Using a drill press to measure valve runout?

[Daedalus]

Anyone ever use a drill press to measure runout of a valve? I need to do a control measurement with a precision CNC cutter, but I bet the bearings are good to within .001. This setup is showing me this valve was probably hit, and has a runout of about .014. Using marks on the valve face, the indications are very repeatable. It does move a few thou when I turn it by hand (flex between the indicator base and the chuck), but when I release it, it settles back to the same spot every time at any given location/rotation. I think the biggest variable is in how true the chuck gets the valve every time. I think I need to measure the stem first to verify it's coaxial with the drill, and then measure the valve.

Was about to buy a couple of V-blocks, but this seemed convenient if it turns out to be accurate.

 

[Old Man Roger]

Maybe mark the highest spot, then rotate 180 in the chuck. If the high spot remains the same, then it is probably pretty good.

 

[BillK]

Most 3 jaw drill press chucks do not have the accuracy needed for this. Even the best ones will have at least .001" which is too much. Get a precision round rod and chuck it up and put your indicator on it.

My valve grinder uses a 6 ball chuck and will repeat within a couple of tenths every time (.0002")

Do you have a set of Vee blocks ? Put the valve in them and try it.

 

[Fuelie]

I agree with Bill, the drill press is not accurate enough for what your doing.
Tom

 

[Byrdnyrd]

Dae,
if you have a gage pin, you could Chuck that in the DP and convince yourself that the jaws in the press are ”True”
Large gage pin would be best to accentuate any runout.

i like your idea BTW! Clever!!

cheers,
BN

 

[larry_g]

Anyone ever use a drill press to measure runout of a valve? I need to do a control measurement with a precision CNC cutter, but I bet the bearings are good to within .001. This setup is showing me this valve was probably hit, and has a runout of about .014. Using marks on the valve face, the indications are very repeatable. It does move a few thou when I turn it by hand (flex between the indicator base and the chuck), but when I release it, it settles back to the same spot every time at any given location/rotation. I think the biggest variable is in how true the chuck gets the valve every time. I think I need to measure the stem first to verify it's coaxial with the drill, and then measure the valve.

Was about to buy a couple of V-blocks, but this seemed convenient if it turns out to be accurate.

If your measuring .014" runout then you are probably correct that the valve is bad. But as others have said this is not a precision measurement. As OMR said try gripping the valve in a couple of different rotations within the jaws and if the high spot follows the mark then mark the valve as bad. I'm guessing here that this valve is from a single cylinder engine? If not then how do all the other valves do in this setup? If multiple valves fail in this setup then suspect the setup. If only this one valve shows the runout and the others way better then it's a good guess the valve is bad. Suspect the valve is bad then put a bit of bluing on the valve face and put it back in its bore and spin it lightly. Does the bluing mark around the whole valve face and mark most of the seat? Does it visually wobble if your turn on the drill press? All tests to use for making an informed decision before you go to the next and more detailed accurate test.

So just what are you going to do for the next step? If you think it is good just throwing it back in the engine? Is it good enough to send out to the valve grind shop? Is it easier to just replace it with a new valve?

What is the intention of the engine after is is put back together? Is this something your playing with just to see if you can make it run, or something that is going to be used and has to be relied on for a long while?

So much plays into what is the correct answer to you question.

lg
no neat sig line

 

[gregs]

If you mark the high spot on the valve space and mark the valve in relation to the dp chuck, then loosen the dp chuck and turn the valve 90 degrees or so and check again. Does the high spot show exactly in the same spot?

 

[Daedalus]

You guys are right. I'm measuring runout of .010 to ,014 across a variety of dril blanks. It seems the bearings are maybe around .004, and the rest of it is in the chuck, and variable.

Most 3 jaw drill press chucks do not have the accuracy needed for this. Even the best ones will have at least .001" which is too much. Get a precision round rod and chuck it up and put your indicator on it.

My valve grinder uses a 6 ball chuck and will repeat within a couple of tenths every time (.0002")

Do you have a set of Vee blocks ? Put the valve in them and try it.

Have them in my cart, but thought I would try this first. Will be buying the blocks!

 

[BillK]

How many are you planning to do ? Unless you are going to do a mess of engines just clean them up and take them to your local Automotive Machine Shop and ask them to reface them. I rarely charge more than a few bucks to do something like that.

