8 cylinder front engine iconic vehicle
  • User avatar
  • User avatar
  • User avatar
User avatar
By Daniel5691
#135063
Hello ! I wanted to post an update on some progress.
I have assembled a little more of the front of my engine.
This is a hybrid build, with my original Euro S heads and camshafts and a newer 5.0L block
Here are some pictures of the old parts I replaced.
Image
Image

Check out the wear / abuse on the drivers side cam spacer...

Image
I had to order a special tool from Scheissco to install the spacers. It was on backorder, and finally arrived.
Image

Here are a few pictures of the assembly to this point. Everything except the cam gear bolts are torqued and triple-checked at this point. I used the WURTH HHS adhesive lubricant to assemble everything. All parts, seals, bolts went in new.
Image
Image
Image

Also, one last question. This is the first 928 engine I have assembled. The block, heads, and cam towers were assembled by the machine shop. Please note the arrows in the next picture, showing the positions of the cam gear marks, and the crankshaft woodruff key. I did not notice the relative positions of these marks until I had the cam gears on today. Do the positions at this point, prior to timing belt installation represent any sort of problem? I just imagined needing to turn the cams or crank relative to each other to set the timing, and running into some sort of problem.
I posted this up on RL 928 as well. There are a lot of great 928 guys still over there that I wish would come over here.
Thanks
Dan
Image
User avatar
By worf
#135087
If the crank is at 45° you can turn the cams all day long. And you will need to do that when you string the belt.

I am confused though, by the concept of allowing a machine shop to assemble a 928 engine.

Did they use the shop manuals? Did they order the trick cam tower seals from Greg Brown?

If the crank is NOT at 45° you may have bent valves.

If it was at 45° and you’ve turned the crank with no t-belt installed then you may have bent valves.

Slip your harmonic balancer on just enough to check the crank angle. Woodruff key should be pointing directly to port.
User avatar
By Daniel5691
#135096
Howdy !

I've not turned anything, cams or crank, at all.

I think a lot of explanation is in order here. This may get really boring.
It started with a running 928S Euro with a simple passenger rear cam seal leak.
I couldn't get to that pesky rear cam seal, so
I pulled the engine and decided to jump off the bridge into While You're At It River.

I wanted to do a reseal and check / replace wear items as needed. Off came the cam towers, off came the heads, and I found damage in the form of deep gouges in a couple of the cylinders. I decided to try to build a 5.0 L hybrid instead of pursuing some sort of repair on the original 4.7 block.

I found a short block on RL which was advertised as good condition. The bores were nice, but the 8 pistons which were sold with it apparently were from some other stupid 5.0 L block and the block needed enough honing on a couple of the cylinders to get the pistons to work.
I found a machine shop in Louisville with the right Sunnen machinery to get it done.

The 5.0 pistons also went out to California to get the relief cuts from a shop recommended by Greg Brown.

The cylinder heads were refurbished at the same time, they got new valves and valve guides, etc, etc, but there was no actual damage to the valves or heads which had to be repaired at the time of refurb.

I was worried about being able to assemble the cradle correctly,
it seemed really overwhelming considering my skillset/ experience,
and I was worried that I would F up all the prior work and new parts that had been invested.

The machine shop recommended a mechanic who had been a Porsche dealership mechanic/ SCCA veteran / European car specialty shop for the assembly. The machine shop didn't actually assemble the engine, sorry for the error on description.

So, the lump eventually showed up back in my garage assembled from short block to cam towers, and I made sure to use all the "newest - type " of gaskets, etc including cam tower gaskets and oil pan gasket.

Everything inside there should be new / fixed / replaced / OK. "Should be. "

Sorry for the rambling reply. I really appreciate your time in reading and responding.
Thanks
Dan
User avatar
By worf
#135217
Greg answered your question in basically the same way I would have: turn the crank backwards to TDC, etc.

However, I would also advise that you pull the spark plugs so that crank is easy to turn and that if gives you “good feel” in the event that you get piston/valve interference.

Actually, what I was going to tell you is to rotate the crank backwards to 45° BTDC, then you can freely turn the cam gears with no worry. Then with the crank as above turn the can gears to 22.5° ‘behind’ the TDC marks. (More or less 11:15 on a clock face looking at the front of the engine. But either use a more accurate measure or do ‘gear teeth’ math.)

