Author Topic: Valve grinding and lapping  (Read 3169 times)

Offline petertha

  • Full Member
  • ****
  • Posts: 237
Re: Valve grinding and lapping
« Reply #15 on: April 17, 2018, 08:02:12 PM »
For model engines some builders prefer the valve cage/seat configuration. The theory is that the whole insert can be cut at one time, guide, pocket and seat so that everything is concentric. There are two catches to this method, one being that a center drill is used to start the hole followed by an undersized drill then a reamer. First of all a reamer only follows the drilled hole so if the drill doesn't drill a perfectly straight, concentric hole then the reamer just follows what is already there. Now the seat is cut with lathe tooling. In most cases the seat is perfectly concentric with the axis of the lathe but the guide and seat might not be concentric to each other. The second problem is that when the valve cage/seat is pressed into the head it can distort and further amplify any misalignment with the seat and guide. I know builders who have had good luck using this method I'm just saying that I prefer the following.

Really good points George. The exact same thoughts crossed my mind. I guess I'll know how this radial engine pans out because integral (bronze) valve guide/port/seat were called for on the plans. The only comfort to me is some YouTubes of them actually running. LOL!

CATCH-1: When I made my prototype valve cup insert for the vacuum testing, I first drilled/bored the larger diameter port ID section of maybe 0.5" depth. As opposed to drilling/reaming through the entire ~1" length of stock corresponding to the much smaller valve stem diameter & then opening up the port diameter as a second operation. The (shorter length) valve stem drilling was still preceded by center drilling inside the port hole with 120-deg spot center drill. But my thinking was minimize the amount of material for the drill to wander.

Catch-1A: Although the valve stem hole is nicely reamed, the valve stems themselves are lapped after turning. They slide pretty nice but I'm guessing they must be 0.0005" undersize (maybe more) just based on my own lapping variation. On one hand this isn't so great, but OTOH might help a bit with valves self-centering & seating where it counts on the valve face itself. Or is that wishful thinking? That's the challenging thing about these little engines isn't it - the same sliding fit annular tolerance on a 1:1 engine valve stem doesn't scale to a 1:4 engine.

CATCH-2: This one worries me a bit. Rather than pressing a delicate valve cup into the head & risking valve seat distortion because the port area wall thickness is quite thin, I am entirely relying on the specs of high temp Loctite. Depending on the grade, an annular clearance gap of .001" is perfectly acceptable. So they slide in on a liquid film & (fingers crossed) stay put once fully cured. Now whether this withstands the head temperature in running conditions is another matter. Again, I am basing this on the designers specs. I made a few slug tests using the same materials though. In one I bored the ID to this .001" clearance slip fit. In another I made with zero clearance but warmed the aluminum so it shrank down on the bronze, again with HT Loctite. Once cured I could not beat them out with moderate hammering. I put them both in my mini oven at 400F & repeated, same result. Actually the shrink fit let go but very unscientific 'beating'. So I'm hoping this will hold true for the actual engine.

My only other data point is Loctite 'gluing' parts to a mandrel as a means of turning on a lathe. It usually takes a butane torch flame plus a whack to release them. I'll keep you all posted.

« Last Edit: April 17, 2018, 08:05:39 PM by petertha »

Offline gbritnell

  • Full Member
  • ****
  • Posts: 1781
Re: Valve grinding and lapping
« Reply #16 on: April 17, 2018, 08:52:19 PM »
All of the processes we use in building engines will work. It just depends on how meticulous we are when machining. The issue with  these miniatures is that .001 in a full size engine with a valve diameter of say 1.50 and .001 with a valve head of .25-.375 makes a world of difference.

I have never personally used Loctite for an end-all assembly although I don't see why it wouldn't work for the valve inserts given the amount of heat needed for disassembly. On a miniature engine I don't think they would get hot enough to release the bond.

One other thing to consider is that when using Loctite with aluminum and accelerator is needed. When using it with another metal, bronze, iron or steel it should work.
gbritnell
Talent unshared is talent wasted.

