Heating a steel spring then quenching in water

[Glenn G]

TL;DR Summary

Grain size and other factors on macroscopic properties

Hi group,

I heated a steel spring until red hot then plunged in water quenching it.
My understanding is that quenching so quickly would have an effect of the grain size such that I would expect it to be much more brittle as with small grain there is much less plastic deformation possible (as the defects can’t travel as far before hitting a grain boundary). However I wasn’t expecting the spring to now become so weak as it just falls apart at very low stresses. I read that quenching can make metals harder and stronger?I was then wondering is it due to small cracks that have maybe been created on the surface due to such rapid cooling down OR is it something to do with oxidation at the surface whilst being heated red hot or something else.

I would love your opinion on this please as I can’t find a definitive answer elsewhere.

Kind regards,
Glenn.

 

[berkeman]

Hi Glenn,

Is there any particular reason why you are doing this experiment? What are you using to heat the spring to red hot? What kind of safety precautions do you have in place for this heating, handling and cooling experiment? Are you doing this in your garage, or in a lab somewhere?

 

[jrmichler]

There are a number of things happening here. Not knowing the particular spring alloy, I'll guess that it is an oil hardening steel. Quenching such a steel "too fast" can create internal stresses due to size changes as the steel makes the austenite to martensite transition. Fully hard steel is also brittle. Brittle materials break very easily, particularly in the presence of surface defects such as a scratch. Internal stresses make it worse. Glass is a good example of a brittle material, and untempered martensite is about as brittle as glass.

Hardened steel is tempered for the purpose of slightly reducing strength to dramatically increase toughness. The end result is that the part is stronger in the presence of real world stress concentrations. The photo below shows some test parts that I had made when I taught a class on tool design. Three screwdrivers were made from annealed oil hardening tool steel. The one on the left was hardened to Rc 59, the one in the middle hardened to Rc 59 and tempered to Rc 38, and the one on the right annealed to Rc 1.5. A screw was clamped in a vise, and these screwdrivers were used to turn the screw using a torque wrench. The Rc 59 screwdriver shattered at 10 in-lbs, the Rc 38 yielded at 65 in-lbs, and the soft Rc 1.5 screwdriver yielded at 45 in-lbs.

If it is an oil hardening steel, then it needs to be quenched (cooled through the martensite transition) fast enough to convert to martensite, and slow enough to minimize internal stresses. Water cools it too fast, so oil is used as a quench medium. Search terms heat treating steel brings up a very good Wikipedia article on heat treating. Keep in mind that the Wikipedia article is a summary of a very thick book...

Quenching hot steel in oil can vaporize just enough oil to make a small fire that goes out by itself. These experiments are best done outside on the ground so that if the quenching oil container gets knocked over, the oil will spread out on the ground, which will cool it and put out any fire. Also, if the container is on the ground, hot oil cannot spill on you. In addition, if you are standing while handling hot steel, and drop the hot steel, it will fall to the ground and not land in your lap. This is not something to work on while sitting at a wood workbench.

 

[marcusl]

Red isn’t hot enough to mobilize carbon (to produce the transition mentioned above) in spring steel like 1095. You need a shade of yellow. Look it up online for your alloy.

 

[Twigg]

Glenn, one more safety tip to add to @jrmichler's list. Wear leather welding gloves and keep your entire body (especially your face) away from the oil when you quench. I've never had it happen, but the vapor off the oil can flash and make a burst of flame above the oil. If it happens, keep calm and don't flinch. Just hold your tongs steady and it'll be gone in a second. If the oil stays on fire, then you have a problem. Have someone else around and have a class B fire extinguisher on hand. Putting a piece of sheet metal or plywood (in a pinch) over the container to smother the flame it also works, but remember having two ways to deal with a fire is better than one.

Also, to anyone reading this thread, never ever put an automotive strut in a fire. If you have a uncompressed coil spring (removed from the strut by someone who knows what they're doing! DO NOT try to DIY it!), that uncompressed spring can go in the fire safely. The struts themselves should never go in the fire because they will explode, and never put a strut with a compressed spring in the fire! Glenn, I realize that this isn't what you're doing, we just also have to think about other people meandering through these threads.Now to your question:

However I wasn’t expecting the spring to now become so weak as it just falls apart at very low stresses.

Look at the interfaces where the steel broke apart. Is it shiny? Break a test piece for comparison. If it's dull and/or you see any hint of rust color, then it was probably already cracked before it broke. As you said, cracking is common in a water quench because water is just too fast as a quenchant for high carbon steels.

Just checking, did you make sure that your spring is not an austenitic stainless steel (like 304, 316, etc.)? (Just check if it's magnetic. If it is, you're fine.)

