What are the rules on significance of digits in numbers in IEC/ISO standards?

Loekje

Involved In Discussions
Hi,

Maybe a lack of search skills, but I cannot find any indication or referral in the standards like the 60601-1 we have right now (we cannot afford to buy them all) on how to interpret the significance of the last digit of values.

My late science teacher hammered on the notion that the significance of the last digit relates to the precision of measurement. A figure of 10.0 could mean a value between 9.95 and 10.05.

If a maximum allowed patient current of 50uA in SFC for CF is defined, does that mean that a measured 50.5uA still complies? This can be important to know because then for instance a 100k 1% tolerance resistor would be allowed to limit a 5V (ehh, 5.00000) voltage. For sake of argument this is disregarding all implications of measurement accuracy/precision, definition of resistor tolerance and all MOPP implications.

Does a collateral standard exist that addresses the interpretation of numbers? Or is it implicitly set in the way the tests must be performed, like requirements on the accuracy of the test setup?
Opinions are plenty available, but can anyone point me to the right standard(s) that have this in black and white? :evidence:


Kind regards,
Loek
 

Stijloor

Leader
Super Moderator
Hi,

Maybe a lack of search skills, but I cannot find any indication or referral in the standards like the 60601-1 we have right now (we cannot afford to buy them all) on how to interpret the significance of the last digit of values.

My late science teacher hammered on the notion that the significance of the last digit relates to the precision of measurement. A figure of 10.0 could mean a value between 9.95 and 10.05.

If a maximum allowed patient current of 50uA in SFC for CF is defined, does that mean that a measured 50.5uA still complies? This can be important to know because then for instance a 100k 1% tolerance resistor would be allowed to limit a 5V (ehh, 5.00000) voltage. For sake of argument this is disregarding all implications of measurement accuracy/precision, definition of resistor tolerance and all MOPP implications.

Does a collateral standard exist that addresses the interpretation of numbers? Or is it implicitly set in the way the tests must be performed, like requirements on the accuracy of the test setup?
Opinions are plenty available, but can anyone point me to the right standard(s) that have this in black and white? :evidence:


Kind regards,
Loek

Net een PM gestuurd!
 

Mike S.

Happy to be Alive
Trusted Information Resource
IMO, if 50 uA is the spec max, that means </= 50.000000.....uA. You assume the zeros extend on forever.

The only place I can think of off-hand where this is explicitly stated vs. just "understood" is in the GD&T world with ASME Y14.5
 

Michael_M

Trusted Information Resource
Hi,

If a maximum allowed patient current of 50uA in SFC for CF is defined, does that mean that a measured 50.5uA still complies?


In this case, max is called out so 50uA max means 50.1 or 50.000001 is out of spec (if you could accurately measure that).
 

Al Rosen

Leader
Super Moderator
Does a collateral standard exist that addresses the interpretation of numbers? Or is it implicitly set in the way the tests must be performed, like requirements on the accuracy of the test setup?
Opinions are plenty available, but can anyone point me to the right standard(s) that have this in black and white? :evidence:
According to IEC60601-1
7.4.3 Units of measurement
Numeric indications of parameters on ME EQUIPMENT shall be expressed in SI units according
to ISO 80000-1 except the base quantities listed in Table 1 may be expressed in the indicated
units, which are outside the SI units system.
For application of SI units, their multiples and certain other units, ISO 80000-1 applies.
I believe annex B addresses rounding.
 

Pads38

Moderator
The Technical Report 'IEC TR 62354 - General testing procedures for medical electrical equipment' includes, at clause 11.2, a list of acceptable accuracies of test equipment during the 60601-1 tests.

eg

Voltage up to 1000V (dc to 1kHz): +-1.5%
current up to 5A: +-1.5%
Leakage current: +-3.5%
 
Top Bottom