Machinist Question

[petejw]

phil, i wouldnt do it with out a press,
u really need to be spot on in drilling,
and rodger outlined how to set it up well.

[oseaghdha]

I would think about spinning the axle down.
That has the least chance of FUBAR.

It's always easier to make a pin fit a bore, rather than the other way around.

Otherwise, you need to heed the advice of shimming the gap up, and using a vise with some way to keep the perpendicularity and location. You are trying to make two aligned bores, not one.

A drill will make an opening but as Ivan stated, it won't be round, will be ugly and have uneven contact when you try to use it. It will wear on the high spots and get loose, or crack the aluminum when you tighten it.

I would hire it out to someone that has the proper equipment to do the job right.

$.02

[daveyg]

I would machine it on a mill, dial in the hole location at the top and bottom, and use an end mill to circular-interpolate it, to ensure that the hole was circular and straight with the original hole. But would be alot less work to throw the 17mm shaft in a lathe and turn it down to 15mm on the ends. I assume the reason behind this project is to fit a different front rim(with different bearings and different brakes, etc)?

[Ivan]

Why use a G81 when you could use a G43, G3, G17, R or I and J and have to screw with D comps and such?

Thats too much work and head scratching.

I have offered to do the work for him, and I would sock it up on the Bridgeport (No G codes! yay), clamp in a spud to dial center and perpendicularity (which would have to be done on a CNC anyway) and use a boring head to size the hole. Both forks would only take about 10 minutes, and wouldn't require programming, tool setup and fixturing.

Also, circular interpolation with an endmill has less accuracy than a 4 foot long twist drill with a flat tip angle and no relief. Not to mention chatter and cutter deflection.

The answer here is simplicity and rigidity. Screwing up the front wheel is playing with someone's life, and it needs to be done right. No hand drills or complex setups.

Oh, and those of you that propose turning down the ends if the axle, how do you propose installation and removal? Take one of the forks off the bike?

[daveyg]

Well circle-interpolation is easy on an accu-rite retrofit sharp mill... That's what I had in mind, and it would be way more accurate than shoving a 17mm drill through it on a drill press. You are correct in that, a boring head would be a little bit more accurate. And with the turning down the dia of the axle, I was envisioning it on my XS650, which has end caps on both forks, my mistake. But, there is more than one way to skin a cat, you could even make a bushing if you wanted to, that would slide onto the stock XS11 axle, and then into the 17mm wheel bearings, but then that would get us back to the precision vs. value of a life topic...

[Schming]

XScellent reply Ivan JAT Would it behoove You to locate bearings that would work on the 15mm axle?

[oseaghdha]

Why use a G81 when you could use a G43, G3, G17, R or I and J and have to screw with D comps and such?

Thats too much work and head scratching.

I have offered to do the work for him, and I would sock it up on the Bridgeport (No G codes! yay), clamp in a spud to dial center and perpendicularity (which would have to be done on a CNC anyway) and use a boring head to size the hole. Both forks would only take about 10 minutes, and wouldn't require programming, tool setup and fixturing.

Also, circular interpolation with an endmill has less accuracy than a 4 foot long twist drill with a flat tip angle and no relief. Not to mention chatter and cutter deflection.

The answer here is simplicity and rigidity. Screwing up the front wheel is playing with someone's life, and it needs to be done right. No hand drills or complex setups.

Oh, and those of you that propose turning down the ends if the axle, how do you propose installation and removal? Take one of the forks off the bike?

Not enough coffee is why. I was thinking the one end was bigger.

I think it should be done in an 800MM HMC on a tombstone with a boring head and a roller burnisher. Use a probe to find the existing and check it with air gages and SPC the results.

Phil needs to write up a justification.

[DGXSER]

all I can say is that I used to think engineers were bad about being set in THEIR way of doing sh$t and no other way can be right. Now I know they can not hold a candle to a machinist!!

[fredintoon]

Buy a drill press.

