Is there a correlation between Brinell hardness and yield strength? 1

[EdDanzer]

We design and built high pressure hydraulic cylinders. On problem area is the rod to piston interface. The threads need to be large enough to handle the pull of the piston, and their needs to be a shoulder to handle the push of the piston. The area where the nut contacts the piston and the piston contacts the rod we see brinelling. Shouldn’t a higher yield material have a higher Brinell hardness? An example is cold finished 12L14, yield 60,000 psi, and 163HB and hot rolled C1045, 58,000 yield, and 201HB.

 

[Metalguy]

Briefly, hardness does not correlate with YS, only UTS.

 

[BillPSU]

Cold finished strength properties are "typical" properties based on sampling a specific diameter. The larger the diameter of a rod the surface will have the higher strength while the core will remain the original strength and hardness of the unworked material.
Point contacts such as surface texture ridges may be experiencing plastic deformation due to compressive strength yielding. Another problem would be the mating surfaces not being perpendicular to the force applied by the thread again causing plastic deformation.
Check the surface conditions of the mounting surfaces. If you are using a coldworked rod there probably is increased tool pressure in the worked area of the rod when facing the piston mounting surface on the rod and thus an out of flat condition.

 

[YLWEA]

Metalguy -
there are empirical relationships that have been established for hardness and yield strength for some steel products. Examples include yield strength of steel pipe established in a study conducted by Battelle for ASME a few years ago (see "Determining the Yield Strength of In-service Pipe" DA Burgoon, OC Chang, et.al., Battelle for ASME Gas Pipeline Safety Research Committee, 1999.)

Also, there have been some papers published relating steel weld metal hardness to yield strength. Those include (for example)
"Yield Strength from Hardness - a Reappraisal for Weld Metal" RJ Pargeter, The Welding Institute Research Bulletin, November 1978.
"Yield Strength from Hardness Data", letters to The Welding Institute Research Bulletin, by GN Rawlings, response by authors PHM Hart and RJ Pargeter, July 1979.

As for all empirical relationships, the data should only be used for materials that were within the scope of the original studies and extrapolations beyond the limits of the presented data are not advised.

 

[Metalguy]

Without reading those papers, I suspect they are attempting to relate the YS/UTS ratio to steel processing. For example, annealed 304/316 ss will have a YS/UTS of ~.3, while the YS/UTS ratio of cold-worked ss will be be more than .5--sometimes a lot higher. Same thing with annealed vs HT steel if the HT causes transformation to martensite or bainite.

Also, the YS/UTS ratio can vary widely with relatively small changes in processing, and a hardness test will not see it.

But if you know the history of the processing and have enough other data to compare it with, you can make a fairly close estimate of what the YS/UTS ratio will be.

 

[CoryPad]

The microstructures are completely different for 12L14 and 1045 steels. What is the size of the steel products, and by what methods were they produced? What ball diameter was used for the hardness test?

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[YLWEA]

I'm not advocating the use of the references I cited to establish the hardness/yield relationship of the componenets that were the subject of this thread. I only meant to illustrate that such relations can be established for steel.

In the case of the ASME study I took a quick look at the summary. It states that the study included review of 4,698 data records relating yield strength to hardness. The data were analyzied to establisha direct relationship between yield strength of the steel pipe and Rockwell B hardness. The statistical relationship was set at a lower 99% confidence bound on 0.5 percentile (99% chance that 995 or 1000 pipes would be at least as strong as the indicated conversion.) The study focuesed on pipeline steels typical of those made before 1980 having a stated grade of X52 (52 KSI SMYS) or less. The goal of the project was to help pipeline operators establish the probable strength of older pipelines for which records were limited.

 

[EdDanzer]

These cylinder see pressure spiked over 15,000 psi and frequently swell the barrels. We make 2 primary sizes, 5”OD barrels honed 4.025/4.027 ID and 6 ¼”OD honed 5.025/5.027 ID with 80,000 min yield tubing. With the tubing it seems once the yield strength is exceeded the bore is oversize. If we use contact area times yield we should not see the nut deform into the piston.
BillPSU,
The parts are CNC turned, the rods typically are 100,000 min. yield induction hardened (.050 deep case) C1045. The pistons are mostly 4” and 5” diameter and have had good luck with C1040, the nuts grade C stover.
CoryPad,
The brinell hardness quoted is from a book and does not give the diameter, but it is probably the standard steel ball.
Maybe I should rephrase the question some. Is there a difference between compressive yield strength and tension yield strength?

 

[unclesyd]

If I'm reading this right you are exceeding the elastic limit of the cylinder material if it stays oversize after a pressure excursion.

Are you designing your cylinders for 15,000 psi?

Is the 15,000 psi loads via the hydraulic system or by a shock load on the piston?

