Yea good point, not worth the risk of leaving those in there and having a problem, so I pulled them last night. I was just liking the idea of working on some of the part removal from the engine stand where it seems so much easier.



I think I have most of those other concerns planned for... I have this scissor style lift, I think marketed for motorcycles, for my trans. I used it when I did my torque converter last year. Maybe I'm thinking about it wrong, but I have the three tc bolts off the flex plate and have slid the tc all the way into the trans leaving a small gap. The two are free of eachother and I'm planning on just leaving the torque converter mounted to the trans input shaft, so I shouldn't have to worry about trans fluid...



I did a dry run of the engine lift last night, this is how I plan to do it, from the side. Front is not an option in my garage. I shortened most of the chains to minimize the extra height needed to lift it up. And fully extended the boom. It is still rated for 1.125 ton at that length. It does look like the hood will have to be lifted higher or removed, which is fine because I'm actually planning to make the hood blisters into functional heat extractors anyway over the winter. I'll definitly want another strong adult on hand but I think it looks like it should work, right?

Yes, I do have an extension on the hoist boom. Like you said, the long front ends of these cars don't allow the use of engine cranes that you get today. At least not the "Harbour Freight" version I have now. It's about a 7/8ths scale copy of the full sized crane I used to have. Damn imported junk.
Luckily, with the extended boom out so the hook is centred over the intake, the balance is just enough so that it doesn't want to tip. That's with the engine alone though.
As for over the side, yes, I've been cornered into doing that once before. Same issues as with coming out the front, just presented in a different way.
My last install (the one in the picture) has the trans propped up like you have yours. Torque converter still in place and no leaks. I have been that guy that ended up laying in a puddle of fluid because I wasn't paying attention!
Good luck.

Very good @raptere , now you are ready for the removal.

Curious: Have you managed to remove all of the bellhousing bolts with the transmission tucked up as it is?
One thing comes to mind with the setup you have. The engine has to move in two directions for removal. One is up to clear the engine mount while at the same time (or nearly so) moving forward to get off the dowel pins that align the engine and transmission.
Although I re and re'd my engine on its own last fall, the transmission was on my homemade trans cradle but with no driveshaft or torque arm. This allowed me to drop the engine onto the mounts, then push the transmission into place.

My plan is to lower the trans far enough to get to the bell housing bolts, then pull the trans rearward to separate from the engine. Only thing is, once I separate them, I need to support the engine from the pan, so I was waiting till I was ready to pull it...maybe even till I have it hanging from the hoist.

Understood. I was under the impression that you were to try and not deal the torque arm but I see now that your approach will be similar to my last go around.
I had the engine sitting on it's mounts with my mini floor jack supporting the oil pan. This allowed for some tilting action along with lowering the transmission and that tilt provided the requisite room to get those bolts. Once the transmission was pulled free, I leveled the engine and properly blocked up the oil pan and let the engine sit while the transmission was getting rebuilt.

Good news and bad news... The short block shipped at last, but for some reason, the soonest date FedEx would allow me to schedule delivery is next wed the 8th. Even though it is sitting in a FedEx facility an hour south of me already...

Guess it will give me time to get the old engine out, and the K member installed.

You make a good point that I think I made a mistake on with separating the engine from the trans... I'll have to see how low the trans will tip without taking apart the torque arm bushing... the whole thing with the torque arm and drive shaft should be able to slide back an inch or so, but pivoting to let the tail shaft of the trans down far enough to remove the trans to engine bolts could be a problem... I may have to take apart that bushing mount and drop the drive shaft after all... we shall see...

Regarding your engine sitting in limbo...I wouldn't sweat it. No matter how you you plan, there'll be something to slow you down. Just put that into the point of view of being a hobby and with that, there's no timetable really. I only say that because I tried that and failed. Your intentions and urgency could obviously be different than mine.
As for your transmission and the TA and DS still in place...that'll be a tough one. I found that with those bits out of the picture, I could to drop the rear of the engine substantially to gain access to the bolts. Otherwise, it was impossible.

