Framebuilders - Tolerance for error in custom frames

I have a couple questions about custom frame "margins of error" - and thought this might be a good place to ask.

In measuring a custom frame I recently received I noticed a few misalignment issues that have me concerned. Especially because this frame cost so much more than if I had just bought a frame off the shelf. If any of you more experienced than I can offer your insight as to whether these misalignments are real issues or "within standard build tolerance" I'd greatly appreciate it.

All measurements made twice with a digital caliper as carefully aligned as I can manage it.

1) The rear cantilever brake mounts are misaligned by approximately three millimeters.
2) The seat stays are joined at slightly different spots - and one ends approximately two millimeters higher than the other
3) The rear drop outs appear to be joined at a slight angle - and measure ~133.5mm, when 135mm was the specified width.

Are these kinds of things typical for a hand built frame or are these kinds of inconsistencies considered errors worth bringing up with the builder? I don't want to be accusatory but I measured two other lugged frames sitting around my home (not custom) and they came up consistent down to the millimeter.

Customs frames are typically held to much higher tolerances than factory stuff, it's one of the things that set them apart.
How much your issues matter is really up to you, they aren't things that will affect the ride or fit. I'm not sure what you mean by #1 or the "angle" thing in #3 but I can see how something like mismatched seatstays would be annoying.
How do you like the bike otherwise?

as carefully aligned as I can manage it.

If your frameset were measured on a surface table using appropriate tools and care, the measurements may reveal insignificant misalignment (that which does not affect the function, longevity, or appearance) or even more than you were able to measure without a flat, square reference. Without a fixed reference and repeatable measurements, it is difficult to gauge precisely what is misaligned and by how much.

Many framesets, custom and production, have varying degrees of misalignment which do not affect the function of the bicycle or visually detract from it. For example, fork blades may have differing amounts of camber (curvature), but have matching rake. Or, as you noticed on your frame, one of the seat stays may be at a different level from the other. Or dropouts may not be parallel to each other or with the centerline of the frame. However, the bicycle may track and handle just fine; the wheels center correctly in the frame and fork; the brake pads align on the rim.

However, even though your measurements may not be exact, the simple fact that you visually perceived inconsistencies, made the effort to take measurements, and were motivated to post to this forum suggests dissatisfaction with the final product. Since your expectations don't seem to have been met, I suggest you open a polite dialogue with your framebuilder, detailing your questions and concerns.

Presumably, you made the decision to order a custom frameset from the builder you chose based on examples of her or his previous work, or his or her reputation, and had expectations of what you would receive based on that (I'm assuming lowest cost wasn't the sole deciding factor in your purchasing decision). Your framebuilder is in the business of framebuilding to make money (I presume): you don't make money from dissatisfied customers. Also, I imagine (and hope) that your framebuilder takes pride in his or her work and will be quite willing to work with you until you are satisfied.

The magnitude of these misalignments is small. Just ride the bike.

I wouldn't be at all happy with my custom frame if it was as described, we aren't talking about a Soma, Surley, etc. The rear dropouts should fit the axle very near perfect, this is not a topic for debate IMHO.

Sorry about brevity of my last post. If you see movement of the stays when clamping the QR down, that's not good, obviously it's not going to happen if you have to lightly jam the wheel in. 1.5 mm may be attributed to paint build up??? 3 mm difference in position of the canti mounts does not sound awful, but I think most builders do much better. Also I know how to spell Surly.

Without showing anything that would give away the identity of the builder, it would be helpful if you could post close-up photos of the rear brake bosses showing the 3mm misalignment, and the seat cluster from the rear showing the 2mm misalignment of the seat stays.

I agree with Chauncey; close tolerance is one of the things you should expect when you pay a premium for a custom frame. A 2mm vertical difference in where the seat stays join the seat cluster is pretty obvious, as is a 3mm vertical difference where the two rear brake bosses are mounted on the stays. If they really are that far off and it were my bike, I'd be an unhappy customer.

I would try to amicably resolve your disappointments with the framebuilder. He has a reputation to protect and should make every effort to fix the problems.

