Precision cross feed setting

[Gordon]

How do you set the lathe cross feed to a precise angle? Seem like after all this time I would have figured out the correct method. I have to machine a long valve seat and the mating valve to a precise 45° angle. Eyeballing it obviously is not the correct method. The problem is that for one piece the compound must be set at 45° right and the other piece must be 45° left. I have a precision triangle but holding it against the chuck face while I move the compound and work an indicator leaves me short on hands. I am not even sure that the chuck face is rally the proper surface to index from.

 

[Cogsy]

If you use the George Britnell valve seat cutter method, you don't need to alter the compound angle from the first setting and you're guaranteed to have perfectly matching angles. Basically you just make an extra valve with a bit of extra length on the head end, then mill a few straight flutes into it to make it a cutter. The stem of the cutter ensures it is concentric to your valve cage and being that it's exactly the same angle as your valve then the seat it cuts will match up perfectly. I make a new cutter every time I make valves so it really doesn't matter if my valves aren't precisely 45 degrees, the seats will still match. For cutting seats in aluminum, brass or bronze I don't bother with hardening the cutter.

 

[Gordon]

If you use the George Britnell valve seat cutter method, you don't need to alter the compound angle from the first setting and you're guaranteed to have perfectly matching angles. Basically you just make an extra valve with a bit of extra length on the head end, then mill a few straight flutes into it to make it a cutter. The stem of the cutter ensures it is concentric to your valve cage and being that it's exactly the same angle as your valve then the seat it cuts will match up perfectly. I make a new cutter every time I make valves so it really doesn't matter if my valves aren't precisely 45 degrees, the seats will still match. For cutting seats in aluminum, brass or bronze I don't bother with hardening the cutter.

The problem is that it is a wide valve seat. I am trying to make a good seal on a Lunkinheimer mixer and the seat is about 1/4" wide. I have made a Britnell type valve seat cutter but I am having a problem getting the valve itself to fit for that wide seat. I have tried lapping but it does not have to be off much to make a poor seal which is beyond what can be straightened with grinding compound.

 

[Cogsy]

I'm not familiar with the Lunkinheimer design so I'm not really sure what the problem is to be honest. I would assume a G.B. cutter would be able to cut a seat as wide as the valve face itself, if that's what is required, but obviously there's some confounding issue I don't know about. As you're no doubt aware, a smaller seat is much easier to achieve a seal with so your wide seat is going to need to be near perfect to get a good seal.

So, plan B I guess (and this is a bit of a guess but it's how I'd do it). Chuck up something with a decent diameter and face it absolutely as flat as you can get it. Then set up a sine bar (or a known accurate triangle) at 45 degrees across the face and adjust the compound to the reference angle. Not sure why you'd need an indicator as well here? Cut your valve then rechuck the faced piece and take another skim, then repeat for the opposite angle. I still don't like your chances of having them 100% perfect but they should be very close if done with care.

 

[clockworkcheval]

I agree with step 1 of plan B of Cogsy - face a solid disc with the precision your cross slide affords. As step 2 I use myself a simple triangulation.First I set the compound at 45 degrees using a sine bar. Second I move the topslide over at say 100 mm. An indicator on the topslide at a right angle against the reference face should then go from zero to 70,71 mm. Adjust till it does.

 

[Charles Lamont]

If you chuck the finished valve by the stem in a collet with the head outwards, you can use a dial gauge to set the topslide parallel to the seating cone. You can then cut the seat at this setting. It won't matter whether the valve cone is at 45° or some other angle; it will transfer. The advantage is that no fancy angle measurement is required. The disadvantage is that the measurement is being taken across a narrow surface, so you will need a good surface finish and ideally a clock reading to 0.0001".

For setting with your triangle, the chuck face should be OK. If you have a faceplate that should be better. If your topslide body does not have (accurately) machined sides then you can remove it and use the machined dovetail on the topslide base as a reference surface. That, after all, is what you actually want to set.

 

[fcheslop]

Machine both parts with the same topslide setting
At least thats the way I do it and dont have to worry about resetting
Cut the valve then reverse the lathe and use a boring tool to cut the seat from the backside the opposite to the valve
cheers

 

[petertha]

This reminds me of something I was going to make but never got around to. Sketch shows the general idea.
- chuck holds a piece of round stock that has a slot milled in the end
- the slot acts as a clamp to grip a precision angle gage block
- get the gage block horizontal & engage the drive so chuck stays put
- traverse across base of angle block with DTI mounted on carriage. Adjust angle block rotation setting so DTI runs 0-0 true (green dots). Fix this position with clamp screw (black dot)
- now with compound set to approximate angle, run DTI down the angle face of gage block (orange dots) until DTI runs 0-0 true
- if you want the mirrored angle, just rotate the gage block 180-deg

This probably would not work great for odd angles where normally you would stack up blocks or shallow angles where the base is very short, but maybe for basic 0-30-60 or 90-45-45 could be a viable method?

 

[Gordon]

This reminds me of something I was going to make but never got around to. Sketch shows the general idea.
- chuck holds a piece of round stock that has a slot milled in the end
- the slot acts as a clamp to grip a precision angle gage block
- get the gage block horizontal & engage the drive so chuck stays put
- traverse across base of angle block with DTI mounted on carriage. Adjust angle block rotation setting so DTI runs 0-0 true (green dots). Fix this position with clamp screw (black dot)
- now with compound set to approximate angle, run DTI down the angle face of gage block (orange dots) until DTI runs 0-0 true
- if you want the mirrored angle, just rotate the gage block 180-deg

This probably would not work great for odd angles where normally you would stack up blocks or shallow angles where the base is very short, but maybe for basic 0-30-60 or 90-45-45 could be a viable method?

That looks like a good solution for my needing three hands to hold everything. I will probably make something like that. Thanks

 

[Gordon]

I am presently making some new parts. I have made a Britnell type seat reamer and I am in the process of making a new valve which I can machine while the working face is facing out and can be machined to the same angle as the seat reamer regardless of whether it is exactly 45°.

I do not have a sine bar and have never used one. I understand the principle but I am not sure that I would have enough use for one to make it worth the expense. Trying to hold a sine bar setup on the face of the chuck (or faceplate) could get pretty extra hands intensive.

 

[justisla]

Try

He also has vids on angles for internal turning

 

[tornitore45]

How you delete a post? I posted erroneous info and do not want to be told I am an ***.

 

[XD351]

How you delete a post? I posted erroneous info and do not want to be told I am an ***.

Click on your post and it will change colour and a info bar will open up with an edit option , just delete the words within the post and put deleted post in their place .
I’m not sure you can remove the post yourself - probably need a moderator to do it

 

[Cogsy]

How you delete a post? I posted erroneous info and do not want to be told I am an ***.

No-one can be right all the time - don't be afraid to be wrong, just be open-minded enough to realise when you are and then you learn something which makes you smarter. It's a win-win.

 

[minh-thanh]

No-one can be right all the time - don't be afraid to be wrong, just be open-minded enough to realise when you are and then you learn something which makes you smarter. It's a win-win.

" No-one can be right all the time "

 

[bobden72]

" No-one can be right all the time "

My wife is

 

[tornitore45]

My wife is

They must be sisters.
Bob, Is that an Edward 5 radial engine in your avatar? I am 1/2 way into one myself. Making the rings now.

 

[bobden72]

Yes one Edwards Five, it was the second I/C engine I built about eight years ago. Built strictly to the drawings turned out fine and runs well, think it must be due a strip down and clean soon.