PDA

View Full Version : Q of the week: Bent tail dogs Vs straight tail



Forrest Addy
06-15-2010, 01:58 AM
General question for the forum inspired by another posting:

There's a trap associated with using bent tail lathe dogs that's not present when using straight tail dogs. What is it?

Hint: think skinny work.

Winner gets a warm fuzzy feeling.

dp
06-15-2010, 02:08 AM
Assuming it has nothing to do with symptoms of Peyronie's disease, there is a bending moment applied to the work when using a bent dog that will cause the work to bow and possibly whip.

A straight dog applies only rotational force to the work.

Peter S
06-15-2010, 08:46 PM
Forrest,

I am not quite sure what the bent dogs are for (I used to avoid them), I presumed it was for when you use a chuck jaw to drive the dog, rather than a carrier plate and stud? I preferred to drill and tap the straight leg and put in a bolt if using a chuck jaw for drive (bent leg always in the wrong place).

Anyway...this raises something from the past which may be related.

Back in the early 1800's, Joseph Clement (1779-1844) invented the "Clement driver", it was a double-ended dog (or carrier as some call them) using two driving pins.

Here is the description I have:

"The effect of having a single point of contact was to bend the work at the point it met the tool tip, a result of the twisting action of the force applied. Clement overcame this by designing a double-driver with two pins, which equalised the forces tending to make the lathe cut eccentrically."

I have asked this question before and I don't think anyone believes this explanation, perhaps the description is not correct. Mr Clement was an extremely clever and accomplished machinist, e.g. making the Babagge calcualting engines, making one of the first planers etc.

So any comments welcomed!

Clement's driver illustration taken from Henry Maudslay & the Pioneers of the Machine Age:

http://img.photobucket.com/albums/1003/PeterS/clementsdriver.jpg

An earlier attempt to get some ideas:

http://www.practicalmachinist.com/vb/general-archive/clements-driver-87454/

John Stevenson
06-15-2010, 08:57 PM
I have found from painful personal experience that small terriers with bent tails are nasty little bastards and are to be avoided or booted with hob nailed boots on at the first opportunity.

Some years ago I was riding down one of our local streets on ower kids Honda 50 step thu, absolutely flat out in second as it was that clapped out it wouldn't pull third.

This bloody bent tailed terrier came rushing out this garden to chase the bike, not the first time this had happened but today I was determined it wasn't going to happen again, I drew my leg back and as it approached I lashed out but unfortunatly I lashed out that hard I threw myself off the bike and into a heap in the middle of the road when the blasted dog ran up and bit me on the arse.

So what was the question again ?

.

reggie_obe
06-15-2010, 11:10 PM
Straight tail dogs are normally associated with grinding setups on grinding machines. Bent tail dogs are normally used on a lathe with a dog plate. Lacking the proper equipment, they can be driven via the slot in a faceplate or via a chuck jaw. With skinny work, the mass of the dog could create an imbalance and exert a bending force while turning. The added mass of the bent tail would amplify this effect.

Mcgyver
06-16-2010, 12:06 AM
I have drive plates with factory mounted posts to catch straight dogs...saw a drive plate the other day where they post traveled in a T slot which i thought clever. 9/10 times i turn between centres using the three jaw so a bent dog is needed

As was mentioned, with straight dogs the work just sees torque vs seeing a force pulling it off axis with the bent dog, ie there's a bit of torque being applied perpendicular to the lathe axis with a bent dog.

why do i suspect there's some offbeat reason none of thinking of that's going to make us look goofy :D

Paul Alciatore
06-16-2010, 12:20 AM
On the question of why the double tailed dog would be superior to a single tail, both being straight tails, I would offer the following.

First, remember that all machines are made out of rubber and thus WILL distort.

OK, the single tailed dog will apply the torque from one side. This side will rotate with respect to the point where the tool is also applying torque to the work piece. This alone will create a changing situation as regards the stresses and therefore the distortions in the work piece. But this would be a minor effect compared to what I talk about below.

Considering further, even if the work is still, but is being torqued by the single tail dog it can move sideways at the head stock end: here's how. The machine (ways, headstock, and tailstock) is made of rubber and the work is mounted on centers that are cones. Torque can not be applied by a single point, so the torque transmitted at the headstock end is SHARED by the single point of contact at the end of the dog and by the headstock center. If you don't believe this, try to turn a wrench with one side of one finger. Thus, there is a SIDEWAYS load on the headstock center. So due to the angle of the center's cone, that sideways force will be translated to a axial force directed toward the headstock and toward the tailstock throught he work piece. Hence the two centers will be pushed a bit apart - remember, it's a rubber machine so it bends and flexes. When they move apart, there will be some clearance created between the conical hole in the work piece and the headstock center. This will allow the headstock end of the work to move sideways. Thus, the work will not be centered at the headstock and will not cut true. Perhaps we are only talking tenths, but it is real.

A two tailed dog will equalize this torque between the two tails and will take it almost completely off of the center so the work would turn a lot truer. Even if the two tails were not exactly equal, it would greatly reduce this torque as an equlibrium point would be reached fairly quickly if it is a small error.

But, as one of the posters on the other board noticed, Mr. Clement's drive plate seemed to have a floating element that would equalize any slight difference. This would be an important part of his system. This system does seem to offer definite advantages for obtaining the best accuracy.

What is curious is the use of a threaded center on that drive plate. This would be far inferior to a tapered center unless it was ground in place after being mounted there. Perhaps this is what he did.

