Talk about engine building here. New products, tricky questions or showcase your work - If it's engine building related it's welcome here.
If anyone is reading this forum please skip towards the bottom (post #14) to save yourself having to read all of this. That's an updated roundup of where this is at. (2021-03-10).
Well, I've finally got the data and I'm confident in it now, but it's not what I expected. I know for certain people run much more aggressive setups than me and based on my numbers either I don't understand the limits or I don't know how they do it if 0.100" is truly the minimum safe clearance for exhaust and 0.080" for intake.
At the bottom of this post is my very complete listing of p2v clearances using my Skunk2 Tuner Stage 1 Cams, stock valves and stock pistons on my b18c1 with a p72 head (which is the stock head). This was a lot of work to get. I measured the clearances using the check-springs method and locked vtec mechanically. The measurements were taken at 0.000 lash, my running lash is 0.007 for intake and 0.008 for exhaust (in a previous post I misspoke stating that running lash is .08, it is not). My rocker ratio is 1.55. I went back and re-checked some of the clearances again with the running lash set and I found under those conditions there is about ~0.015" additional clearance gained with the running lash. I think that makes sense given the rocker ratio and margin for error in having the lash set perfect and variations on lining up the markings on the cam gears etc...
I was also able to measure the mill indicators on the head and the block to get an idea of how much my head and block have been milled. The head's mill indicator is at 0.035" depth and I know stock it starts out at 0.040". This indicates the head has been milled 0.005", so nothing too crazy. The block mill indicator has a depth of 0.033". I don't know the stock depth for the block indicator but I assume it's the same as the head's which would indicate the block has been milled 0.007" which also doesn't seem too crazy. Cumulative milling comes to 0.012".
Measuring the timing changes needed to bring my cam timing up to the manufacturer's recommended specs indicates that the exhaust cam should be retarded 6 degrees (-6) and the intake cam should be advanced 2 degrees (+2).
Looking at my clearances below, even after adding in .015 for the running valve lash after rocker ratio, I am nowhere near the commonly accepted minimum clearance of 0.100". I'm nowhere near it with the cam gear at zero. I'm very surprised by this given most people doing this stuff run JDM Type-R pistons where are domed and have shallower valve reliefs and they often run oversized valves and much bigger cams than me! I'm also surprised that my exhaust clearances are so much tighter than my intake. Looking through posts on all sorts of Honda forums from other people's measured p2v I'm the first time I've seen the exhaust clearance less than the intake, usually it's the other way around. In fact my intake clearances look easy to work with.
Only other data I can gather would be to try again using the play-doh method and see if that is for some reason completely different (shouldn't be) or to put back in the stock cams and check a few of the clearances with the check-springs on those to see where I'd be at stock. I have a feeling even with stock cams my exhaust clearances will be less than the commonly accepted safe minimum.
So what's the word on the real deal here from HPAcademy on how far I can go with my cam timing?
It should be noted that I've already run the engine for 1,000 miles with these cams with the cam gears set at 0 Intake, 0 Exhaust, and I've run it quite hard for those 1,000 miles. Redline 8,400 and many times I ran it all the way up to there.
+Adv/-Ret | Exhaust | Intake
-15 | 000 | 159
-14 | 000 | 155
-13 | 000 | 149
-12 | 000 | 145
-11 | 004 | 136
-10 | 009 | 132
-9 | 014 | 129
-8 | 017 | 122
-7 | 022 | 116
-6 | 027 | 110
-5 | 032 | 107
-4 | 036 | 100
-3 | 038 | 095
-2 | 042 | 090
-1 | 048 | 089
0 | 052 | 083
+1 | 056 | 076
+2 | 060 | 073
+3 | 065 | 066
+4 | 070 | 062
+5 | 073 | 057
+6 | 079 | 053
+7 | 083 | 047
+8 | 086 | 043
+9 | 092 | 040
+10 | 096 | 036
+11 | 100 | 031
+12 | 106 | 027
+13 | 113 | 022
+14 | 121 | 019
+15 | 123 | 015
Not sure what those numbers represent. It also seems you're measuring the "lash" between the follower/rocker and the camshaft, instead of the valve and the valve/rocker?
It's a bit of messing around, but as you seem to be rather paranoid about it, I would suggest this process.