 

[Daedalus]

How many are you planning to do ? Unless you are going to do a mess of engines just clean them up and take them to your local Automotive Machine Shop and ask them to reface them. I rarely charge more than a few bucks to do something like that.

I know some of them were hit and some are very obvious. I want to make sure there aren't some that are slightly bent before I even consider using them. What's the limit on this? At what point do you say the valves are beyond refacing? I was hoping to find enough in the parts bin where a little hand lapping of the mating surfaces would be enough. Beyond that, and I would probably also have to consider tipping the valves or machining down the lifters to keep the lash in spec.

One of the donor engines came out of storage from a garage where the young owner took on a project and might have been in over his head, as evidenced by the chipped camshaft end (rookie mistake when installing the t-belt on this engine). I fear he may have also inadvertently turned the motor over into some valves by hand. as leakdown test was pretty bad on a couple cylinders, even though the valves all look ok.

 

[BillK]

I know some of them were hit and some are very obvious. I want to make sure there aren't some that are slightly bent before I even consider using them. What's the limit on this? At what point do you say the valves are beyond refacing? I was hoping to find enough in the parts bin where a little hand lapping of the mating surfaces would be enough. Beyond that, and I would probably also have to consider tipping the valves or machining down the lifters to keep the lash in spec.

One of the donor engines came out of storage from a garage where the young owner took on a project and might have been in over his head, as evidenced by the chipped camshaft end (rookie mistake when installing the t-belt on this engine). I fear he may have also inadvertently turned the motor over into some valves by hand.

If they are bent, they are bent. Period in my opinion. If they are bent .001" , they are going in the scrap bin. If I can see that they have hit, they are scrap. Its not worth saving $5 and taking the chance that it will break.

Sometimes its hard to tell if they are actually bent or just wearing to one side, which they will do.

 

[Daedalus]

If they are bent, they are bent. Period in my opinion. If they are bent .001" , they are going in the scrap bin. If I can see that they have hit, they are scrap. Its not worth saving $5 and taking the chance that it will break.

Sometimes its hard to tell if they are actually bent or just wearing to one side, which they will do.

Nothing's perfect, even for valves that weren't hit. I would like to establish they're all good to within .002 before I use them. That's a challenge, but I have another idea. Parts will be here Wednesday. Stay tuned!

I wish these were $5, I would buy a whole set at that price. OEM parts are NLA, and pretty rare even used.

 

[RoninB4]

Nothing's perfect, even for valves that weren't hit. I would like to establish they're all good to within .002 before I use them.

-For .002 TIR (.001 for both high/low) you'll really need the V-blocks and an indicator. The stems need to be immaculate and so should your set-up. It really doesn't take much to throw an indicted value off by +/- .001. Chucking a valve stem, or any precision surface, in a drill chuck is a bad idea unless the stem is wrapped in shim stock to avoid marring the surface. Not trying to be critical of your methods.

 

[BillK]

Nothing's perfect, even for valves that weren't hit. I would like to establish they're all good to within .002 before I use them. That's a challenge, but I have another idea. Parts will be here Wednesday. Stay tuned!

I wish these were $5, I would buy a whole set at that price. OEM parts are NLA, and pretty rare even used.

What engine is it ? Very high quality valves are available for most domestic engines for less than $10 Small block Chevy $5

 

[Firebrick43]

checking runout on valves can easily be done on a v block and a 123 block as a stop for the stem.

 

[Daedalus]

What engine is it ? Very high quality valves are available for most domestic engines for less than $10 Small block Chevy $5

02-04 Focus SVT 2M5Z-6507-AA, 2M5Z-6505-AA

My general observation is that aftermarket valves are often heavier than factory ones, or subpar in some other way. I would rather keep it OEM, but I also don't want to pay $600 for a set of valves, even if I could find them. I have enough donor parts that I should easily be able to field a full set. I just need to verify what I have.

 

[Daedalus]

Chucking a valve stem, or any precision surface, in a drill chuck is a bad idea unless the stem is wrapped in shim stock to avoid marring the surface. Not trying to be critical of your methods.

