Engine and Transmission Tech

Thanks Jim, not bad. Would need to cut the seats .134"

Jim

Jim, My math was wrong . I used the lower coil bind of the other springs.1.1 coil bind. 1.68 installed gives .030 until coil bind so would have to cut .030 deeper.



Edited 2 time(s). Last edit at 02/03/2023 12:52PM by mgb260.

Please beware of Comp springs on Amazon.
I ordered up a set of 26981's from what looked like a legit seller.
They arrived in a used comp box with photo copied instructions.
A quick check showed that they were nowhere near what they should be.
They were returned and reordered. Same result.
Ordered finally from Summit. Got what I actually ordered cheaper than the big A.

Cheers
Fred

Good point.
These are a lot cheaper than anywhere else. Amazon really sucks about this kind of stuff.

[www.amazon.com]

Figured I'd throw this up here as I'd already created it as a reference of the PAC-R310 to the Comp 795.

I would say the Comp 795 is the better option for what we are trying to do (machine out to 11 degrees), especially as the rover keepers would be hanging out the bottom of both, but will be twice the stick-out on the PAC.

I also screen-shotted this useful side-by-side comparison and figured I'd post it.

Also this:
[www.lsxceleration.com]

PAC-1276X
Type:
Beehive Valve Springs
Spring Diameter Bottom (in.):
1.290
Spring Diameter Top (in.):
1.040
Spring I.D. Bottom (in.):
0.906
Spring I.D. Top (in.):
0.650
Max Lift (in.):
0.650
Load (lbs) @ Installed Height (in):
150 @ 1.800
Open Load (lbs) @ Open Height (in.):
420 @ 1.140
Coil Bind (in.):
1.070
Rate (lbs/in.sq):
409

So far the best prospect I've run across for the solid roller cam, installed at 1.664 for .544" lift and .050" from coil bind, 205 lbs seated, 475 lbs open. Its a bit more than I was shooting for but right in the range my machinist thinks is appropriate for this cam. Requires the spring pocket to be cut .104" deeper and .060" larger in diameter.

Dave, the dimension I am the most curious about on those retainers is the thickness of material at the thinnest point of the inner wall of the taper, where the shoulder of the spring seat most closely approaches it. Does your cad program allow you to easily dimension that thickness? If we are able to thin that section of the web any at all it will decrease the depth of cut in the spring pocket at a ratio of roughly 3:1 so that's worth keeping in mind. Probably more critical not to go too thin in the retainer but if we had over .050" (I doubt that) we might chance it.

Jim



Edited 1 time(s). Last edit at 02/04/2023 12:05PM by BlownMGB-V8.

Also Dave as I'm sure you are aware, you will have to allow for the radius in the retainer where it meets the collet in your lathe to get good repeatability on your depth of cut. Just a friendly reminder. If you are using a depth stop in the back of the collet it shouldn't matter of course.

Jim

JimB
Scroll back to page 8 and look at the post from January 23, 2023 10:34AM.
I have detail on the current (10 deg) and reduced (11 deg) wall thickness and percentage reduction on cross sectional area.

As regards the retainer machining (from 10 to 11 degree)… it will be done on a Haas CNC lathe.
The first one will be Dykem’d and set in the Chuck gripping on the .640 dia and pressed into the chuck while tightening - basically axially seated on the surface the top of the valve spring contacts. I will use a bestest indicator to zero the tru-adjust chuck so it has no run-out, and likely try a few more in and out to see how well they all seat in the chuck and evaluate the run-out.

On the first one. I’ll take very slight cuts (at exactly 11 degrees) and increase the X (cross feed) offset till it cleans up. I am about to purchase a selection of ball bearings to use as a way of very accurately measuring the gauge depth of the taper, and will use these as a checker; and also use them to more perfectly measure the existing taper gauge depth (I’ve been seating pin gauges in there thus far).

Once the first one is right - we will be able to repeatedly machine all the others exactly the same - as well as they are repeatable to each other - and assuming the chucking is repeatable. Hopefully the earlier test on run-out repeatability will determine I can hold an acceptable amount and not have to adjust the chuck on each one!

I see. 5 thou isn't likely to matter but I wouldn't want to go any thinner.

