Machinist Question - Measuring Valve to Valve Guide Clearance

[bulletpruf]

Fellas - I'm building a 7.3 IDI diesel for my crew cab dually project. I'm working on the valves and valve guides and I'm getting some strange results.

I measured the valve guide ID with a split ball/small hole gauge and a mic (down to the tenth thousandth) and felt pretty good about my results - valve to valve guide clearance was right around .002" which is where I expected it to be on a low-mile engine.

However, next I measured it by installing the valve in the guide, pushing it out .400" (to simulate valve being open) and then using dial indicator on the valve to measure amount of wiggle. When I was measuring side to side (perpendicular to rocker arm) I was getting about .004" of movement, indicating .002" of clearance. However, when I moved the dial indicator parallel with the rocker arm, I got .010" of movement, for .005" clearance, which is a big difference. I realize valve guides were oblong, but I didn't expect this much of a variation.

I then switched to a snap gauge, which allows me to measure a bit more accurately than the split ball gauge because I can measure the larger axis of the valve guide as well as the narrow axis. I got measurements on the larger axis (parallel to the rocker arm) that were larger than what I got with the split ball gauge, but only by a few tenth of a hundredth thousandth (example - .3747" instead of .3743"). However, I didn't see any readings that indicated .005" of clearance like I saw with the dial indicator method.

Any ideas on what I should be using as the most accurate method?

Thanks

Scott

 

[saon]

most accurate is a dial bore gauge made for checking guides. you clamp two valves in the set up fixture, then adjust yr gauge to match the valve stem thickness and measure yr clearance.
sunnen brand is what I have used.
other than that, I got nothing else for ya, sorry.

 

[MBfreak]

The valve guide and valve stem wear may be elliptic or close to that.
Replace both.

Ola

 

[bugnut]

You have evidently found ovality or some other wear pattern. The gages you are using lack the discernment to determine by how much. @saon is spot on in a better device with higher capability would be a indicating dial bore gage as it will provide a direct measurement, and can provide some measurement to show ovality, and not require the feel required for a small hole gage (reference measurement tool) and micrometer.

Dial Bore Gage, 0.0001" Graduation, 0.24 - 0.40" Range

www.cylinderheadsupply.com

 

[bulletpruf]

most accurate is a dial bore gauge made for checking guides. you clamp two valves in the set up fixture, then adjust yr gauge to match the valve stem thickness and measure yr clearance.
sunnen brand is what I have used.
other than that, I got nothing else for ya, sorry.

Yeah, I'd love to have one of those, but at $2,600 it's not in the tool budget now - https://goodson.com/products/p-310-sunnen-valve-guide-gauge-set

Have tried to find a used one, but no luck yet.

Thanks

 

[bulletpruf]

The valve guide and valve stem wear may be elliptic or close to that.
Replace both.

Ola

Yes, the valve guide is elliptic or oblong. The valves are not.

Thanks

 

[bulletpruf]

You have evidently found ovality or some other wear pattern. The gages you are using lack the discernment to determine by how much. @saon is spot on in a better device with higher capability would be a indicating dial bore gage as it will provide a direct measurement, and can provide some measurement to show ovality, and not require the feel required for a small hole gage (reference measurement tool) and micrometer.

Dial Bore Gage, 0.0001" Graduation, 0.24 - 0.40" Range

www.cylinderheadsupply.com

Ok, looks like I need to add that to the ever-growing list of tools to purchase. Thanks

 

[RoninB4]

Not to throw doubt on your set-up (since I wasn't there) but using a dial indicator doesn't always mean it shows an actual number value. If the indicator you were using is a plunger type it absolutely must be set perpendicular in 2 axis to the direction of movement or the plunger will travel at an angle relative to the movement, numerical error results. If the indicator is a lever type (DTI) then the potential for error is even worse because the movement is linear and the lever swings in an arc. There is an angle stated to set a DTI to for better results but we won't go into that here. In precision machining an indicator is most often used for a comparative value rather than a quantitative value. The "zero" is set to a reference and the movement is compared against that. There's still an error factor doing this but it is usually such a small increment as to be negligible.

