Home Depot LS Head Bolts? 12ton Dyno Charts of ARP, OEM, L19 & More

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
[Music] welcome back to ttc and another one of our four science episodes this is the humble oem torque to yield cylinder head bolt this being a chevy one from an ls v8 recently we tested various grades of bolts like grade two five and eight for how much impacting beans they can take and many of you asked what about arp well there's probably a few cases where you would need to make the decision between a custom engine bolt and a hardware store one but the silliness of that scenario is right up our alley so today we test as many different head studs and head bolts as we could acquire with the help of some viewers in the form of arp8740 arp2000 a very affordable amazon alternative fancy l19 vel pro and because we can grade 8 to measure the stretch under load and with load up to 24 000 pounds just to complete failure which will in turn also allow us to do some math to compare our results with those huge 200 000 psi or whatever tensile strength claims from each let's hop in this is a drill bit and a rare thread pitch high speed steel tap and here we have gray cast iron slugs we had made for us in the same alloy as a standard gm cast iron block and then this is a block of 6061 t6 aluminum which will act as the cylinder head i think you get the idea tap the block in this case with as much thread as arp told us was in the ls engine block drill through the cylinder head and we have a loosely representative test case the hydraulic punishment for today will be coming from a single acting hollow ram cylinder one that is as short as we could find but still delivers 12 plus tons of force this size allows us to test different stud lengths from ls engines coming up one thing i was curious about is clamping load like the starting off load each bolt or stud applies down onto the head doing the correct torque sequence will give us an idea which this is an oem style tty bolt that's going to be 22 foot-pounds then 90 degrees and then another 90 degrees i say oem style because this is a fel pro replacement cylinder head bolt but we noticed these look identical to the oem gm ones same markings and thread lock applied to them so we're going to compare them as this velcro set is just 43 dollars for both heads while the oem set is 101 but we have a lot of fell pros so we're starting off briefly with that this gauge is in psi given the surface area of the ram you can multiply this by 2.77 if you want to know pounds we're seeing 2600 psi once torqued which settles down to 25 50 or around 7 000 pounds of clamping force when torqued we're setting up a dial indicator here to measure bolt stretch but on this first test we're mounting it sort of in the incorrect location for a reason first we test with a hand manual hydraulic pump to reach around 6 500 psi which is its limit or 18 000 pounds as you can see here the bolt is not showing stretch but negative dial movement it's being sucked down two thousands because the dial indicator's base is mounted to the moving in this case moving up cylinder head but we theorize this sucking down or cylinder head i don't know compressing under the bolt might happen and we don't want to measure that but no way to see all of it or if it increases until we bring it to failure with the help of our compressed air powered hydraulic pump so the most this one ever saw is 6500 the whole cylinder head and dial indicator took quite a ride now for the oem bolts same torquing procedure but we'll be moving the dial indicator remote of the whole assembly then zeroing it this gauge looks awfully snug up against that head but you can pass paper between these two so the oem tty bolt clamps down with about three thousand it goes down to 2940 psi 400 up from the fel pro maybe they're different torque to heal bolts aren't known for bringing the most consistent or highest clamping loads though these style bolts you torque them with a degree of rotation until they yield or basically see plastic deformation giving them a nice wide window of it's probably this much tight but also can't be used again because they are past being elastically deformed or going back to their original shape and length but when we tested bolt grades you could see this shape play out in real time the climb up is a bolt that is happy and can be reused that cresting over meant you're increasing output without getting much more of that tension or tightening plastic deformation that's it yielding basically on a chart which you'll be seeing soon if force is on the bottom and stretches on the y-axis a perfect bolt or head stud would do this if you can appreciate this piece of art i'm creating for you stretch very little as forces applied then slowly and predictably yield at the end a bad bolt would stretch quite a bit early then fail suddenly lifting your head suddenly as well and a high performance stud would do something in between perhaps stretch a bit slower but mainly in its failure zone would be further out and not off a cliff in the first stage of our testing we're going to be looking at 6 500 psi 18 000 pounds based on some over simplified math we did on the ls piston diameter and number of bolts basically that's cylinder pressures of pretty spicy boosted engines then we're just going to use compressed air pump to blow these apart and find out their ultimate tensile strength the oem gm bolt starts at zero at its torqued clamping load then climbs up at a good clip not really comparing it to anything yet but accelerates quite a bit at the end 35 thousandths which is going for broke you'll notice that the dial indicator is measuring just behind the bolt now since we want to know how much just the cylinder head itself raises whereas the bolt itself measured differently for us we also hope this might save the dial indicator from an early demise so with the air pump hydraulically doing it what it does we see a lot more stretch and basically it keeps stretching and fails but the most it ever saw was 6600 and yeah we're going to be going through some of those dial indicators we already know the failure point of the velcro so here's what it does stretched when we measured it a more gradual line for sure but then seems to be equally unhappy past 6k which is confirmed here by bumping things up quite a bit more seeing