With the bare head, fit the valves for #1 cylinder using the lightest springs you can find - they don't need to be valve springs, just hold the valve closed - them carefully measure the valve position. Attach and align a degree wheel to the crankshaft. Fit your head to the block using the head gasket type you intend to run - you don't need a 'new' one - and run the bolts/nuts down a little more than finger tight - you don't need to torque it, just so it's firmly in place.
OK, bring the crank to, say, 30 degrees BTDC and depress the intake valve until it makes light contact with the piston top or contacts the valve guide/seal and measure the position. Now do the exhaust.
If, as I expect, there is no contact bring the crankshaft to 25 BTDC and repeat. If there is contact with the piston you may wish to back up a litle and recheck, if not repeat at 20 BTDC, until contact is made.
Once you know where initial contact is made, record the valve travel for 1 degree increments until there is no piston contact.
So, now you know the TOTAL valve to piston clearance at all points between the intial and final points. You can now make a graph, or chart, - either on graph paper or preferred software program - of the total available clearances to the piston at those points.
Next, if the camshaft supplier doesn't provide it, you need to establish exactly what the valve lift is as the camshaft(s) are in the previously established ranges.
Easiest way to do this will be to remove the head and use a smaller degree wheel (modified protractor may be more easily available) fitted to the camshaft(s) and zeroed to a known datun (measuring reference point) - I would suggest the either the maximum lift, split open-close points, or something based on their timing proceedure - the important thing is that you KNOW the camshaft position. From the nominal "zero" point you're using, measure the actual valve lift as the camshaft is rotated every degree, or +/- 30 degrees of the nominal crankshaft TDC point.
This will give you the actual valve lift you need to account for. You should now make a graph, or graphs, of the lift to camshaft position.
You can now overlay the charts to see exactly what you have to work with , and the advantage is that by moving the camshaft graphs to the left or right you can see what that does to the available clearances.
It's been a while, so I may have missed, or glossed over, some parts but I trust you have the gist of it.
Hi Gord, you've pretty much described exactly what I've already done. It's been a lot of work.
To answer some of your questions:
You say: "Not sure what those numbers represent" . Those are the clearances. For example at -15 (both exhaust and and intake at -15 that is) the exhaust clearance is 0.000" (or no clearance) and the intake clearance is 0.159". That's an extreme example from the list in the first post. That is the first row in the list so I picked that first to explain. For an easier example where the left column says "0" that is 0 degrees advanced/retarded for both exhaust and intake, or as some people put it the cam gears at 0,0. For 0 degrees advanced/retarded the exhaust clearance is 0.052" and the intake clearance is 0.083". Hopefully this makes sense of it for you, sorry if that was confusing.
Your other question: It also seems you're measuring the "lash" between the follower/rocker and the camshaft, instead of the valve and the valve/rocker?. Correct, that is how it's valve lash is measured this engine as per the factory service manual.
TL:DR? Nope, the process I was describing is much, much more involved and produces much more useful data than the simple process you used - it's about five times the work, though.
I expect we just have different ways of looking at the clearance data you posted, it makes sense to you and that's the important thing. The reason I had/have a problem was that I was expecting actual checked V2P clearances, with the distance available from the lift point to the piston, and zero values aren't what was expected.
Most engines have the clearance at the valve, some have at the valve and camshaft - so long as you use the correct method it's all sweet.
Oh, going back to your concern about valve clash - their hitting each other - I expect that would only be a potential problem with oversized valves and/or much higher lift than you're using, as the 'cylinders' formed by the perimeters of the valves during their travel would be significantly larger in diameter and length.
I know you gone through a lengthy explanation but could a I ask you to unpack the problem a little. Not to be patronising but going back to basics with a problem can often help identify any areas of weakness.
where are you at now with it?
where would you like to be?
what do you consider your blockers to be?
Answer these concisely then we can go from there.
Skunk2 stage1 tuner cams are designed to be drop in replacements and both the b-series’s and cams have been around for a very long time now so we should be able to get you sorted.
First the answers to your specific questions:
I'm glad you asked and not patronizing at all. I've spent a disgusting amount of time working on this since my last post so an update is in order. Where I'm at now is however unfortunately not far from where I was at that last post :(
Where I'd like to be: Either believing my measurements are accurate or figure out precisely why they are not, and if they do turn out to be accurate then how can I work with them?
My blockers: clearances on exhaust side are well below the safety zone. Catch 22 situation now. I'm half a mind to just put it all back together and run it how it was but now I have this new "data" that tells me if I do that the risk is substantial, even though I was running it that way before without ill effect (and it ran well).
