Matronics Email Lists :: View topic

Matronics Email Lists
Web Forum Interface to the Matronics Email Lists

Get Email Distribution Too! FAQ Search Memberlist Usergroups Register

Profile Log in to check your private messages Log in

More torque wrench
Goto page 1, 2 Next

Matronics Email Lists Forum Index -> Zenith-List

View previous topic :: View next topic

Author Message

zsmith3rd(at)earthlink.ne
Guest

Posted: Wed Jul 25, 2007 9:57 am Post subject: More torque wrench

do not archive



Mention was made earlier as to accuracy of various torque wrenches.



Some years ago a field mechanic for Caterpillar told me that their wrenches were tested occasionally.

Said they had, and most larger Cat distributors also had, a real test fixture for this purpose.



When they were in the field and were in doubt, or just wanted to check one, they clamped a known good one in a vise and used a double-female very short "extension" to attach the unknown to the known and pulled on the free handle until the known read full scale....then checked the reading on the suspect wrench.



Said they could check at any point across the range of a given wrench.



This method may not be suitable for NASA, but he swears it works very well for checking (not calibrating), provided a known "good" wrench is used.



Let the flames begin.



Regards to all,



Zed/701/R912/90+%/etc/do not archive.

ashontz

Joined: 27 Dec 2006
Posts: 723

Posted: Wed Jul 25, 2007 1:28 pm Post subject: Re: More torque wrench

How did they know the good one was good? I guess a better method would be to clamp the torque wrench perpendicular to the floor and then add a given weight (say 80 lbs) to the handle of the torque wrench at the 1 foot mark on the handle from the center of the socket fitting and see that the wrench reads 80ft-lbs. That's probably how NASA would do it. Actually, NASA would do it that way, but it would involve 14 engineers each making $180,000/yr, 2 project plans, 4 revisions, a committee meeting, a follow up meeting, a meeting with the dept of weights and measures, a $1,200 dinner and drinks working meeting with the head of the dept of weights and measures, another project revision, and finally wrapped up with acceptance testing, only to find out they did it in metric rather than US English measurements.



 	  zsmith3rd(at)earthlink.ne wrote: 	
		  do not archive



Mention was made earlier as to accuracy of various torque wrenches.



Some years ago a field mechanic for Caterpillar told me that their wrenches were tested occasionally.

Said they had, and most larger Cat distributors also had, a real test fixture for this purpose.



When they were in the field and were in doubt, or just wanted to check one, they clamped a known good one in a vise and used a double-female very short "extension" to attach the unknown to the known and pulled on the free handle until the known read full scale....then checked the reading on the suspect wrench.



Said they could check at any point across the range of a given wrench.



This method may not be suitable for NASA, but he swears it works very well for checking (not calibrating), provided a known "good" wrench is used.



Let the flames begin.



Regards to all,



Zed/701/R912/90+%/etc/do not archive.

planecrazydld(at)yahoo.co
Guest

Posted: Thu Jul 26, 2007 2:26 am Post subject: More torque wrench

no reason for flames. Dead weight testing is one of the most stable methods known (as long as the reference is accurate). Just make sure that the socket is not off perpendicular and that there is no applied force to keep it that way.



Zed Smith <zsmith3rd(at)earthlink.net> wrote:[quote] --> Zenith-List message posted by: Zed Smith 



do not archive



Mention was made earlier as to accuracy of various torque wrenches.



Some years ago a field mechanic for Caterpillar told me that their wrenches were tested occasionally.

Said they had, and most larger Cat distributors also had, a real test fixture for this purpose.



When they were in the field and were in doubt, or just wanted to check one, they clamped a known good one in a vise and used a double-female very short "extension" to attach the unknown to amazing places on Yahoo! Travel.  [quote][b]

ashontz

Joined: 27 Dec 2006
Posts: 723

Posted: Thu Jul 26, 2007 4:17 am Post subject: Re: More torque wrench

No flames. Actually, the metric vs English thing was an actual incident that occurred and caused them to lose a 150 million dollar Mars lander a few years back. I'm sure the project plan stuff is pretty accurate too. LOL



do not archive



[quote="planecrazydld(at)yahoo.co"]no reason for flames. Dead weight testing is one of the most stable methods known (as long as the reference is accurate). Just make sure that the socket is not off perpendicular and that there is no applied force to keep it that way.