Offline petertha

  • Full Member
  • ****
  • Posts: 237
Re: Valve grinding and lapping
« Reply #17 on: April 18, 2018, 07:47:45 PM »
Here is a recent integral valve cage example, Terry's Knucklehead (post #2). I recall he employed similar method on his radials. Also some useful trials & tribulations regarding specific bronze alloys. http://www.homemodelenginemachinist.com/showthread.php?t=27584&highlight=loctite

I reluctantly ordered the C544 from MSC, and what arrived appeared to be identical to the smaller diameter material I was used to working with. The cages machined beautifully with no internal chatter, and the test seats cut buttery smooth. The cages were sized for a light press fit in the aluminum heads, and high temperature Loctite retaining compound was used to insure they stay in place. A 9/16" standard end mill was used to plunge cut the major bore in the head for the cage, and its actual measured i.d. was used to set the target o.d. for the turned cages.

Offline steamer

  • Global Moderator
  • Full Member
  • *****
  • Posts: 9276
  • Central Massachusetts, USA
Re: Valve grinding and lapping
« Reply #18 on: April 30, 2018, 05:38:50 PM »
Actually this has been one of the most informative and useful posts I've seen in a while!    Well done everyone!

Dave
"Mister M'Andrew, don't you think steam spoils romance at sea?"
Damned ijjit!

Offline Vixen

  • Full Member
  • ****
  • Posts: 902
  • Hampshire UK
Re: Valve grinding and lapping
« Reply #19 on: May 02, 2018, 06:58:01 PM »
Here are some of my findings which relate directly to steel valve seats. Some of this information may be relevant to combined valve seats and valve guides made from bronze. However, there is a greater risk of abrasive becoming embedded in the softer bronze valve seat, so beware. Parts of this post are copied directly from my earlier By Jupiter post.

Any valve seat cutter, no matter how well made, will always leave minute radial ridges and grooves on the surface, you may need to use a X5 eyeglass to see them, but they are there. It takes a lot of extra time to lap away these marks. I followed Terry Mayhughs method of using two cutters one after the other to produce a finer surface to the valve seat. The idea being the second cutter should flatten any ridges left by the first. Only the lightest of pressure is used with the second cutter.




The easiest way I found was to grip the valve stems with a small drill chuck. There is a short piece of soft aluminium tube between the chuck jaws and the valve stem to prevent the valve being marked.





In order to test the quality of the valve seat lapping it purchased this little hand pump vacuum tester. It can get to about  400 mm Hg (-15 inch Hg) that's about  -8 psi without too much problem from internal leaks. I cap off the top of the valve guide with a short length of silicon tube, and insert a rubber cork into the inlet/ exhaust port. I can create the vacuum by squeezing the trigger a couple of times. The internal volume of the volume under test is approx 3.9 cc. I can make the volume bigger (20cc) by connecting a longer pipe, but the short pipe is the most sensitive. I have set an arbitrary leak rate pass criteria of 20 to 30 seconds for the -8 psi to leak to zero. Consider the engine running at 2000 rpm, the combustion chamber pressure should be at over 200 psi during the compression stroke and much higher still during the firing cycle. At 2000 RPM that's a duration of somewhere between 20 to 30 milliseconds. So my leak test criteria (8 psi over 20 to 30 seconds) has an enormous margin compared to the running environment.




I experimented with several types and grades of lapping compound. I tried various grades of metal polish (as suggested by Terry Mayhugh) as well as different grades of fine silicon carbide powder made up as a lapping slurry  It quickly became obvious that it would become a trade off between time and quality of the gas seal.

Solvol Autosol uses pumice as the abrasive, Given time (hours) it will lap the valve and seats to perfect finish which can achieve my 20 to 30 second test criteria. This shows that polished valve seats will provide the best sealing possible, but it takes forever. However some valve seats, for whatever reason, just would not ever achieve a perfect seal, no matter how long I lapped with the metal polish, On these problem valves I tried ultra fine 600 grit silicon carbide. The seats lapped in quickly, but this lapping compound leaves a dull surface rather than a polished surface, sufficiently that a vacuum leak rate much over 10 seconds could not be achieved, seems you need polished surfaces for a perfect gas seal. That was rather unexpected, as I have always used grinding paste on full size engines in the past to get an even all round matt finish, which always appeared to be adequate, I never had a vacuum tester. Our model engines have much small capacity, so will be more susceptible to any leaks. Especially an engine with four valves per cylinder.