And now the updates as to what's been done since. Boy I wish I could call you, would be a lot less typing, and interactive dialogue has a better chance of sorting this. If you prefer to jump on a call and avoid reading all this I'm game.
1) I started to question if maybe I had bent a valve in the process and that could be causing the low clearances. The reason I thought this was a) the clearances were so tight, and b) I ran a leakdown test and it was showing a meaningful leak in cyl #1. Pulled the head off (again) and found no obviously bent valves. Did not pull the valves out at this time but all looked well seated. I did however find that regardless I need a valve job as an aside as cyl #1's valve seats are rather worn and leaked when I put water in the port, a slow drip. Frankly all cylinders valve seats are rather worn but cyl #1 is the worst of the lot. Put the head back on, re-checked all the clearances, results came back the same again, at least we're consistent.
2) Searched and searched the internet and found I'm the only example I can find of a person with wide open intake clearances but ultra thin exhaust clearances, not just on Hondas but any engine. This doesn't sit well with me, I start doing all sorts of experiments on the exhaust vales to find something that would cause that. In one of my experiments I suspect I really did bend them this time. No big deal as I need a valve job anyways and have already ordered a full set of new valves from Honda.
3) Pulled the head off yet again. Both exhaust valves now very bent. At this point I decide to simplify things and cut down cyl #1's valves to just 1 intake and 1 exhaust valve. One of the variables I have been experimenting with was lash and how it affects the measurements and a precise lash has been difficult to achieve with check-springs and vtec locked since you've got 2 rocker arms each with their own lash but they are all locked together mechanically for vtec. I also removed all of the valves from cyl #4 while I was at it for use in cyl #1 to replace the bent valves.I verified the #4 valves are not bent by spinning them with a dial indicator. Neither intake valve was bent in cyl #1 but I swapped in the valve from #4 anyway just because I've been messing with #1 so much I wanted a fresh-ish valve to work with. I also removed the lost motion assemblies from cyl #1 as I found these are probably what caused me to bend the exhaust valves since I have to fight the lost motion assembly when pressing down against the check springs.
4) Re-assembled the engine again and re-checked clearances now with just a single exhaust and intake valve in cyl #1. I only checked with the cam gears set at 0,0 to get an idea of how a definitely not bent exhaust valve would measure up. To my pleasant surprise I gained 0.015" in clearance so now the p2v for exhaust at 0 degrees adv/ret is .067" at 0.000 lash. Since it's much easier now to set lash with just 1 valve in the cylinder and the LMA's not pressing back up against the vtec lobe I was able to measure with the running lash of 0.008 and the same clearance came out 0.080". It's worth mentioning that on this engine lash is measured between the base circle and the rocker pad and the rocker ratio with vtec locked is 1.55. Getting closer to good clearances but still not the generally accepted safe minimum of .100" and also noteworthy if I were to degree the cams to Skunk2's specs I'd have to retard the exhaust cam 6 degrees and the clearances will be much less then. To my unpleasant surprise I re-checked the intake valve clearance again, which was supposed to be a sanity check since it should not have changed, and unfortunately I gained another .020" of clearance on the intake valve so now my intake clearance is almost large enough to drive a bus through it (I'm exaggerating but hopefully you get my sarcasm). So that makes for a failed sanity check because the intake clearance should not have changed. The changes I made were remove 1 of the 2 intake valves, swap in an intake valve from cyl #4 however the valve that was in there was not bent anyway.
5) Further experimenting with lash, I find that if I do back off the lash adjustment screw all the way then I can get achieve a nice exhaust clearance well larger than the .100" however obviously I'm not going to run the engine with a huge lash like that, just happy to see the numbers possible in any way shape or form.