Is this a single or double acting cylinder?

It sounds like you are flexing the threads on the rod, off hand I would look for a deep thread. One way to design out is with a "SuperBolt" type connection.

 

[BillPSU]

The best answer I could find is compressive and tensile strength are the same.
The calculated burst pressure for your different cylinders is about 19,500 psi. Your margin of safety is limited in this design if you are seeing 15,000 psi spikes. The barrels of the cylinders are probably yielding when the spikes occur.
Which side of the cylinder is seeing the spike? The rod side or the base side?
The threaded connection is probably elasticly stretching if the spike is on the rod side and lifing the piston from the rod seat and hammering it down when the spike disappears. The situation of seeing this large of a spike in a cylinder would cause me to review the addition of a pressure relief valve on the lines feeding the cylinders.

 

[EdDanzer]

These pressure spikes are from work induced loads. The cylinders are used in log loading grapples, for safety they have a pilot operated check valve to lock the cylinders from opening if a hose breaks. The operating pressure cannot exceed the excavator pressure (5,000 psi) but our customers pull stumps with them to build roads, and some time hit the grapple on things when open. These loads create high spikes in both directions, damage parts include splitting the barrels, blowing the PO check out, seal failure, and cracking welds. If we increase the thread size keeping the thread from failing the piston cold forms over the rod, if the thread is smaller the rod breaks at the start of the thread. If the piston is threaded on the rod, clearance in the thread allows movement and sealing the piston to the rod is a major problem.

 

[EdStainless]

Is there no way to put dampeners or snubbers into the system?
The problem with the rod is that if you go to higher strength materials you will be up against toughness issues, since this is shock loading.

Even though compressive and tensile yields are not strictly the same it usually doesn't matter. If you are using reported engineering yield strength to design remember that this is at 0.2% strain. (4.026" id becomes 4.034") If you want less deformation you need to de-rate.

My biggest concern with your system is making sure that you are comfortable with what part will fail first. There has to be a 'weak link' and you need to control it so you can minimize the risk with failures.
A 20ksi burst rating on a 5ksi system isn't much of a safty factor when you know that you will see spikes.

= = = = = = = = = = = = = = = = = = = =
Corrosion never sleeps, but it can be managed.

http://www.trenttube.com/Trent/tech_form.htm

 

[BillPSU]

EdDanzer
In the aerial device and digger derrick industry they are faced with some of the same issues you are facing. Loss of pressure to the cylinder and overloads. The solution utilized at previous company was a move away from PO check valves and toward counterbalance valves. Sun Hydraulic counterbalance valves were used directly in the cylinder.

http://www.sunhydraulics.com/cmsnet/cbcsp.aspx?Cat_Level_0=1&Cat_Level_1=2&Cat_Level_2=6

 

[EdDanzer]

BillPSU,
If we ship cylinders with counterbalance valves, the customer will replace them with PO checks. Our customers are willing to pay for maximum performance even if some failure occurs.
EdStainless,
There is not an excavator hydraulic cylinder built with a barrel that will handle a 15,000 psi pressure spike. Excavator boom cylinders all have port relief valves to limit the cylinder pressure to 6500 psi max. Most we have worked on only have a 2 to 1 safety factor.

The reason for all the questions. We sold our original design information to a company we made cylinders for. They have been subcontracting out the manufacturing, but quality problems have shown up that we have not had. We are staring to manufacture cylinders again and need to understand why these problems exist.

Thanks to all that have responded,
Ed Danzer

 

[EdDanzer]

We did a Brinell test on our competitors failed piston and it is softer than the material we use, 183 BNH instead of 200+ BNH.
I would like to do a non-linear FEA on the interface between the rod and piston, but need accurate values for C1040, C1045, 4140, 4150, and 4340. Since this is my first attempt at non-linear, I think the information needed is Young’s modulus, and the stress strain curve?
I will be posting on the Finite Element Analysis Forum also.

 

[unclesyd]

This website will have a lot of the information you will need.

Go the down the Technical Content to the Bar Database.

http://www.steel.org/autosteel/bar_rod/index.htm

 

[EdDanzer]

Unclesyd,
Thanks for the link. There is no information on C1040 4150 or 4340, but the information on the grades they have is excellent.
I’ll give you a star for the help.

Thanks again,
Ed Danzer

 

[TVP]

Ed,

MIL-HDBK-5H has stress-strain curves for 4340 (chapter 2.3). I have no idea where to get one for 4150. It usually costs less than $100/sample to have tensile testing performed by accredited laboratories. Use the following link to access MIL-HDBK-5H from Granta:

http://www.grantadesign.com/cgi-bin/stat-tree.cgi?src=MIL5-H.xml&br=0