I had the same concern, but valves are hardened steel. I went "firm snug" on the chuck. Didn't crank on it too much. I don't see any marring of the surface. I really can't see it affecting the endurance limit, and it's not where the seal engages.

 

[nadogail]

I would not trust my low bucks Taiwan made Drill Press for anything requiring a percussion measured in Thousandths of an inch. It was purchased by Mail Order from an advertisement on the back cover of a magazine. If I remember correctly it was in 1978.
It still does what I bought it for and I have replaced the chuck and the arbor. It came with an Eagle brand chuck and I treated myself to a Jacobs Chuck and an imported arbor from MSC about 35 years ago.

 

[cannuck]

Everyone has covered the drill press thing, but IMHO one of the best tools to measure with is the chuck of a valve grinder Even my lathe is not that accurate. Before doing so, though, check it's runout with a G&P reference bar. On top of measuring runout, it is the a perfect time to very lightly reface the valve even if already "assumed" to be true.

 

[RoninB4]

I had the same concern, but valves are hardened steel. I went "firm snug" on the chuck. Didn't crank on it too much. I don't see any marring of the surface. I really can't see it affecting the endurance limit, and it's not where the seal engages.

-You may already know this but for others that may not know. The term "hardened steel" is not an absolute value that applies equally to all components. If measured on Rc scale it could be from Rc 35 to Rc 65 and still qualify as hardened steel. There's steel, pre-hardened 4140 for example, that qualifies as hardened yet is still plenty soft enough to cut/dent with HSS. All I'm saying is that whether something is supposed to be hardened, you don't know how hard it is, and it can easily be damaged by 3 points of contact. The expected runout from a drill chuck, even when new, makes it a poor choice for checking relative concentricity. Not bashing the OP, trying to inform others.

 

[Daedalus]

Super happy with this latest solution and amazed it worked as well as it did. I ordered some 1/2" G25 bearings off Amazon, which showed up yesterday and which I bonded into 4 of the countersinks in a 1-2-3 block with epoxy, after cleaning all surfaces with Acetone. I let them cure under a heavy weight (and with a soft material between them, to soak up any tolerances), so the bearings are in good metal contact with the countersinks, and the bond is more of a fillet stake, just so the bearings don't pop out during testing.

The 1-2-3 block is fastened down on the underside to the drill press baseplate with a 3/8 bolt and large washer. I also clamped a metal block to 1 end as a stop. That eliminates play on 5 degrees of freedom for the valve. The rubberband is just there to hold the valve in place until I can get my hands on it, so the dial gauge doesn't push the valve out of place. The 4 spherical points of contact means not having to deal with the variability that might be introduced with machining and positioning tolerances of a v-block.

With a bit of grease and a very light twirl with my fingers on this first test, the gauge moved less than .001 total, and it's repeatable on where the high and low spots are. Confident I can lap that in!

Not bad for $3.20 worth of bearings (though with a min buy of $8). The only improvement I might consider making is to sub a .008" total travel "federal" style gauge in for the dial gauge, to greatly increase the resolution and accuracy of the measurement.

https://www.amazon.com/gp/product/B079HFGD7Y/?tag=atomicindus08-20

 

[Old Man Roger]

How did the valve you tested in the drill press compare when tested in the new rig?

 

[Daedalus]

How did the valve you tested in the drill press compare when tested in the new rig?

I've only tested 2 valves in the drill press (quickly realizing it wasn't good enough) and this valve was tested the most of the 2. The first time I used the drill press I got .014 runout. Every time I re-checked this valve I got a different value. With the drill press I was fighting error in the bearings, and error in the chuck. Even with the most careful centering of the valve in the chuck I could almost never get better than .004, which tells me that's probably about what the bearing contribution is. The rest is in the chuck. This setup is better in every way. It's quicker to set up (what set up?), has much less error and is dead repeatable...pretty much foolproof.

 

[Old Man Roger]

Nice. For the drill to work you would probably have to create some type of roller contraption to push the chuck to one side.

 

[BurtEggley]

lap in? You are going to all this trouble to lap them in? What a total waste of energy. They get ground a touch on a valve grinding machine, and the seats get cut with three angles. You can use your drill chuck and a good file to back cut the intakes for better flow. Your pistons should tell you which ones struck, unless you had valve to valve interference. The guides have more play in them than your drill chuck probably and they might need to be replaced too while the head is off.