I'm trying to remember which framebuilder posted a rant about this recently. Definitely contact the builder and see what they have to say for themselves. It's interesting that all three of these issues may have started at the dropouts.

Given that the tubing that the bike was built with probably has 2-3mm bows in it, you can always find some "misalignment". Handling can easily cause "misalignment." I'm curious how you would measure the position of seat stays with any accuracy. I can see if you have a full alignment table and a square. Does a properly dished wheel sit square in the dropouts?

Personally, I expect that a perfectionist with a ruler would probably find some measurements they could disagree with among the winners at NAHBS.

I wouldn't be at all happy with my custom frame if it was as described, we aren't talking about a Soma, Surley, etc. The rear dropouts should fit the axle very near perfect, this is not a topic for debate IMHO.

I completely agree. I bet the OP looked at his bike and saw the inconsistencies before measuring it and knowing for sure. If I'm buying a custom bike, then I'm paying someone to be _more_ of a perfectionist than I am. I would contact the builder and raise the issues in a respectful manner.

Heck, I build frames in my garage using homemade jigs and a Home Depot torch, and am still capable of building a frame with the seatstays lined up and the dropouts spaced correctly. If I spent good money on a custom frame from a "name" builder, I would expect at least that much. Of course, if you went bargain basement an ordered a frame from, well, a guy like me, then you probably have to expect a lower level of quality.

I have a couple questions about custom frame "margins of error" - and thought this might be a good place to ask.

In measuring a custom frame I recently received I noticed a few misalignment issues that have me concerned. Especially because this frame cost so much more than if I had just bought a frame off the shelf. If any of you more experienced than I can offer your insight as to whether these misalignments are real issues or "within standard build tolerance" I'd greatly appreciate it.

All measurements made twice with a digital caliper as carefully aligned as I can manage it.

1) The rear cantilever brake mounts are misaligned by approximately three millimeters.
2) The seat stays are joined at slightly different spots - and one ends approximately two millimeters higher than the other
3) The rear drop outs appear to be joined at a slight angle - and measure ~133.5mm, when 135mm was the specified width.

Are these kinds of things typical for a hand built frame or are these kinds of inconsistencies considered errors worth bringing up with the builder? I don't want to be accusatory but I measured two other lugged frames sitting around my home (not custom) and they came up consistent down to the millimeter.

Other important questions...Is this a very new or established builder? If it's a new builder, did you get a bro deal? If so, it's part of the learning curve and you should just deal with it. If you're actually able to measure that stuff accurately and you got it from and established builder OR paid full price for it, talk to the builder and he or she should sort you out.

"1) The rear cantilever brake mounts are misaligned by approximately three millimeters.

In what way? If you mean one is 3 mm higher than the other, or further from the centerline than the other, that is grotesque, and would bug the heck out of me. I can't imagine a person making a mistake like that on their first frame let alone an established person. This is not the kind of eror that happens just due to heat pulling, that is a different kind of misalignment.

2) The seat stays are joined at slightly different spots - and one ends approximately two millimeters higher than the other

That probably doesn't do much harm, but it should never happen to that degree. If it's visible and this is a style type frame, vs. one that is mostly about fit and function, then it is a serious problem. Like if you paid for 30 hours of paint work and polished lugs and they slipped this kind of stuff in there, what is the point?

3) The rear drop outs appear to be joined at a slight angle - and measure ~133.5mm, when 135mm was the specified width.

This wouldn't bother me if the angle was consistent and slight and was in the direction where spreading apart again would correct the fault, or reduce it a lot. It is normal for a fixtured rear end to narrow after the joinery is done. My rear end spacing is wider than 135 to allow for this, though that isn't a necesarry thing. It shoudn't be a deal where the skewer is expected to bring the plates into plane since a skewer is really too light for that.

I would sight over the head tube and seat tube to see whether there is twist in the frame, also.

Thanks to everyone for the replies - I've contacted the builder and we're working things out. They're being extremely professional and courteous about it and presented me with a number of options.

As a quick aside - I'm aware my measurements can't have been perfect, I was just trying to quantify a visual difference I noticed and was curious about.