Thursday, December 1, 2016

The 3 Key Changes to ISO 9001 Compliance and What this Means for Force and Torque Instrumentation






1.  Documented Procedures are No Longer Required.   - Documented procedures are helpful in regards to how to calibrate equipment and they may be useful to you.  However, you are no longer required to keep a physical manual on site.  There is going to be a lot more risk when equipment is not used properly or written procedures are not documented and followed.   There is an alternative to some of this with verifying the proper measurements via calibration, repeatability and reproducibility studies and proficiency testing. In addition, calibration data (records) will still be required (ISO9001 Clause 7.1.5.1 (b) are maintained to ensure their continuing fitness for their purpose.
The organization shall retain appropriate documented information as evidence for purpose of the monitoring and measurement resources.).


Image above is a normal distribution showing 50.1 % probability of of calling the measurement good, when it is not. The uncertainty of the measurement must be considered when evaluating risk.


2.   Preventative Action Clause No Longer Exists - Everything has been focused on risk-based thinking.  This requires the organization to start taking actions for the following:

Avoiding risk, both consumer and producer should be considered.   Measurement Uncertainty and the resolution of the Unit being tested should be considered when making statements of compliance. (ISO 9001 Clause 9.1.3 Analysis and Evaluation: The organization shall analyze and evaluate appropriate data and information arising from monitoring and measurement.
The results of analysis shall be used to evaluate;
a) conformity of products and services;
e) the effectiveness of actions taken to address risks and opportunities;

Also, see ISO 9001 Clause 10.2 Nonconformity and corrective action



Eliminating the source of risk - this is where SPC and good verification standards as shown above, may help.

Changing the likelihood or consequences of risk.

Making informed decision regarding elimination of risk versus managing the risk




3. Top Management is now held more responsible than before - The expectation is for Top Management to start risk-based thinking and may bear the responsibility for failures.    Remember that device that should have been calibrated and was not or its calibration interval was extended to save money? If there was no risk based evidence to support the decision, top management may be accountable. Use of Control Charts (SPC) is one of the tools to manage risk and calibration interval.   



Morehouse offers a SPC workshop that focuses on mitigating and managing risk in calibration.    Our full training schedule can be found here.



written by Henry Zumbrun with help from Dilip Shah
Dilip Shah teaches our SPC course
www.mhforce.com



Monday, November 7, 2016

Top 5 Common Force Measurement Errors


Top 5 Common Force Measurement Errors


At Morehouse we strive to replicate the loading conditions a customer may be using.  Some other labs, may not follow this process and the additional measurement uncertainty from not replicating use, may create significant error source.  Below are the top 5 common force measurement errors that may invalidate the calibration results.

#1 Pin Size is Critical with Tension Links
  • Errors of up to 20 times manufacturer's specification are possible, if the proper pin size is not used.
  • Sending the pin used with the instrument is recommended, to best replicate use during calibration.
  • If the pin is not available, Morehouse will list the pin size used to calibrate the device on the certificate of calibration.
  • If a pin of different size or material is used, it is quite possible that the instrument will not meet manufacturer's specification.

#2 Loading the Shoulder vs. Thread Loading
  • Load Cells will behave differently depending on whether they are shoulder loaded, or thread loaded.
  • Varying the thread depth on a shear-web type load cell can result in measurement errors of 0.5 %, or more.
  • Locking an integral threaded adapter into shear-web cells is the recommended practice to avoid these errors. If an integral adapter cannot be locked into place, we recommend the cell be shoulder loaded, and any adapters placed between the cell and force applied also be supplied for calibration. 
Note: Non shear-web type cells may have additional errors from shoulder loading. The end-user's adapters should be supplied for calibration, we have observed errors up to 5 % when different adapters are used during calibration.


#3 Load Cell Bolting, Thread Class, and Applied Torque
  • If a load cell is bolted to base or adapter of similar size, we recommend sending the cell attached to it's base for calibration.
  • If the load cell is attached to a larger machine or device, error sources from different classes of threads, materials, or bolting procedures (including torque applied) can be quite large. The bolts used to hold the cell in-place should be sent with the cell, and if possible, even a piece of similar material and flatness as the machine should be sent to best replicate use
  • Variability between bolting techniques, material hardness, and flatness can all contribute to significant error, which can be difficult to identify away from the point of use.