On the other board, someone said that the dog would be about an inch out from the headstock center. I don't see that as an essential part of this or any set-up with a dog and it would be bad practice in any case. I would place the dog as close to the end as possible. But even it it was that way, I don't believe it would have as much of an effect as the mechanism that I explained above. It would require a thin work piece to make much difference that way whereas my mechanism works for all sizes.

wendtmk
06-16-2010, 10:29 AM
And here I thought Forrest was talking about Chihuahua's...

Mark

Arthur.Marks
06-16-2010, 12:19 PM
RUFF!

haha. If we are to use only empirical evidence... it would seem the issues raised here are really a non-issue. Has anyone ever seen a double-leg dog (straight or bent) in use or for sale at any time in their life? Summarily, how many times have you known a shop to work between centers---whether grinding or turning?

dp
06-16-2010, 12:26 PM
RUFF!

haha. If we are to use only empirical evidence... it would seem the issues raised here are really a non-issue. Has anyone ever seen a double-leg dog (straight or bent) in use or for sale at any time in their life? Summarily, how many times have you known a shop to work between centers---whether grinding or turning?


I turn between centers. I prefer it as I have crappy chucks :)

Mcgyver
06-16-2010, 01:06 PM
Summarily, how many times have you known a shop to work between centers---whether grinding or turning?


I turn between centers. I prefer it as I have crappy chucks :)

so do I. I get why from a production viewpoint why its less desirable, but sometimes its the best way when the are lots of operations, concentricity must be maintained and you either don't have collets or the piece won't fit in a collet.....throws on cranks are an obvious one




Has anyone ever seen a double-leg dog (straight or bent) in use or for sale at any time in their life?

http://i785.photobucket.com/albums/yy132/michael01000/DSC_4206Large.jpg

Arthur.Marks
06-16-2010, 01:06 PM
That was my point. Plenty of shops routinely work between centers. Nearly all cylindrical grinding, for example, is between centers; yet I've never heard of or seen someone using a double leg dog.

Arthur.Marks
06-16-2010, 01:07 PM
Thank you, Mcgyver :) Now we're getting somewhere ;)

3jaw
06-16-2010, 01:09 PM
Assuming it has nothing to do with symptoms of Peyronie's disease, there is a bending moment applied to the work when using a bent dog that will cause the work to bow and possibly whip.

A straight dog applies only rotational force to the work.

I think we have a winner.

Mcgyver
06-16-2010, 01:14 PM
That was my point. Plenty of shops routinely work between centers. Nearly all cylindrical grinding, for example, is between centers; yet I've never heard of or seen someone using a double leg dog.

I'm sorry, I misunderstood - thought you were say working between centres was rare implying the question of bent/straight was moot, which it is some circles, ie screw machine operation.

the double sided dog is the only one i have, tons of single leg straight and bent but that's the only double i've seen

Peter S
06-16-2010, 10:16 PM
All the grinding carriers/dogs I have used (J&S) were double-ended (and straight), but I have never seen twin driving studs on a grinder, so I figured it was done for balance, but perhaps it was done so you could use twin driving studs.


Bent tail dogs are normally used on a lathe with a dog plate.

Reggie_obe,

All the lathes I have used came with a small driving plate fitted with a stud intended for driving straight dogs.

------
Getting back to Mr Clement, remember he was using this back around 1820-40, my guess is that most lathe work was done between centres, and the lathe would probably have a solid spindle, no internal taper to fit a centre etc (hence the threaded taper)?

Astronowanabe
06-17-2010, 12:05 AM
General question for the forum inspired by another posting:

There's a trap associated with using bent tail lathe dogs that's not present when using straight tail dogs. What is it?

Hint: think skinny work.

Winner gets a warm fuzzy feeling.


Thanks for asking this question Forrest,
I haven't been doing this long enough for this to have been a problem and now, without any effort on my part, it is so obvious, it never will be.

luthor
06-17-2010, 07:55 PM
General question for the forum inspired by another posting:

There's a trap associated with using bent tail lathe dogs that's not present when using straight tail dogs. What is it?

Hint: think skinny work.

Winner gets a warm fuzzy feeling.


I wonder when Forrest is going to get back to us with an answer?
How skinny are we talking here Forrest?

dp
06-17-2010, 07:57 PM
I got a PM from him telling me I busted the bubble on the first post :)

J Tiers
06-17-2010, 10:15 PM
There is ANOTHER one then......

if the bent tail jams in the slot, it can apply a bending moment to the work that is NOT associated with torque....... it's just the tail "lifting" towards teh T/S and bending the work.

luthor
06-17-2010, 10:26 PM
I have ground a lot of long,slender reamers and other jobs between centers using bent and straight carriers and never had a problem.

Mcgyver
06-17-2010, 11:42 PM
I have ground a lot of long,slender reamers and other jobs between centers using bent and straight carriers and never had a problem.

Nor have i.....however if was a issue, i'd guess it would be more likely when turning where cutting forces are likely higher than grinding...

I like JT's answer best, better than mine actually :D if you're not careful tightening the dog leg can bottom out in the drive plate slot.....post on drive and plate and straight leg alleviates that.

I have turn some small stuff between centres with a homemade bent leg dog...to prevent/minimize the unwanted force direction, make sure its mounted as close as possible to the centre

oldtiffie
06-17-2010, 11:51 PM
Most dogs I've seen have four legs - "double-double"???.

I've never asked a dog why he a has a straight or a bent tail - or in some cases, why he has no tail at all.

On the other hand I have heard of people who were so pleased about something that they were said to be "as happy as a dog with two tails".

Is that what a double tail dog is?

On lathe or a grinder??