Zed Smith <zsmith3rd> wrote:[quote] --> Zenith-List message posted by: Zed Smith 



do not archive



Mention was made earlier as to accuracy of various torque wrenches.



Some years ago a field mechanic for Caterpillar told me that their wrenches were tested occasionally.

Said they had, and most larger Cat distributors also had, a real test fixture for this purpose.



When they were in the field and were in doubt, or just wanted to check one, they clamped a known good one in a vise and used a double-female very short "extension" to attach the unknown to amazing places on Yahoo! Travel.   	  Quote: 	
		  [b]

DanielBK

Joined: 26 Jul 2007
Posts: 2

Posted: Thu Jul 26, 2007 8:06 am Post subject: Re: More torque wrench

 	  ashontz wrote: 	
		  ... I guess a better method would be to clamp the torque wrench perpendicular to the floor and then add a given weight (say 80 lbs) to the handle of the torque wrench at the 1 foot mark on the handle from the center of the socket fitting and see that the wrench reads 80ft-lbs. ...

	




Actually, this would be a terrible method that is commonly thought to be a good method.  Think vectors.  Or for a practical example, think of two extremes, then deduce the situation between extremes.  For the example, we'll use an accurate and new wrench, fresh from the factory calibration bench.  We'll use a special wrench that allows 90° of movement between zero and max torque for no better reason than it makes for an easier demonstration of principle.



If we begin with the setup suggested, with the business end in the vice, everything true, level and orthogonal, the handle horizontal, and no weight attached, we get an accurate reading of zero lb-ft (not "ft-lb", a common error).  That's the first extreme.



Next, apply a whole bunch of weight, such that the handle swings down to a near vertical position.  The reading on the wrench will be considerably less than the weight applied.  Why so?  Because the weight is pulling straight down, not tangent to the arc of the handle swing.  Only force applied on a tangent to the arc (perpendicular to the handle shaft) produces torque.  We lost that perpendicularity with the very first movement of the handle, thus when we reach this near vertical condition, most of the weight is simply pulling straight out on the handle.  That is the second extreme.



It's self evident that the error produced by this method is minor at first and escalates the further we swing the handle.  A tip though:  The progression is not linear.



Three solutions come to mind that would allow us to use a dead weight test.  One would be to dust off a geometry text and compute the effect at any angular displacement.  Sines, cosines, arctangents, etc.  If that makes us groan, an alternative would be to simply draw the vectors (lines showing direction and scale) and measure the resultant.  The third solution would be the dirty fingernails approach (perfectly appropriate):  Build our clamping rig rig such that we clamp the wrench as before,on the socket stud, apply the known weight, then rotate the clamp and stud to the degree that the handle is once again horizontal.  Then read the torque scale on the wrench.  The weight is applied perpendicular to the handle in that situation & thus is all used to produce torque.  For a clicker wrench, simply perform the process twice, once with just a tad less weight than the sought after torque (no click) and once with just a tad more (click).



A final caveat.  These tests only tell us if the wrench is accurate at the tested data point.  If we want to know if the wrench is accurate over the full range, then we must perform a full range of tests.  A wrench is commonly spot on at two points, one low and one high, with a negligible error in between and beyond each of those two points.  The thing is, we don't know if that's the case without testing throughout the range.

_________________
Daniel

ashontz

Joined: 27 Dec 2006
Posts: 723

Posted: Thu Jul 26, 2007 8:10 am Post subject: Re: More torque wrench

Why would it swing down to a near vertical position, it's clamped in the vise. The only movement you'll see is about 1 or 2 degrees as indicated on the torque wrench scale. If you want to get really accurate you could also factor in the 1 or 2 degree movement but it's negligible. This is a static test, not a dynamic test. Not the same as trying to measure the torque of a running engine, which is actually done the same way really, you just measure the perpendicular force at a certain distance and the brake (be it water or asbetos brake pads) adds the force to that arm.



