[TheForge] Heat Treating 5160 was Welding 5160

[David E. Smucker]

Tom, Grant, Ralph and others -- still trying to learn here, so please take
my comments in that vain.

The question seems to be "how long does it take for the transformation to
take place -- once we have reached critical temperature?"  It is clear at
least to me that the center of a given part -- if it is to be heat treated
has to reach -- and go somewhat above the critical temperature for the
transformation to take place in all of the steel.  Once the material at the
center has reached critical temperature or slightly above -- how long does
the transformation take, and why should it take more time (the
transformation not heating) than a thin section?  I don't believe the
transformation to be instantiations -- but I do believe it to be somewhat
fast -- and the higher the temperature is above critical the faster it would
take place.  I can find lots of technical information on the rate of
transformation in cooling (quenching) but no data on the rate direction of
heating up or the transformation of  pearlite and ferrite to austenite.

I have watch the transformation take place before your very eyes on cooling
as product from a hot strip mill is run out on to the run out table (and
again it is thin at this point).  Because of the volume change the strip
will form a loop or hump up on the table as it cools through the
transformation temperature.  This has always been a low carbon or low alloy
product.

I am also not suggesting that the transformation on heating, would in fact
occur at the same rate as on cooling.  I simply don't know.  But wouldn't it
be similar?

I still go back to my conversation with Jim Batson -- that the prime purpose
of the soak time was to insure that the center of the object reach critical
temperature and than the transformation took place throughout the piece.
Question is how much of this soak time was to insure full heating and how
much was to insure transformation?  What steel were we talking about ? 1095
?  not sure.

Most heat treating manuals I have read, assume that you have a heat treat
furnace with an automatic controller and suggest you set the operating
temperature at the prescribed heat treat temperature (and not above as in
our forges).  They then tell you to heat until the surface color of the part
is equal to the color of the furnace and to then start counting soak time.

It also would appear from the soak times suggested that the more "complex"
the steel i.e. the high the alloy content, requires longer soak times -- 
hence longer transformation times.

Bill Bryson list the following soak times based on various steels.  I reread
that tonight and noted that they were not all 1 hour per inch!!   I missed
this before -- having often heard the 1 hour per inch rule.

W1 -1095     5 minutes per inch
4140 -           5 minutes per inch
O1-               5 minutes per inch
H-13             30 minutes per inch
S-7                 1 hour per inch
A-2                 1 hour per inch
D-2                 1 hour per inch

This didn't sink in until rereading Bryson tonight.   Looking at these it
would seem to me that shorter time alloys 1095, 4140, and O1 show mostly
heat transfer time with a rather fast transformation time -- while the more
"complex" alloys require a much longer transformation time.  (Heat transfer
time for all of these will be similar but longer in the high alloy steels -- 
i.e. stainless steels as a class have a lower heat transfer rate than carbon
steels)

No data on soak time for 5160 -- but it would seem to be more like 4140 than
D-2   so I would be think in terms of a 5 minute per inch soak time.

As I reread this post before sending I realize I may have answer my own
question -- those steels designed by alloy additions to have great depth of
hardness or hardenablity by their very nature have a much slower
transformation reactions.  And this slower transformation works in both
directions.  In the quench direction it gives time to get the cooling deep
into to the object -- hence the depth of hardening -- or use a slow quench
medium such as Air to lower the amount of distortion -- in the heating
direction it means longer "soak" times.