Then string your belt and proceed as Greg directed.
Last edited by worf on Mon Feb 21, 2022 9:09 pm, edited 1 time in total.
User avatar
By worf
#135261
worf wrote: Mon Feb 21, 2022 5:18 pm Then with the crank as above turn the can gears to 22.5° ‘behind’ the TDC marks. (More or less 11:15 on a clock face looking at the front of the engine. But either use a more accurate measure or do ‘gear teeth’ math.)
Strike that. Let me be more clear:

Then with the crank as above turn the can gears to 22.5° ‘behind’ the TDC marks. (More or less 11:15 - relative to the TDC marks on the belt backing plate - on a clock face looking at the front of the engine. But either use a more accurate measure or do ‘gear teeth’ math.)

I add that clarification because the TDC marks for the cam gears are NOT at "12 o-clock."

If I remember correctly the cam gear has 48 teeth so 22.5° is 3 teeth.

But, don't trust my memory. Count your teeth and do your own "degree-per-tooth" math.
User avatar
By worf
#135263
Like Greg, I've never been "in" this situation.

After thinking about this again, do not take my advice and attempt to got backwards to 45° BTDC. 'Cause you have to go through TDC to get there when you are going backwards and that will almost certainly result in valve contact.

I have a short block on a stand where I can see the pistons.

Let me go play with it.

Stay tuned.
Daniel5691 liked this
User avatar
By Daniel5691
#135265
Thanks so much.
I'll be really interested in what you observe.

I am giving serious thought to investing in a borescope so that I can observe what's happening as I turn the crank. A borescope might at least give me a chance to adjust and move the camshafts independently to minimize or avoid contact...
Dan
User avatar
By worf
#135269
Daniel5691 wrote: Mon Feb 21, 2022 9:07 pm Thanks so much.
I'll be really interested in what you observe.

I am giving serious thought to investing in a borescope so that I can observe what's happening as I turn the crank. A borescope might at least give me a chance to adjust and move the camshafts independently to minimize or avoid contact...
Dan
You won't see shit with a borescope unless it has a mirror adapter so that you can look sideways at the valves.

mKay... here's the deal... based upon eyeballing the approximate location of your crank's Woodruff key.

#2 is a BDC
#3 and #5 are at TDC

Your cam gear - at least the starboard side (passenger side LHD) - appears to be 5 teeth BTDC which is 37.5°-ish BTDC and that is 75° BTDC in crank units.

This assumes 48 teeth on the cam gears. Count them. It might be 52. I forget. But it's double the number of crank gear teeth. Do the counting and the math.

I'm not going to think about where the valves are. Hopefully valves for #3 and #5 aren't bent already.

I can't count the other side's cam gear teeth from you picture.

Your crank gear appears to be more-or-less about 90°-ish ATDC.

You have about 40° of backward rotation before #1 and #6 get too close to TDC that you shouldn't turn the cams.

Mark your balancer - if it's not an new ATI unit - for 30° ATDC and every 10° after until the balancer tells you where the crank is at. If you're using an ATI balancer it's already marked.

Turn the crank backwards 35 to 40° until you get to a mark on your balancer. At least 20°. 30° is better.

Every cam gear tooth is 15° of crank gear rotation. So, if you stop at 60° ATDC math becomes easy.

At this point you should still be able to turn the cams without hitting pistons with the valves.

Do the "tooth math" based upon where the crank's at and rotate the cams as indicated. And that will be passed the tdc marks.

Frankly if it was "me" I'd pull the cam towers and heads, check the valves, raise the head studs and put everything back together at 45° the way it should have been done in the first place.
User avatar
By Geza-aka-Zombo
#135305
Being a 90 deg. cross plane crank V8, if there is no valve interference when at 45 BTDC, there would be no interference at 135, 225 and 315 BTDC, no? Pistons would be located differently, but none would be further up the cylinder as when at 45 BTDC.

Image

The crank key looks like it's at ~ 225 BTDC. Couldn't you just mark the damper at 180 from the 45 mark to check and set the crank to 225, then back off the cams the correct number of teeth CCW from their mark, and string the belt?
User avatar
By worf
#135314
Geza-aka-Zombo wrote: Tue Feb 22, 2022 7:32 am Being a 90 deg. cross plane crank V8, if there is no valve interference when at 45 BTDC, there would be no interference at 135, 225 and 315 BTDC, no?
There is no interference at 45° + N x 90° for integer values of N.

From “eyeballing” my short block, the window of non-interference is roughly 40° wide. So 20° on either side.