 	  DanielBK wrote: 	
		   	  ashontz wrote: 	
		  ... I guess a better method would be to clamp the torque wrench perpendicular to the floor and then add a given weight (say 80 lbs) to the handle of the torque wrench at the 1 foot mark on the handle from the center of the socket fitting and see that the wrench reads 80ft-lbs. ...

	




Actually, this would be a terrible method that is commonly thought to be a good method.  Think vectors.  Or for a practical example, think of two extremes, then deduce the situation between extremes.  For the example, we'll use an accurate and new wrench, fresh from the factory calibration bench.  We'll use a special wrench that allows 90° of movement between zero and max torque for no better reason than it makes for an easier demonstration of principle.



If we begin with the setup suggested, with the business end in the vice, everything true, level and orthogonal, the handle horizontal, and no weight attached, we get an accurate reading of zero lb-ft (not "ft-lb", a common error).  That's the first extreme.



Next, apply a whole bunch of weight, such that the handle swings down to a near vertical position.  The reading on the wrench will be considerably less than the weight applied.  Why so?  Because the weight is pulling straight down, not tangent to the arc of the handle swing.  Only force applied on a tangent to the arc (perpendicular to the handle shaft) produces torque.  We lost that perpendicularity with the very first movement of the handle, thus when we reach this near vertical condition, most of the weight is simply pulling straight out on the handle.  That is the second extreme.



It's self evident that the error produced by this method is minor at first and escalates the further we swing the handle.  A tip though:  The progression is not linear.



Three solutions come to mind that would allow us to use a dead weight test.  One would be to dust off a geometry text and compute the effect at any angular displacement.  Sines, cosines, arctangents, etc.  If that makes us groan, an alternative would be to simply draw the vectors (lines showing direction and scale) and measure the resultant.  The third solution would be the dirty fingernails approach (perfectly appropriate):  Build our clamping rig rig such that we clamp the wrench as before,on the socket stud, apply the known weight, then rotate the clamp and stud to the degree that the handle is once again horizontal.  Then read the torque scale on the wrench.  The weight is applied perpendicular to the handle in that situation & thus is all used to produce torque.  For a clicker wrench, simply perform the process twice, once with just a tad less weight than the sought after torque (no click) and once with just a tad more (click).



A final caveat.  These tests only tell us if the wrench is accurate at the tested data point.  If we want to know if the wrench is accurate over the full range, then we must perform a full range of tests.  A wrench is commonly spot on at two points, one low and one high, with a negligible error in between and beyond each of those two points.  The thing is, we don't know if that's the case without testing throughout the range.

purplemoon99(at)bellsouth
Guest

Posted: Thu Jul 26, 2007 3:25 pm Post subject: More torque wrench

I wonder if WW would agree with you? 

Joe

---

daveaustin2(at)primus.ca
Guest

Posted: Thu Jul 26, 2007 4:03 pm Post subject: More torque wrench

Are the torque specs so close that it is necessary to go to the extremes you 

folks have written of?  How about torqueing up bolts that, by definition, 

use castle nuts?  Do you go up in torque to the next slot, or down to the 

previous slot?

Dave Austin  601HDS - 912, Spitfire Mk VIII

larry(at)macsmachine.com
Guest

Posted: Thu Jul 26, 2007 4:36 pm Post subject: More torque wrench

Dave,

If you're talking propeller hub bolts, I don't use castle nuts. I use 

Nylock nuts to avoid over-tightening.  Otherwise you'll be fighting this 

battle forever. They are acceptable for metal props and don't come 

loose.  I don't go past the sound of the click on the torque wrench and 

wouldn't pass the indicator of that type.



If you're talking wood props, you need a thick metal faceplate, thin 

washers, castle nuts and better guidance from the mfr of the prop. 



Larry McFarland 601HDS at www.macsmachine.com



Dave Austin wrote:

 	  Quote: 	
		   



 Are the torque specs so close that it is necessary to go to the 

 extremes you folks have written of?  How about torqueing up bolts 

 that, by definition, use castle nuts?  Do you go up in torque to the 

 next slot, or down to the previous slot?