May have had an aha moment on a simplified half lift valvetrain alignment concept...

Does this make sense???
1. Install all valve train, and go around two full rotations of the crank, quarter turn at a time, tightening any loose rockers untill all valves achieve zero lash.
2. If you have the correct pushrod length, tighten your first intake and exhaust valve using using the ice method, to verify 90 deg angle between valve and theoretical line from centerline of roller tip and roller trunion. Question here, if it is not 90, can i adjust and set the rest of my intake and exhaust rockers like the first two. Or is this when I need to get different length pushrods?
3. Then tighten down wach adjuster half the lift (plus .020" to account for the hydraulic tappet compression), different for intake and exhaust, based on thread pitch of rocker stud, in my case 3/8 - 24 (or .042" a turn) effectively setting the rocker so the angle relation will be back at 90 deg at actual half lift.
4. Lock it all down.

Please confirm or correct. I'm tired, and this may make little sense, but it seems like it does while I'm here I'm my shop, fidgeting with the heads and rockers I can't Install for another week...


Terrible picture, but it gives a bit of an idea... of course im not to 90 deg, and i havent turned down the adjuster for half lift so the roller tip is nowhere near the middle of the valve...

You can't determine geometry without the heads installed and the entire valvetrain in place. Although I see you have the basic concept.
It'll work like this, and I apologize if I'm repeating myself or going over what you already know.
Short story is this:
Heads on.
You only need to reference one cylinder.
Lifter, pushrod (whatever you have or an adjustable one), rocker sitting on the pushrod and valve tip. No polylock needed and no preload.
With the lifter on the cam base circle use your ruler across the retainer and this will establish the reference plane you'll use. That being the edge of the ruler that's resting across the retainer.
Measure the distance from that plane to the centre of the trunnion.
Measure the distance from that plane to the centre of the roller tip.




The red line is bottom edge of your ruler (reference).
Blue line is roller tip axis distance to the reference.
Green is trunnion axis distance to the reference.
Subtract green from blue. Green never changes however the blue point will move up or down depending on the pushrod length (also with cam rotation but that's not in play here). Given that you're on the base circle, you want the trunnion centre below the roller axis and exactly half your net valve lift.
This is easily accomplished with an adjustable pushrod.

Like I said though, you have to have the engine assembled (or mocked up) to do this.
There are the elements of the lifter preload and the head gasket thickness (if measured without one) that'll have to be incorporated into your calculations but you have to be setup first. I've done this exact method in the car. On a stand it's much easier get in for a closer look.

I did these illustrations way back when I was sorting through my previous installation errors. Replacing worn out valve guides due to incorrect geometry gets really old really fast.

This what we're trying to achieve.


Valve is closed with the lifter on the base circle. The roller tip axis is exactly half the net valve lift above the trunnion axis. The red lines represent the valve centreline and a 90° reference.


The valve is now at mid-lift. Notice the tip axis and the trunnion axis are "level" with each other.



The valve is fully open. The tip axis is now half the valve lift below the trunnion axis.

I used checking springs for the sake of demonstration.
When you do the procedure described in the post above, your full running spring is in place and no engine rotation is required. Just a static measurement and adjustments to dial it in.

Ha, yes, I understand this all must be done with the engine valve components and heads in place, my photo was just a general visual. 😉

I guess I'm trying to mix different strategies together. I have a hard time understanding yow ypu use a ruler as a straight edge and accurately measure the distance to your theoretical centers when you often can't even see them on a well turned component. That's why I liked the idea of establishing your 90 deg geometry at 0 lash then just running the adjuster down half lift so that ypu result in having the 90 deg geometry at half lift...

I can appreciate what you're saying and in reality, my retainer didn't have flat register to lay a ruler on. So I improvised with a little jig. I have assembled other engines that have your valvetrain components (perhaps exactly) and the visual is surprisingly easy. In your case, the trunnion centre is well indicated. The roller tip is a little more difficult but one can measure the diameter and put a dot at it's centre.

I'll post pictures of my "jig" and you'll see how it's more obvious than it might appear.