 

[Daedalus]

lap in? You are going to all this trouble to lap them in? What a total waste of energy. They get ground a touch on a valve grinding machine, and the seats get cut with three angles. You can use your drill chuck and a good file to back cut the intakes for better flow. Your pistons should tell you which ones struck, unless you had valve to valve interference. The guides have more play in them than your drill chuck probably and they might need to be replaced too while the head is off.

I'm not willing to just assume the valves are good. Seriously doubt the guides have that much play in them, but I'll verify that too. What do you know about these parts that I don't?

 

[BurtEggley]

I'm not willing to just assume the valves are good. Seriously doubt the guides have that much play in them, but I'll verify that too. What do you know about these parts that I don't?

I spent a couple years as an automotive machinist machining and building 1000's of engines including some high end ones that were used at Indy, Formula 1, and in Nascar.

Last night I tried to find some high quality valves for that car but you are right, they are mostly discontinued and aftermarket seem to be unknowable quality. A full set of Zetec DOHC intakes made by Supertech performance are about $200. Individuals can be bought too if you look at their site. If I was looking to do it inexpensively and make the car nothing more than a daily beater, I'd be looking at yards like pick and pull, you pull etc., to see if I could get a replacement head from a wreck. The car is 20 years+ old so there should be some yards within 100 miles of you that have something you can use.

or look to see what else used that head, Maybe the Escape and Tribute both shared that head based on what Summit racing is showing. You'd have to do some homework to compare them. I am not a Ford guy although we have an older Sable Wagon we keep around as a spare car.

 

[Daedalus]

I spent a couple years as an automotive machinist machining and building 1000's of engines including some high end ones that were used at Indy, Formula 1, and in Nascar.

Last night I tried to find some high quality valves for that car but you are right, they are mostly discontinued and aftermarket seem to be unknowable quality. A full set of Zetec DOHC intakes made by Supertech performance are about $200. Individuals can be bought too if you look at their site. If I was looking to do it inexpensively and make the car nothing more than a daily beater, I'd be looking at yards like pick and pull, you pull etc., to see if I could get a replacement head from a wreck. The car is 20 years+ old so there should be some yards within 100 miles of you that have something you can use.

or look to see what else used that head, Maybe the Escape and Tribute both shared that head based on what Summit racing is showing. You'd have to do some homework to compare them. I am not a Ford guy although we have an older Sable Wagon we keep around as a spare car.

I appreciate the effort, but here are no SVT engines in the local yards. They're just too rare. They made about 8000 of those cars total, as you stated, over 20 years ago. Any that pop up are rare and usually pretty costly, often with questionable history. Many are just beat to **** at this point. Now that I have enough donor parts, I want to do this right, and do it once. It's just too much work to have to do it again. Just FYI, Zetec valves will not work with a SVT head. Yes, I have spent years searching, and "doing my homework".

 

[BurtEggley]

I appreciate the effort, but here are no SVT engines in the local yards. They're just too rare. They made about 8000 of those cars total, as you stated, over 20 years ago. Any that pop up are rare and usually pretty costly, often with questionable history. Many are just beat to **** at this point. Now that I have enough donor parts, I want to do this right, and do it once. It's just too much work to have to do it again. Just FYI, Zetec valves will not work with a SVT head. Yes, I have spent years searching, and "doing my homework".

I believe you about the head. The searches I did with what you provided led me to three engines that fit in the Focus of those years, and a couple forum threads implied it is the DOHC version. which came back as the Zetec DOHC head. I know that Ford is kinda goofy on parts, there are two automatic transmissions, near identical, in our 2004 Sable wagon, and one needs the actual build sheet for the car, or the full serial on the transmission to order even a pan gasket. Even though the Sable wagon and Taurus wagon are essentially the same, millions were made, three years after the car was made, things like brake caliper kits went obsolete. The Acura's I have had, mechanical parts are available from Honda for sometimes 20 - 30 years - except electronics. Good luck rebuilding the head.

this is an example of what historians say about your car "The Focus SVT featured a 2.0-liter Zetec engine, tuned by Cosworth, producing 170 horsepower and 145 lb-ft of torque. It was paired with a 6-speed manual transmission"