"Jim Wilkins" writes:

...

Would fatigue cracking in oil be similar enough to cracking in air,
which oxidizes freshly exposed steel?

I think so - very little difference.
Well my instinct says for environment in general that
* negligible difference at high stresses
* possible significant effects at low stresses / huge numbers of
cycles to crack / break
I suspect the difference between in-air and in-oil would be
negligible.

Another thing I must test though.

Rich S

"Jim Wilkins" writes:

...
As for the eccentric, I considered it because you could vary the
piston stroke, but you'd need a lathe to make it, the pump drive is
more than a friend could whip out in a spare hour. A pump built into
the cylinder end might give the fastest cycle rate because there's no
flow restriction. You could fine tune the peak pressure during
operation with a screw that displaces oil. A cheap used tie-rod
cylinder with a scratched bore could be cut down to be your pressure
chamber. Cylinder rebuilders can provide the tubing in any length.

On my tractor's homebrew bucket loader attachment hydraulics I turned
the head of a bolt round and grooved it for an O ring, so it can screw
into or out of the oil space without leaking. It operates the variable
pressure relief valve I made to replace the fixed relief the control
valve came with. The tractor's front tires turned out to be the
weakest link that limited how high I could set the pressure.


Variable - you have many eccentrics, each with a slightly different
throw?

I thought - one eccentric, but many different "pistons" with their
"collar" they go through into the fluid volume.

Yes one thought is that the mechanism could be built onto the lower cylinder end.

With the "bar" driven by the eccentric simply pushing in and out of
the cylinder volume.
For the "170Tonne-force" test it would need to displace about 270cc -
would be 70mm diameter and stroke (or some other combination of
diameter and stroke which gives that displacement).

One advantage of this arrangement is, seeing as it's so stiff, plus
bits can't fly around with being inside the cylinder, the test rate
could be high. Fastest induction motor speed? 3000rpm on 50Hz supply
= 50Hz test rate :-)
That would be 60Hz on N.Am. supply.





The decision comes down to what you can build or buy. I've spent
significant time and money becoming able to build what I or the
customer wanted, electrical, optical and mechanical.

Knowing the metallurgical and fatigue stuff has cost me a lot - money
in various ways and time ...

Leon Fisk writes:


And what about heat? Flexing metal creates heat. If you accelerate
flexing to speed up failure detection you will likely create heat that
would not be present in its actual use...

--
Leon Fisk
Grand Rapids MI

Heating is pretty negligible.
On a resonant machine you can test at up to around 300Hz and there is
no drift in values obtained compared to at much slower rates.
You have full elastic energy recovery.

Servo-hydraulic - not recovering elastic energy with the drive heats
the oil away from the sample where you dump the pressure - and you
dump that heat through "raditiators" (forced convectors transfering
oil-to-air).

But the sample is seeing full energy recovery / negligible energy
loss.

"Richard Smith" wrote in message ...

"Jim Wilkins" writes:

...eccentric...

Variable - you have many eccentrics, each with a slightly different
throw?

I thought - one eccentric, but many different "pistons" with their
"collar" they go through into the fluid volume.

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

One circular eccentric disk, mounted on an offset pivot pin so it can be
centered or swung out as needed. The clamp for the swinging side might have
to include a custom stepped bushing to withstand the torque, rather than
just a bolt that holds by friction, but its surface finishes aren't critical
like the eccentric's. Lathes don't necessarily leave surfaces good enough to
be running bearings, that's extra hand work.

It's easy to turn two cylindrical surfaces with different centers when
holding the work in a 4-jaw lathe chuck.
http://s3.cnccookbook.com/CCLatheEccentricTurning.htm

The eccentric disk could be a slice of hydraulic cylinder rod with a
case-hardened, chromed and polished surface. The strap could be lined with
replaceable slices of Oilite bushing. You don't need the historical accuracy
many British model engineers strive for.
http://www.stockbridgelocomotivework...nd-valve-gear/

You have ideas that need machine tools to create. Good new ones and hired
custom shop work are quite expensive, so I saved by finding older industrial
machines which had become obsolete and too worn to be economical in a
production shop, like a lathe made in 1965. You have an excellent
equivalent:
http://www.myford-lathes.com/used.htm
I have no experience with the current imports.

