Steel Hardening Deformation 2

[reploglejj]

Just joined this forum hoping to get some input from someone who has some experience about this topic. Brief intro, I am a mechanical manufacturing engineer co-op at a stainless steel mill. I work in the hot plate area where we get plates from the melt area already rolled and cut into sections. We then reheat them in a furnace and immediately quench them in a cooling chamber to harden. I've done the research and understand the theory behind the process but there's an issue the area is having with plates deforming more than they have in the past.

The plates have always had some deformation from the hardening process and that's why the next step is a leveling process. But from what I've been told its a lot worse now to where sometimes its too much for the leveler. The plates are getting extreme "crossbow" across the width of the plate. I have also noticed that thicker plate (usually around 1") seems to bow length wise so bad that it gets hung up op on the roller lines. Its bowing down and causing the plate to have one large arch.

Getting to some details about the quenching process, I am completely new to this process and still learning every day but the cooling chamber doesn't seem "ideal." I've seen videos of other mills that have a giant tank the a crane lowers the heat treated plates into to uniformly quench them. Our chamber has a system of headers above and below the plate with full-cone- spray nozzles designed for this purpose. I have been given the task to research some other brand of nozzles and figure out which have the most uniform coverage but I worry that the problem is deeper than that. I've searched and searched for information on this topic and all I can really find is the basics : "if the quenching is not uniform, the steel will warp", "if the quenching is too severe, it could change the dimensions of the plate", "but if quenching isn't strong enough to cool the plate down quick enough, it may not harden."

My main questions:

1. Why is the thicker steel bowing down? Cooling off the bottom at a slower rate than the top creating uneven heat transfer?
2. In general how does the heat transfer rate in the steel define how the plate will deform? I know there's going to be deformation with the structure of the steel changing but I just don't know what would cause the extreme dimension changes.

If anyone could point me in the right direction of where to find the answer to this I would greatly appreciate it. I know its next to impossible to figure out just by what I've typed and not seeing more but I'm just looking for ideas on what could be going on. Thanks in advance!

 

[kumkumvijay]

Hi Reploglejj,

1. If you get U shape bow while cooling , the reason should be different cooling rate between top and bootm. Severity of quenching in top is more than bottom.
2.
Heat transfer:
The spray system create two kinds of heat transfer mode. 1. wet, 2. not wet.
Wet: The heated surface is wet by the spraying coolant which means there will be a lot droplet to droplet contact and they fall over the plate surface.
The wet mode will create a vapour blanket on the plate surface after first fall and this blanket will be prevent further falling liquid cooling the surface. This is less effective cooling mode. To avoid this effect, during induction hardening process , a polymer is mixed.

Non wet: In this mode, the droplet to droplet contact should not be there and continuous impingement of the droplet on the hot surface create more effective and uniform cooling.

To have a effective cooling, impinging weber number ( amount of water droplet) non wet mode spray are important with other parameters.


Why bottom spray is less effective:
The bottom spray tend to have problem due to gravity. the water droplet net incoming momentum and gravitation force act in opposite direction, so the droplet can sometime will not reach the quenching surface at certain parameters which leads to less sever cooling on bottom side. The optimized cooling from bottom side may have smaller to wider size of droplet with optimized distance from the nozzle to plat. this will ensure the small droplet hit the further away from the impinge point and big droplets hit the center points.

Hope my reply can help you a bit.

 

[EdStainless]

What grade of stainless is this? That is very critical to the situation.
In general metals will bow toward the side with the least cooling.

Quench is usually done with either very high pressure water (>3,000psi) or an air+water system.
The goal is to 'blow away' the surface film of steam that forms. The steam at the surface of the metal actually insulates it and slows the cooling.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[reploglejj]

kumkumvijay,

Thank you for that information that will defined some good points to bring up and look closer at. I had a feeling about the bottom spray possibly not being adequate because of the effect with gravity but also I believe one of the engineers attempted to turn down the water valve that hits the bottom of the plate believing that this was causing the bow.

Currently what I am looking at are full cone nozzles that will maximize cover but what you said definitely leads me to believe there's more to look into. I will look into the amount of water droplet the nozzles put out. I believe I saw calculations on each company's website for their nozzle's droplets dimensions. Also when you spoke of the "wet mode" you said in order to avoid the blanket effect a polymer is mixed in with the liquid? Am I understanding that correctly? As far as I know the water is treated but not with any polymers or chemicals I will have to ask. That was going to be my next assumption was that the current nozzles were covering the plate thoroughly but not having enough force to push off the previously sprayed water, also making the blanket effect.

EdStainless,

Its hard to narrow down the type of stainless because of how the mill runs. We have no middle man involved so our sales works directly with the customer and we could be running a thin gauge 309 one day then the next shift could be running thick 64mm 304. But on average, we seem to run a lot of 304 stainless. The cooling process has low pressure headers that spray onto the plate from top and bottom with a blower at one end (do not have specs on blower) that blows the vapor out. When the plate is quenched (timed program depending on type and gauge) it continues on forward and passes through high pressure nozzles that are meant to descale. Even with the blower there does seem to be a good amount of steam in the chamber. Might look into a stronger blower to move that steam out more efficiently.

I appreciate both of your responses! I'm just a co-op, but trying to make a good impression with these projects for when I graduate soon. This has already been a great learning experience. Thank you.