Drilled the oil manifold drain back holes today. Used the lower crank case as a template and just ran the drill through on a press. All the holes in the manifold can be safely enlarged to 9/16", but the one immediately forward of the oil ports on the left side of the engine is VERY close to the seal... probably only 0.020 away...
Tomorrow evening I'm going to get it clean enough to bolt the bottom end together and measure the main bearing ID's with the new bearings. The clearances were too wide with the old bearings so I ordered another set, as well as a 0.010 under set if necessary. Alan Johnson likes to see 0.0023 to 0.0025 clearance on all bearings, main and rod. It would be an absolute shame to turn that carnk though... it's absolutely pristine and within 0.0002 of spec on all journals with basically no out-of-roundness at all... after 100,000 miles.
Sealed Power at one time made rings for a 262 Chevy, but that set has been discontinued.
Hastings, however, makes a 1/16" racing piston ring in that bore size. They also have a standard cast 2nd ring as well. Hastings is the OE supplier for Cadillac oil rings, so I went ahead and ordered their oil rings as well.
Since the Northstar uses 1.5mm rings with 0.135 radial depth, and the Hastings rings are 1/16" with 0.166 radial depth, I have sent the pistons (along with the rings) back to Ross to have the grooves re-cut for the wider, deeper rings.
The '99 and older N*'s use a direct acting hydraulic bucket tappet. The Y2K and newer engines use rocker arms with rollers and the hydraulic lash adjusters in the cylinder head itself. Overall it should be a reduction in valvetrain mass and friction.
I had hoped it would be pretty obvious that different cams would go along with conversion from flat tappet to roller lifter
I just realised it when i was comparing my 97 motor and my "supposed" 99 motor. Apparently its a y2k so I am kind of upset that I cant swap cams if I wanted to
After dorking around with my block all morning, I've decided to have it align honed.
The front four main bores and bearings vary from 0.0002 to 0.0005 out of round. Thanks to an incident with flywheel bolts too long, the rear main is 0.0013 out of round. Numbers 5 and 4 are also tapered about 0.0004.
The clearances vary from 0.0034 to 0.0038, which is WAY too wide. Alan Johnson recommends 0.0023 to 0.0025, which I think is still a bit wide. I'm going to call him and ask him about that.
I could fix the clearance and out of round of #'s 1-4 with 0.001 shim under the lower insert... But #5 is just too hosed to be fixed that way, so I'm going back to the machine shop with it. Might be breaking new ground align boring this engine, as it doesn't have conventional caps. I might end up doing some of the work myself--clamping the lower crank case in a mill and skimming a couple of thou off the top in order for the shop to finish with the boring.
Rings are coming back to me. I will be hand lapping them to get the thickness tolerance where I want it, then the rings and pistons will go to Ron Baxter at Rebco in Kansas to have the grooves cut with .0001" flatness tolerance.
Just need to find a free weekend to spend on setting up the crank for balancing.
Does anyone know how to get in touch with Hank the Crank (HTC)?
[This message has been edited by Will (edited 09-09-2005).]
I was just reminded that we have a surface grinder... I'm going to save what I've learned from this build for the next, and go ahead and use the Ross grooves, in the interest of getting it together sooner and cheaper.
I'm still around... I've just been very busy and haven't been on this forum as much (62 hours on the time card so far this week).
Plans are a changin' Money's available now and I've decided to build this engine so that I won't have a good reason later to say "well, I could have done such and such..."
[This message has been edited by Will (edited 12-08-2005).]
I'm looking at Total Seal diamond lapped rings and CP pistons for the build. I wanted ceramic piston pins but HTC is having some "issues". Anyone else know about those?
Ross pistons will be for sale once I get the CP's in hand.
The VVT Northstars have the same intake ports (300+ CFM with mild porting) as the Y2K engines, except that the VVT heads have exhaust ports enlarged to match. This is enough port flow for over 700 HP N/A, and the VVT can be used to make it streetable at the same time. The potential problem with using a longitudinal engine in a transverse app is the same as for SBC... waterpump clearance. The VVT heads will require welding and machining to work with the transverse block and waterpump.
The transmission will be scratch built transverse 6 speed with sequential shifting.
Contrary to popular belief the Northstar head bolts are really not torque to yield fasteners..... The bolts will stretch very slightly (permanently) but they are good for probably 10 rundowns before any perceptable yield would occur that would render them unusable.
In fact, new bolts are run down and then loosened in the plant in the normal operation tensioning the head bolts. The pre-tensioning step actuall subjects the bolts to more tension then the final tightening step.... This is done to burnish the threads in the block as they have never seen bolts in them before the heads are installed and the head bolts tensioned. The aluminum threads need to be "worked" once before the final tensioning step so the head bolts are run down, loosened and then re-tensioned. If you started with new bolts on your reassembly then you have only done the equivalent of the first pre-tensioning step on the bolts. Use them, they will be fine. Re-using a new bolt that has been run down one time is not the same as re-using a bolt that was in the engine for 100K.......