Geza-aka-Zombo wrote: Tue Feb 22, 2022 7:32 am The crank key looks like it's at ~ 225 BTDC.
When #1 is at TDC the Woodruff key is at the 4:30 clock position. Pretty sure. But I might have that mixed up with 45° BTDC. It’s early (for me) and my first cup of coffee isn’t empty yet.

Geza-aka-Zombo wrote: Tue Feb 22, 2022 7:32 am Couldn't you just mark the damper at 180 from the 45 mark to check and set the crank to 225, then back off the cams the correct number of teeth CCW from their mark, and string the belt?
The key here is to first mark the balancer/damper sufficiently to determine the actual position of the crank.

Given the posted picture, #3 and #5 are within a few degrees of TDC. If BTDC we want to rotate backwards. If ATDC we want to rotate “forwards” (clockwise when looking a crank nose.)

Yes?
Last edited by worf on Tue Feb 22, 2022 9:20 am, edited 1 time in total.
User avatar
By Daniel5691
#135315
Hello gentlemen ! Wow, thanks so much for the time and attention to this issue. I will install the HB and post up some pictures this PM for a better reference point. The photos above aren't really taken from a vantage point that offers much accuracy...

Dan
User avatar
By worf
#135323
Daniel5691 wrote: Tue Feb 22, 2022 9:07 am I will install the HB and post up some pictures this PM for a better reference point.
At the risk of being considered a pedantic asshole when I just want to be a friendly pedant: Make sure whatever balancer you’re using is marked sufficiently so that we can determine the absolute crank position.
User avatar
By Geza-aka-Zombo
#135326
And don't put the damper on backwards...

Looking thru my stuff, this pic is from Duane's timing belt write-up - shows crank at 45 BTDC (~3:00)

Image
User avatar
By worf
#135375
Geza-aka-Zombo wrote: Tue Feb 22, 2022 9:30 am Looking thru my stuff, this pic is from Duane's timing belt write-up - shows crank at 45 BTDC (~3:00)
Yup. Thus, TDC is 4:30.

Based upon me screwing around with the short block I have on a stand, Dan’s crank is about 90° ATDC with #3 and #5 in valve killing position.
Geza-aka-Zombo liked this
User avatar
By Daniel5691
#135616
A quick update.
First of all, thanks so much for the amazing analysis and diagnosis of my predicament.
The level of knowledge is just outstanding.

My mechanical experience prior to the 928 was limited to simple repairs on a 1970 Opel GT, a 1978 MG Midget, a 1980 VW Wabbit Diesel C (!!!), and helping my sons on their 1991 BMW 328 IS and 944S. Then plopped right down into the icy water of the 928.

Anyway, I have decided to pull the cam towers, at a minimum. With the cam lobes not in contact, there's nothing pushing down on the valves.

I have Greg Brown's "super cam tower gaskets " installed. I don't know if they can/should be re-installed.

I am unclear about pulling the heads after that. I understand that the head studs are installed too low, but I can't wrap my mind around the concept of 'stretching' the head nuts, and how much / to what degree the engine is in danger of failure the way it currently sits. If the nuts were torqued on the studs to the correct spec, would that mean that the twisting/stretching force on the nuts is sufficient? I apologize for being dense here, I am honestly trying to learn, and I am simply ignorant at this point.
I know I need to replace the head nuts, but what about the head gaskets? can they be re-installed or should they be replaced also ? The gaskets all have 0 running time on them.

Thanks for the help. I work 2 jobs and kind of dive in and out of the 928 project. it's frustrating. But I want to come out of this not only with a decent running 928, but as an educated and competent owner as well.

have a great day,
Dan
User avatar
By worf
#135672
Daniel5691 wrote: Wed Feb 23, 2022 3:46 pm Anyway, I have decided to pull the cam towers, at a minimum. With the cam lobes not in contact, there's nothing pushing down on the valves.
Since you have to pull the heads anyway...

Daniel5691 wrote: Wed Feb 23, 2022 3:46 pm I have Greg Brown's "super cam tower gaskets " installed. I don't know if they can/should be re-installed.
You can double-check with Greg, but I'm gonna bet he'll tell you to throw'em away and get new.

About the only "reusable" gaskets on a 928 engine are the big thick rubber ones and only for the first year or three.

Daniel5691 wrote: Wed Feb 23, 2022 3:46 pmI am unclear about pulling the heads after that. I understand that the head studs are installed too low, but I can't wrap my mind around the concept of 'stretching' the head nuts,
Yeah. Don't fixate on that.

