Marking Legibility (7.1.2 of IEC 60601-1): Size of Letters

As a test engineer I've already come across a couple of cases with very small letters. According to IEC 60601-1, marking should be visible at 1m by a person with 20/20 vision, in light varying from 100-1500 lx.

Many people already know this test is very subjective, so I did some research to try and get some more objectivity. Here are the results, just for reference.

From wikipedia and other sources, the definition of 20/20 vision is the ability to differentiate lines which are 1 "minute angle" apart. For a distance of 1m, this equates to 0.29mm, or 1.45mm height for the classic E shape on an eye chart.

So, any marking less than 1.5mm height clearly fails the standard.

But there are also modifiers to this basic calculation:

If the marking is mixed capitals and smalls, then you would need to make sure that the smalls are 1.5mm (making the capitals around 2.0mm).

If the font is heavy bold or another font that makes the gaps between appear smaller (e.g. an E with the legs not evenly spaced) the size should be increase to make sure gaps are at least 0.3mm.

Finally, the marking should have good contrast and good quality printing (sharp edges), otherwise again the size should be increased to compensate. For example, if the printing resolution is limited to 0.05mm, then you would need to increase the size to around 2.0mm to make sure the gaps are 0.3mm.

All in all, it seems to me that:
<1.5mm is clear fail
1.5 - 2.5 is subjective and should be checked
>2.5mm is a clear pass

The standard mentions a range of 100-1500 lx. Obviously, the low end seems to be the worst case. The higher value of 1500 lx seems to be more important for indicator lights, as per IEC 60601-1-8 (alarm indicators).

Peter Selvey said:

The standard mentions a range of 100-1500 lx. Obviously, the low end seems to be the worst case. The higher value of 1500 lx seems to be more important for indicator lights, as per IEC 60601-1-8 (alarm indicators).

Click to expand...

I think that 1500 lx would also apply to any information shown on a display (LCD / CRT) screen.

When it came to this clause I set up an adjustable up-lighter in a small darkened room so that I could set the lighting to the required 100 lx and then got 5 or 6 colleagues to correctly read/verify the smallest markings (this equipment didn't have a display screen).

Yes, visibility of displays is important, and could be called up for example under 7.4.2 (controls).

It would be good if the standard made it clear this applies to indicator lights and displays, at least for higher risk devices (e.g. Class IIb devices in Europe) because wording of the clause could be interpreted to apply to fixed marking only. IEC 60601-1-8 directly calls out visibility of indicators under varying light conditions, but this is only for alarms.

Also, 1500 lx might not be enough if a high risk device is intended for outdoor use. Recently when doing sound level tests outside, I found that indicator lights and displays were difficult to see. When I measured the light it was well above 10000 lx, and matches with literature which says the light levels can be 10000 ~ 30000 lx outside of direct sunlight. So an AED, transport incubator, patient worn insulin infusion pump might need brighter displays and indicators, or better still automatic brightness control.

But this obviously depends on the intended user, right? Meaning, the info would be right if you would use the test as defined in the standard (which was based on an average user with good vision).

The test specifies that:

The observer has a visual acuity of 0 on the log Minimum Angle of Resolution (log MAR) scale or 6/6 (20/20), corrected if necessary.

Click to expand...

So in effect this creates an objective requirement independent of real people. The MAR, 6/6 or 20/20 are all defined as the same thing, the person can perceive a gap of 1 minute arc = 1/60 degrees. For a distance of 1m, the gap is:

sin (1/60 degrees) * 1000mm = 0.29mm

Assuming a normal E shape, there are 3 "legs" and 2 spaces. Assuming they are evenly spaced, perfect printing, high contrast (black on white) means the height should be 1.45mm.

This is all well documented in literature, although the sizes are usually shown for the classic 20 feet or 6 meters. All I have done is to re-calculate for 1m as required in IEC 60601-1:2005.

It seems to make sense to simplify the test and just specify a minimum height (e.g. 2.5mm), but font, contrast, printing quality etc could still make something unclear. So, I guess the way the requirement is written is perhaps the best way.

For a conclusive test, the marking could be photographed with a reference scale, and determined if any gaps smaller than 0.29mm exist in the marking.

Ok, but if my intended user is not expected to have a 20/20 vision (older people in a home care setting or something like that), is this

<1.5mm is clear fail
1.5 - 2.5 is subjective and should be checked
>2.5mm is a clear pass

Click to expand...

still true?

I think that's a separate issue.

The point here is to try and and get some objectivity behind 7.1.2 criteria, as stated in the standard.

Currently I have a project where some characters (e.g. model name) are 2.0mm and others (e.g. rated voltage) are 1.2mm high. The smaller ones seem to be not visible.

To test properly, I need to get a pair of glasses that adjusts my eyes to exactly 20/20 vision. To do it properly, you really have to have an ophthalmic device to measure your vision at the time of test. Eyes vary with age, time of day, temperature, stresses (I know, because I'm on the edge of needing glasses, sometimes I don't need them, other time I do).

Also you need to be careful that your vision is not over corrected - usually glasses are designed to achieve 20/20 or better, and many people naturally have better than 20/20. These people would need a special pair of glasses to downgrade their vision to 20/20.

So ... you can see why I'm looking for more objectivity.

Your research result is close to the size of warning markings specified in IEC 61010-1 (I have some experience on the IVD equipment).

The 61010-1 says that:

The size of warning markings shall be as follows.
a) Symbols shall be at least 2,75 mm high. Text shall be at least 1,5 mm high and contrast in colour with the background.
b) Symbols or text moulded, stamped or engraved in a material shall be at least 2,0 mm high.
If not contrasting in colour, they shall have a depth or raised height of at least 0,5 mm.

Click to expand...

I could not understand very well for your calculation for E shape.

Assuming a normal E shape, there are 3 "legs" and 2 spaces. Assuming they are evenly spaced, perfect printing, high contrast (black on white) means the height should be 1.45mm.

This is all well documented in literature, although the sizes are usually shown for the classic 20 feet or 6 meters. All I have done is to re-calculate for 1m as required in IEC 60601-1:2005.

Click to expand...

Could you explain it more detailedly?

Thanks for the IEC 61010 reference

The E character seems to be often referred to in literature.

The character has 4 strokes: 1 vertical and 3 horizontal.

The gap between the horizontal strokes is the usual point of reference, which should be 0.29mm for the 20/20 test at 1m.

Assuming the width of the strokes is the same as the gap, there are a total of 3 strokes and 2 gaps, so the calculation is 0.29 x (3 + 2) = 1.45mm.

But you can see that it depends on the font: EEE EEE EEE

Why did you assume the width of strockes is same as the gap? It is obvious that the width is far less than that gap.