I don't like posting this much as it looks pretty hacked however that "jig" which is pinched under the bottom of the retainer provides the reference plane mentioned in this discussion. My retainer as you can see doesn't have a flat surface to lay a ruler on. I used a checking spring in this case because I had to insert the jig between the retainer and spring. For a flat topped retainer, the full spring is used and a ruler is the reference.


All this old caliper is doing is providing a flat, parallel surface to the retainer to measure from. A ruler laid on top of a flat retainer will provide a similar data point. It doesn't matter where that point is, as long as it's the same for all measurements.


This is a view trying to capture the relationship between the reference plane and the trunnion centreline. I laid a piece of known bar stock across the jig (in this case it's an Allen key) and could now see a measurable distance between those two points.
That's measurement 1.
Remember that this is totally static with no preload and with the cam on the base circle. No rotation. You can see how the roller tip centre is higher than the trunnion centre. That's the value we're looking to ascertain.


From that same plane, I could get a reasonable measurement to the roller centre. (Excuse the cat hair!) . That's measurement 2.
Now I have the numbers to work with. Subtracting one from the other should equal half the net valve lift.
In my case half the net lift is (.575" / 2 = .2875"). The roller tip is .530" above the plane (as per the digital read out on the vernier used). The bar stock was 1/4" thick (.250").
.530 - .250 = .280". That's pretty darn close to the target. You have to take into account whatever amount of preload you'll add to the lifter which in turn will require a slightly longer pushrod.
I had worked with the adjustable pushrod until the trunnion axis (which moves up or down with varying pushrod lengths relative to the roller tip which is always at a fixed distance) was at the desired value. Reading the scale on the adjustable pushrod then gave me the length of pushrod I had to get.

Now I feel like I'm making some progress! Engine is out, hood still bolted, and done solo, with radiator in place. Hardest part was getting the ground straps on the back of the driver's side head off... didn't notice I had the whole stud turning and not just the nut... There is certainly something slightly unsettling looking at the car you love with no engine in it!

Next up, swap in the tubular K member! Should me much simpler considering I can litterally stand in the engine bay!

What to do with the old long block? It ran fine just wasn't about to ask it to handle 470 hp. There a market for them?

It's here and looks pretty nice!

Now... back to the non stop questions...

Is this a problem with the discrepancy in the shape of the block and head gasket opening for the center upper coolant passage??? At least I think that is coolant...

There is also a slight misalignment with the two openings all the way to the right, I don't think that is an issue, is it?

EverythEng else look good, I've got heads and fasteners ready to go!

Another interesting note, I'll use thread sealant anyway, but is it possible the special Smeding block has things rearranged so the head bolt holes dont break through into anything? I looked with a flashlight and couldn't find any that seemed to break through into any other passages...

Those passages look nearly identical to those on the actual chevy gasket, so maybe it is by design?


Top is mine, bottom is the chevy one.

I really want to install the heads, anyone have any thoughts on the gasket holes?

I was reading that the coolant passages are large to allow sand to come out of the casting easier, but many gaskets intensionally use smaller holes to balance flow and reduce hot spots. I'm leaning towards using it as is... about half of each hole is blocked, so it should give about a net one hole worth of flow, and that is the size single hole I see in that position in some other gaskets...

Or... do I just drill one more hole the same size right between and slightly above the existing holes? To basically give the intended two-hole flow.

Or... Just use the Fel-Pro 1003's I already have sitting in my garage that AFR suggested in the first place. They have much larger holes there in the center that may locate better. It will be giving up a little bit of compression and quench though because it is a 0.041 instead of an 0.032.

Original Plan with strange coolant hole alignment: Mahle .032 Gasket, Upgraded 9cc dished piston - Quench .042" - CR 10.46
Secondary Option I already have on hand: Fel-Pro .041 Gasket, upgraded 9cc dished piston - Quench .051" - CR 10.25 I would probably be fine with the slightly lower compression, it's still over ten, but what am I giving up with the .051" quench? Isn't that pretty high?
Third option: I just order a pair of the Cometic 0.032 gaskets that have a single larger hole in the location where the two smaller are currently on the Mahle.