#4 Top Adapter Hardness and Flatness
  • Top adapters with different hardness values may affect the strain level in the load cell column or web, and result in different measurement outputs. We have observed errors of up to 0.15 % from varying just the material on top compression pads.
  • Adapters and bases that are not flat may produce additional errors.  We have conducted several tests, and found repeat-ability errors to be about three times higher when the surfaces interfacing with the load cell are not flat.
  • We highly recommend the end-user send us any top adapter which is used with the load cell.



#5 Not Following Published Standards 

  • Our last blog dealt with the top 3 ASTM E74 load cell calibration mistakes  and can be found here. There are several examples where calibration labs simply do not follow what is published in the standard.
  • We have observed several laboratories violating the ASTM E74 standard.  Specifically in not taking a non zero test point below 10 % and assigning a loading range below the first calibrated test point. Per Section 7.2.1 of ASTM E74-13a states  “In no case should the smallest force applied be below the lower limit of the instrument as defined by the values: 400 x resolution for Class A loading range & 2000 x resolution for Class AA loading range”    Per Section 8.6 of ASTM E74-13a  “The loading range shall not include forces outside the range of forces applied during the calibration”   This means, Zero cannot be the first test point.
  • There are other documents like ISO 17025 that are not followed.   We have seen numerous calibration reports mentioning trace-ability to N.I.S.T. without any consideration of measurement uncertainty.    "Traceable to N.I.S.T." is not correct.  The proper statement should be traceable to SI Units, through N.I.S.T.    This topic will be further expanded upon, in a future blog.    








written by Henry Zumbrun
www.mhforce.com

Tuesday, November 1, 2016

Top 3 ASTM E74 Load Cell Calibrations Mistakes

We are writing this post to clear some common ASTM E74 misconceptions.   We have written two early blogs on the ASTM E74 standard ASTM E74 Simplified, Calculating CMC using ASTM E74 standard and have a slide share presentation detailing the ASTM E74 standard. 

This blog post is to help clear the three top misconceptions commonly observed in industry.


Misconception #1: Zero can be used as the first calibrated test point. 

This is not true in anyway possible.    In the ASTM E74-13a standard the following sections point to this not being allowed.


Per Section 8.6 of ASTM E74-13a  “The loading range shall not include forces outside the range of forces applied during the calibration”
Per Section 7.2.1 of ASTM E74-13a states  “In no case should the smallest force applied be below the lower limit of the instrument as defined by the values: 400 x resolution for Class A loading range & 2000 x resolution for Class AA loading range” 

Do Not assign a Class A or Class AA loading range below the first non-zero force point.   Note:  We have observed numerous labs violating this rule!  If your loading range is less than the first non zero test point, your calibration provider is not following ASTM E74.






Misconception #2: Designation of loading ranges.  

Some labs think a Class AA loading range can be used to assign a Class AA loading range.  This is not true.  

A force measuring device with a Class AA loading range cannot assign another Class AA loading Range; A force measuring device with a Class A loading range cannot assign another Class A loading Range.       

Do Not Assign a Class AA loading range, unless you are calibrating with primary standards accurate to better than 0.005 %


Do Not Assign a Class A loading range, unless you are calibrating the device using a secondary standard that was calibrated directly by primary standards.




Misconception #3:  A Calibration interval of one year is required for all force measuring devices not meeting the stability criteria set forth in ASTM E74.

(Note:  The maximum calibration interval is two years and this includes any force device)  

Calibration Intervals Per ASTM E74-13a:

New Devices are calibrated at a one year interval per Section 11.2.1  "New devices shall be calibrated at an interval not exceeding 1 year to determine stability" 

"The calibration intervals for force-measuring instruments and systems used as secondary force standards or for the verification of force indication of testing machines shall be calibrated at intervals not exceeding two years after demonstration of stability supporting the adopted recalibration interval" 

If the force measuring instrument does not demonstrate changes in the calibration values over the range used during the calibration of more 0.032 % of reading for Class AA loading range or 0.16 % of reading for a Class A range, A two year calibration interval can be assigned.