I have just obtained some 1200 grip abrasive powder and will experiment further (there are plenty of valves in the Jupiter)  to find a workable compromise between the time require for a highly polished valves seal using metal polish against the quicker, but less perfect seal that come with abrasive lapping media.




Some suggest that fine diamond lapping compounds are the answer, but I worry about the diamond becoming embedded in the valves or seat faces and the long term effects on the seats and the rest of the engine.

Mike
« Last Edit: May 02, 2018, 09:44:42 PM by Vixen »
It is the journey that matters, not the destination

Offline petertha

  • Full Member
  • ****
  • Posts: 237
Re: Valve grinding and lapping
« Reply #20 on: May 04, 2018, 07:38:46 PM »
Beautiful!
(Sorry I replied to your Jupiter post, let me know if you want to confine my other question to this post.)

What do you figure the width of the actual valve seat is?

Another tidbit I can pass on from my limited testing: the presence of any liquid, even petroleum based cleaning spirits, can give you false optimistic pressure drawdown readings over dry. Even WD40 with even slight viscosity can influence the numbers. Oil film itself will make you look like a hero. So I washed the abrasive grit with whatever works best & then a final dunk in alcohol or acetone, blow dry & then test. One may argue the running engine is oily so base the seal results on that. I'm just saying for apples to apples comparison, might help to keep everything as consistent as possible.

I also found surfacing blue pretty much useless. The good seals looked about the same 'shiny ring' as so-so seals. There just isn't very much surface area to look at



Offline Vixen

  • Full Member
  • ****
  • Posts: 902
  • Hampshire UK
Re: Valve grinding and lapping
« Reply #21 on: May 04, 2018, 08:52:42 PM »
Hi Petertha,

It may be more appropriate to keep all the valve lapping discussion to this topic rather than the Jupiter build.

I looked at the Brownells' gun muzzle cutters, but the cost, shipping from the US and then the UK import taxes etc. can double the price, which makes it unattractive to me. So I machined my two sets of cutters from drill rod and hardened them with a water quench. I did not bother to temper them. The cutting edges were lightly stoned, by hand. The raw machined valve face was originally 1.5 mm wide, I reduced them to approx 0.75 mm by opening the throat with the 60 degree cutter. Measuring the face width is tricky, I found visual comparison with the end of a steel wire of known diameter (0.75mm) was the most reliable. As you found, only the lightest pressure is required, The objective is to dress the previously machined surface rather than cut a new surface. The second cutter serves to refine the surface finish.

If I ever make another set of seat cutters, I will try to slightly stagger the cutting faces, each by a few degrees. This may reduce any tendance towards harmonic chatter and a wavy seat face.

Agree with you about not using oil, it does give an over optimistic reading. I do all my testing dry to be consistent. If the engine later wets the seals with oil, so much the better. Also, marker pen, engineers blue, or visual inspections of the seat faces don't tell you much. The vacuum test is the only practical way to measure the seal quality.

These are my findings which relate directly to steel valve seats. Some of this information may be relevant to combined valve seats and valve guides made from bronze. However, there is a greater risk of abrasive becoming embedded in the softer bronze valve seat, so beware.

However, we should remember that most car (auto) DIY enthusiasts and most model engineers appear to get acceptable results when they simply lap their valves with an abrasive, cross their fingers and hope. Valve seat polishing and vacuum testing may not be essential, but it is the only way I know of, to measure the quality of the seal and so eliminate one possible cause of of an engine's failure to start or run consistently.

Mike
It is the journey that matters, not the destination

Offline petertha

  • Full Member
  • ****
  • Posts: 237
Re: Valve grinding and lapping
« Reply #22 on: May 05, 2018, 02:52:46 AM »
Great info Mike. Maybe this is a good point for a dumb question I've been harboring. What is it about bronze that makes a good valve seat in our application? Attaching my informal cheat sheet of specs pulled from various sources. My limited understanding is it makes for good bearing material against rotating shafts & such. The valve stems see a bit of that but mostly its face to face slapping relationship that defines the seal. The hardness varies by alloy, but in all cases the copper content is quite high. Unfortunately the same element that makes for good laps because abrasive grit likes to embed in there & not let go easily. Full confession, the testing mucking around I was doing thus far was on 12L14 cages, prototyping the heads. Now with some bronze stock in hand I'm concerned I'll have this grit embedding problem when the time comes. I opened up an RC engine OS 4-stroke & sure enough, copper-y looking valve seat in there. So I would have to think there is a good reason even if its not obvious to me.