6) I have 2 current suspicions /theories that I think may cause the bizarrely tight exhaust clearances.
a) First theory is perhaps a previous owner of this engine or the machine shop that did the rebuild for me last year before I started doing this stuff myself may have mixed up all the rocker arms and just put them in wherever. One thing I found is that I am unable to lock vtec with shop air, in fact that was my plan originally but I ended up using mechanical locking pins instead since the shop air trick didn't work. Regarding the shop air trick, I was using the special tool from Honda and doing it right (I believe). The air would quickly eject the vtec pin with the middle rocker arm lifted up so it wasn't for lack of air pressure. I am thinking they could just be mixed up and because the bore for the locking pin is align honed from the factory if they are mixed up those pins might not line up nicely. Also I find inconsistencies with lash at the 3 lobes of cyl #1. Now that I'm just running 1 valve with a check-spring, and no LMAs, and with vtec locked, given that the base circle is the same on all 3 lobes I'd think that lash would be the same on all 3 rocker arms. As you probably know lash is measured between the base circle and the rocker arm on this engine. Well if I set the rocker with the valve installed to .008 I see a lash of .012 at the middle rocker and an even larger lash as the 3rd rocker. So that doesn't match up. Meanwhile when the engine is on the road in operation I have no perceivable problems with vtec engaging / it seems to work fine.
b) The other theory I have that may be contributing to my clearance issue is (and I posted about this in a different thread) that my Skunk2 Exhaust cam does not match the cam card exactly. When I measure where the events occur in comparison to the cam card (the the cam gear at 0) I come up with the opening event happening 6 degrees early, the peak lift happening 6 degrees early but with 0.012" (or 0.3mm) more lift than advertised, and the closing event occurring only 1 degree early. You'd think the closing event would be 6 degrees early like the other two events but it's not. This makes the duration 5 degrees greater than advertised as well. I'm working with Skunk2 to see if I can experiment with another exhaust cam, they are only a 1.5 hour drive from me so don't have to wait for shipping. They haven't decided yet, they are trying to say they can't warranty a product that's already been run, I'm saying warranty or not, running the cams would not increase their lift and duration.... if my measurements are correct it's a defective product that would carry product liability for damaged engine (if I'm right). I did measure with a vernier caliper all the vtec lobes and base circles of the Skunk2 cams and although this is very hard to measure accurately like this, the measurements did point to peak lift matching or being very close to what the cam card says. With these measurements I got the vtec lobes all being around 36.54mm and the base circle being around 29.6mm so subtract that and then multiple by the 1.55 rocker ratio and you get a peak lift of 10.76mm which is very close to the advertised 10.8mm. Of course since it's a hard thing to measure kind of have to take this with a grain of salt.
7) Next things I can think to experiment with are swapping back in the original stock cams to see if they also have thinner than safe exhaust clearances. I was on my way to do that when I saw your message come in. And the only other "brilliant" idea I have is to clay cyl #1 and cyl #2 at the same time and see how they measure up. I figure cyl #2 is still virgin / I haven't screwed with it in this process, so if for whatever reason cyl #2's clearances with clay look good but cyl #1's don't then whatever the cause my be it's only cyl #1. So far I've only been working with the checking-springs method of measuring these clearances. It was supposed to be a little easier, more precise, and supposed to not require taking the head off.... Oh well.
At any time in the past were new valve seats installed in the head? Perhaps they just aren't cut deep enough.
@David, never to my knowledge nor by inspection does it seem they have ever been done. I recieved this motor with 250k original miles on it and from what I was told and as I far as I can tell I’m the first person to ever send it to a machine shop. Not surprising the valve seats are leaky that said.
If the 'clearance' seems to have changed, carefully review what you've done and whether it is actually possible for the components to have altered their positions - if you decide they could, see what you can do to prevent this - if you conclude they couldn't, it may be a problem with your measuring procedure, or technique.
You mention a camshaft not being correct in it's timing compared to the card - this is more likely (but not 100%) to be an error in your setting up as even though you may be 'aligning' everything you may not be correcting for stacked tolerances that may throw alignment off, TDC may not be accurately established, or there is some other issue.
It's a little unclear - I know it can be a bit confusing reading, and writing, information as it is editted - but it seems like you may have both camshafts a little out in their timing - are you SURE you have the TDC position correct? With intakes, advancing them will increase valve to piston clearance BUT with exhausts it will reduce it. Retarding the camshafts will be the opposite.