I can't easily hit the tolerances on a customer's drawing but I can still
make two pieces fit each other although they may not be quite to spec, so my
antique machines are fine for making single devices for my own use (and for
fixing each other). From the reference above:
"I turned the final diameter on a good portion of the bar and then machined
each eccentric one at a time, individually match-fitting each eccentric
strap."

The trick is that two parts of a complementary operation may not be equally
difficult, for example the piston is easier to turn and finish than the
cylinder, so make the difficult one first and fit the easier one to it. The
boring head he used on the eccentric has a micrometer adjusting screw to
change size, I have an identical one. Many shortcuts are possible when you
control the design.

I made the prototype of an inch-long diode laser and lens mount in a few
evenings that later cost $4000 apiece from a job shop that normally made
parts for BAE. Mine wasn't quite as well finished but it worked and proved
my ideas. First I needed approval to charge it as overtime, but the project
engineer knew how expensive the company's main machine shop was. I suspect
part of the high cost was due to the electrical engineers' inexperience with
mechanical design and machining.

"Richard Smith" wrote in message ...

"Jim Wilkins" writes:

...

Would fatigue cracking in oil be similar enough to cracking in air,
which oxidizes freshly exposed steel?

I think so - very little difference.
Well my instinct says for environment in general that
* negligible difference at high stresses
* possible significant effects at low stresses / huge numbers of
cycles to crack / break
I suspect the difference between in-air and in-oil would be
negligible.

Another thing I must test though.

Rich S

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

Racing engine builders might know.

"Jim Wilkins" wrote in message ...

....
http://s3.cnccookbook.com/CCLatheEccentricTurning.htm

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

I made a split bushing to hold short 3/8" Grade 8 bolts in a 5C collet like
in the "I love this tooling." example shown. Grooves near the ends hold O
rings that keep it together. Neither the hex head, shank nor threads run
true to each other so I have to clamp the bolt by the section I'm modifying.
OK, they are technically hex head cap screws when they go in a tapped hole
but 'bolt' is their common name.

The bushing's center hole is 0.370" diameter to match the shanks of the
bolts at the store, and before splitting I tapped it 3/8-16 to grab the
thread crests without damaging them, as rolled threads can be larger than
the shank. It lets me modify bolts to within as little as 1/2" from the
head, for instance to trim back the slightly-long shank so a nut bears on
the parts being clamped, or to cut a root-diameter pilot and pointed end on
the threads for simultaneous alignment of multiple 40 Lb parts with 0.370"
bolts in 0.375" holes. Those piloted bolts are longer and go at the outer
ends to stop the trolley after using them to align the center splice. Most
of the bolts joining my gantry track sections had to be modified to put
their shanks in the shear planes and clear the moving trolley.

My previous 5C bolt-holding fixtures are bushings with tapped holes, but
this version is more versatile and clamps tight on either shank or threads.
I split it with a hacksaw after scribing the end with a tool point aligned
with the 5C splits.

...
One circular eccentric disk, mounted on an offset pivot pin so it can
be centered or swung out as needed. The clamp for the swinging side
might have to include a custom stepped bushing to withstand the
torque, rather than just a bolt that holds by friction, but its
surface finishes aren't critical like the eccentric's. ...
...

I haven't got this for certain.
Not grasped the idea for sure, yet.

Forces would be in Tonnes to tens of Tonnes.
The size of the eccentric - both what it will withstand as a bend or
shear, and preesenting enough bearing surface for the bearing to take
the load - can make these things quite big. Maybe 200mm diameter with
190mm shaft for 10mm "throw", for some enormous tests in high yield
steels.
As I calculate / estimate as best I can.

If it came to making one of these things, there would have been a lot
of proof-of-principle with "testing lab. scale" samples, and the
stakes would be quite remarkable by that stage, if we got there.

The main point is to have plans in place.
In a political world, you have to have everything covered, so every
interjection, objection, etc. is smoothly put in its place.
As I've experienced.
I've certainly had the skill of predicting what the ploys might be
tested. Having big efforts to derail the plan slapped down in
seconds.

So it's about being able to see a way ahead, far along a perceived
path. What would actually be going on, where you would actually be by
then, what method you would use given experience had by then but ahead
of you now - that might be a different story.
But for now - you are "covered"...