Offline zeeprogrammer

  • Full Member
  • *****
  • Posts: 6232
  • West Chester, PA, USA
Re: Valve grinding and lapping
« Reply #23 on: May 05, 2018, 02:54:41 AM »
Just letting you know I've been watching...but woohee...way beyond me.
I'm sure some of it will stick.
Carl (aka Zee) Will sometimes respond to 'hey' but never 'hey you'.
"To work. To work."
Zee-Another Thread Trasher.

Offline Vixen

  • Full Member
  • ****
  • Posts: 902
  • Hampshire UK
Re: Valve grinding and lapping
« Reply #24 on: May 05, 2018, 09:28:10 AM »
Maybe this is a good point for a dumb question I've been harboring. What is it about bronze that makes a good valve seat in our application?

Hi Petertha, That's a very good question, Why do we use bronze?

I guess for the majority of model engine makers use bronze because that is what our drawings tell us to use, rarely have I seen a specific grade of bronze specified.

In our miniature engines, a combined valve cage incorporating the valve seat and valve guide as a single item, is a very practical approach. It is so much easier to achieve the all important concentricity of the valve seat to the valve guide bore, in a single bronze valve cage compared with juggling three separate parts (the head, the seat and the valve guide). The combined valve cage is also easier to install and retain in the cylinder head, no shrink fit valve seats to come loose and wreck the engine. Also the high copper content of the bronze efficiently conducts heat away from the valve components and into the cylinder head.

What grade of bronze should we use? A much more difficult question. The valve guide requires excellent bearing properties, it needs to work at high temperature without additional lubrication. It needs to endure these poor conditions for it's life time. This suggests the use of a bronze with excellent bearing properties. Machinability is another criteria, the valve guide needs to be reamed and finished to fine tolerances. I, for one, have had some scary experiences with some grades of phosphor bronze pinching and seizing on a drill or reamer. so I select a bearing bronze with known machinability.

The valves on my Mercedes Benz W165 engines are only 7.5mm in diameter. There are four valves per cylinder and eight cylinders per engine. So I needed a minimum of 64 valves for the two engines. I designed a combined valve cage in bronze for all the reasons identified above. I selected C932 (SAE 660) bearing bronze because of it's bearing and machinability properties. However, I feel that a "harder" grade would be better for the valve seating. Too late now, I have made them and have no intention of repeating the exercise.


This is what the screw in Mercedes bronze valve cages look like


This is what 6 foot (2 metres) of bronze bar becomes


This is the four valve cages filling the combustion chamber. There is just enough room in the center for the spark plug.

You will notice that I have not yet cut the 45 degree valve seats. I plan to cut the narrowest possible seat with a piloted seat cutter. That will wait until I have made the 64 valves to fill those holes.

The bronze is softer than the steel valve, therefore the lapping compound will want to embed itself into the softer material. It cannot be removed or washed of with a solvent. Therefore we need a lapping paste which breaks down quickly, we want a lapping compound which effectively self-destructs. I plan to use 'Autosol' metal polish, which contains a 'soft' abrasive, it's probably pumice or something similar. The abrasive quickly breaks down into smaller and smaller particles and stops working. It should be possible to polish the bronze valve seat for a good gas seal without long lasting side effects. This soft abrasive break down is one reason why metal polish takes so long to polish hard steel valve seats, bronze seats should be quicker.

Hi Zee, How's the retirement going? You could now  find the time to change your signature from "to work, to work" into "to play, to play". Just thinkin.

Mike

 
It is the journey that matters, not the destination

Offline petertha

  • Full Member
  • ****
  • Posts: 237
Re: Valve grinding and lapping
« Reply #25 on: May 25, 2018, 08:13:21 PM »
Here's another thought I had, maybe worthy of a chat. For conventional countersinking I've used the typical fluted cutters over the years with so-so success. Depending on ?? cutter quality, tooth sharpness, alloy being cut, rpm, number of teeth to hole circumference, phase of moon... cutting can go smoothly. Other times (usually correlated to how important the part is) it can start developing a kind of undesirable pattern harmonic. The result is a weird, symmetrical but bumpy path syndrome that looks like ring of fine teeth under magnification.