Per Sections 11.2.2 Devices not meeting the stability criteria of 11.2.1 shall be recalibrated at intervals that shall ensure the stability criteria are not exceeded during the recalibration interval.   This means the user needs to shorten the calibration interval to ensure the device will meet the stability.   This could mean 16 months or it could mean 10 days.  It all depends on the quality of the instrument.

For Class AA force measuring devices ASTM recommends in note 16 - " For secondary force standards, it is recommended that cross-checking be performed at periodic intervals using other standards to help ensure that standards are performing as expected."   If you cannot cross check your instrumentation, it may be best practice to have the force measuring instrument calibrated on a periodic basis.

This  post covers the basics.  Anyone calibrating in accordance with the ASTM E74 standard should  purchase a full copy of the standard here http://www.astm.org/Standards/E74.htm
Morehouse Developed a Calibration and Measurement Capability worksheet for instruments calibrated in accordance with the ASTM E74 standard.   This sheet can be downloaded at  http://www.mhforce.com/s/CMC-CALCULATIONS-FOR-FORCE-MEASUREMENTS.xlsx



written by Henry Zumbrun
www.mhforce.com

Thursday, October 6, 2016

Excellence In Innovation - Morehouse Instrument Company 2016 Finalist



Morehouse Instrument Company was honored to be selected as a finalist for Excellence In Innovation.

Excellence In Innovation - A culture of innovation is necessary to seize greater opportunities in the global economy.   Finalist were selected for having a strong record and solid reputation for developing new products, adopting new technologies, and improving manufacturing processes leading to customer retention and expansion into new markets.

Morehouse is excited to be nominated for this award and has just started to really innovate.   We believe in our people and feel very privileged to have such an excellent staff, as well as some of the best customers, a company could possibly hope for.  

We have just started on a long continuous improvement journey to provide our customers the best possible service, we can offer.   Morehouse welcomes the future, and looks forward to announcing some new products, and service upgrades soon.





.


written by Henry Zumbrun
www.mhforce.com

Saturday, October 1, 2016

AC Versus DC (mV/V) Differences in Load Cell Meters Using a Load Cell Simulator



There is a difference in output from Alternating Current (AC) measurements and Direct Current (DC).  To show the difference, we used a load cell simulator on two different meters.   We used a simulator that was tested at the National Institute of Standards and Technology (N.I.S.T) as the reference.  This simulator was utilized to accurately replicate the excitation and output response of a load cell when connected to the meters in the experiments.  On the DC meter side, a Fluke 8505A Reference Multimeter was used, and on the AC side, a HBM DMP40 Precision Measuring Instrument was used.   The differences between the simulator set point value and measure values by the meters are reported in the table below. In this table, the first column represents the set point values.

Note: If we wanted to standardize a Morehouse 4215, or Morehouse DSC meter, we would use the N.I.S.T. values.  At -3.00000 mV/V we would enter -3.00010 as we want to standardize the meter to repeat the N.I.S.T. value of -3.00010 when -3.00000 set point, is selected.
Looking at the test data above, it appears the difference between AC and DC mV/V can be quantified between these two very high end meters and the difference is about 0.003 %. As depicted in the chart below, the DC meter output consistently involved higher differences when compared to the AC meter. 


The importance of this blog is to show an AC meter cannot be interchanged with a DC meter as the difference between AC and DC measurements, is most likely not linear.   If a lab is using a DC or AC meter as a reference, the measurement traceability can only be derived from the type of current used by the reference lab.  AC and DC meters are not interchangeable and one cannot be substituted in lieu of another one without requiring calibration of the entire system.