plenty of stretch 43 thousands here on the leaderboard we can see their stats there the oem total stretch needed to subtract a starting value we didn't zero at four but here's that comparison once we did the math they're close here and here and then sort of close here we'll have to see how that boils down versus the others with you familiar with the test setup we can just start throwing some bolts at you here's the pqy racing head stud that we got off of amazon for sixty dollars a set is one of the cheaper higher reviewed sets and one of the few ones with an actual psi tensile strength rating 220 000 in this case so basically saying they are an arp 2000 type stud for less than oem replacement torque to yield bolts no instructions came with these but reviewers of these bolts said they just followed the arp torquing procedures so we're going to do that here and it's pretty clear that we're going to need a lot more dial indicators to sacrifice the science gods i don't enjoy killing precision tools but it looks like we're going to be in for a wild ride no matter how we do this on all these studs today we're going to be using arp's thread lube to keep the torque delivery consistent the hardware on the pqia racing feels lightweight and thin we'll have to see if that makes for some decent head clamping the benefits of a stud over a bolt are theoretically that the nut is the only thing spinning not the whole length of bolt so no lost torsion transmission and extra torque being spent on that and also the bolt head itself doesn't need to turn against the cylinder head it has its own theoretically equal hardness nut and lube to work against plus you can reuse it as it's not torqued to yield nothing should be getting yielded in normal use so let's see looks like 22 thousandths definitely climbing up at the end but less sharply than stock and here's its fatality point 7621 and another dial indicator sadly meets its end and at that 7621 that was 52 000 total not bad though this performance especially for the price we'll be doing some math to figure out how this compares to that advertised psi of tensile strength coming up next up in our long bolts category we're going to be jumping right up to arp2000 this is an arp bolt not a stud still reusable as it's not torque to yield cheaper than studs so we're curious how it does higher torque procedure than erp's 8740 variety which we couldn't source in bolt form at this time because of the national shortage going on we got all of our torque procedures directly from arp when we called them to order parts and this one clamps down to a massive 4000 psi 11 000 pounds when bumping it up to 6500 psi manually to stretch it its curve stays quite low the entire duration maxing out at 17 and a half foul hour graphic here rounding that to 18. definitely the best so far here's this max beans and that's how it yielded its stretch spiked up to the point of it being no longer happy but we weren't able to send this one off into outer space so that's seventeen and a half thou nine thousand one hundred and forty psi plus since we didn't completely bugger it and sixty four thousands at that ninety one forty that's very good and the best so far now for some really fancy stuff l19 l19 is a 260 000 psi tensile strength stud material it compares directly to arp's custom mage 625 plus studs that are also custom order and several months out so we bought these l19 studs from speed factory basically same length same body diameter and has 11 millimeter threads like the ls bolts but made for hondas these as a comparable set as in like the number of bolts you need to build an ls would be a 628 set an arp custom age set is 820. though those arp studs are ones you can actually touch with your hands these l19 studs are so fancy you're not allowed to touch them with your hands moisture from your hands can cause hydrogen and brittleman i guess so with the stud nut and washer and torque sequence from speedfactory we're left with about 3600 psi let's see if they stretch [Music] and they stretch but not a lot similar to the arp 2000 so far 18 thousands and here's maxing out our ram 9529 psi and it's not jumping huge amounts of stretch with each increase 51 000 total when you look at that amount of stretch at that pressure level that's 26 400 pounds of force exceeding our rams max and still just less stretch than these that's pretty stout not going to be lifting heads on these without a ton of heat and boost and maybe some pre-detonation skipping grade 8 for a moment we want to test some arp stuff some studs and shorter ones at that later model gen 4 ls's use shorter bolts still same m11 size threads but around 4 inches rather than around 6 inches also it's some of the only ones we could source and borrow in the flavors we wanted first up another pqi racing stud this time is just their shorter variety for gen 4 ls engines is shorter actually better a debate that's is oldest time i'm sure this is why we needed a ram as small and as short as this hollow one we got the cylinder head has shrunk quite a bit now to fit these shorter head studs just this bit of billet aluminum the amazon stud sees 3650 psi worth of clamping 250 up from the long stud but also as close as any two we've tested so far here's how they compare to each other in stretch and here's it to ultimate failure a whole lot of stretch on this one as it passes thousand and ouchy that's seven thousand nine hundred sixty peak it saw and sixty two thousands of stretch not exactly way up on the long stud from the same place more stretch from getting to higher numbers they both lift ahead probably around the same time i feel and now for a new arp alloy 8740 their bread and butter one of the most popular types 170 000 psi we have this one in our arp 2000 stud coming up thanks to michigan motorsports who answered the call when several of our orders were on multiple months worth of back orders and offered to send these to us appreciate that our nuts and washers were ordered directly from arp and they are much more substantial feeling in finish weight and size wise over the amazon pqi ones and they take a larger socket for it as well but that comes with a 338 dollar price tag as well let's see with the suggested torque sequence it sees 2 900 psi clamping force not altogether very high when stretched up to 6500 psi it matches the amazon for a