My understanding is that most factory style gaskets had two similar sized holes in that center location and only one to the outsides because of the two side by side exhaust ports there which is where a lot of heat builds up. Many performance head gaskets have those center openings even larger....

I have a Mahle 5776 gasket. .032" thick and rated as a marine application.
In the past, and it appears now since discontinued, I used a Victor Reinz 5746 @ .026". They appear identical including the centre coolant holes. Chevrolet Performance looks the same as well. Zero issues to report in tens of thousands of miles.

As for the head bolts, I didn't think the Smedding block was anything our of the ordinary but, sealant never hurts. I use it always on all head bolts. I also use ARP fasteners with their recommended lube and torque spec. Once torqued, I revisit that spec the next day to possibly compensate for any slack that might occur. I can't say I've ever had to add another 5 ft-lbs but it's cheap insurance.

PS. You might want to soften the sharp edges on those pistons. I'm surprised they weren't touched a little with some Scotch-Brite.

Further to head gasket question. I went back through my build pictures.



I never saw anything unusual regarding the gasket match however it would appear that this old school Chevy block (~ 1979) has a larger coolant port that the block you have.

Check out this thread at Speed-Talk. It may may add to the confusion although your research does show that the exhaust side is the critical area. A restriction at the top will force more through the bottom (as it's orientated in the picture you posted).
FTR, there's a collection of head gaskets in one picture. I can't say I've seen a "group photo" with so many. Yours and mine are in the bunch as well.

https://www.speed-talk.com/forum/viewtopic.php?t=65492

Hope that helps.

So what would ypu do about the head gasket and how the coolant passage lines up with it in my case? Use as is? Modify the gasket? The Block?

Side note, I am using all ARP hardware, and the right fastener lubes and sealants. Only the best, right?

I've never heard of braking edges on piston features... I'll have to do some reading...

I reached out to an engine building forum and the reference to the FelPro gasket with the larger hole came up. Seems the thinking is to add a hole between the two that are there. FelPro obviously had their reasoning. Looks like yours has more too do with the Smedding block and how it's cast.
I'd add a similarly sized hole between the two and call it a day.
Yes sir, only the best as in ARP hardware. My engines are full of them. I like too that they have specific lubes and torques for everything they manufacture.
As for the pistons, it's good building practice to smooth out any sharp edges so as to eliminate a potential hot spot in the combustion chamber. This applies to the heads too. Something like that can act as a glow plug and start combustion before you want it. The pictures makes them look very pronounced but it may not be that way in person.

So hopefully this should do it... think I'm good to torque them down???

Also, I ran my fingers over the edges of the piston. Not sharp, looks like there are actually scratches from a scotchbright pad...

I'd say you've done your due diligence. One way or the other I can't see it making a difference. That at least one manufacturer had a reason to do their gaskets different from just about everything else out there at least suggests that you've done no harm. So, yes. Good to go with peace of mind.
As the pistons. Perfect. It's just one of those good practices that could keep you out of trouble. Well done Smedding.
There's a couple of things I always do before buttoning down the heads. I like to verify my piston below deck value at TDC. I also like to determine TDC and how it is relative to the timing pointer. It easier with the heads off but can be accomplished after the fact.

Heads are on and torqued, starting to look like an engine! Tomorrow evening I plan to move on to valve train!

Just to confirm, on cast aluminum valve covers, the continuous inner lip goes towards the bottom or exhaust side of the head right?

Thought I'd share my trick for cutting the holes in the gaskets, it's just a variation of something I saw on another forum. They suggest to use a piece of steel tubing, but I ended up using a spare air chuck. You chuck it up in a drill press and while spinning at the slowest speed use an angle grinder to taper the bottom edge to a sharp blade, then use that to slowly cut the holes in the drill press with a hard wood or aluminum backer. I think there is a good chance a standard twist bit may just grab and shread up the gasket...

Raptere, this is a very good build. I enjoy following it. I like the Smeding 383 short block. Any particular reason why you choose Smeding? Reputation? Quality? I have been shopping around for a 383.