The instructions to not reuse old head bolts is primarily because the bolts when new have a special microencapsulated coating on the threads and under the head of the bolt. The coatings act as a high pressure lubricant during tensioning and then a thread locker once installed. On a simple run down and loosening without running the engine the bolts can be "used" several times. Once the bolts see a lot of time and thermal cycling in the engine the coatings are rendered unusable again so the bolts have to be replaced as there is no repeatable means or reapplying the special coatings in the field.
If you simply installed the head with a new gasket and new bolts and pulled a timesert out and dissassembled the head the bolts are fine to use again and so is the gasket. If the head gasket was not used in the running engine and subjected to any thermal cycles it is fine. The gasket will compress permanently somewhat when torqued into place....that does not ruin it. I have seen LOTS of head gaskets run down and loosened and re-run down and continued on test fine. As long as the gasket did not stick and tear when dissassembly the gasket is perfectly fine to re-use. It has just been "pre-compressed" much as it is done in the above mentioned pre-tensioning step to condition the head bolt holes. That is done with the gasket in place so the gasket sees the compression and then relaxation in production.
With the compacted graphite gaskets it is sometimes necessary to pre-compress the gasket and even heat it during compression prior to installing it into the engine. So, simply compressing the gasket to the installed load does not hurt or ruin it. If, however, the gasket is held under load and thermal cycled in the engine it will not be reusable. That is because the thermal cycling subjects the gasket to even more load that would cause it to be deformed beyond recovery if relaxed.
At least you can reuse the new gasket and bolts with no concerns.
What's the compressed thickness of the N* head gaskets? It's close to time to tell CP what compression height I want and I would like to set quench appropriately with the stock head gaskets.
Update: per conversation with Kevin at TS, he can get a better ring package into 3.670 bore than 3.667, so the block will be honed again... still only 0.008 over, though.
Sample stock piston is at CP for measuring. I will be going with their "X" style forging for light weight. The only number we really lack is final compression height. To set that I will need to talk to Eagle about rod stretch, get the compressed thickness for the head gaskets and take a measurement AFTER the blok is align honed and (possibly) decked of the final deak height.
Brought the block back from the machine shop. The machinist can't align hone the mains because he isn't equipped to skim a lower crank case, just main caps. I'm going to skim the lower case half, then button it back up and take it back to him for align honing.
Considering having the block decked just to be sure, although I'm reasonably confident it is not required.
Will have to take a weekend to balance the crank, and then another to assemble the engine... AFTER the pistons are ready, of course.
How is that mean? nothing is bulletproof you know, I was just askin because thats alotta tedious work on a motor to miss a shift or something go wrong. Ya know it isn't always human error all the time.
[This message has been edited by 86 FIERO GT (edited 01-11-2006).]
Rotating assembly and valvetrain will both be good to at least 8500. The rev limiter in the factory based chip is in the 6400-6700 range. A 3-2 upshift at 6400 would give me 9500, but I think I'm a good enough driver to not complete a 3-2 upshift.
Besides, you can blow up just about anything by over revving it enough. A journalist even managed to toast a Carerra GT engine by picking out the wrong gear.
Cross sectional area of the beam: 0.240305 in^2 For estimation purposes, I'm going to use 0.240 in^2 in my calculations.
I will calculate piston acceleration at 8500 RPM with 3.307" stroke and 5.943" connecting rods. I will use the mass of the piston, pin, rings and small end of the rod to estimate tensile stress on the rod shank. Using 30-33 million psi as the elastic modulus of 4340 steel, I will estimate rod stretch at my intended redline RPM.
I will use measured block deck height and head gasket compressed thickness to determine how much room I have and tell CP what the compression height of the pistons should be so that I have a VERY tight quench at redline RPM.
It's nice to see someone going the extra mile, instead of just saying, "zero deck pistons, .040 compressed HG, I'm good to go" (which I did, btw). I'm assuming you're going to account for the thermal expansion of the rod and piston, as well as the block and head? Everything's aluminum except the rods, pins, crank, and sleeves (those may not make a difference), so that shouldn't be too difficult. Excellent job! Post up the math, if you don't mind!
With a bit of work on my part, I can tell CP one number that can help my engine make a handful more HP and be more detonation resistant... Of course it will also be more susceptible to over-rev, but we won't think about that...
LS1 block is 319-T5, heads are 356-T6. Supercharged Northstar block and heads are 319-T7... references to the SC lower crank case have not mentioned the alloy. I haven't found a page that specifically calls out the alloy used in the naturally aspirated Northstars, but I will @$$ume that it's 319-T7 just like the SC ones.
However, this site: http://www.key-to-metals.com/Article106.htm mentions that different alloys are preferred for sand vs die casting. The SC block is sand cast, but all lower crank cases and all N/A blocks are die cast... Hmm... I found another page that mentions 380 as being good for die casting and widely used in the automotive industry.