Maybe its a form of chatter kind of analogous to a large contact line cutter in a lathe. But my experience is the toothed countersinks behave worse in thin sheet metal & shallow depth, so that doesn't quite fly either. Well, for basic flathead screw countersinking, its not the end of the world. Sometimes you can get the cutting under control again with a quick downfeed stroke to cut across the hilltops & get things back on track. Or proceed with very slow & light pressure down feed.

What I find works much better is these 'single lip' type cutters. They just seem so much more controlled & I have not seen this harmonic issue develop. I had to buy some 90-deg versions for metric fasteners & I was impressed with how they cut. Nothing even special, just happened to be AliExpress sourced. In my pictured gear ring, I had to take down the chamfer for M3 bolt in small increment so the screw was just under flush (because very little material in the underside flange to work with). I was impressed with how easy it was to take a couple thou off in a controlled manner, test fit, try again. The fluted cutters never gave me this kind of control. 

But the reason for mentioning this is maybe its the same issue with valve seat cutting?. We are only taking a couple thou off, but maybe that's analogous to the sheet metal example. When I used the fluted muzzle chambering tool on my valve seat, I learned to only apply super light pressure. If you press a bit too hard to cut, it developed what I think is 'bumpy pattern syndrome' described as above. Those valve seats are hard or impossible to ever lap back into condition. I suspect because it takes a lot of abrasive work to undue the damage & to wear down the hilltops back into a uniform sealing surface.

So.... I wonder if these single lip cutters (or a carefully made home brew) might help us? At the time it never even occurred to me because the ones I had were for N-Am 82-deg flathead screws. But if 90-deg are available, might be worth some testing.

« Last Edit: May 25, 2018, 08:19:47 PM by petertha »

Offline Vixen

  • Full Member
  • ****
  • Posts: 902
  • Hampshire UK
Re: Valve grinding and lapping
« Reply #26 on: May 25, 2018, 10:51:16 PM »
A conventional, multi fluted, countersink cutters will produce perfect holes when used with a milling machine with controlled down feed. The chatter marks are more likely to occur with manual downfeed in a drill press and even more likely to occur if done completely by hand or hand drill.

The valve seats on the Jupiter were cut perfectly smooth and concentric with the screw threads for the valve guides, on a very solid Bridgeport. The problem was the use of screw-in valve guides, concentricity could not be maintained when the guides were later screwed home. A lateral displacement ( non concentricity) of only one thou, would require two thou to be removed from one side of the valve seat and none from the other. You can see the contact patch extending round the valve seat as the valve is progressively lapped in.

The Jupiter has 36 valves and I learned the hard lesson about screw in valve guides. Each of the 36 valves was lapped in and tested with a vacuum leak tester. I started by using Autosolv metal polish, This proved to be excessively time consuming. I next tried 600 grit silicon carbide, which was much quicker but did not give consistant leak test results. Finally, I obtained some 1200 grit green silicon carbide powder from a jem stone polisher. The ultra fine powder was made into a paste with mineral oil and a little white spirit. The ultra fine abrasive powder worked well, each of the valves was lapped reasonably quickly and all were able to pass a full 30 second leak down test.

At first, I tried to correct any misalignment using the hand powered 45 degree seat cutters. Eventually I found it was much quicker and less frustrating to simply let the 1200 grit lapping compound do the work. You can see the contact patch extending round the valve seat as the valve is progressively lapped in. Most valves lapped in reasonably quickly and all were able to pass a full 30 second leak down test.

The lesson from all this is to use press in valve guides or one piece valve guides/seats. The 45 degree valve seat should be cut on a milling machine with a piloted seat cutter after the valve guide has been pressed home. 1200 grit lapping compound works reasonably quickly and a 30 second lead down test can be reliable achieved time and time again.

Well that's what worked for me, others may have different experiences to share.

Mike
« Last Edit: May 25, 2018, 11:23:04 PM by Vixen »
It is the journey that matters, not the destination

Offline petertha

  • Full Member
  • ****
  • Posts: 237
Re: Valve grinding and lapping
« Reply #27 on: May 26, 2018, 05:35:01 AM »
A conventional, multi fluted, countersink cutters will produce perfect holes when used with a milling machine with controlled down feed. The chatter marks are more likely to occur with manual downfeed in a drill press....