Thanks,
Fred

Correct.

Yea I did a lot of reading on various forums. I saw good experiences people have had with smeding, no nightmarestories. I heard to stay away from a few others. Blueprint sounded like a good choice too, but I liked the cam offerings from smedding better, plus you get mostly all forged internals for the same price or less than the blueprint option, which I believe does not have all forged parts. I liked that you also had the option to tweak or upgrade the build, like I did with my pistons.

Lastly, I really liked Smeding's communication. About half the time I called, it was answered by the same guy that was very knowledgeable and paitient with my many questions. The other half of the time I reliably got a call back the same day, or the next business day. This is huge to me. The decent website is a plus for me too.

Alright, I used all the strategies I've seen, and came up with some on my own, and I'm pretty sure considering only the half lift method, I should be using 7.500 pushrods, BUT that moves my contact patch pretty far out on the valve stem. My previous pushrods, that I was actually planning to reuse, are 7.200, which should be too short per the half lift method, but the contact is perfectly centered on the valve...

Which is more important??? Or, do I split the difference?


7.500" Pushrod


7.400" Pushrod


7.200" Pushrod

Just also noticed the AFR heads manual, says typically standard length (7.200") or 0.100" longer pushrods work...

I'd love to be able to reuse the pushrods I have, they are chromo for guide plates, but not if it is going to significantly negatively effect valve train stability...

SkinnyZ is going to yell at me, but 7.200" unless you want to be replacing valve guides in those beautiful heads.

Thank You Raptere for your very detailed response. Smeding is at the top of my list. Also, their prices are reasonable not outrageous. As a comparison, Chevy's 383 short block just like Smeding's (4.000" bore and 3.800" stroke crank) is now over $6K. Smeding's 383 short block, with the extra's, is about $4,800.

Fred

Minimal sweep and mid lift accuracy is more important than being centred.
I made the mistake of making centred my priority and I wiped out the guides in a few thousand miles. Now, keep in mind that going off the exhaust side of the valve stem isn't something you want to have happen and there are remedies for that but from what I see, you're nowhere near that edge.

AFR has no idea what your stack up of parts is so they've no way of knowing what pushrod length you need. Decked block (or the block itself for that matter), lifter seat height, head gasket thickness, rocker arm used, all affect the length needed.
FTR, 3/10's of an inch is fairly big swing.

There's no yelling. All I can say that if you want the least amount of stress on the valve guides, then you want the narrowest sweep possible. This is what occurs when mid-lift accuracy is achieved. The roller tip moves an equal distance "in" and "out" relative to the valve tip. With too short a pushrod, there'll be greater travel across as the valve before it's half opened and that greater travel equals more angularity with respect to the rocker and valve. That's the killer. And reducing that angularity comes at the "peril" of getting close to the tip edge. It looks to me though that the roller tip is still fully in contact with the valve in the pictures posted. So if it was pile of parts, I'm not concerned with that.
But that's just me.

Here are a couple of visuals.
First up is what amounted to too short a pushrod.
This is 7.2".


Notice the steep angle the rocker arm has relative to valve. This is with the valve seated.


Mid lift.


Fully open.
You can see how there's greater movement at the start of the cycle and therefor greater stress on the parts in play. Plus you're not communicating the full information from the cam lobe.

After several trials I arrived at 7.45".


Closed.


Mid lift.


Fully opened.
The travel and angularity is equal and the stress on the guide is minimized. This is what results in the narrowest contact path.

I made up these pictures as a visual aid to what I was trying to accomplish. Now I just use a straight edge and calipers and measure with the valve closed and do the math to see which direction I have to go. Far and away the simplest and quickest way to get accurate results.