I'm assuming that HAAS uses 5C collets which have a very sharp corner at the nose. Fine except for when you have a radius like the 795's do. That can throw out your repeatable clamping enough to matter.
But, all 5C collets accept an adjustable back stop and most people who use them have at least one of those so there should be one lying about in a handy spot. Just have to make sure the end matches the bore and leaves room for the boring tool. Sometimes an intermediate plug is used and would likely be handy here.

That machine should do a fine job and be quick about it as well. Pierre had a Hardinge chucker and later traded it for the CNC version. I honestly think he liked the old mechanical one better.

Jim

Good call Jim.
Suffice to say, although I did an apprenticeship before moving pretty quickly into engineering (but still do a lot of machining at home) , I will be relying on the journeymen machinist/owners of this tool & die shop to be making the setup for this operation. They understand I’ll be making quite a few pieces, but that they need to be accurate and repeatable.

Personally, I feel like the best place to be locating axially against is the spring contact face.
I may just machine & surface grind a thin washer (in advance of going to the shop) with a chamfer on the ID that I can use on each piece to keep away from that radius.


Edited to add a little more:
As I mentioned, I've got some ball bearings coming my way that I'll use to verify dimensions on the existing 795 retainers to set the gauge dia of the taper. The way I've measured them thus far is to use a pin gauge on the 795 (and all the other retainers I've drawn) at the .4560 reference dimension, but this is not assured obviously as there may - or may not - be a relief in the angle at the end. It also assumes the stated angle they are machined at, is perfect....

I'll measure at least 10 of the 795 retainers to get a good data set and perfect the drawing as needed. This technique will obviously allow me to verify the post-machined 795 pieces.

I will also use this "two ball" method to verify the exact angle of the rover keepers - both the pre 1980 machined units and the sintered ones from the 1994/1995 serp heads i have in my collection.




Edited 2 time(s). Last edit at 02/05/2023 12:40PM by turbodave.

Sounds reasonable. If you were to check the dimension from spring seat to the bottom end I'd be willing to bet you won't see more than a thou or two in a 32 piece batch which would make the back stop perfectly fine to use. It does allow for a stack up of tolerances though. Your washer takes more time to load and unload and it has to locate on a taper (the nose of the collet) unless it is a very close fit. It could work OK but if there's an indicator stand available it would be quicker to just inspect that dimension and eliminate the extra step. I have no doubt those guys will get you set up the best way.

I'll wait to see how this all turns out before I order the springs. Hopefully the end result will be as good as we are hoping.

Jim

Looking for input from you guys...

What do you REALLY think we need to run for seat / open pressures on a 6400 RPM Rover V8 using a Crower 50233.

Crower states exactly the same 120lb / 290lb numbers for everything from a 50227 with 402/420 lift and 4000rpm redline to the 50233 with .408/501 lift and 6500 redline... So it seems obvious they just pencil-whip these numbers.

Most of the published data points we see regarding flat-tappet valve train spring pressures are related to the SBC. Of course the SBC has valves (I'm guessing), 3/8" larger dia than the largest oversized 1.63/1.40 valves that can be installed into the seats of the stock heads. Those larger valves have to be at least 50% more mass of the Rv8 valves? I'm pretty sure the historic SBC numbers were asol with stock (heavy) retainers and non-beehive springs...

Anyway,
These are the springs I was looking at for my engine. I'm now erring back towards the Comp 26981 - if anything because they clearly WILL work, and I already have the Manley seat cutter that'll work for them - and I just need to get moving on this!!!! Any other thoughts or suggestions?


oh, and if anyone is interested, I'm happy to share my excel file so you can play with the spring installed heights of the five comp beehive springs pictured several posts above in the image (along with a few PAC springs) to easily see the seat and open forces.




Edited 3 time(s). Last edit at 02/07/2023 09:48AM by turbodave.

Dave, I think the 26981 at 1.7 is perfect for the cam you have. When you cut the retainers you may have less seat cutting to do if you can get +.050.

Well I've decided to pull the trigger on the PAC-1276X springs unless anyone has seen a better candidate.

