X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Sender: To: lml@lancaironline.net Date: Fri, 06 Oct 2006 23:17:07 -0400 Message-ID: X-Original-Return-Path: Received: from mta9.adelphia.net ([68.168.78.199] verified) by logan.com (CommuniGate Pro SMTP 5.1c.5) with ESMTP id 1445919 for lml@lancaironline.net; Fri, 06 Oct 2006 09:34:19 -0400 Received-SPF: pass receiver=logan.com; client-ip=68.168.78.199; envelope-from=glcasey@adelphia.net Received: from [70.32.170.45] by mta9.adelphia.net (InterMail vM.6.01.05.02 201-2131-123-102-20050715) with ESMTP id <20061006133347.DKIT23918.mta9.adelphia.net@[70.32.170.45]> for ; Fri, 6 Oct 2006 09:33:47 -0400 Mime-Version: 1.0 (Apple Message framework v752.2) In-Reply-To: References: Content-Type: multipart/alternative; boundary=Apple-Mail-6--701683516 X-Original-Message-Id: <30385816-356E-4F76-81DF-9E84155567E9@adelphia.net> From: Gary Casey Subject: Re: Torque specs X-Original-Date: Fri, 6 Oct 2006 06:33:46 -0700 X-Original-To: "Lancair Mailing List" X-Mailer: Apple Mail (2.752.2) --Apple-Mail-6--701683516 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed I generally agree with your comments, but there are engineers out there that have spent their careers working with fastener technology and one of them would certainly be more qualified to comment than I. One could argue that while the torque from the locking feature creates an error it could be a small one compared to the torque spec, although in my experience using typical published torque values on locking nuts will result in a significant loss of clamp force. Normally the torque spec is there to ensure an adequate clamping load while avoiding overstressing the bolt. However, this is all about "normal" joints where the bolt is the most elastic element in the joint. Standard bolting methodology designs the bolted joint so that once the bolt is torqued and "stretched" its stress remains constant and the bolt won't fail from fatigue. Imagine wrapping a rubber band around to parts - as long as you don't pull on them hard enough to separate them the rubber band doesn't see the load. A lot (most?) of the joints on the Lancair don't follow this methodology - the bolt clamps the composite structure, which is much lower in stiffness than the bolt. Consequently, when the engine pulls on the mount the firewall material flexes and the bolt absorbs the variable load. This increases the fatigue (variable) loading on the bolt and makes it more likely that the nut could come loose. Solution? Use oversize bolts so that the stress in the bolt is very low. My load calculations on the firewall joint imply that 1/4-inch bolts would probably be adequate to hold the static load, so the strength of the existing bolts is far in excess of what would be required, a very good thing. The only way to ensure the nuts don't come loose is then to use some sort of lock. A cotter pin would work, but would be very difficult to install. The temperature of the bolt doesn't get high enough to require a metal locknut so the plastic ("Nylock") nuts are a reasonable choice. How to torque them? As good a method as any might be to tightened until the toque goes up, indicating that the joint is tight, and then continue for a certain angle after that. I do it by feel - not very scientific - but the torque usually ends up higher that what the handbook would indicate. After all that discussion, I'm with you - I'd like better guidance as to the proper torque. Gary Casey On Oct 6, 2006, at 3:02 AM, Lancair Mailing List wrote: > The recent discussions about proper torque on such things as engine > mount bolts reminds me that over the years I have heard from > several normally reliable sources that using a torque wrench on > locking nuts (either plastic or all metal) is not only a waste of > time but possibly even a bad idea. The logic is that because of > the built in interference of the nut either a lower than desired > reading will be read or upon approaching the desired torque the nut > will "slip" and result in higher than desired torque. It's been > pointed out that truly torque-critical items, like connecting rod > bolts and case half connecting bolts never use locking nuts. In > other words, if the torque is really important don't use lock > nuts. I am curious what others on the list might have to say about > this. I am not an engineer, nor do I play one on TV, but I've been > using nuts and bolts since way back in the Erector Set days and > have absorbed "do" and "don't do" advice from people whose > expertise I respect, including the above observations on torquing > fasteners. I probably won't change what's worked well for me over > the decades but perhaps someone has some compelling argument to the > contrary, or perhaps confirm what I've heard. The floor is open. > > Leighton Mangels --Apple-Mail-6--701683516 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=ISO-8859-1 I generally agree with your = comments, but there are engineers out there that have spent their = careers working with fastener technology and one of them would certainly = be more qualified to comment than I.=A0 One could argue that while the = torque from the locking feature creates an error it could be a small one = compared to the torque spec, although in my experience using typical = published torque values on locking nuts will result in a significant = loss of clamp force.=A0 Normally the torque spec is there to ensure an = adequate clamping load while avoiding overstressing the bolt.=A0 = However, this is all about "normal" joints where the bolt is the most = elastic element in the joint.=A0 Standard bolting methodology designs = the bolted joint so that once the bolt is torqued and "stretched" its = stress remains constant and the bolt won't fail from fatigue.=A0 Imagine = wrapping a rubber band around to parts - as long as you don't pull on = them hard enough to separate them the rubber band doesn't see the load.=A0= A lot (most?) of the joints on the Lancair don't follow this = methodology - the bolt clamps the composite structure, which is much = lower in stiffness than the bolt.=A0 Consequently, when the engine pulls = on the mount the firewall material flexes and the bolt absorbs the = variable load.=A0 This increases the fatigue (variable) loading on the = bolt and makes it more likely that the nut could come loose.=A0 = Solution?=A0 Use oversize bolts so that the stress in the bolt is very = low.=A0 My load calculations on the firewall joint imply that 1/4-inch = bolts would probably be adequate to hold the static load, so the = strength of the existing bolts is far in excess of what would be = required, a very good thing.=A0 The only way to ensure the nuts don't = come loose is then to use some sort of lock.=A0 A cotter pin would work, = but would be very difficult to install.=A0 The temperature of the bolt = doesn't get high enough to require a metal locknut so the plastic = ("Nylock") nuts are a reasonable choice.=A0 How to torque them?=A0 As = good a method as any might be to tightened until the toque goes up, = indicating that the joint is tight, and then continue for a certain = angle after that.=A0 I do it by feel - not very scientific - but the = torque usually ends up higher that what the handbook would indicate.=A0 = After all that discussion, I'm with you - I'd like better guidance as to = the proper torque.=A0

Gary = Casey

On Oct 6, 2006, at 3:02 AM, Lancair = Mailing List wrote:

=A0
Leighton = Mangels

= --Apple-Mail-6--701683516--