So, I've been thinking about pushrod length and valve tip roller to valve stem interactions for the last day now, and while I do understand the half lift method, I'm not sure I'm convinced it is the best route to go... I can understand it being more critical for stamped steel style rockers there the tip needs to slide across the valve stem and having a wider contact patch, means you are putting more lateral forces on the valve, but when you have a roller tip rocker, like I do, the only force exerted on the valve is perpendicular to the tangent the roller makes in contact with the valve, so perfectly in line with the valve, regardless of where it contacts the valve. The thing is, when you apply pressure off the center line of the valve, that WOULD induce some bending force on the valve, so I would think that is what you would want to minimize...

When it comes to roller tip rockers, I have a hard time understanding why strictly adhering to the half lift method, would be better than trying to center your contact patch on the center of the valve...

Maybe I'm thinking about this one wrong, or maybe I'm just thinking is way more detail than many people do, but I'm struggling to determine if I stick with the nicely centered 7.200 rods, or if I go with the 7.500 rods that give me the nice 90 deg relation at half lift. Or, if I go somewhere in-between...

Found an interesting read on the topic, its related to a ford engine, but the geometry and physics should all be the same...

Setting Shaft Rocker Geometry

My takeaway is this:

"Patterns should never be centered by changing shaft height without considering any effects on pattern width, nor should they be made narrowest without consideration of centering. A wide pattern will wear parts (including the valve guides) prematurely and can also reduce lift, while a narrow but offset pattern can also increase wear on valve guides (although less than a wide pattern). The ideal pattern would be the narrowest possible pattern at or very close to valve center."

This makes me think getting new rods somewhere in the middle, may be my best bet... 7.350 or 7.400...

Centered is not the criteria to shoot for. And your thinking that when centered the force is only perpendicular is incorrect. Regardless of there being a roller tip, there's a lot of pressure applied in the small contract area. It pushes across the valve as well as pushing the valve down.
Half lift, for ease on valvetrain parts, is what to shoot for.
Now, in my case, I also saw the contact patch very close to the edge. And on more than one occasion. When I chose to settle on centered rather than the correct geometry, the guides got wiped prematurely. I've documented proof of that. And it happened to me more than once (I was a slow learner).
To correct this I found that Crower (among others) makes a roller rocker that has the stud hole in the trunnion backset by .050". This pulls the contact patch away from the exhaust side, allowed me to get proper geometry and be centered.
I'm not the first person to encounter this problem. There are so many variables that stack up which in turn has created these solutions.




Ultimately the decision is yours. Your valve guides will tell you if they're happy or not. But you're not going to know overnight.

This is exactly what I'm saying. Notice what they say, " offset pattern can also increase wear on valve guides (although less than a wide pattern). The ideal pattern would be the narrowest possible pattern at or very close to valve center."
But again, within reason, narrow/correct geometry trumps centered. You now have it from another source.
And like my reply above states, "the ideal pattern would be the narrowest possible pattern at or very close to valve center." Also quoted from your post.
But to get there, you've got to fork out a grand for new rocker arms! This hobby is expensive. I've got well over $3000 (USD) invested in cam, lifters, pushrods, rocker arms, springs, retainers, locks, shims and locators alone. All top shelf parts though. I paid a premium for the cam and lifters I wanted. The rockers were a significant hit as well. But I did end up with a sweep of about .030" and dead nuts on centre.

I've a question regarding how you established that contact on the valve tip?
You obviously rotated the engine through a cycle or two correct? Did you use checking springs or make up a solid lifter in place of the hydraulic?

Ok, you got me there... I turned it over two cycles, with the full load springs, and I did notice the plungers were compressing slightly, so the pattern is not perfect, but I think I can use it as a reasonable perportional comparison of the various pushrod lengths and their resulting patterns, though they may in reality be a bit wider. I actually ment to mention this before, but forgot...

The actual pushrod length establishment process does not involve rotating the engine, so i dont think that is in question now, right?

In the end this started as a budget build, but I started stretching my budget a bit where I could, and where the benefit seemed greatest. At this point I prioritize reliability over all out performance. If I don't want to switch rockers at this point which I just bought for the build a few months ago, which length would you go with? I'm leaning towards 7.400"...