With the beam tests, you can tune the testing force by moving the beam
end supports in and out - present different spans. So one fixed throw
/ movement drive would cover all purposes.

For the "hydraulic inner fatigue test", I think different diameter
"spuds" sliding in and out of the fluid volume, on a fixed eccentric
drive, might be easiest?

"Richard Smith" wrote in message ...

...
One circular eccentric disk, mounted on an offset pivot pin so it can
be centered or swung out as needed. The clamp for the swinging side
might have to include a custom stepped bushing to withstand the
torque, rather than just a bolt that holds by friction, but its
surface finishes aren't critical like the eccentric's. ...
jsw

I haven't got this for certain.
Not grasped the idea for sure, yet.

Forces would be in Tonnes to tens of Tonnes.
The size of the eccentric - both what it will withstand as a bend or
shear, and preesenting enough bearing surface for the bearing to take
the load - can make these things quite big. Maybe 200mm diameter with
190mm shaft for 10mm "throw", for some enormous tests in high yield
steels.
As I calculate / estimate as best I can.
RS

[[[[[[[[[[[[[[[[[[[
Lets say the 200mm eccentric is pinned to its mounting plate with two
diametrically opposed 25mm round bushings. If the bushings are straight
cylinders the eccentric disk runs in a true circle. If both bushings have a
5mm offset step in the middle to shift the eccentric disk sideways the throw
is 10mm. The lathe setup to turn the bushings is the 5C collet block in a
4-jaw chuck, as in the posted reference.

The bushings provide high shear and bearing strength. They would be slotted
and keyed to prevent rotation. Making them is a reasonably simple lathe
operation. To change the throw you make bushings with different offsets
rather than changing the eccentric disk.

Two identical stepped bushings is simpler to describe, one stepped and one
parallel is more rigid, triangle vs parallelogram.
jsw
]]]]]]]]]]]]]]]]]]]

If it came to making one of these things, there would have been a lot
of proof-of-principle with "testing lab. scale" samples, and the
stakes would be quite remarkable by that stage, if we got there.

The main point is to have plans in place.
In a political world, you have to have everything covered, so every
interjection, objection, etc. is smoothly put in its place.
As I've experienced.
I've certainly had the skill of predicting what the ploys might be
tested. Having big efforts to derail the plan slapped down in
seconds.

So it's about being able to see a way ahead, far along a perceived
path. What would actually be going on, where you would actually be by
then, what method you would use given experience had by then but ahead
of you now - that might be a different story.
But for now - you are "covered"...
RS

[[[[[[[[[[[[[[[[[[[
Been there, done that, I was the lab rat who built the proof-of-principles
from sketches and gave the engineers simple, practical and economical ways
to convert their concepts into products. I provided you several different
approaches you could pursue, and mentioned possible sticking points like
difficulty of fabrication or programming, and the skills and equipment to
moved past them.

A very handy trick I picked up to counter objections is to memorize the
squares and reciprocals of numbers up to 32 so I could do engineering math
rapidly in my head and present an off-the-cuff mathematical basis for my
arguments. Above 31 [sqrt(1000)] they pair with lower numbers, 1/25 =
0.040, 1/40 = 0.025. Engineers not from the slide rule era don't learn the
simplifying tricks we had to.

For example I was running an error rate test on a satellite link when the
chief engineer demanded to know how much longer I would be tying up the
channel. The test was 10 million bits at 2400 per second. Knowing that 1/24
is 0.041666... I mentally figured the test duration as 4166.7 seconds, then
converted that to 1 hour (3600), 9 minutes (540) and 26.7 seconds and gave
him the answer. He pulled out his calculator, hesitated, then admitted he
didn't know where to start and left me alone to complete the test.

I understand the physics but it's still weird to see something stored and
later retrieved intact from empty space.
jsw
]]]]]]]]]]]]]]]]]]]

With the beam tests, you can tune the testing force by moving the beam
end supports in and out - present different spans. So one fixed throw
/ movement drive would cover all purposes.

