Skip to main content

Good morning!

Can anyone recommend a good quality, high-accuracy steel tape to use for calibration course set up?

I've been reading up on tape accuracy ( and it seems like going with a class 1 tape would best (+/- 1cm on 100 meters) but I can't seem to find one in the wild.

I'm just curious what others are using.

My best,


Original Post

Replies sorted oldest to newest

Tim, somewhere in the archives of this forum, there is a body of info on steel tapes. One report that interested me is the revelation that decent-quality steel tapes sold in hardware stores aren't terribly accurate. Dave Katz paid to have his steel tape checked by NIST. It was found to be a little short - I do not remember how short.

I looked into getting a "certified accurate" 300-foot tape, but the cost was several hundred dollars. Had the cost been say, $150.00, I would have purchased it.

My assumption is that the magnitude of the length error in well-known brand hardware store steel tapes isn't typically something that would cause a calibration course to be so far off that it would materially affect a certification measurement. Otherwise, we'd have a lot of re-certification work on our hands...

Can anyone else weigh in on Tim's question?

I tend to agree with Lyman's last paragraph. I recently bought a 100ft steel tape because it's easier to take with me on every course measurement than my old 100-meter tape (very heavy). I pounded a nail in the pavement, hooked the end of both tapes on it, and took a measurement about 100 feet away. The difference between the two tapes was quite small, and definitely smaller than the difference I saw depending on whether I pulled with a little less, or a little more tension.

I can think of several sources of error in laying out a calibration course:

1) Accuracy of steel measuring tape itself.

2) Accuracy of tension applied to tape.

3) Accuracy of tape temperature.

4) Accuracy of holding the zero point of the tape at the mark as it is being tensioned by the other person.

My guess is that the combination of #2 thru #4 are a good bit bigger than #1.

All, thanks for the measurement suggestions and also the link to the Keson tape.

I spent the morning doing a bit more research and ended up on the NIST:

It's my understanding that the Tape manufacturers regularly certify their production against the NIST.  The page above has a link to the NIST Handbook 44 which indicates that a metal tape between 81 and 100 feet that is used for commerce (buying and selling goods) should have a tolerance of 1/4inch. Assuming your tape is off by 1/4 inch for each 100 feet, a 1000 ft calibration course could be off by 2.5 inches. That's 13.2 inches per mile or 81.84 inches for a 10K  --- or 2.08 meters of potential error. 

Regarding tape temperature, the USATF course measurement manual says you should use a non-contact thermometer and measure the actual tape temperature. The interesting thing here is that a 12F increase in temperature from 68F to 80F results in only a .9 inch correction on a 1000ft calibration course.  Ignoring that would result in less error than is possible from an inaccurate tape.

Of course, the specific tape you buy may not be 1/4 inch off. It could be exactly as stated or some number between zero and 1/4 inch which would result in a potential error less that 2.08 meters.

At any rate, I appreciate the input! This is a fun and certainly interesting topic and I'm looking forward to starting to measure.

My best, Tim


The link below shows the specs for Class 1 and Class 2 measuring tapes.'s%20compare,accurate%20to%20within%200.7mm.

The difference between Class 1 and Class 2 is about 30mm over 300m, which is about a third of 1 Jones count. This is much less than the variability we see in multiple calibration rides, and an order of magnitude less than we often see between pre and post calibrations. I don't think the added expense and trouble of finding a Class I tape is worth it.

Thanks for sharing that table and working through this.

Based on that and assuming that all other potential errors in the measuring process were zero, using a class II tape with a tolerance of +/- 20.3mm on 100m, would result in a potential error of +/- 2 meters on a 10k course. So, somewhere between 998m and 10,002m.

Reading back through the manual, it sounds like the short course prevention factor, accounts for any of the above error: "This lengthens the course by one meter per kilometer or 5.28 feet per mile."  That's an addition of a whopping 10 meters on our 10K!

Still, I would think we would want to be as accurate as we could.

I did find that you can buy calibrated tape measures traceable to the NIST. They would come For example:

Thanks again!

My best,


Add Reply

Link copied to your clipboard.