Something I have found useful, in the past, is to do what Scott suggested - take a step back, do some deep breathing to relax, maybe going for a walk, and going right back to the basics - sometimes one overlooks something, has accidentally altered something, mis-read something, or is operating under a false assumption - that last has cost me a lot of time and money over the years, but I'm getting better ;-)
Well, I've done all I can do on this. Somebody tell me if I'm out of line here but best I can figure at this point is fretting over some sort of magical number for a safe minimum clearance is futile. It's really not cool that people all over the internet post info and videos on this and talk about measuring clearances like you can take these measurements to the bank, that's just not the case. I think the deal is you're either worried or you're not and at the small scale of measurements being considered here accuracy is futile unless you have some very expensive and specific instrumentation. Just for example I went and measured all the valves total length with my digital caliper. I found they vary +- 0.03" in length. This makes the largest variation less than 1mm. Meanwhile (for example) when you measure a clearance at cyl #1 of .08" and the safe magic number is .1" you can easily have that on a different valve that just happens to be .02" longer. If you were to take any 2 random valves and balance a level on the stems the level would still be level even though the 2 valves aren't the same height, they are all just that close and the measurements being taken are that tiny.
The new things I've done are:
1) Measured clearances with the stock cams back in. I found that regardless of how imprecise the measurements are there are generous clearances available. You could advance or retard the stock cams a lot without concern. I also found that the same pattern exists the stock cams which is the exhaust clearances are less than the intake clearances. This makes me think my measurements I took of the Skunk2 cams are accurate (depending on how you define accurate).
2) I clayed the clearances with play-doh on both Cyl #1 and Cyl#2 at running lash. The difference between cyl #1 and #2 are that #2 is still pure, I haven't touched it in all this and still has the real springs installed. Cyl #1 has been touched a lot and has the checking springs installed. When claying the motor I found much more forgiving clearances than using the check springs at the same cam timing. Which one is more accurate? Who really knows. The checking springs have their issues and so does the play-doh. I've attached some photos of the play-doh, to me I thought it looked like there is plenty of clearance there. The tightest clearance I measured with the play-doh on the exhaust valves was .1" which happens to also be the "safe" magic number. Of course measuring play-doh when you are worried about very small distances like .01" is kind of silly.
To illustrate my conclusion after spending way too much time on this here are several different measurements of the exact same thing but with different variables stacking like spring deflection, small variations in valve dimensions, small variation from lobe to lobe and base circle to base circle, lash variations between the two rocker arms that are locked together by vtec, deflection from LMAs, variations in the amount of pressure you push down on the check springs or the digital caliper when measuring with that, etc...
Measurement of Exhaust piston to valve clearance at .008" valve lash with the same Skunk2 cams:
Cyl #1 measured with dial indicator on retainer, check-spring and only 1 valve installed: .08"
Cyl #1 measured with dial indicator on retainer, check-spring and 2 valves installed: .072"
Cyl #1 measured with digital caliper on impression in play-doh, check-spring and only 1 valve installed: .1075"
Cyl #2 measured with digital caliper on impression in play-doh, real valve-springs with 2 valves installed: .100"
Thec dial indicator with check-spring measurements are repeatable +- .002" after several tear down and re-builds. The margin of error is much larger when measuring play-doh and a digital caliper for obvious reasons.
So here we have 4 different measurements of the exact same concept but with results that vary as much as .035" trying to evaluate risk regarding a space as small as .1". The variation is 35% of the magic number we strive for.
I had a good friend come over to help and go over all of this with me yesterday who is a machinist and engineer. He checked over all my work and we did several of these measurements again together and he saw the same thing as me.
At the end of the day my feel is that if your measurements show generous clearances like I got with the stock cams then even though the measurements lack real precision this tells you you don't really have to worry about it. If your measurements are paper thin then regardless of real precision you've got a problem. If your measurements are within 35% of the magic number you are after then it's a bit of a throw of the dice and given the measurements don't have the precision needed for this scenario best you can do is lay your eyeballs on a cross section of play-doh and decide good-enough or not good enough and remember what that play-doh looked like when you're on the dyno and considering adjusting cam timing.
Funny thing is I started with check-springs because I thought that was more accurate and gave the ability to measure clearance at a variety of timing settings without having to dear down and re-do the dummy assembly over-and-over-and-over. However precision (unless you have some very high tech equipment) is kind of a misnomer in the scenario and you're better off just doing some play-doh with your cams degreed and using that as a barometer on what you're working with.
Now I'm going to go and measure some v2v with this in mind. Hopefully the clearances are generous enough to outweigh the precision discrepancies.
OK... That first paragraph, with your thoughts on valve length and it's supposed affect, clearly demonstrates you don't quite understand the basics. Some other comments also suggest this.
Nowt wrong with that - I'd repeat my earlier suggestion of walking away for a bit, relaxing, and coming back fresh and going back to basics.