Note: This blog is comparing two very different meters and does not represent what is likely to happen in all AC versus DC comparisons.  The point of this blog is to demonstrate that any indicator being substituted may need to be thoroughly tested to know the additional contribution to measurement uncertainty and that AC meters may produce entirely different results than DC meters.    



written by Henry Zumbrun
www.mhforce.com

Wednesday, September 14, 2016

Battery life and voltage may impact your calibration results

It is becoming more and more common to have portable devices that use batteries.   This can be a great benefit to someone who wants ultimate portability.   However, this portability can come with additional errors, if the portable device is not a true ratio metric device, or if the batteries voltages discharge below the allowable operating range. In such cases, the output of a force measuring device may change based upon the batteries conditions. Knowing these issues may exist, Morehouse decided to do some testing on a Crane Scale submitted for calibration (Figure 1) ...   


Figure 1: 250K Crane Scale being calibrated in our 5 MN UCM

A 250,000 lbf crane scale was submitted to Morehouse laboratory for calibration along with a handheld load indicator.   The meter uses three AA batteries and for this test, we chose not to calibrate with a new set of batteries.   We decided to do the calibration exactly how it was submitted to the laboratory for calibration.   We exercised the crane scale and recorded the "As Received" calibration values (Figure 2).    We did not move the device from the original position and simply replaced the batteries with new ones.   The test was repeated and the second set of data is displayed  in the "As Returned" column in Figure 2.  


Figure 2: Calibration Results

As Figure 2 indicates, we observed a 700 lbf difference at the maximum capacity point.   This equated to a 0.28 % difference in load reading when the batteries were replaced with new ones.   The issue was that the device submitted for calibration had a tolerance of +/- 0.1 % of Full Scale or +/- 250 lbf.  The Morehouse Calibration and Measurement Capability (CMC) for this test was  59 lbf (This includes contributions from the expanded uncertainty of our system of 13 lbf and the resolution of the UUT which was 100 lbf).    For the end user, there would not be a good way to determine when the device went out of tolerance as these numbers clearly show the device to be "In Tolerance" with a new set of batteries.   Chances are the device will stay "In Tolerance", until the batteries discharge to a certain point over time.    The problem is, it is very difficult to know the exact voltage level where this will happen and the device will once again go out of tolerance.     Monitoring this effect could be very difficult for the end user as frequent testing of the battery voltage may not be practical, nor may be replacing the batteries after several hours of use.   This problem is clearly one the end user will have to deal with.    Luckily, not all portable devices have these issues.  There are several portable instruments that maintain accuracy specification, display a low battery warning, or will cease to operate when voltage is lower than the predetermined range.    



In an effort to produce more confidence in our measurements, Morehouse has adopted a new policy to calibrate instruments with a new set of fully charged batteries.  These batteries are shipped back with your instruments, as well as any batteries provided.    Most instruments will operate fine with a lesser charge; the word “most” is what concerns us.   The Morehouse mission is to be regarded as the best independent force calibration resource in the world.  In keeping with our mission, Morehouse provides a new set of batteries to ensure we can provide meaningful measurement results with the lowest uncertainties possible.   


Figure 3: Charge Levels of Batteries
Remember, confidence in your test and measurement results starts with your calibration provider.

We will be running an additional blog post on effects of 5 Volt versus 10 Volt excitation voltage as well as post on AC versus DC meters before years end.


Written by Henry Zumbrun

Sunday, August 21, 2016

Load Cell Alignment Plugs Proper Use

Morehouse load cell alignment plugs are used to help center the load cell in calibrating machines. They are used in combination with the other adapters below.   More information on load cell adapters and the mounting accessories can be found here.

Morehouse Reference Standard Mounting Kits for Universal Calibrating Machines


Thread is past flush and into the cell.  


When using your alignment plugs that thread into the bottom of your load cells, make sure they are threaded flush to the bottom of the cell.  Once they are flush, thread the adapter an extra turn into the cell.  You want to make sure that none of the threads are exposed below the base of the cell.   If there is a thread or more exposed, the load will be generated through the internal threads of the cell and not its base.  This will result in an additional calibration error of about 0.012 %.  It will often result in damage to the alignment plug.


Alignment plug is not threaded past the base of the cell.

Written by Henry Zumbrun with help from the calibration lab technicians,  who have received numerous load cells with damaged alignment plugs.