OK, that's encouraging. I wont give up on them. I don't have power feed in my RF45 mill but I can manually fine feed the quill vs. lever drill press action for sure. Good info.

Re your screw threaded valve guides, was it done purely for retention (as opposed to Loctite 'glue' mode). Or did it facilitate tweaking the depth setting maybe?

Offline Vixen

  • Full Member
  • ****
  • Posts: 902
  • Hampshire UK
Re: Valve grinding and lapping
« Reply #28 on: May 26, 2018, 08:49:36 AM »
The screw threaded valve guides were used in preference to an interference press fit because they were thought to be easier to make accurately in the large quantity required. The effect of the screw thread, on concentricity, had not been anticipated.

We all have perfect 20:20 hindsight  :facepalm:

Mike
« Last Edit: May 26, 2018, 11:31:27 PM by Vixen »
It is the journey that matters, not the destination

Offline steamer

  • Global Moderator
  • Full Member
  • *****
  • Posts: 9276
  • Central Massachusetts, USA
Re: Valve grinding and lapping
« Reply #29 on: May 26, 2018, 12:27:43 PM »
Maybe this is a good point for a dumb question I've been harboring. What is it about bronze that makes a good valve seat in our application?

Hi Petertha, That's a very good question, Why do we use bronze?

I guess for the majority of model engine makers use bronze because that is what our drawings tell us to use, rarely have I seen a specific grade of bronze specified.

In our miniature engines, a combined valve cage incorporating the valve seat and valve guide as a single item, is a very practical approach. It is so much easier to achieve the all important concentricity of the valve seat to the valve guide bore, in a single bronze valve cage compared with juggling three separate parts (the head, the seat and the valve guide). The combined valve cage is also easier to install and retain in the cylinder head, no shrink fit valve seats to come loose and wreck the engine. Also the high copper content of the bronze efficiently conducts heat away from the valve components and into the cylinder head.

What grade of bronze should we use? A much more difficult question. The valve guide requires excellent bearing properties, it needs to work at high temperature without additional lubrication. It needs to endure these poor conditions for it's life time. This suggests the use of a bronze with excellent bearing properties. Machinability is another criteria, the valve guide needs to be reamed and finished to fine tolerances. I, for one, have had some scary experiences with some grades of phosphor bronze pinching and seizing on a drill or reamer. so I select a bearing bronze with known machinability.

The valves on my Mercedes Benz W165 engines are only 7.5mm in diameter. There are four valves per cylinder and eight cylinders per engine. So I needed a minimum of 64 valves for the two engines. I designed a combined valve cage in bronze for all the reasons identified above. I selected C932 (SAE 660) bearing bronze because of it's bearing and machinability properties. However, I feel that a "harder" grade would be better for the valve seating. Too late now, I have made them and have no intention of repeating the exercise.


This is what the screw in Mercedes bronze valve cages look like


This is what 6 foot (2 metres) of bronze bar becomes


This is the four valve cages filling the combustion chamber. There is just enough room in the center for the spark plug.

You will notice that I have not yet cut the 45 degree valve seats. I plan to cut the narrowest possible seat with a piloted seat cutter. That will wait until I have made the 64 valves to fill those holes.

The bronze is softer than the steel valve, therefore the lapping compound will want to embed itself into the softer material. It cannot be removed or washed of with a solvent. Therefore we need a lapping paste which breaks down quickly, we want a lapping compound which effectively self-destructs. I plan to use 'Autosol' metal polish, which contains a 'soft' abrasive, it's probably pumice or something similar. The abrasive quickly breaks down into smaller and smaller particles and stops working. It should be possible to polish the bronze valve seat for a good gas seal without long lasting side effects. This soft abrasive break down is one reason why metal polish takes so long to polish hard steel valve seats, bronze seats should be quicker.

Hi Zee, How's the retirement going? You could now  find the time to change your signature from "to work, to work" into "to play, to play". Just thinkin.

Mike

Mike....did you put in some way to drive those home?    A spanner feature?

Dave
"Mister M'Andrew, don't you think steam spoils romance at sea?"
Damned ijjit!