronic, I was wondering about vacuum advance; if you are using the 'can', and if it's ported or manifold. It can affect idle and temperature.

Your pictures don't show a vacuum pipe.

The HIF is manifold vaccum unless you re-drill it for ported.

Sorry Nigel, I misunderstood. Yes I am using the can, and I have drilled the carb for ported vacuum. There was another long thread about this. I tried running without vacuum and it didn't help.
Les Rose

Then maybe an induction air leak caused by a crack that only opens when the temperature is up?

Fair question Anam, but the idle is stable without the PCV and connecting the breather to the carb. The problem with that is that the leak is worse.
Les Rose

perhaps if you've not all ready tried you could experiment with getting more air through the cap or less air air through the cap to see if that makes any odds at all.
Nigel Atkins


In conclusion then, I think it's reasonable to say, that you cannot resolve this problem with the pcv valve you have. Not because the valve is faulty, but because it is specified for an engine with different parameters, to how your engine is now configured. And you can't easily adjust/alter how THAT pcv valve performs, to make it operate properly on your engine.

So, get one that you CAN adjust.

Introducing the Dual Flow ADJUSTABLE PCV Valve.

" -- splits airflow into independent idle and cruise circuits, giving you the ultimate in control over your tuneup. Each circuit is independently tunable to allow for full customization to your specific application. Flow rate as well as transition vacuum level can be adjusted with the turn of a screw."

$129.00 US.

And they deliver to the uk.

Subtotal $129.00
Estimated Shipping $24.95
Tax $0.00
Total $153.95

Owz that?


"Understanding how a PCV valve may be causing your engine’s problems"

"Our fully adjustable Dual Flow PCV valve may be the cure for your engine’s chronic engine oil leaks, in-car oil smell, a non-responsive carburetor or EFI system, idle concerns and prematurely dirty crankcase oil."

"The Dual Flow Adjustable PCV valve was born from the same frustrations that you have endured trying to solve some nagging engine problems… first and foremost we are hot rodders too!"

"The Dual Flow PCV valve is the result of us trying to diagnose unrelated engine issues that did not make sense but would usually happen at the same time.

We spent many sleepless nights trying to figure out why the rear main seal was leaking when it was carefully installed, how three different carburetors could respond incorrectly on the same engine, and why we smelled oil while driving."

Read the rest of it.

"If you have any questions we welcome the opportunity to assist in solving your engine’s crankcase ventilation issues. Please contact us at:


Read all about it.

In depth magazine artical about pcv and valve function, as suggeted by ME Wagner.

Sound familiar Les?


Not too bulky either.


Says it all eh Les?.


Have y'all realised this is related to oil sucking? 🤪.

Seriously, I feel a diversion coming on. 🤣🤣

Anam, "The HIF is manifold vaccum unless you re-drill it for ported." Not necessarily so. My HIF, unmodified, has a ported take off. I think it depends what car it originates from. Mine was off a Maestro, as far as I remember.

Anyway, that's a nicely engineered little PCV device. Pricey, but will appeal to many I imagine.

Ah right, the maestro being a later car. That would have been handy to know when I went looking for my hif.

Yep it does look well made. Pity someone more local hasn't done the same thing. If cheaper I'd consider being tempted too, not that I need it.

This thread was discussed between 21/06/2022 and 26/08/2022

MG Midget and Sprite Technical index

This thread is from the archive. The Live MG Midget and Sprite Technical BBS is active now.