Dave Smucker




----- Original Message ----- 
From: "Thomas A. Troszak" <tom@tomtroszak.com>
To: <theforge@mailman.qth.net>
Sent: Monday, August 25, 2003 6:58 PM
Subject: Re: Re: [TheForge] Welding 5160


> > From: "David E. Smucker" <davesmucker@hotmail.com>
> > Subject: Re: [TheForge] Welding 5160
> > Date: Mon, 25 Aug 2003 08:43:47 -0400
>
> > For most Blacksmithing applications the "hold for one hour per inch of
> > thickness" doesn't make sense.
> > The reason it is required is that to control the temperature in a
industrial
> > heat treat furnace the operator adjusts the set point of the furnace to
the
> > desired critical temperature -- and then expects the work piece to come
up
> > to that temperature over time.  To be sure that the piece fully gets to
the
> > temperature we use that 1 hour per inch of thickness rule of thumb.  "We
let
> > it soak" to be sure that the whole piece reaches critical temperature.
>
> Dave,
>
> There is more to it than that. Even when the entire piece is finally at
> critical temperature, the transformation takes a measurable amount of time
> to spread through a thick section.
>
> Soak times are measured from the point at which the entire piece arrives
at
> critical temp., not from when the piece is tossed in the furnace.
>
> I have used a small volume (about 1 cu ft.), high input (300,000-500,000
> btu.) gas furnace to harden large blocks of tool steel (more than 3 inches
> thick) in 90-120 lb. batches, with manual gas and air controls. In this
> situation, it is possible to set the gas throttle to a fixed heat input.
>
> With fixed heat input, the temperature in the furnace remains relatively
> constant throughout the heating phase, even after the surface of the work
> shows a color matching the color of the furnace interior. This indicates
to
> me that the blocks are still absorbing heat.
>
> 20 minutes or so after the blocks first show color, the furnace temp
> suddenly starts to rise, indicating to me that the blocks are now heated
all
> the way through.  The operator must then throttle back the input, over
> overheat the blocks. I count the "soak time" from the temp spike, not the
> first color.
>
> Rough generalization:
> Think of ice melting in a bucket of water on a hot day. The temp of the
> water will not rise until all of the ice is melted, as all the thermal
input
> is absorbed in melting the ice. With the blocks of steel in the furnace,
the
> temp of the furnace remains remains constant (with a constant heat input)
> until the center of the blocks reach the furnace temp, then the furnace
temp
> suddenly starts to rise.
>
> In actual practice, blocks of H-13, 3 inches thick pulled and quenched as
> soon as color matched furnace temp, rarely got above R 40. Blocks soaked
for
> an hour after the temp spike and quenched get to R 50 or R 55 no problem,
so
> "something important" definitely happens during the soaking, long after
the
> outside of the blocks reached critical temp. Perhaps someone on the list
can
> explain "why", I just know "what happened".
>
> Below one inch of thickness, however, I can find little reason (in my
> experience) to soak any appreciable length of time once the critical temp
is
> reached. I have quenched 1 inch dia. punches as soon as they reached
> critical temp, and I have let them soak for 20 minutes at critical temp,
and
> could not discern a difference in the final product.
>
> Also, according to the tech guys at Uddeholm (whom I hold in high regard),
a
> lot of action takes place between 350-150 degrees F. during the quench, so
> they advise quenching all steels to room temp. before tempering.
>
> I am not an expert, I have just hardened hundreds and hundreds (and
> hundreds) of pieces of tool steel, and I keep good notes, and this is some
> of what I have found out.
>
> Also, the recommended critical temp for a given grade of steel varies
> noticeably from brand to brand, even from batch to batch. The product I
got
> from Crucible was vary variable, and even though it was cheap, it was too
> frustrating to deal with on a production basis. The product I got from
> Uddeholm was incredibly consistent, and never varied from batch to batch.
> If you are buying steel from a reputable manufacturer, ask then for the
heat
> treatment data, and follow it to the letter. If you are using steel from
> scrap, you will have to experiment to find the sweet spot. If you only
have
> one piece, you cross your fingers and hope for the best.
>
> Tom Troszak
>
>
>
>
>
> _______________________________________________
> http://mailman.qth.net/mailman/listinfo/theforge
> theforge mail list group photo site is
> http://www.photoaccess.com
> Login:  blacksmithblacksmith@hotmail.com
> password:  anvil
> ___________
>
>
>