Calibrating scales

[Steve Ekstrand]

CASOC has club scales. The are longacre wired scales. Pretty standard set with setup stands.
Heavy and bulky....

Here's the question...

Some have said that they are read quite different than Topeka's scales. I weighed the car yesterday and it was quite a bit lighter (35lbs) than it weighed last time we corner weighted the car with these scales.

So, how do we calibrate longacre scales???

Not that I mind my car weighing 2024lbs... I just don't think it does.

[Bill Schenker]

Isn't the answer obvious? Put a series of know weights on it; if they all read correct, your golden, if not, but they still have a liner deviation, your still good - just calculate/extrapolate what it would be for your expected weight.

All IMHO.

[Bob Beamesderfer]

The problem is whether there's a linear deviation or not, and whether it can be determined. The scales can be zeroed on the display, but is that really zero? If you put 100 lbs. on and reads 100.3, that's pretty accurate. But if you put 500 lbs. on and it reads 505 ...

Sounds like some questions for Longacre.

[Will Kalman]

The problem is whether there's a linear deviation or not, and whether it can be determined. The scales can be zeroed on the display, but is that really zero? If you put 100 lbs. on and reads 100.3, that's pretty accurate. But if you put 500 lbs. on and it reads 505 ...

Sounds like some questions for Longacre.

I was going to say the same thing but Bill did say a "series of weights" which would allow one to understand the nature of the error as he also pointed out. Oh, and if you have nothing on the scales and zero it, that's zero - it's the tare weight you're interested in.

[Bob Beamesderfer]

The problem is whether there's a linear deviation or not, and whether it can be determined. The scales can be zeroed on the display, but is that really zero? If you put 100 lbs. on and reads 100.3, that's pretty accurate. But if you put 500 lbs. on and it reads 505 ...

Sounds like some questions for Longacre.

I was going to say the same thing but Bill did say a "series of weights" which would allow one to understand the nature of the error as he also pointed out. Oh, and if you have nothing on the scales and zero it, that's zero - it's the tare weight you're interested in.

Yeah, good point. Zero is .... That series of weights will need to approach a number close to a typical corner weight, me thinks. Maybe 75%?

[Thomas Smith]

I think for corner weighting you could simply stand on each scale pad and verify if the readings are the same. If they are the same or within a couple pounds, the scales should be fine for corner weighting.

[Steve Ekstrand]

I'm 295-305 on any given day and dress. I'll have the scales out on Friday and I'll try standing on them to compare each pad. I guess I could also bring my bathroom scale out to compare. Its a pretty high end one that is supposed to read with in +/- .1%.

[Jonathan Lugod]

I'm 295-305 on any given day and dress. I'll have the scales out on Friday and I'll try standing on them to compare each pad. I guess I could also bring my bathroom scale out to compare. Its a pretty high end one that is supposed to read with in +/- .1%.

shoot... buy 4 of those scales and put the metrosexual on them.

[Jayson Woodruff]

Hey, that was pretty smart Bill.

To 'do it right', you do just what Bill said. If the system has an adjustment then trim it out with known rates. If not, Chart known rates vs displayed weights to make a 'factor'. It's unlikely the discrepancy is non-linear.

Jay W

Isn't the answer obvious? Put a series of know weights on it; if they all read correct, your golden, if not, but they still have a liner deviation, your still good - just calculate/extrapolate what it would be for your expected weight.

All IMHO.

[Bill Schenker]

Hey, that was pretty smart Bill.

To 'do it right', you do just what Bill said. If the system has an adjustment then trim it out with known rates. If not, Chart known rates vs displayed weights to make a 'factor'. It's unlikely the discrepancy is non-linear.

Jay W

Isn't the answer obvious? Put a series of know weights on it; if they all read correct, your golden, if not, but they still have a liner deviation, your still good - just calculate/extrapolate what it would be for your expected weight.

All IMHO.

I have my moments.

[Steve Ekstrand]

Any ideas on "known" weights???

[Christine Grice]

Any ideas on "known" weights???

Right off the bat I can think of a weight lifting weight, but I don't know what kind of tolerance they hold on gym equipment.

[Will Kalman]

The only time you really need really accurate absolute weight is when you're in a class where you can be disqualified for being underweight. And at those times, it's prudent to weigh the car on the official event scales beforehand.

Other than that, as Tom Smith said, you only need the pads to be equally (in)accurate.

For setup, it's not like you're going to pick different springs because you're +-10 pounds.

[Steve Ekstrand]

yeah but I lightened the car and I want to race scales!

[Tadd Ekstrand]

Any ideas on "known" weights???

Ask your PT if you could borrow a dumbbell.

[David Avard]

Just don't do what I once saw SCCA staff members do. They have two 50lbs weights that they use to calibrate their scales. One weighed 48 lbs, and so did the other one. So, according to them, the scales were 2lbs off. They never did put both weights on a scale together to find out if they were indeed 2lbs off, and not 4%. And I'm not sure if they checked all 4 scales, either.

Oh, and Frank Stagnaro got dsq'd for Saturday at that Atwater Tour for being 6lbs underweight on the SCCA scales.

That is of course more proof that you should weigh beforehand on the "official" scales.