Checking side to side movement is an angle, the longer distance the valve is out from the seat the greater the movement from side to side still showing the angle it can travel. None of this may mean anything, I wasn't there to see what you were doing. IF the guide ID is truly round (not oblong) then figure your clearance by the ID of the guide and the OD of the valve stem. That's what we do in machine shops. If you have a manual that states to check by your method with an indicator then ignore what I've posted.

I should have asked this first. Are the guides new or what's been run for a while? If not new then, as others have pointed out, you may have a wear pattern in the guide.

 

[bulletpruf]

Not to throw doubt on your set-up (since I wasn't there) but using a dial indicator doesn't always mean it shows an actual number value. If the indicator you were using is a plunger type it absolutely must be set perpendicular in 2 axis to the direction of movement or the plunger will travel at an angle relative to the movement, numerical error results. If the indicator is a lever type (DTI) then the potential for error is even worse because the movement is linear and the lever swings in an arc. There is an angle stated to set a DTI to for better results but we won't go into that here. In precision machining an indicator is most often used for a comparative value rather than a quantitative value. The "zero" is set to a reference and the movement is compared against that. There's still an error factor doing this but it is usually such a small increment as to be negligible.

Checking side to side movement is an angle, the longer distance the valve is out from the seat the greater the movement from side to side still showing the angle it can travel. None of this may mean anything, I wasn't there to see what you were doing. IF the guide ID is truly round (not oblong) then figure your clearance by the ID of the guide and the OD of the valve stem. That's what we do in machine shops. If you have a manual that states to check by your method with an indicator then ignore what I've posted.

I was using a plunger type dial indicator and it was set at 90 degrees to the valve. The valve was opened to .400" which is the max lift on the camshaft.

The guides are definitely oblong which is why I was getting a significant variance in the up/down (parallel to rocker axis) measurement versus the side to side (perpendicular to rocker axis) measurement.

Thanks for the input.

 

[RoninB4]

I was using a plunger type dial indicator and it was set at 90 degrees to the valve.

-Plunger type is ok, nothing wrong with those. Being 90 degrees to the valve axis is good but you also have to be at 90 degrees to the movement, two directions required here or the plunger will travel at an angle to one. the other, or both. Small bores like that are also difficult to get a good set of readings from. Three readings (120 degrees apart) minimum should be taken for any given length down the bore. Multiple distances down the bore should also be taken to "map" how much wear/variation there is from the bore being a true cylinder. This is what's done in precision machining.

If you already know this then all my replies concerning your bore.....now I've become the bore. Apologies if I'm preaching to the choir.

 

[bulletpruf]

-Plunger type is ok, nothing wrong with those. Being 90 degrees to the valve axis is good but you also have to be at 90 degrees to the movement, two directions required here or the plunger will travel at an angle to one. the other, or both. Small bores like that are also difficult to get a good set of readings from. Three readings (120 degrees apart) minimum should be taken for any given length down the bore. Multiple distances down the bore should also be taken to "map" how much wear/variation there is from the bore being a true cylinder. This is what's done in precision machining.

If you already know this then all my replies concerning your bore.....now I've become the bore. Apologies if I'm preaching to the choir.

No, I definitely don't know all this. Not my first time with precision tools, but it is my first time checking valve guides. Appears that it's a bit of an art as well as a science. Thanks for the input.

 

[ez-duzit]

Ok, looks like I need to add that to the ever-growing list of tools to purchase. Thanks

Not really. You just need to have this checked by someone who already has the right equipment.

 

[larry_g]

If you already know this then all my replies concerning your bore.....now I've become the bore. Apologies if I'm preaching to the choir.

I am of the opinion that even if the OP knows what is right many others reading your post will learn something new. An informative and educational post is never wrong.

lg
no neat sig line

 

[bulletpruf]

Not really. You just need to have this checked by someone who already has the right equipment.

True. But I'd like to be able to do this myself if possible.