For the "hydraulic inner fatigue test", I think different diameter
"spuds" sliding in and out of the fluid volume, on a fixed eccentric
drive, might be easiest?
RS

[[[[[[[[[[[[[[[[[[[
The most demanding part to make is the piston/cylinder sealing surface,
especially if you don't have machine tools. Any spud screwed into the end of
the piston displaces a fixed volume of oil. There may be another way but I
think the easy answer is to change the stroke length (lever?) unless you
collect an appropriate assortment of polished rod stock and seals (car shock
absorbers, gas cylinders etc). Modifying the rods into pistons and making
packing glands to retain the seals are lathe operations.
jsw

... Modifying the rods into pistons and making packing glands to
retain the seals are lathe operations. ...

I took it a lathe would be available.
Cylindrical "spud", cylindrical bush.

Maybe the "hydraulic fluid" could be grease - thick - tranmitting to
oil or water through a rubber membrane. So the piston/bush sees
viscous grease not thin oil.

"Richard Smith" wrote in message ...

... Modifying the rods into pistons and making packing glands to
retain the seals are lathe operations. ...

I took it a lathe would be available.
Cylindrical "spud", cylindrical bush.

[[[[[[[[[[[[[[[[[
It's difficult to create innovative mechanical solutions with only stock
parts meant to solve standard problems. At Segway which was an engineer's
playground by official policy the CNC lathe and milling machine were almost
always tied up making someone's wild idea. I used their manual lathe or my
home shop machines but my main task was custom electronics.

https://news.yale.edu/2008/12/05/stu...orrell-s-class
Electric model airplanes were practically a second product there, and they
taught me the care and feeding of Lithium batteries.
Unfortunately like bridge building the team dissolved and sought new
challenges once the project was complete.
]]]]]]]]]]]]]]]]]

Maybe the "hydraulic fluid" could be grease - thick - tranmitting to
oil or water through a rubber membrane. So the piston/bush sees
viscous grease not thin oil.

[[[[[[[[[[[[
It's known that high pressure oil seals work if done right, so I'd bin that
with the minor issues that can be solved later by throwing enough money at
them, and concentrate on identifying and resolving the show-stoppers.

I'm glad I read the Amazon reviews before applying the thick black tire bead
sealer goop. A reviewer advised to let it set somewhat before inflating the
tire lest it spray out. I did, and only soapy water rubber lube sprayed me
before the bead seated when I inflated it.

Here is a racing engineer and technology historian who might know about
metal fatigue in oil:
https://www.calum-douglas.com/

...
It's known that high pressure oil seals work if done right, so I'd bin
that with the minor issues that can be solved later by throwing enough
money at them, ...

I'd wish - but lots of slightly different sizes - and you are going to
be able to find a seal for each diameter?

Maybe one could get a list of stock seal diameters, and find that yes,
the pressure and therefore test force you'd end up with would be just
fine - sits just where it needs to be to plot the S-N curve for
fatigue performance. Throwing aside minor vanities like getting the
stress which should mean the mean-average sample break is at exactly
250k cycles, etc.

Rich S

"Jim Wilkins" writes:

...

[[[[[[[[[[[[[[[[[
It's difficult to create innovative mechanical solutions with only
stock parts meant to solve standard problems. At Segway which was an
engineer's playground by official policy the CNC lathe and milling
machine were almost always tied up making someone's wild idea. I used
their manual lathe or my home shop machines but my main task was
custom electronics.

https://news.yale.edu/2008/12/05/stu...orrell-s-class
Electric model airplanes were practically a second product there, and
they taught me the care and feeding of Lithium batteries.
Unfortunately like bridge building the team dissolved and sought new
challenges once the project was complete.
]]]]]]]]]]]]]]]]]

Maybe the "hydraulic fluid" could be grease - thick - tranmitting to
oil or water through a rubber membrane. So the piston/bush sees
viscous grease not thin oil.

[[[[[[[[[[[[
It's known that high pressure oil seals work if done right, so I'd bin
that with the minor issues that can be solved later by throwing enough
money at them, and concentrate on identifying and resolving the
show-stoppers.

I'm glad I read the Amazon reviews before applying the thick black
tire bead sealer goop. A reviewer advised to let it set somewhat
before inflating the tire lest it spray out. I did, and only soapy
water rubber lube sprayed me before the bead seated when I inflated
it.

Here is a racing engineer and technology historian who might know
about metal fatigue in oil:
https://www.calum-douglas.com/

Off-topic but...

Two
great leads in there!

Engine Development and how done.
Empires were at stake then and technical folk got resources - amazing
what they did.

