Quantifying feeler gauge drag

[Luca]

I was talking to my brother the other day about not having a good standard for how much drag a feeler gauge should have when adjusting valves.  After calling me a "Nervous Nellie" he offered a pretty good suggestion: set a micrometer to a certain gap and run a feeler gauge through it, dialing the mic in and out a touch to note the change in feel.

Finally, a better way than guessing at "slight drag."

[Justin B.]

That's not a bad idea, I think I'll try it tomorrow and see.

[Barry]

Never thought of that one before although I have measured feeler gauges and was surprised to find either the feeler or the micrometer was not accurate.  Not that it matters that much for valve clearances where it's much more important that both cylinders are the same than that they are dead accurate. And that's where the feel comes in. I'll have to give it a try.

[Tony Smith]

Guys.

You do realize that the actual dimensional difference between "loose as a goose" and "so tight you can barely move the feeler" is not going to be enough to make any difference don't you?
 
For what it is worth. the "correct" level of drag is a clearance that is such that you can both draw the feeler gauge through and "push" it back (assuming for the moment we are using the thinnest gauge in your set), without distorting or "bunching" the gauge.

I generally do not use feelers to set valve clearance anyway. I won't give the whole game away, but follow this logic and your problems are behind you.

Firstly, what we need to measure is the running clearance of the rocker to valve. As our wonderful machines age you are not really able to accurately measure this running clearance with a feeler gauge as the end of the rocker wears a slight hollow where it constantly hits the valve head. As the gauge is much wider than the contact area you are not measuring the actual clearance anyway.
 
Now you can rush out and have the tappets refaced, but the hardening on them is very thin, so budget on getting that redone as well - the price should be about 50% of the price of a new rocker, but this is needless expense, my own view is that you only have that job done when you fit new valves and then you get the lash-pad built up by weld (not ground down) and then deep hardened.

The clearance is adjusted by a screw and locknut. The thread is a known number of turns per millimetre. The amount of clearance required is a fraction of a turn, as in a quarter og a turn, 1/2 a turn or by way of visualising a "clock". If you have a rocker ready to set clearance, all you need do is wind it in until it gently contacts the valve, call the position of the slot "12 o'clock" then wind back to 10 to 12, 1/4 to 12, half past etc to accurately and repeatable set the actually running clearance to whatever is required.

Let's say that our valve adjuster has a 1:00mm thread pitch (I am embarrassed as I no longer recall what standard pitch on a BMW adjuster is, I know mine are 1mm, but I made them myself years ago and can't recall if that was stock or whether stock is 1.25mm, in any event the principle remains the same).
 
So, if the clearance required is .25mm, or 1/4mm, then to achieve that clearance involved rotating the adjuster 1/4 turn, or taking our imaginary 12:00 start point - rotate back to 11:45.

.15mm equals a rotation of 54 degrees - near enough to 60degrees in my book as makes no difference or "10 to 12" in round numbers.

There you go, I just set your valve clearances without a feeler gauge, and probably set them more accurately than you can with a feeler,

[Barry]

There you go, I just set your valve clearances without a feeler gauge, and probably set them more accurately than you can with a feeler,

The whole subject of accuracy and precision is interesting. True accuracy is impossible i.e. the gap will never be exactly 0.25 mm as there will always be an error no matter how small.

In the scheme of things accuracy (achieving a dead on 0.25mm)  is less important than  precision or repeatability (achieving the same result every time) on both cylinders. I guess your method should achieve good precision providing you can turn the adjusting screw by exactly 1/4 of a turn each time.

There was a tool called the SPQR tappet adjuster based on the same principle. I tried one back in the 70's and the problem with it will be familiar to everyone. Setting the adjusting screw by a number of clicks was easy. Keeping it in that exact same position while tightening the lock nut and at the same time allowing for the gap changing as the slack in the threads is taken up was not.