 

[saon]

I understand wanting to do it your self.
I try to do everything myself.
I dont own 1 either, but the machine shop I spent 25 years working at had one.

some things pay for themselves by having it done with the correct equipment.
if they are bad, do you plan to replace and size the guides yourself? or will you have to have a engine machine shop replace them then broach and hone them?
I dont know your milage but anything over 100k most likely will need atleast honed back straight. and even new heads run on the tight side if your looking for performance use.

 

[bulletpruf]

I understand wanting to do it your self.
I try to do everything myself.
I dont own 1 either, but the machine shop I spent 25 years working at had one.

some things pay for themselves by having it done with the correct equipment.
if they are bad, do you plan to replace and size the guides yourself? or will you have to have a engine machine shop replace them then broach and hone them?
I dont know your milage but anything over 100k most likely will need atleast honed back straight. and even new heads run on the tight side if your looking for performance use.

These engines actually don't use replacement guides, they just get the guides reamed/honed oversize and then use valves with oversize stems.

The guides on one set of heads are too tight - .00005" - that's not a typo! Most of the exhaust valves had some galling and several were difficult to remove. I'll defer to the machine shop to hone to correct size.

 

[RoninB4]

I am of the opinion that even if the OP knows what is right many others reading your post will learn something new. An informative and educational post is never wrong.

lg
no neat sig line

-This is what I always hope for in sharing my limited experiences with others, even at the risk of sounding like Mister- Know-It-All or boring the audience. Thanks for posting this.

 

[RoninB4]

The guides on one set of heads are too tight - .00005" - that's not a typo!

-Not to doubt your readings or instruments but I don't think I can get a reliable/trusted number down to half a tenth and I've got several indicators reading in tenths. For accuracy there's a standard rule in metrology that says your instrument should have a resolution capability 10x greater than the smallest dimension you're shooting for. If you want to work (reliably) to .001 then you should have an instrument that reads to .0001. The reading you have (.00005) should only be trusted if you have an instrument resolution of .127 microns (also read as .005 millionths). Getting reliable/trusted dimensions closer than +/-.0005 is sketchy and the temperatures of everything have to be factored in. Even holding something relatively stable like a micrometer in your hands for a minute or two can easily change the reading a tenth or two due to thermal expansion. Temperatures above/below 20 Centigrade must be factored for accuracy. This is what I learned while working at Midwest Gage Laboratory (Chicago).

I'm going to presume you had an extra "0" in there.

 

[bulletpruf]

-Not to doubt your readings or instruments but I don't think I can get a reliable/trusted number down to half a tenth and I've got several indicators reading in tenths. For accuracy there's a standard rule in metrology that says your instrument should have a resolution capability 10x greater than the smallest dimension you're shooting for. If you want to work (reliably) to .001 then you should have an instrument that reads to .0001. The reading you have (.00005) should only be trusted if you have an instrument resolution of .127 microns (also read as .005 millionths). Getting reliable/trusted dimensions closer than +/-.0005 is sketchy and the temperatures of everything have to be factored in. Even holding something relatively stable like a micrometer in your hands for a minute or two can easily change the reading a tenth or two due to thermal expansion. Temperatures above/below 20 Centigrade must be factored for accuracy. This is what I learned while working at Midwest Gage Laboratory (Chicago).

I'm going to presume you had an extra "0" in there.

Yes, I did have an extra 0 in there! The clearance I measured was .0005". I realize it's tough to get an accurate measurement at that level, but I'm confident that it was less than .001".

Thanks for clarifying that!

 

[joe49]

You do know that if you replace or knurl the guides they will be sized as needed. I'm more worried about the valve stem wear. Replacing them as needed. Knurling at home is easier and cheaper then installing new guides.

 

[bulletpruf]

You do know that if you replace or knurl the guides they will be sized as needed. I'm more worried about the valve stem wear. Replacing them as needed. Knurling at home is easier and cheaper then installing new guides.

The problem with my guides is that they're too tight, not too loose, so knurling would be counterproductive.

Thanks

 

[joe49]

You said they move more in line with the rocker arms. Means oblong. Means knurling and sizing with the proper size ream will take you to wear you need to be.