Geza will probably take me to the woodshed for this explanation, but I think it will get the concept across adequately:

The purpose of tightening to a low torque and then turning 90° (180°, 270°, etc.) is to apply a specific clamping force. Torque is a measure of thread friction only. It isn't directly a measure of clamping force. However, once you've got a fastener in contact with whatever's being clamped (the initial torque), when you turn a fastener X° you can calculate the clamping force given knowledge of the materials involved.

A metric pitch of 1.0 means 1 mm of 'compression' per rotation. So, given Y rotations you can calculate deformation of the fasteners and what's being clamped, and then derive a clamping force. This "angle method" is used in situations were clamping force must be within a narrow window lest "bad things happen" whereas the actual clamping force for a "torqued" bolt can vary considerably due to surface conditions of the fasteners, moisture, etc.

This all, of course, depends upon relative material composition but lets assume for purposes of understanding that the nuts are far less "stretchy" than the studs. When you turn the nut 90° on 10 threads you are spreading the force from .25mm of compression (assuming 1.0 pitch) on 10 threads. When you turn the nut 90° on 5 threads you spread that compression on 5 threads. So you are putting that same clamping force on half as many threads.

(insert discussion of material elasticity, cycle fatigue and other stuff but the basic issue is too much force over a too small area.)

Now...

Daniel5691 wrote: Wed Feb 23, 2022 3:46 pm and how much / to what degree the engine is in danger of failure the way it currently sits.
None. As long as you never heat it up. Say... from running it.

Daniel5691 wrote: Wed Feb 23, 2022 3:46 pm If the nuts were torqued on the studs to the correct spec, would that mean that the twisting/stretching force on the nuts is sufficient?
Yes. Until those big chunks of aluminum that are the heads and cylinder block expand under the heat of operation. That expansion is more "compression" on the fasteners and that force is getting spread over fewer threads than as-designed.

The failure mode here is that the block gets hot and the top of the stud let's go and your head nut flies around the engine compartment. Then the clamping force on that specific portion of the block/head is close to zero. A head gasket leak would be the *best* failure mode but warping of the head and block is possible if not absolutely probable.

Furthermore realize that some fraction of the first thread on a threaded surface is "tapered" and thus not a "real" thread. And the stud ends are possibly slightly weaker than the body. Two reasons you want to see a thread or three poking up past the nut.

Daniel5691 wrote: Wed Feb 23, 2022 3:46 pm what about the head gaskets? can they be re-installed or should they be replaced also ?
The silver rings on the head gaskets are "crush rings." They crush once. And only once. Reliably that is.

Daniel5691 wrote: Wed Feb 23, 2022 3:46 pm The gaskets all have 0 running time on them.
Yup. Sucks. BTDT: Had to through away a custom head gasket set once due to issues not of my own making that required re-removal and rework of the heads on a client's build.

Cheaper now, than after the engine is running in the car...
User avatar
By Geza-aka-Zombo
#135758
Brown’s “nut stretch” – that paints a picture – is basically incorrect, but does have a slight element of truth to it. There is stretching involved, but that occurs in the stud, which should be thought of as a very stiff spring. What’s happening in the nut is not stretching, but there is some deformation, which may be temporary (elastic) or permanent (plastic) depending upon the nut material and clamping force. Basically, when a nut is first tightened, the first engaged thread takes up all the load. As the torque is increased, the clamping force increases proportionally and the first thread starts to deform and some load gets transferred to the second thread, and this cascades further into the nut. Fully torqued, the thread loads follow roughly per this chart: thread 1- 34%, 2- 22%, 3- 17%, 4- 11%, 5- 9% & 6- 7%. So, what happens if that 6th thread isn’t there? I suppose that 7% of load will just get distributed on the other 5 threads – no biggie, assuming the nut isn’t made of a material on the hairy edge of failure; I suspect this is not the case considering the other end if the stud (assuming it has the same thread) threads into tapped cast aluminum (generally not as strong as steel).

Image

All this said, your condition is still not ideal, as I’d like to see the stud at least flush with the nut (I really like to see at least 1 thread protruding in my designs – it’s more of a visual thing to know you have full engagement whether needed or not). One thought I had was to replace the washer you have with a thinner version. In this instance, I’m thinking the washer is employed to distribute the clamping load to a greater area to reduce what is called bearing stress, and to protect the aluminum head from scoring by the nut as it tightens. I think both of these functions, in this instance, can be handled by a thinner washer of the same diameter. So an option would be to replace the washer, one at a time. The water pooling corrosion angle is a distraction, which could be easily addressed by smearing a bit of RTV on the nut, if really concerned.