[georgesgiralt]

Hello !
You are trying to split hairs here.
When BMW specifies a play between the valve head and the rocker it is a compromise between not enough and no play when hot and too noisy or too much for the life of the rocker. Remember the early settings from BMW for the play on the R65 ?  
So if you aim for 0.20 mm and achieve 0.18 or 0.22 it does not matter. Of course it is better to err on the "too much" side but a feeler gauge is quite overkill for the job. (remember the old car engine where the play was adjusted by ear engine running ? A young mechanic set too low a play and an old and deaf one set for excessive play... Both engine ran fine even if noisy...)
The situation is different when you look at the main bearing play or at the con rod adjustment. Here precision is mandatory if you want the pieces to run on the oil film and not metal on metal... But none of us will use feeler gauges for this job, do we ?

[Barry]

Georges,

I agree the exact gap is not critical. I choose to run them slightly wider than spec with out even any noticeable increase in noise.

That both cylinders are as close to the same as you can get them is important as otherwise you could have different valve timing and lift on each cylinder. Maybe only a few degrees different but it matters. Valve clearances have a real and demonstrable effect on idle speed and idle mixture settings. This is why it's always recommended to set valve clearances before tuning the carbs.

[Luca]

I know that valve adjustment isn't overly critical, as long as there is some lash, but it's still nice to have a way to quantify feeler gauge drag... and "how much drag is the right amount?" is a question you find all over the internet.  This is an easy way to answer that.

Tuning up your feel for clearance can be helpful for measuring valve recession if you want to keep track of that in your maintenance log.  Also, having a standard for feeler gauge drag sitting on your bench will help you consistently set the valves on both sides of the engine.  Unnecessary?  Probably.  But it doesn't hurt and it might help some newbies do the job with confidence.

P.S. A long handled wrench on the adjuster will give you much more torque to hold it in place while you tighten the locknut.

[Barry]

Tuning up your feel for clearance can be helpful for measuring valve recession if you want to keep track of that in your maintenance log.

That's a very good point.

I don't get any noticeable recession between 2000 miles checks but I only know that because I set the gaps as accurately as I could on the previous service and  recorded the settings. It reminds me of Duane Aushermans advice on valve clearances

"The main reason to do the valve adjustment is to check for a change in settings from the previous time"

http://w6rec.com/duane/bmw/valve/index.htm

[Justin B.]

Why not use a dial gauge and measure the amount you can move the rocker?

[Tony Smith]

Why not use a dial gauge and measure the amount you can move the rocker?

For mine, simplicity is the key, hence my method of valve adjustment. Mind you, for the reasons I outlined, I reckon my method is more accurate anyway.

You probably wouldn't approve of my using a packet of matches instead of a scale ruler when flying either, but then once you have a 50, 25 and 5 mile rule, you can mentally break down other distances with more than sufficient accuracy, which is the key.

I also own dial gauges, but it would never occur to me to use one for setting valve lash. Firstly you (with standard BMW adjusters) cannot even begin to achieve the necessary accuracy when lock-nutting the adjusters, and secondly, even if it were achievable, you simply do not require that degree of accuracy.


And to further offend accuracy Nazis, I used to set timing with a "roll your own" cigarette paper and points gap with a Coke can ring-pull - in both cases the extra "accuracy" achievable by using more sophisticated tools is either illusory or simply not worth the effort.

[Burt]

Ah Tony, I love it.  Your bush engineering skills are coming through.  

I have had a couple of breakdowns in the middle of nowhere with one resulting in being trailered home.  Some folks here (no disrespect intended) do not fully appreciate some of the remote places we travel to here in Oz.  

It is one of the reasons airheads were so popular for travelling "out back".

Dave

[AlfromNH]

Tony, your method strikes me as brilliant! I'm trying to find a reason it wouldn't be accurate, because its definitely too easy!  

I may hafta try it out today, thanks!

[montmil]

I've never been a smoker but a bit of cigarette package cellophane was always in my tool kit. Easy way to tell when the points break.

[Luca]

Nothin wrong with odd styles of measurement if they work, and I'm not against being resourceful.  Nor did I mean to start an Accuracy Nazi thread; just help people quantify feeler gauge drag if they're so inclined.

FWIW, I find feeler gauges to be about as simple as it gets for setting valves.  Get the adjustor set while dragging your gauge and lock it down.  When using feelers you can double check your adjustment, and there are no calculations to perform or degrees of rotation to monitor.

Two feeler gauges in your kit (three if you've got points to set) will take up about as much space as a stick of gum.  If I was going to motor out into the middle of nowhere, I'd probably just make space in my luggage for them.