[www.lsxceleration.com]

I'm counting on Dave and Dustin to make them work. How is it going on the retainers, anything new for us on that front?

Jim

Well, I'm back off my honeymoon, but work was crazy busy first week back, so I'll be going to the machine shop this coming week to machine the 795 from 10 to 11 degrees.

In the meantime, I decided to do a little more investigation into the angles of retainers in the evenings...

I used the ball bearing height approach as shown below, using "amazon special" metric and inch assortment boxes. It was interesting seeing just how far some bearings can deviate from their supposed perfect diameter, but actually used the true diameters in CAD, noit the theoretical diameter.

For the retainers, I lightly cleaned the taper using very fine emery to remove any high spots, then cleaned with scotchbrite. Obviously, the rover retainers were used, but the Comp ones were brand-new. I measured four samples of each; and the dimensions were all very similar across each sample. I also re-verified the zero at the end, and re-verified the dimensions on random samples to validate the Gauge R&R and each time was no more than 1.5 thou max variation (usually less than 1 thou).

Anyway, this is pretty interesting. The "machined" rover one had a lot of variation, but this could be seen as the various machining marks and past use keeper contact points. The Comp 795 was amazingly consistent, but the Comp 712 was confusing; how can something CNC machined be so poor? Maybe this is made on the older machines at Comp?, or outsourced? Regardless, it's a little disappointing.

But even if you look at the angle measurements (oops, except for the 795 white angle, - its X to B in that example as the .437 ball is too small)) the angles are all over the place.
And these angles deviate from the 22 degree included angle (11 degree single angle) of the Rover valve keepers......
EDIT -just to clarify, I show all these dimensions as "included angle", not the one-side angle as these are typically reffered to; just half the numbers if you prefer)....






Edited 4 time(s). Last edit at 03/12/2023 04:54PM by turbodave.

So, now I wanted to verify for sure, what the angle is for the Rover Keepers / locks / collets (depending on what part of the world you live)!!!

As I have a DRO on my mill, I decided to make use of that, so set a paralell in the vise, and verified it was paralell to the indicator over two planes.
I then selected 12 different keepers (four each from pre 1980, 1980 ish (first year of sintered retainers) and 1995. These were lightlty cleaned with very fine emery, then scotcbrite to remove any burrs.

To the mill, and setting them each on the top of the paralell, and using a .030 height reference (total swing of my bestest was .040"), I ran the table along until the full .030 was seen, then back to verify return to zero - per the image below.
I actually needed to put some blocks of steel either side of it to stop it sliding, but got very repeatable results over all the samples.

I started drawing it up in cad, both as a min, average, and max, but then realised a schoolboy error - that i didn't have my indicator pointer at 90deg, so went back (luckily it was all still set up) and instead of repeating everything, verified the heigh against two feeler gauges, and sure enough, I was only seeing 29 thou instead of the indicated 30.... No matter, so I just drew them up using the same DRO axial readout, but made them for 29 thou.

The end result (these are included angles - just as with the retainers above), was 21.956 average, with a max angle of 22.47 and min angle of 21.67. none of the locks seemed any different from the next over the different years either, which is good to know.

The one big takeaway here, is that the lock / retainer intraction (given the info here and the post above) clearly function with angles of greater tolerance mis-match, than many of us would have expected - at least in stock applications....




Edited 2 time(s). Last edit at 03/12/2023 04:49PM by turbodave.

Not surprising that there would be some variation. Clearly if it was an exact match the parts would lock up more tightly but given existing machining tolerances absolute perfection is often impossible and we have to accept "good enough".

Nice work anyway.

Jim

I just stopped by the shop today; they are finishing off a large thread cutting job on the lathe tomorrow morning and I should hopefully be able to knock these out Wednesday afternoon.

Sounds great.

Jim

Well, we are ready to move!
17 modified retainers headed to Jim today. I have two sets (and a couple spares) for my own engines as well.

These are a thing of beauty for sure!

Just in the retainer, they are a 10g saving over the pre 1980 Rover steel retainers, 11g saving over the Comp 712.
And that's not including the reduced mass of the top few coils of the spring either....







Edited 1 time(s). Last edit at 03/23/2023 10:37AM by turbodave.