Image

Wouldn’t extending the studs further out of the block reduce the number of aluminum threads engaged? Knowing how the thread load distribution on a nut looks, wouldn’t a stronger material (grade) shorter nut work? Just some thoughts.

Regarding WORF’s torque discussion, as I bolded above, tightening torque directly relates to load (tension) on a bolt. It follows a simple equation: T = c*d*F, where T is applied torque (lb-in), d is nominal bolt diameter (inches), F is the preload (lbs) and c is the torque coefficient (.20 for dry, unlubricated and .15 for lubricated). When tightening the head, it is done in stages and in a specific bolt order to insure an even load distribution to compress the soft (relative to the other materials) gasket, while keeping the head flat.

I’m not an engine builder, but I assume that if you are tightening the head in stages using the ¼ turn approach, you first tighten the head to an initial torque value (in the proper bolting order), then set your torque wrench to the final torque value. At this final value, you do a ¼ turn on each bolt in the proper order, enough times until you reach the final torque setting, and the nut will no longer turn. You’re basically doing multiple torqueing stages without having to change your torque wrench for each. Each time you turn the nut the torque is increasing as is the load (tension) in the stud. The thread pitch WORF mentioned is indeed a factor, in that if you turn the 1mm pitch thread 1 turn, the stackup is being stretched (the stud) & compressed (gasket) by 1mm total.
User avatar
By worf
#135815
Geza :banghead:

Read the shop manuals’ sections on engine assembly. You obviously haven’t.

Then re-read Dans’ posts. In particular, read the first post that describes the context. You clearly do not have that context.

Then go revise your last post accordingly.

Torque as displayed on a torque wrench is a measure of thread friction when in the elastic region of the fastener. But, it is not as accurate in predicting actual clamping force as can be done with the torque-by-angle method.
User avatar
By Geza-aka-Zombo
#135831
worf wrote: Thu Feb 24, 2022 11:52 am Geza :banghead:

Read the shop manuals’ sections on engine assembly. You obviously haven’t.

Then re-read Dans’ posts. In particular, read the first post that describes the context. You clearly do not have that context.

Then go revise your last post accordingly.

Torque as displayed on a torque wrench is a measure of thread friction when in the elastic region of the fastener. But, it is not as accurate in predicting actual clamping force as can be done with the torque-by-angle method.
Don't hurt your head too hard. I think I can agree that the torque wrench indicates thread friction (and head bearing friction), both of which are a function on the tension on the stud. That being said, since the torque coefficient can vary by up to 33% based on the conditions (lubrication), I suppose the Porsche engineers removed that variable by using the torque by angle approach. If that's what they specified (you are correct, I didn't read the shop manual on engine build) by all means follow it. I'm sure they did a lot of testing to verify their recommended procedure gets repeatable results under varying conditions. Thanks for the info.
User avatar
By Daniel5691
#136111
Hello !
Thanks again so much for all the help thus far.
Dual-posting this here and RL, wish those guys would just come over here with the Cool Guys.

When removing the cylinder heads, WSM manual refers to "reverse order"
Does this mean that the torque / tightness of the CH nuts needs to be sequentially relieved, in the same way that it is tightened... or are they just referring to the removal order of the nuts themselves?
Sorry for being so ignorant, I just want to do the best possible job and not miss anything.
Here's the WSM reference....
PS I have the 16V not the 32 shown in my WSM...
Image
Thanks in advance,

Dan
World's Dumbest Home Mechanic
User avatar
By worf
#136177
Daniel5691 wrote: Fri Feb 25, 2022 2:55 pm When removing the cylinder heads, WSM manual refers to "reverse order"
Does this mean that the torque / tightness of the CH nuts needs to be sequentially relieved, in the same way that it is tightened... or are they just referring to the removal order of the nuts themselves?
I pedantically do the procedure in reverse. So: progressively loosening the bolts/nuts.
User avatar
By worf
#136182
Geza-aka-Zombo wrote: Fri Feb 25, 2022 6:12 pm I would think that the progressive loosening would start with #10. Is this correct?
That is the way I have interpreted the shop manuals.

That's got to be about as fun as my LR2 that has a[…]

The 928 Photo Thread

Hi there. image.png

Widowmaker Project

A teaser https://i.imgur.com/JxgfLt0.jpg https:[…]

Just watched a recent Doug Demuro video re[…]