The best advice I can give you for finding the right temperature for heat treating is to do your own experimentation. Don't rely on the color of the steel in online videos. Cameras increase the apparent color temperature of the steel (for example, a red hot piece of steel might look orange on video). Even with the naked eye, the apparent color of the steel will depend on the lighting of your room (and not in a small way). If you don't believe me, try brazing in the sunlight (it sucks!). The best thing you can do is have some test pieces to play around with. Try heating them to different color temperatures and quenching them. Vary your process, and compare the results. If you don't have the luxury of test pieces for whatever reason, you can heat the steels to the critical point. When the spring is hot enough that it stops being magnetic, it is hot enough to quench (actually it'll be a bit too hot IIRC, but it'll probably be ok).

 

[sophiecentaur]

Quenching hot steel in oil can vaporize just enough oil to make a small fire that goes out by itself.

Definitely a safety consideration. I have 'blued' a number of small mild steel parts and it's quite dramatic; loads of grey smoke and bubbles (outside, on the concrete apron in front of the shed) and I could believe you could start a fire with a big item. Tempering would be similar and more critical than just cosmetic stuff.

 

[Rive]

However I wasn’t expecting the spring to now become so weak as it just falls apart at very low stresses.

Based on my childhood experiences that may be the problem of heating up and not the problem of quenching.
Guess it was some thin wire spring? Those are tricky to heat up since usual (household) flames are quick to eat away the actual steel. Did you see some residues at the bottom of the water after the quenching?

 

[sophiecentaur]

When heating up small items it’s a good idea to put them in a larger container to avoid hotspots. Also good for slowing down the cooling if needed (sand box).

 

[Rive]

I don't know if an (open) container alone would work in this case.
It is kind of a shock to everybody for first that steel actually may be a flammable material. If heated up, given the chance it burns away. Hence, some kind of inert atmosphere would be required.
As far as I know, small springs are usually made cold to avoid all the fuss.

 

[Glenn G]

Glenn, one more safety tip to add to @jrmichler's list. Wear leather welding gloves and keep your entire body (especially your face) away from the oil when you quench. I've never had it happen, but the vapor off the oil can flash and make a burst of flame above the oil. If it happens, keep calm and don't flinch. Just hold your tongs steady and it'll be gone in a second. If the oil stays on fire, then you have a problem. Have someone else around and have a class B fire extinguisher on hand. Putting a piece of sheet metal or plywood (in a pinch) over the container to smother the flame it also works, but remember having two ways to deal with a fire is better than one.

Also, to anyone reading this thread, never ever put an automotive strut in a fire. If you have a uncompressed coil spring (removed from the strut by someone who knows what they're doing! DO NOT try to DIY it!), that uncompressed spring can go in the fire safely. The struts themselves should never go in the fire because they will explode, and never put a strut with a compressed spring in the fire! Glenn, I realize that this isn't what you're doing, we just also have to think about other people meandering through these threads.Now to your question:Look at the interfaces where the steel broke apart. Is it shiny? Break a test piece for comparison. If it's dull and/or you see any hint of rust color, then it was probably already cracked before it broke. As you said, cracking is common in a water quench because water is just too fast as a quenchant for high carbon steels.

Just checking, did you make sure that your spring is not an austenitic stainless steel (like 304, 316, etc.)? (Just check if it's magnetic. If it is, you're fine.)

The best advice I can give you for finding the right temperature for heat treating is to do your own experimentation. Don't rely on the color of the steel in online videos. Cameras increase the apparent color temperature of the steel (for example, a red hot piece of steel might look orange on video). Even with the naked eye, the apparent color of the steel will depend on the lighting of your room (and not in a small way). If you don't believe me, try brazing in the sunlight (it sucks!). The best thing you can do is have some test pieces to play around with. Try heating them to different color temperatures and quenching them. Vary your process, and compare the results. If you don't have the luxury of test pieces for whatever reason, you can heat the steels to the critical point. When the spring is hot enough that it stops being magnetic, it is hot enough to quench (actually it'll be a bit too hot IIRC, but it'll probably be ok).

Thanks for the reply.

 

[Glenn G]

Hi Glenn,

Is there any particular reason why you are doing this experiment? What are you using to heat the spring to red hot? What kind of safety precautions do you have in place for this heating, handling and cooling experiment? Are you doing this in your garage, or in a lab somewhere?

I had a number of lab steel springs (ones used in science class) - interested to see what happened when I heated and quenched - had seen a video about it becoming brittle (it did) but my question was really about how weak it became. Thanks for reply.

 

[sophiecentaur]

my question was really about how weak it became.

For a spring wire of , perhaps 1.5mm, I'd suggest you oxidised sufficient material to reduce its effective cross sectional area.

If you repeated the experiment with the spring buried in sand and cooled slowly then the steel would become softened by a process of annealing.
Someone already mentioned that springs are usually cold-formed to produce consistent results.

 

[Jodo]

Quenching does make certain metals harder/stronger. Carbon/carbon equivalent is the major player for hardness in metals.
Resultant microstructure you get when heating/quenching is time/temperature dependant. The faster it cools though certain temperature ranges yields higher hardness/strength within limits ( note, mild steel doesn't get very hard when quenched, just not enough carbon to produce bainite or martensite ).
Hardenabilities of metals is a very involved science. Please consult your local metallurgist for more details