Vega 4 cylinder blocks were 390 aluminum. Hypereutectic aluminum alloys have >12% Silicon.
a(t) = 2 Pi S t for 8500 RPM, S = 141.7 a(t) = 890.1t which is an angular velocity of 890 rad/s
I'm only concerned about acceleration at TDC, so I set t=0 (which means a(t) = 0) so that all the sin(a(t)) terms magically go away and all the cos(a(t)) terms go to 1. What remains is:
A(t)|t=0 = (da(t)/dt)2(-T - T2/R)
da(t)/dt = 890 (da(t)/dt)2 = 792,309.9
So piston acceleration at TDC in a Northstar at 8500 RPM is...
The above is (should be) the *correct* piston acceleration, with rod length taken into account.
This is the "conventional" estimate of piston acceleration taking only stroke and angular velocity into account. P(t) = Tcosa(t)
V(t) = -(da(t)/dt)Tsina(t)
A(t) = -((d2a(t)/dt2)Tsina(t)+(da(t)/dt)2Tcosa(t)
Again, the sine term goes to zero and the cosine term goes to 1 at TDC and we have
A(t)|t=0 = -(da(t)/dt)2T
which from above is 792,309.9 * 1.654 = 1,310,480 in/sec2 or 3,412 g's, which differs significantly (22% low!) from the *correct* number above.
I picked up on the fact that I'd made a sign error by doing a sanity check and noticing that the number taking rod length into account was less than the one without... the opposite should be true. By not taking rod ratio length into account, the equation is implicitly for an infinitely long rod, and lengthening the rod reduces piston acceleration. With the sign error, lengthening the rod increased piston acceleration.
[This message has been edited by Will (edited 02-07-2006).]
With a (estimated) 0.400 kg piston, this comes to 16,924 N of tension on the rod, which is in the neighborhood of 4200#. On a 0.240 in2 cross section, this is a stress of about 17,600 psi.
This obviously neglects the mass of the rings, oil, pin and small end of the conrod, which I will add in later (when I have the balance sheet in front of me).
[This message has been edited by Will (edited 02-07-2006).]
Block deck height is 8.848". With a coefficient of expansion of 1.21E-5/F, the deck height will grow by 0.0139 going from 70*F to 200*F operating temp. I guess I need to figure the same data for the rod, as well... Mental note, look up thermal expansion coefficient for 4340 steel...
Piston weight is TBD but in the 375-400 g range Stock pins weigh 112 g Locks weigh 2 g Rings weigh 30 g Rod small end weighs 158 g
So total weight less the small end is 544 g. This leads to tension of ~5750#. The tension from the mass of the small end can not be added in as easily, because that mass is distributed along the shank of the rod. Thus the rod is under greater tension close to the big end and less close to the small end. I am surprised that the rods are not tapered. The difference in stress leads to a difference in strain and fatigue life across the rod. The small end's 158 g's translates to a strain of 1670# at the big end and 0# at the small end. Thus the average stress through the rod, and the number that should be used in the calculation of total rod stretch is 835#.
So the total rod tension for the purposes of calculating stress & strain is 6585#. Using a 0.240 cross section, this is a stress of 27,437 psi. With elastic modulus numbers between 30 & 33 Mpsi, this translates to a strain of 0.00091 - 0.00083 and elongation of 0.0049-0.0054". Wow. That's less than I expected...
But quench needs to be more than that. We need to add ~0.003-4 for bearing and piston pin clearances, which will drop to microinches with this much tension on things. Crank stretch needs to be taken into account, although that will surely be VERY minimal due to the MUCH larger cross sectional area of the sides of the crank throw. Piston stretch may account for a good chunk, but I'll need to talk to CP about that...
Of course all those numbers are just shank stretch... I have neglected the deformation of the big and small ends, about which it would appear I need to be more concerned.
a(t) = 2 Pi S t for 8500 RPM, S = 141.7 a(t) = 890.1t which is an angular velocity of 890 rad/s
I'm only concerned about acceleration at TDC, so I set t=0 (which means a(t) = 0) so that all the sin(a(t)) terms magically go away and all the cos(a(t)) terms go to 1. What remains is:
A(t)|t=0 = (da(t)/dt)2(-T - T2/R)
da(t)/dt = 890 (da(t)/dt)2 = 792,309.9
So piston acceleration at TDC in a Northstar at 8500 RPM is...
If you want to be really exact don't forget about bearing clearance and the thickness of the oil film. It is not uniform under load.
Yes, that's on my list, as well as oblonging of the big end and small end bores.
quote
Originally posted by Will: But quench needs to be more than that. We need to add ~0.003-4 for bearing and piston pin clearances, which will drop to microinches with this much tension on things. Crank stretch needs to be taken into account, although that will surely be VERY minimal due to the MUCH larger cross sectional area of the sides of the crank throw. Piston stretch may account for a good chunk, but I'll need to talk to CP about that...
quote
Originally posted by Steven Snyder: I checked it up through V(t) but I didnt feel like differentiating any beyond that.. A(t) is ugly. Anyway, P(t) and V(t) are certainly right.
Thanks. Yeah, a triple product derivative nested in a quotient derivative gets ugly fast.
