The real scoop on what happens when you order an Anvil Frame Fixture…and some other stuff too.
Contrary to popular belief, our frame fixtures are not sitting in a giant warehouse all boxed up waiting for a shipping label. Nope, we don’t have that much room so they’re sitting in pieces on the shelves awaiting your order. Once we receive your order, we pull the fixture parts from the shelves and then assemble them. Once the fixture is fully assembled, it’s set up on one of our surface plates and prepared for proofing.
“Proofing” means that we measure the fixture’s critical dimensions and then make any corrections to bring it to our specifications, in the case of the Super Master and the Journeyman frame fixtures this is .003″ across its range of adjustment. We start the proofing process by determining the actual bottom bracket standoff as installed on the fixture and then match all other towers to that standard. We do so by measuring the cone diameters and the rear dropout dummy axle holder and recording their centerlines. Using these numbers we determine the target height for each tower and then compare that target height to the actual height we measure on the fixture.
I should note here that all our towers are machined .005″ to .010″ longer than what their nominal height should be in addition to the dimensional growth we get from hard anodizing fixture components. This allows us to always match tower heights by remachining the tower bases instead of the much less acceptable method of shimming. We don’t shim. Any discrepancy between the fixture’s target heights and the actual height is then corrected on the mill. The fixture is then reassembled and its critical dimensions are re-measured to prove they’re within tolerance and the whole process is documented. Proofing fixtures like this takes a while but it takes the fixture out of the variables when building frames and lets you focus on your technique instead of worrying about what your fixture is doing.
A few words about measuring and why it’s important.
OK, so after reading our proofing process, are you wondering what we use to measure with while proofing our fixtures? I’ll tell you, but first let’s back up a minute and talk about tolerances, resolution, and accuracy.
First, all metrology equipment (tools used for measuring) have their own level of accuracy and resolution they are designed to operate in. Most errors are broken down by a unit of measure. So, if you have a set of calipers that has a .001” resolution with an accuracy of 1 unit of measure, this means when you measure something, your result can be its exact length to .001” resolution, or it can be .001” undersized or .001” oversized (just so you know, most 6” calipers have an accuracy of about .002” over their range). The take away from this is you don’t know whether that part you just measured is 5.999, 6.000, or 6.001 unless you have a known standard to measure against and compare results (gauge blocks, height masters, etc.).
So, if you want to measure to within .005” (five-thousandths of an inch) reliably, your tools have to be able to accurately measure to .001” (one-thousandths of an inch) resolution. With a 1 unit of measure accuracy, this means when you measure to .005” you might not know if you’re actually measuring .004” or .005” or .006” but that is an order of magnitude better than not knowing if you’re measuring is -.005” or +.005” on either side of your target. Hopefully I’m not confusing you.
Anyway, a thousandth is pretty small; an average human hair is .004” (four-thousandths of an inch) just to put it in perspective. Likewise, if you want to accurately measure .001”, you need an ability to accurately measure to .0001” (one-ten thousandths of an inch, commonly called a “tenth” in the machining world; don’t forget this.).
At Anvil Bikeworks, we have height gages, Height Masters, micrometers, and gage blocks that resolve to, or are accurate to, .00005” (fifty millionths) depending on shop temperature (and we know that because we actually have our Met tools professionally calibrated) and we do most of our measuring on a Grade A surface plate. But, and it’s a big but, we can’t maintain that nor do we have the ability to produce parts to that tolerance with the machinery we have and the temperature control we have in the shop. We don’t even try. Honestly, 50-millionths is basically a ghost in the machine. You can hold a room temperature 6” steel rod in your hand for just a few moments and it will grow by more than 50-millionths. It’s so small it might as well not exist.
What we can do is produce parts within a few tenths (+/- .0002) when necessary and we need to be able to measure those tenths accurately to work within our specified tolerances. If you can’t measure it accurately, you don’t know what you’re producing, and you certainly can’t make claims of accuracy for your products. We’ve invested a significant amount of money in our Met equipment and because it’s something I’m “into” we’ll continue to do so. Though our own tolerance for our frame fixtures is within .003” across their range of adjustment, they’re typically more accurate than that.
So, if you want to be a real ass, the next time someone claims how accurately something is made, ask how they measure it, and how they confirm their accuracy. Or, to save time, ask when they last had their tools calibrated.
Unless they have the Met tools and the processes to back it up, they’re just blowing smoke.