Well, considering that the lifter could collapse something approaching .200", you can see where that gets you. Now, if you think about, that nearly 2/10ths (if that amount is actually realized when cycling the engine) might bring your 7.20" pushrod right into play. What you have now are dubious results.
That said, checking springs will also skew the results as they don't keep all the bits tightly held together.
So in the end, measuring those points of reference in the mid-lift article will tell you unequivocally if the pushrod is right. Then with that bit of information you can make an informed decision as to whether that 7.20", while maybe not exact (like a tenth off) would be something that you'll settle on.
As for 7.40"? No idea really. I'd have to get in there and measure. Or find another way to get to where you're going. There's more than one way to go about it obviously. I've tried a few different approaches with unsatisfactory results. Whether that's me or the method is open to debate. There are more than a few videos that describe the process in one manner or another but they all seemed to me to be somewhat flawed. One was measuring how much travel there is in the lifter plunger to bottom it out. Do the testing as you've done with it fully compressed (bottomed) and get your witness mark. Measure the pushrod, factor in plunger compression and there's your length. Don't forget to add in the amount of actual preload you'll get when lashing the valves. Which, incidentally, is something you should be doing anyway (if your lifter wasn't compressing as it is).
But in the end, I'd be dropping the rocker on the stud, ensuring the lifter is on the base circle (no rotation), no preload (no polylock for that matter), establish a reference plane (using the retainer top if they're flat and lay a straight edge across that), measure to the roller tip axis (from the reference), measure to the trunnion axis and do the math. Might take you 20 minutes. Especially if it's on a stand.

@skinny z

I must not have explained myself well. The lifter compressing only effects a portion of the contact patch location on the stud. I should also clarify that I did tighten down the adjuster for the 0.020" predoad, before checking the patterns...

My calculations for 90 deg mid lift were all static, lifter plungers seated against their retainer, so I don't think the 7.500 is in question for that method. I can share some calculations in a bit... I used the stem of another caliper that is known to be rigid and flat for measurements.

My Calcs in inches:
Roller height = 0.540/2 = 0.270
Tip height = 0.152
Int Half lift + lifter preload = 0.272 + 0.020 = 0.292
Ext half lift + lifter preload = 0.278 + 0.020 = 0.298
Avg lift + lifter preload = 0.295
Trunion Reference + tip height + roller height = half lift + preload
Trunion reference + 0.270 + 0.152 = 0.295
Trunion reference = -0.127 (meaning the trunion axis needs to actually be 0.127 above the top of the spring retainer.

I achieved this by laying a caliper on the retainer, then using a second caliper to measure the distance to the trunion axis. In my case my caliper was conveniently 0.127, so I adjusted the rocker till the trunion axis lined up with my caliper. I then lengthened my adjustable pushrod until just the slightest resistance could be felt while turning it. This pushrod length was 7.500.

You look to have the math well sorted. So much so that I'm now compelled to revisit my own measurements. I may have neglected to factor in the tip height in which means I could be off a substantial amount.
Hmm.
In your case, you have the trunnion axis in a position where it's exactly half the valve lift below the roller axis. That would be correct mid lift geometry.
If the valve were to be fully opened (without compressing the lifter plunger) then the opposite would be true in that the roller tip axis is now the half lift amount below the trunnion axis.
What do you think?

Funny how this can play out.
Looking over your way at gathering the numbers was a little different than mine.
While you look to have arrived at a satisfactory conclusion, it got me to thinking about my own approach and maybe I'd left out something in the math chain.
But, looking back, seeing as I was measuring from the same reference, (despite it being tucked underneath the retainer,) simply allowed me to measure differences in axis height above the reference. It doesn't appear as though the tip height enters into it this way.
Anyway, I think you've got it.

SkinnyZ, you said that you achieved a perfectly centered .030" sweep on the valve tip. DO NOT start second guessing yourself. Leave well enough alone.

"The ideal pattern would be the narrowest possible pattern at or very close to valve center."

And no one has mentioned seat and valve open pressures and where those occur on the valve tip.
If the sweep pattern must be off in a direction, I would want it to be towards the intake manifold side so that the rocker is pushing down on the valve tip in the center when the pressure is greatest.