Valves Leaking

[bobden72]

A few days ago there was a discussion on valves leaking / seats, I have this problem with my Cirrus and joined in the conversation, but my problem was sort of side stepped with no suggestions. So I thought if I started a new discussion some one might have an ideal.
I have just finished a Cirrus and can not get any compression as the valves are leaking. I have cut the seats to be 1/32 wide as per the drawings and ground them in by hand, but found them leaking. Recut the seats and ground the valves to a 1 deg shallower than the seats in my lathe using collets to ensure everything was true. I then blued the seats turned the valve by hand again and got a nice clear seat, but they still leaks. I made the lapping tool as described in SIC mag again no joy. Each time I try another ideal I am aware of the seats getting wider, from what you say the seats should be narrow, is there any remedy to get these valves and seats to seal. As there is about a months worth of work in this cylinder head and don't want to scrap it if it can be helped.
This is my sixth I/C engine using the same methods as before and they worked but this one has beaten me.

 

[tornitore45]

OK Bob I give it a shot.
Once the seat and the valve are machined following the technique of using different angles, like 45 and 44 degrees, then the necessity of keeping the seat land to a minimum is eliminated. The first contact is a circumference of zero width IF, and here is the rub, the valve and seat are concentric. This is a long winded way of saying that your issue may be concentricity.
The seat must be concentric with the stem bore, for this one make a tool with a pilot dowel. Let assume you got this licked.

Now, I love my lathe but accuracy is not its best quality.
I make my valves as follow:
Let's say the stem is 1/8" diameter and the head is 3/8" dia. and 1/8" thick
I braze a 1/4" long 1/2" dia cylinder near the end of a long 1/8" drill rod leaving about 1/8" stick out.
Chuck the stem close to the valve head and center drill the stick out piece as accurate as I can.
At this point I chuck the stem around 3" from the valve head and center the stem near the chuck.
The stick out piece is now kept by a dead center.

There may be error at the chuck end, the dead center may not be perfectly on center but one thing is sure any cut on the head is very much equidistant from the fixed dead center. Circular and on center. The long stem centering error is cut down by the ratio between the 3" length and the 1/8" stick out.
I machine the head toward on the stem side leaving the other end for later.
Part off and use the long stem stock for the next valve.
The flat part of the valve head can be faced and brought to size chucked close, concentrity at this point is not important since is a facing job.

The advantage of making valves in two pieces is that the requirement to cut a long aspect ratio stem parallel is eliminated by using a ground drill rod.

 

[bobden72]

.
Mauro thanks for your input and your time just to clarify my seat cutting tool and the lapping tool both have a pilot dowel to keep them concentric as per SIC. I also make my valves as do you in two parts silver soldered together then machined again as per SIC.
My next plan is have some 600 grit lapping compound on order and I will try lapping with that. Before I have been using automotive fine grinding past which I have found out is usually 350 grit. If that fails I have in mind to make a tool with a pilot to cut out the valve seat to a depth of 1/8" and insert a bronze seat and start the lapping again, but only using the lapping compound to cut the seat thus ensuring a narrow seat. Or maybe make new valves with a few degrees difference from the seat to establish a narrow seat. What do you think.
Bob

 

[goldstar31]

Having modified a few full sized cylinder heads, I would be loathe to have valve seats of anything other than cast iron for leaded fuel and hard steel inserts for unleaded stuff. I was taught to use a full coverage of the exhaust valves to dissipate the exhaust heat but a mere ring for inlets is satisfactory.

Yonks ago, when I was doing a City and Guilds in Motor Vehicle Restoration as a manure student , I was told that Ford Motors did not grind in their valves. Maybe, letting the valves hammer in will be sufficient because of this.

My thoughts, of course

Norman

 

[Cogsy]

Personally, I make 1 piece valves at a nominal 45 degrees and make a seat cutter at the same time, so I know it's identical to the valves. I then cut the (thin) seats and lap the valves by hand for roughly 2-4 minutes using regular toothpaste and the standard oscillating motion plus small rotation every once in a while. Of course once lapped each valve is now 'matched' to a particular seat. This method gives me enough of a seal to generate suitable compression to fire and the combustion pressures quickly seal the valves fully. Your mileage may vary but it has always worked for me (drill rod valves in brass cages) and I never use gritty lapping compounds.

 

[mayhugh1]

Have a look here starting with post 222:

https://www.homemodelenginemachinist.com/threads/quarter-scale-merlin-v-12.24153/page-12

Terry

 

[mayhugh1]

Also here starting with post 185:

https://www.homemodelenginemachinist.com/threads/another-radial-this-time-18-cylinders.21601/page-10

Terry

 

[bobden72]

Have a look here starting with post 222:

https://www.homemodelenginemachinist.com/threads/quarter-scale-merlin-v-12.24153/page-12

Terry

Very informative thanks Terry.

 

[mnay]

this suggestion may shock the precision machinists. I had trouble with valves seating on a couple of engines and one of the old timers suggested that I make the valve stem and guide a little looser.
This allowed my valves to seat and worked great. I don't know that this would work as well on a high speed engine, but it sure worked great for me.