The Calum Douglas lead - never met or seen a pic. of the brittle lacquer
method for stress distribution in-action. ****! it's effective!
The "perspex / polarised-light" method - I wanted to use that when I
knew a Finite Element Analysis modelling engineer must be talking
...nonsense!..., before I could FEA. I actually blagged some
"perspex" from a nearby company - but the Finite Element Analysis
engineer then made such a bad political slip-up that we lost the job
it was for anyway.

"Yale" / Segway lead...
The one about engineers "kissing a lot of frogs to find a
prince". Exactly! Yes, that's the reality of a scientist too!
You have to be able to think and process things that way, to be a
human having an effective channel of perception into the Natural
World.
I almost envy people living in what they think is a determinate world.

"Richard Smith" wrote in message ...

...
It's known that high pressure oil seals work if done right, so I'd bin
that with the minor issues that can be solved later by throwing enough
money at them, ...

I'd wish - but lots of slightly different sizes - and you are going to
be able to find a seal for each diameter?

Maybe one could get a list of stock seal diameters, and find that yes,
the pressure and therefore test force you'd end up with would be just
fine - sits just where it needs to be to plot the S-N curve for
fatigue performance. Throwing aside minor vanities like getting the
stress which should mean the mean-average sample break is at exactly
250k cycles, etc.

Rich S

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

This is a US company selling inch-sized products, but it shows what to look
for.
https://www.baileyhydraulics.com/Pro...er-Components-

When I needed new seals for leaky used Porta-Power-type cylinders I went to
a pump rebuilder who found seals that were close enough in the catalog of a
supplier I don't remember. My cheaply made Chinese cylinders were somewhat
non-standard so I had to machine the pistons a bit. I've been told they are
typical of import hydraulics that can't be repaired (without remachining).

The cylinders were $10 and $15 each and the right very compact size for my
bucket loader design so they were worth the trouble.

I suspect that seals are available for all cataloged sizes of chromed and
polished hydraulic rod stock, a part I can't make. Then you only need to
make (or buy) the packing gland that adapts the seal to the cylinder end
cap, and a crosshead to keep the piston running straight, both fairly easy
on a thread-cutting lathe. They could be combined into one part. Instead of
machining a hex you can drill two holes for a pin spanner wrench to tighten
it.

Last night the air cleaner cover latch on the Chinese engine of my sawmill
broke off so I made a stainless sheet metal replacement. I've stopped
hunting for spare parts that are difficult or impossible to find and no
better than the originals.

A -good- lathe is an investment that may appreciate. Collectors restore the
better examples of mine, like classic cars.
https://www.tinshackrestoration.com/...vy-10-lathe-2/
"Too funny! You bought a great lathe, and with all of those vintage Rovers
you better learn to make parts quick! Thanks for the note!"

"Richard Smith" wrote in message ...

"Yale" / Segway lead...
The one about engineers "kissing a lot of frogs to find a
prince". Exactly! Yes, that's the reality of a scientist too!
You have to be able to think and process things that way, to be a
human having an effective channel of perception into the Natural
World.
I almost envy people living in what they think is a determinate world.

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

I have a BS in Chemistry so I understand how a scientist must think and
operate. It's a very hands-on discipline that applies to almost any industry
and we received a broader training than I've seen in other types of engineer
and scientist, such as Materials Science (the properties of steel) and two
years of Physics.

The profs told us a BS degree doesn't prepares us to do useful work
immediately, only to understand the explanations wherever we go. Advanced
degrees were essential, they said self-servingly. (Mitre was like that too.)
I was chosen for summer research grants and did learn more about real life
applications, and what it's like to hole up in the lab all night and not see
other humans for weeks at a time. I was disturbingly comfortable with that.

At graduation time I still needed 4 more credits in any subject and, being
somewhat burned out by then, signed up for 6 credit, certain-to-pass summer
theatre as a carpenter, and was packed in with as many touchy and demanding
humans as I could stand for 12-16 hours a day, good training I suppose. I
was dancing on stage when Armstrong landed on the moon.

When I graduated the grad school draft deferment had ended, taking with it
my Chemistry career, but the Army was glad to find someone they could train
to maintain complex electronics along the lines of your Colossus machine.
The few who survived that school all had technical degrees. The integrated
circuit was enabling electronics' Great Leap Forward and I got in it at the
start.