 Dave Austin  601HDS - 912, Spitfire Mk VIII

planecrazydld(at)yahoo.co
Guest

Posted: Thu Jul 26, 2007 5:02 pm Post subject: More torque wrench

to amazing places on Yahoo! Travel. [quote][b]

planecrazydld(at)yahoo.co
Guest

Posted: Fri Jul 27, 2007 2:09 am Post subject: More torque wrench

[quote][b]

Jaybannist(at)cs.com
Guest

Posted: Fri Jul 27, 2007 5:25 am Post subject: More torque wrench

Dave,



I don't know what you do differently, but about half the time I click on your message in the forum, it is blank.  Sometimes, like today, I go to reply, and your message comes up in the reply box.  Strange.



Jay

David Downey <planecrazydld(at)yahoo.com> wrote:



 	  Quote: 	
		  torque to the low limit and tighten to the next slot/hole combination. If the bolt has 4 holes and the nut has 3 slots it is very easy to make the middle of the range. If the bolt has 1 hole and the nut has 3 slots you may need an AN960 "L" washer to adjust the length under the nut.

David Downey <planecrazydld(at)yahoo.com> wrote: to amazing places on Yahoo! Travel.  



  Dave Downey

  Harleysville (SE) PA

  100 HP Corvair

       

---------------------------------

Moody friends. Drama queens. Your life? Nope! - their life, your story.

dkbrooks(at)gmail.com
Guest

Posted: Fri Jul 27, 2007 5:37 am Post subject: More torque wrench

Dave, Jay - 

  

 Dave's messages come up blank for me as well but I don't see the message when I hit Reply.  Kind of a bummer too, when I do see his messages quoted in replies they look pretty informative and useful.

  

 I am reading the messages in gmail, in Firefox 2.0.0.5 or IE7 on XP Pro.  If that matters.  

  

 Dave

  

  

 do not archive



 

 On 7/27/07, Jaybannist(at)cs.com (Jaybannist(at)cs.com) <Jaybannist(at)cs.com (Jaybannist(at)cs.com)> wrote: [quote]--> Zenith-List message posted by: Jaybannist(at)cs.com (Jaybannist(at)cs.com)



Dave, 



I don't know what you do differently, but about half the time I click on your message in the forum, it is blank.  Sometimes, like today, I go to reply, and your message comes up in the reply box.  Strange.

 

Jay

David Downey <planecrazydld(at)yahoo.com (planecrazydld(at)yahoo.com)> wrote:



 	  Quote: 	
		  torque to the low limit and tighten to the next slot/hole combination. If the bolt has 4 holes and the nut has 3 slots it is very easy to make the middle of the range. If the bolt has 1 hole and the nut has 3 slots you may need an AN960 "L" washer to adjust the length under the nut. 

David Downey <planecrazydld(at)yahoo.com (planecrazydld(at)yahoo.com)> wrote: to amazing places on Yahoo! Travel.



  Dave Downey

  Harleysville (SE) PA 


	


[b]

carlossa52(at)gmail.com
Guest

Posted: Fri Jul 27, 2007 5:50 am Post subject: More torque wrench

Same happens to me, but Dave is not the only source of apparently empty messages.

I am using gmail, and can see the msg if I select "show original".

My environment is similar: XP Pro, Firefox  2.0.0.5 - but I think this is independent from the reader environment. More likely it is a Matronics glitch: This did not happen before HTML was allowed.



Carlos

CH601-HD, plans



 do not archive



On 27/07/07, David Brooks <dkbrooks(at)gmail.com (dkbrooks(at)gmail.com)> wrote:[quote] Dave, Jay - 

  

 Dave's messages come up blank for me as well but I don't see the message when I hit Reply.  Kind of a bummer too, when I do see his messages quoted in replies they look pretty informative and useful.

  

 I am reading the messages in gmail, in Firefox 2.0.0.5 or IE7 on XP Pro.  If that matters.  

  

 Dave

[b]

daveaustin2(at)primus.ca
Guest

Posted: Fri Jul 27, 2007 6:23 am Post subject: More torque wrench

Carlos,

 Which "Dave" is it you can't get the msg body  from?  There are too many of us!