bit but then does quite well and not fall off a cliff at the end just 17 and a half thousands we see again the young's modulus between the alloys when not yielded should be fairly similar after all so it shows that this one is not yielding yet this 170 000 psi rated stud doing better than the quote unquote 220 000 psi pqi ones so far under max beans the 8740 stud sees a similar fate though at 8000 psi the stretch figures spike without the pressure side doing so and that's at yielding failing at a max of 8068 which looks like this so the amazon pqi ones start to yield sooner but their ultimate strength is pretty comparable between these the pqi more like an 8740 than an arp 2000 if you had to pick one for sure but if saving money is your aim maybe try arp bolts instead of studs same alloy but 173 dollars a set instead of 338 let's take a look at the 8740 bolt we have thanks to our viewer patrick daly 2620 psi clamping so a bit under the stud and similar to just regular oem tty bolts now the 8740 bolt just stretches quite a bit and early it just barely beats the oem stock one at the end of this test but not a stellar performance compared to its stud twin of the same alloy the bolt ultimately can quickly get up to 7 000 7 500 or so just piles on the thousands from there though is thoroughly yielded at that point and not making any more so that gives us the opportunity to switch out cameras point 4 000 lumens at this thing to get some high speed footage of that carnage this is at 1 500 frames per second or what will be called exhibit a in my future trial of crimes against dial indicators blink and you'll miss it you have to artificially slow this down to 5000 frames per second to really see its acceleration which is hard to measure but seems to likely be over 100 miles per hour it's like beautiful terrifying and disgusting all at the same time okay arp 2000 stud for a gen 4 ls one of the things we wanted to see the most but also about 400 a set of these studs in borderline impossible to find in stock these days so we saw around 3300 starting clamping load and its stretch got up to 18 thousandths a familiar level we've been seeing among bolts and studs that are not starting to yield at this level the arp2000 stud took a lot of air pressure and hydraulic psi before seeing some serious stretch and top things out over 9000 before activating the ejection seat and here's what that looks like 18th house stretch at 6 500 psi 54 all the way up to 9233 psi that puts it above all of these matching maybe slightly besting the arp 2000 bolt and only being passed by the l19 stud which makes sense that's sort of how these things are advertised and sold to you but good to see that difference and what it looks like and maybe indicates our setup isn't totally worthless i guess all right so grade eight you find yourself in the odd situation where you need head bolts like right now the dealership doesn't have them and you can find the correct thread pitching length at your hardware store will grade 8 or class 10.9 equivalent such as up here do the job let's find out we started out at around 2600 psi to match the oem ones since there's really no procedure for this the grade 8 matches the oem trajectory for quite a while but finishes this first step under the factory class 10.9 ttypo not too bad but it was under max torture that we saw the real difference this bolt goes to around 7000 then just spun and spun the dial indicator stretching to infinity before letting go [Music] [Applause] [Music] you could basically slide some full-size washers between the head gasket and the block before this thing let go plenty of water in those cylinders now so without pre-yielding the bolt i guess you see some serious stretch and no way to know how to torque a grade eight until it's in that goldilocks zone of stretch but not failing when you tighten it so maybe stick to oem stuff despite the small advantage you see here these grade eight and class 10.9 bolts come with some inherent tensile strength ratings well all these bolts advertise them so let's see them using this peak psi converted to pounds at the ram versus the cross section of the minor diameter of their thread pitches we get these values the grade 8 rated 450k saw 157 makes some sense these two fairly close as well but we're also pre-yielded too if that makes a difference then the pqi racing definitely not 220 000 psi but pretty much 170 000 psi yeah and around the same as an 8740 hardware down here we saw 200k and over that from arp 2000 stuff but we did max out the ram here though with plenty of stretch happening in the meantime the l19 seeing the highest around 230 000 and did not meet its max it could probably keep going for the price of even the amount of grade eight bolts you would need to build a cylinder head pqy racing not bad perform better than the stock felpro ones nearly two or about the same as arp 8740 levels for 60 bucks and you can reuse them we found when reusing them you see an additional five to seven thousand stretch once torqued to normal levels but otherwise work the same it's elastically deforming it's not permanent but when we're talking performance 8740 studs seem to be not the huge increase for the money that we were expecting arp 2000 bolts are cheaper quite a bit higher performance may not deliver as consistent clamping loads when torqued as studs but that resistance to lift seems to be there we would have loved to test other brands as well in this category but nothing was really available individually and air p2000 studs and l19 studs that's some serious stuff most of the time we were not able to blow these up into pieces with around 26 000 pounds 11 800 kilograms and you'd have 10 of this size to hold down each head plus some other m8 ones now that we have all these very not happy simulation engine block threads maybe we'll do a thread inserts challenge on one of our next episodes let us know if you'd like to see that appreciate you joining us for this one hopefully you found it informative or at least entertaining click subscribe to see stuff like this every friday and thanks for watching [Music] you
Info
Channel: Torque Test Channel
Views: 291,989
Rating: undefined out of 5
Keywords:
Id: VlfVt1wqr40
Channel Id: undefined
Length: 21min 34sec (1294 seconds)
Published: Fri Jun 17 2022
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.