That said, I'll remind you of this paragraph - "At the end of the day my feel is that if your measurements show generous clearances like I got with the stock cams then even though the measurements lack real precision this tells you you don't really have to worry about it. If your measurements are paper thin then regardless of real precision you've got a problem. If your measurements are within 35% of the magic number you are after then it's a bit of a throw of the dice and given the measurements don't have the precision needed for this scenario best you can do is lay your eyeballs on a cross section of play-doh and decide good-enough or not good enough and remember what that play-doh looked like when you're on the dyno and considering adjusting cam timing." That looks like good self-advice - remember, you were running this set-up without problems before you decided to check on that valve leakage.
I'm going to leave my comments there.
Hi Gord, many thanks again for your reply! I’m not trying to be all negative here. I just find that there are so many stacking issues involved and dial indicators (at least the cheap ones I have) are less than perfect and a digital vernier caliper is even worse as that depends on how much pressure you apply. I can measure a valve’s length 10 times and get 10 different answers all very very close but when we’re stacking up errors against very small clearances it’s challenging. Glad you liked my self advice, I’ve put a lot of effort in for something I think most people would not even consider.
I’m definitely taking a breather from all this measuring for a bit. I may try and clay it again degreed to spec before final assembly. I like the results of claying it because they gave me the most favorable results. I can’t say they (or any of the other approaches) were the most accurate but there is an old addage about a picture being worth 1000 words.
To my own disgust and curiosity I put the head back on and measured again. At this point the head is all torn down except for cyl1. So no other valves, no LMAs, no other springs in other cylinders. Only 1 valve and check-spring on the exhaust side and 1 valve and check spring on the intake side. Nothing else. Assembly was much easier this time with nothing else fighting me.
I gained .010” of clearance.
The measured maximum lift of the cams also reduced by the same .010” so now the cam’s peak lift is almost matching the advertised values.
I think deflection is a meaningful factor.
What I can’t figure out is why still are the clearances better with play-doh than with the dial indicator. Now they are seperated by about .010”. I wonder if the fact that the cams are in motion the whole time against the play-doh vs the cams are static while I measure with the dial indicator is where that last .010 is hiding?
Again caveats, measuring play-doh with a caliper is hard to call the gospel and in general how much can you rely on your measurement instrumentation at these small amounts anyway.
Sorry to add to more reading to this thread. In an attempt to make this less of an effort to read everything above this I'm putting a more concise, updated roundup of where I'm at at the moment. Just ignore all of the above unless you're brave enough to read it all:
I'm running Skunk1 Tuner1 Cams on b18c1 with stock pistons and stock valves. Using Crower springs and retainers. Car has actually been running great for about 1k miles (dyno'd at 173whp, it's NA) just I had never degreed the cams when I installed and after watching the HPA videos I thought I might as well go ahead and "do the job right". I know the videos recommend the play-doh method, but I went with check-springs and dial indicator method so that way I could get the clearance "easily" for a variety of cam timing combinations rather than just one. There is a popular fixture kit for this engine that seems like the whole Honda world uses so sounded like a win.
1) I swapped in the check springs on cyl1 and mechanically locked vtec and installed the fixture for mounting the dial indicators and measured my p2v clearances. I have good clearance on the intake but scary clearance on the exhaust. That alone is surprising to me with these mild cams with stock pistons and valves. For instance, at 0 degrees advanced/retarded my exhaust clearance @0.000 lash was .052". That's a far cry from the rule-of-thumb safe minimum of .100". I checked again with the running lash of .008 and this brought my clearance up to .065" which is still pretty scary. For all the tests I did I used 0 degrees advanced/retarded as the controlled variable.
2) I was worried the clearance issue might be that perhaps I had bent a valve or something on the exhaust side swapping the check-springs in so I pulled off the head and removed the valves from cyl #4 and put them in cyl #1 just to be sure I've got straight valves in there. None of the valves appeared bent visually or when I spun them with an indicator but just to be sure. I also removed the LMAs from cyl1 as I found they were fighting my finger pushing down on the valve and that was a problem. After doing this I put the head back on and re-checked clearances. To my surprise this gave me additional .015" on the exhaust and also an additional .015" on the intake for some reason. Still scary exhaust clearance though.
3) I pulled the head off again and this time I clayed the motor (with play-doh) and using this technique I found my clearances were all fine. The exhaust was .108" at the same cam timing (never moved the cam gears). I actually clayed cylinder #1 and #2 at the same time because I was curious if they would be different since one had the check-springs and no LMA and the other was still in the untouched stock configuration. Both were set at running lash. With both cylinders the clay measurement was more or less the same, of course measuring play-doh leaves a lot to be desired.