 Dave Austin  601HDS - 912, Spitfire Mk  VIII

 Do not archive

    [quote][b]

a.f.rupp(at)ATT.NET
Guest

Posted: Fri Jul 27, 2007 6:30 am Post subject: More torque wrench

To All, 

Do not forget that the main reason for torquing is to get the same stretch on all the fasteners so no one of them is over loaded. Haven't seen that mentioned in the discussion. 

Al Rupp 

601XL done awaiting engine. 

Mark wrote

 If you're talking propeller hub bolts, I don't use castle nuts. I use 

 Nylock nuts to avoid over-tightening. 

 Dave wrote: 

 Are the torque specs so close that it is necessary to go to the 

 extremes you folks have written of? How about torqueing up bolts 

 that, by definition, use castle nuts? Do you go up in torque to the 

 next slot, or down to the previous slot? 

      [quote][b]

dkbrooks(at)gmail.com
Guest

Posted: Fri Jul 27, 2007 6:31 am Post subject: More torque wrench

I can't read David Downey's messages (although Carlos's suggestion of using "Show Original" in gmail does work).  

  

 Dave (Brooks)

 



 

 On 7/27/07, Dave Austin <daveaustin2(at)primus.ca (daveaustin2(at)primus.ca)> wrote: [quote]  Carlos,

 Which "Dave" is it you can't get the msg body from?  There are too many of us!

 Dave Austin  601HDS - 912, Spitfire Mk VIII

 Do not archive

 	  Quote: 	
		  

	


[b]

planecrazydld(at)yahoo.co
Guest

Posted: Fri Jul 27, 2007 7:38 am Post subject: More torque wrench

I know Jay! that is why sometimes you see a posting from me that is blank then another with the content pasted in from the sent file. I do not know what causes this.

   

  do not archive



Jaybannist(at)cs.com wrote:

  [quote]--> Zenith-List message posted by: Jaybannist(at)cs.com



Dave,



I don't know what you do differently, but about half the time I click on your message in the forum, it is blank. Sometimes, like today, I go to reply, and your message comes up in the reply box. Strange.



Jay

David Downey wrote:



[quote]torque to the low limit and tighten to the next slot/hole combination. If the bolt has 4 holes and the nut has 3 slots it is very easy to make the middle of the range. If the bolt has 1 hole and the nut has 3 slots you may need an AN960 "L" washer to adjust       Ready for the edge of your seat?    [quote][b]

purplemoon99(at)bellsouth
Guest

Posted: Fri Jul 27, 2007 2:49 pm Post subject: More torque wrench

It happens to me also....N101HD    Joe

---

Float Flyr

Joined: 19 Jul 2006
Posts: 2704
Location: Campbellton, Newfoundland

Posted: Sun Jul 29, 2007 5:42 am Post subject: More torque wrench

Regarding the considerations about weight testing your torque wrench a few

things come to mind.  



If you are going to torque a nut or bolt for say 240 in. lb. ( the way it

was written when I was in school back when Moses was a boy) then you would

use 20 lb(at)12in.  The wrench can be considered accurate I suppose + - 10%

(216 - 264 in.lb.)  As the nut or bolt you are torquing is at the required

calibration point I can't see a problem. 



If you want to have your torque wrench mapped on an electronic load cell

calibration machine expect to spend some serious dollars.  In Canada any

measurement tools used by an AME (Aircraft Maintenance Engineer) have to be

calibrated on a 24 mo. Cycle and you can believe me it co$ts big time.  Not

only torque wrenches have to be calibrated but everything from micrometers

to scales and magnetic compasses have to be marked as calibrated.



If you can duplicate similar accuracy without the cost of a commercial

calibration I figure it's worth it.  



The idea of using a double female socket sounds viable especially if the

calibration wrench is a beam type wrench.  Using one of the new load cell

electronic torque wrenches will probably give accurate results with the same

double female socket.



Noel

_________________
Noel Loveys
Kitfox III-A
Aerocet 1100 Floats

Display posts from previous:

	Matronics Email Lists Forum Index -> Zenith-List	All times are GMT - 8 Hours Goto page 1, 2 Next
Page 1 of 2

Jump to:

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum