[technical Articles] Steel Heat Treatments: How Carburizing Increases The Hardness Of Steel

On the topic of steel heat treatments, we will discuss the analysis of the problems related to the deformations of the mechanical parts built in 18NiCrMo5 steel subjected to thermochemical carburizing treatment.

Keywords: thermochemical treatments, Carburization, heat treatment deformation.

SUMMARY. The problem of deformation triggered by thermochemical treatments of steel has always been one of the most critical issues for companies involved in the construction of mechanical components, both because of the inevitable technological complications and the economic repercussions this brings.

1. Introduction to Heat Treating Steel

Extensive studies carried out by companies operating in the rolling bearing sector have highlighted that for ow-tech mass production, the heat treatment (in the broadest definition of the term), while affecting a modest percentage of the industrial cost of the finished product, gives rise to a series of problems whose solution, alongside the surface finishing operations, brings these processes to a value of almost (and sometimes higher) than 40% of the industrial cost of the finished product [1].

It is therefore easy to understand that, on complex components and with a high degree of precision required in the finished piece, the control of this aspect is of primary importance especially when the number of batches being processed is very low. It should also be considered that, while in some cases it is possible to remedy this problem by operating with suitable techniques (hardening under press, mechanical processing of a blank piece for the compensation of deformations, mechanical straightening), this is no longer possible where it is due to the complexity of the detail, both for the global metal mass to be treated, the deformations reach very high values ​​and, consequently, the deformed piece must necessarily be rejected. The experience of the company has led to highlight, in this study, the case of toothed shafts for power transmission which, due to their geometry, have statistically high and difficult to control deformations. As already highlighted, the high value of the product, in addition to the fact that it is made in batches with a reduced number of pieces, has led to the search for a correct procedure capable of controlling the onset of deformations.

2. The deformation mechanism in steel heat treatments

In order to intervene and solve the problem related to deformation, first of all, it is necessary to understand the causes that trigger these mechanisms. Although the parameters that come into play are many and often difficult to identify, the fundamental aspect on which to focus attention is the substantial thermal shock that the piece must withstand both in the cooling and heating phases [2]. This rapid temperature variation, generating thermal gradients between the core and the surface, triggers the onset of thermal stresses or microstructural variations and ultimately causes variations in the shape of the piece.

If the intensity of these stresses remained within the elastic limits, the piece would not undergo any alteration because the stresses would cancel as soon as room temperature is reached but this generally does not happen because the yield strength of steel at high temperatures is low and can easily be overcome by thermal stresses.

Heat treatment deformations can be classified into two families [3]:

• The volumetric or dimensional variations linked to the microstructural modifications induced by the heat treatment. It is useful to remember that the density of martensite is lower than that of ferrite and pearlite, the lower the higher the carbon content in the steel. Therefore, for the same mass, a martensitically hardened piece will have a greater volume than the annealed one. Austenite, on the other hand, is decidedly denser than the other phases.
• Distortions or permanent variations in the geometry of the pieces. They depend on many factors, not the least of which are differentiated thermal expansions and contractions, as well as microstructural changes.

As shown by the cause-effect diagram above, it is possible to understand that numerous other parameters influence the success of the treatment in addition to the composition of the steel, the temperature of the quenching bath, the residence time at temperature, the cooling speed and , when foreseen, also the duration of the discovery.

Furthermore, if the piece is characterized by particularly complex geometries or of highly variable thickness, differences in temperature and internal stresses can be determined such as to inevitably lead to the onset of deformations: therefore the heating and cooling phases require greater attention and longer times. long.

Due to the number of factors involved, it can therefore be understood that in reality it is extremely complex to be able to create mathematical models that can predict the onset of deformations with certainty, all the more so because most of the theoretical calculations assume isotropic dimensional variations, which in practice they occur very rarely, both due to the lack of homogeneity of the steel and of the treatment parameters, but above all due to the objective impossibility of obtaining a homogeneous and contemporary structural transformation in all points of the piece, given the different cooling speeds .

Furthermore, the contribution to the deformation linked to the instrumentation used for the heat treatment cannot be underestimated: the control and homogenization of the temperature inside the furnace has always been one of the most pressing problems but which finds a difficult solution due to the impossibility of move the thermal sensors from the walls of the furnace to the center of the charge.

3. Case study on steel heat treatments

As discussed in the previous paragraph, we can firmly say that the problem of deformation is even more accentuated on components with a complex geometry and large dimensions. Furthermore, checking this type of piece is made necessary by the fact that in most cases these deformations can no longer be corrected: it will therefore be inevitable for the company to incur high costs due to piece scrap.

In this study, we want to report the case of a shaft with pinion m=22 Z=14 of which both the construction drawing and the three-dimensional graphic simulation are reported. It should be noted that the weight of the piece in the state of manufacture before heat treatment is 182 kg.

This detail typically used in power transmissions has the free length of the shaft greater than double the largest diameter of the shaft itself; due to this type of geometry, the most common characteristic deformations are the ovalization of the cylindrical section of the shaft, the deflection in the centerline under the head [1] of the modular toothing due to the sudden change in section, the so-called “hourglass” narrowing of the entire ideal cylinder that approximates the figure.

Given the impossibility of being able to control these deformations, the possible solution is to prevent the occurrence of conditions such as to trigger the deformation mechanism itself: for this it is necessary to intervene directly on the morphology of the thermal cycle since this is the only aspect on which to intervene directly and which most influences the phenomenon.

In this regard, the results obtained from three different Heat Treatment suppliers were compared, who processed pieces of the same code taken randomly within the same production batch, punctually controlled during all the previous processing phases.

It is useful to reiterate that batch uniformity means the same basic raw material uniquely identified by the steel supplier both with the same casting number (and therefore reasonably with the same chemical composition) and with the same preliminary heat treatment of isothermal annealing.

After carburizing and obviously before the final grinding operations, a deformation under the head of the modular toothing of 0.1-0.15mm was set as the acceptability limit of the lot (Fig.5).

---

The effects produced by the various suppliers are summarized as follows and reported in Table 1:

• SUPPLIER A: the deformations of all the details fall within the imposed tolerances;
• SUPPLIER B: almost all of the pieces have deformations out of tolerance, reaching in some cases even the value of 1mm;
• SUPPLIER C: is positioned at an intermediate level between the two previous suppliers. Approximately 35% of the parts have bending values ​​beyond the permitted limits

ITEM NO.	SUPPLIER A	SUPPLIER B	SUPPLIER C
1	0.06	0.2	0.1
2	0.1	0.4	0.25
3	0.1	0.1	0.2
4	0.14	0.6	0.12
5	0.09	0.3	0.3
6	0.04	1	0.64
7	0.05	0.35	0.38
8	0.15	0.35	0.11
9	0.12	0.25	0.25
10	0.07	0.27	0.2
11	0.05	0.28	0.04
12	0.04	0.3	0.08
13	0.12	0.95	0.1
14	0.04	0.3	0.26
15	0.08	0.65	0.25
16	0.08	0.3	0.15
17	0.07	0.7	0.17
18	0.18	0.8	0.15
19	0.1	0.4	0.2
20	0.07	1	0.11

Typically, the thermal cycle used (case hardening + tempering + tempering) has a standard morphology which can be schematized as in Fig.6: by acting on the characteristic phases in an appropriate manner, different metallurgical results can be obtained. By adjusting the activation time (BOOST) and diffusion time (DIFFUSE), for example, you will obtain different carburizing depths while by changing the tempering phase (TEMPERING) you can intervene on the final hardness of the piece. The thermal cycle object of our study has the temperature-time characteristics reported in Tab.2. These characteristics allow to obtain a depth of the cemented layer of 2÷2.5mm and a hardness of 58÷60HRC.

OPERATION	TEMPERATURE (°C)	TIME (mins)
ACTIVATION (BOOST)	925±5	1200
DIFFUSION (DIFFUSE)	925±5	600
HARDENING	825±5	30
QUENCH	105	20
TEMPERING	150÷200	120-180

by carefully analyzing the data just reported, it is clear that these specifications do not include the heating phases and the cooling phase following the diffusion process for which no particular indication is envisaged regarding their duration; the choice of these parameters is generally left to the experience of the treatment specialist.

The results obtained from the comparative analysis of the three different cycles confirm what has just been said: in fact, by comparing them with each other, it was possible to notice how the only differences can be found in the duration of the heating and cooling process and in the permanence at the temperature hardening (HARDENING) while for all the other phases reference can be made to the data in Tab.2.

	SUPPLIER A	SUPPLIER B	SUPPLIER C
RAMP	180	45	90
COOL	120	15	30
HARDENING	60-90	15	30

As can be seen from the comparison between Tab.3 and Tab.1, a correlation can be assumed between the speed of execution of the three phases with the bending values ​​measured under the head following the treatment: the greater the angular coefficient of the straight lines HEATING and COOLING therefore the speed of execution of the phases, the greater the deformations.

In the case of suppliers B and C, therefore, it can be said that the temperature-time relationship, which represents the angular coefficient of the straight lines, assumes values ​​that are decidedly too high to allow complete homogenization of the temperature inside the piece, making possible the onset of thermal stresses and microstructural variations responsible for the deformation process.

As far as supplier A is concerned, it is possible to assume that the longer residence time at the temperature guarantees such uniformity of temperature between the core and the surface of the piece as to almost completely avoid deformations or in any case keep them within acceptable tolerances (see calculated values in diagram 1).

Furthermore, the longer duration of the hardening phase should not be underestimated, during which the piece tends to make its temperature stable and homogeneous, thus avoiding possible distortions due to the different thermal gradients in the subsequent hardening phase.

Considering the results obtained by supplier A to be optimal, we proceeded to develop a thermal cycle with the same characteristics as the standard cycle (Tab.2) but to which the following have been changed:

• the duration of the warm-up phase (extended to 180 minutes);
• the duration of the cooling phase (extended to 120 minutes);
• the duration of the hardening phase (from 30 to 90 minutes).

OPERATION	TEMPERATURE(°C)	TIME (mins)
RAMP	30 – 925	180
BOOST	925+/- 5	1200
DIFFUSION (DIFFUSE)	925+/- 5	600
COOL	925 – 825	120
HARDENING	825 +/- 5	90
QUENCH	105	20
TEMPERING	150 – 200	120-180

Once the so-called special heat cycle had been defined, the next step was to test the heat treatment directly on the pinion shaft and, with the help of supplier C, check whether the modifications made allowed values ​​to be obtained within the pre-established tolerances. The strain values ​​measured following the special treatment are shown in the table below.

ITEM NO.	BEND UNDER THE HEAD (mm)
1	0.06
2	0.1
3	0.1
4	0.06
5	0.16
6	0.07
7	0.11
8	0.06
9	0.05
10	0.08
11	0.09
12	0.16
13	0.13
14	0.04
15	0.11
16	0.03
17	0.04
18	0.05
19	0.07
20	0.07

From Tab.5 it can be seen how the modifications made to the heating, cooling and hardening phases considerably reduce the resulting deformations.

4. Conclusions on Steel Heat Treatments

At the end of the work, it is possible to affirm how appropriate control of the heating phases occurring at austenitization temperature and cooling phases at quenching temperature can limit the deformation problem. Specifically, it was possible to ascertain that the thermal cycle used for this particular component can be applied to components with similar geometry with satisfactory results. On the other hand, however, it is not possible to extend this result to components with different geometries, such as thin-walled rings (on which further studies will be carried out). However, it is reasonable to assume similar results.

Furthermore, the results obtained from the experiments can be improved by combining them with appropriate numerical models currently being developed. This process will be helpful in finding the optimal slope of the critical phases mentioned in the publication by reducing the level of deformation and limiting the duration times of the entire thermal cycle and related costs.

Find out how we apply this knowledge to our products, reading other articles from our blog.

5. Bibliography of steel heat treatments

[1] Lübben T., Zoch H.W., “Distortion engeneering – A systematic strategy to control dimensional change, European conference on heat treatment 2008: innovation in heat treatment for industrial competitiveness, 2008.