4) Last test I pulled the head off again and this time removed the entire valve train except cyl #1. So now it's down to only 1 valve each in intake and exhaust in cyl 1, both using check-springs, and the rest of the valve train is vacant so a lot less sources of deflection going on. Stuck the head back on and re-measured my clearances and now I've gained yet another .010" in clearance. With this configuration with running lash set I've got .090" clearance on the exhaust valve. This is still less than the rule-of-thumb .100" safe minimum. This is with cam timing at 0 so I haven't even degreed the cams yet which if I did would have me retard the exhaust timing a few degrees which would reduce clearance so that's scary.
So now I'm left with the problem of trying to understand what all of this means. I've got 4 different tests all telling me different measurements of the exact same thing. Some of them indicate I should never run this engine, yet I ran it hard for 1k miles and it was driving great. The clay test indicates a reasonable margin of safety. All of these measurements are easily repeatable if you re-configure the head the same way it was when each test was taken so they aren't random (I verified this). So what does all this mean? How do I measure a clearance which I can take to the bank?
I figure it's worth updating this thread as I managed to figure out where my clearances went to my satisfaction.
A) I determined my best measurement was #4 from the post above. The less sources of tension on the camshaft the better for taking this measurement. I found also that a big advantage to the play-doh method is that everything is in constant motion when you rotate the crank around and make your impression in the play-doh. Being in constant motion alleviates some of the error introduced by static tension that you can get with check-springs and a load valve train. In #4 however the valve train was all unloaded except for the 1 valve being tested and thus why I believe it is the best of the above choices.
B) I finally managed to come up with a reliable method of measuring the cams (which from some of the posts above I had a suspicion that something wasn't right) to see if they are within advertised spec. Pic is attached of how I did this. I found that the intake cam was pretty much right on advertised spec. Each lobe was at most +- .001" from spec. This amount of error can easily be accounted for. However the exhaust cam was .009" larger than advertised on cyl #1&2 and .008 & .007 respectively on cyls #3 & 4. You may laugh at me about .009" but it all stacks up when you put it together. So where did my clearances go:
.007 to deck milling
.005 to head milling
.009 to exhaust cam being larger than spec
Add these up = .021"
Given the magic number / generally accepted safe minimum is .100" we can see I've lost nearly 1/5 of my safe clearance to all these stacking issues. At least with the cams I may be able to get the .009 back. Can't get the milled material back from the deck or head.
Where did my clearances go? They went to several places that add up to a meaningful amount. .007 to deck milling, .005 to head milling, and head.
There is one caveat worth mentioning which is that I'm asserting that the extra .009" oversize in my cam is uniform across the lobe which it may not be. There is a way to get more precise measuring the exact right spot on the cam using a degree wheel attached to the cam (which Gord suggested in an earlier post). I think I'll just have to go on without that level of precision.
I did just get a valve job out of all this which will have moved the valves up a few thou higher so at least I've gained back a small amount of p2v there.
Now that I finally have measurements I can believe in I just have to decide what sort of minimum p2v clearance is the least I'd run with which boils down to a personal decision.
Hi Joe, I feel that you may be overthinking things since you've already proven that the clearances you've measured are fine, given that the engine has been run hard for some time. Part of the issue is that 'rules of thumb' are just that - A guide. There's very few absolutes in the world of engine building and what is perfectly safe for a small bore, 4 valve engine, would be dangerous for a large bore 2 valve engine. It also depends whether you've got good control of the valves which comes down to your valve spring and installed pressure vs the cam profile.
Ultimately I'd suggest using plasticine/playdoh on the piston crown with your lash set to the normal recommendations. This will give you the most realistic idea of what your clearance is. The technique is demonstrated in this webinar - https://www.hpacademy.com/previous-webinars/260-checking-piston-to-valve-clearance/
Thanks for that Andre. I tend to agree with you after all this and have come to realize that unless I'm building a super precise race-car (which I'm not) then as long as the play-doh doesn't look too tight then there's no need to chase more exacting precision. If I ever need to do this again I'll just do the play-doh method. Sure it's imprecise but it's practical enough.
Not a shocker given you're the expert! I guess I'm just a guy who really likes to understand all the details.