All this data on how Phil should computer-mill that fork leg and the guy don't even have a drill-press?
Phil, trot down to Harbor Freight or other similar establishment and buy a bench-mount drill press on special for ~$50 or so.
It's good for a lifetime of amateur use and you will wonder how you ever lived without one.
Make a sturdy wooden V-block to support the leg. Wood because you are going to drill into it as you enlarge the fork hole.
Fill the sawcut with aluminum sheet. If you can't find the right gauge of sheet use the next thinnest one and pad it with 28ga aluminum flashing that cuts with scissors and you can buy at the building store.
Do the pinch bolt up tight to hold the shim in place as you drill.
Drill the hole in steps, first with a 5/8" or a 16mm then with a 17mm drill bit. With aluminum, high speed, low drill pressure and lots of aluminum-specific cutting fluid.

[Silent]

Why use a G81 when you could use a G43, G3, G17, R or I and J and have to screw with D comps and such?

Thats too much work and head scratching.

I have offered to do the work for him, and I would sock it up on the Bridgeport (No G codes! yay), clamp in a spud to dial center and perpendicularity (which would have to be done on a CNC anyway) and use a boring head to size the hole. Both forks would only take about 10 minutes, and wouldn't require programming, tool setup and fixturing.

Also, circular interpolation with an endmill has less accuracy than a 4 foot long twist drill with a flat tip angle and no relief. Not to mention chatter and cutter deflection.

The answer here is simplicity and rigidity. Screwing up the front wheel is playing with someone's life, and it needs to be done right. No hand drills or complex setups.

Oh, and those of you that propose turning down the ends if the axle, how do you propose installation and removal? Take one of the forks off the bike?

Amen!

The man speaketh the gospel

[Ivan]

Not enough coffee is why. I was thinking the one end was bigger.

I think it should be done in an 800MM HMC on a tombstone with a boring head and a roller burnisher. Use a probe to find the existing and check it with air gages and SPC the results.

Phil needs to write up a justification.

Well, I run an Okuma 800 horisontal, but the 4th axis is only indexing, and not contouring. I could use the probe, but I would rather monkey around inside the machine with all the soured coolant and chips stuck under my fingernails while I dial the hole and each axis with a test indicator. Also, I would need to employ the skills of an inept programmer that will need a solid model supplied so he doesn't have to read the print or make a wire frame. Then I will need to tell him what a boring head is and make sure he knows not to run it in G83. WHen I get done teaching him how to do that, well start working on the negative so we can build a vacuum fixture. when all that has been accomplished, we'll have to redo the whole damn setup since the -18 is a different profile than the -57, which looks like the same part three rev levels back, so it should have worked. Really. Although SPC is normal practice, on a short run, I'll have to do a 100% acceptance, and report all dimensions but the critical dimensions that can actually be controlled on the machine. Half way through completion of the first article there should be an ADCN in my email that will cause us to go back to square one because there should be a 44 degree chamfer on the hole instead of a 45. By this time it will be 10 am on Monday and the engineering department's 2 hour work week is finished, so we are screwed until next Monday, so pray like hell they have used up all their vacation. Like that really matters.

Oh, and for that, we will make $.17 in profits for a run of 60,000 parts.

I love machining.

[Silent]

You sound like you're working at my shop, funny how I never see you here tho.

[bikerphil]

Ok, thanks everyone for your replies. I know the correct way to go is to have an expert do the work. Gonna look into that tomorrow. My original plan was to leave the holes alone at 15mm and just go to a 15X47X14 wheel bearing. The problem is that this bearing exists, but is very hard to track down. I decided to enlarge the holes to 17mm because those bearings are readily available, the same as the XS. I could have some spacers made to reduce the inner dimension down to 15mm, but that would be a spacer only 1mm thick. There is a 20X47X14 bearing that is easy to get, and that spacer(s) might be a better route. Any thoughts on this approach?

[daveyg]

Ok, thanks everyone for your replies. I know the correct way to go is to have an expert do the work. Gonna look into that tomorrow. My original plan was to leave the holes alone at 15mm and just go to a 15X47X14 wheel bearing. The problem is that this bearing exists, but is very hard to track down. I decided to enlarge the holes to 17mm because those bearings are readily available, the same as the XS. I could have some spacers made to reduce the inner dimension down to 15mm, but that would be a spacer only 1mm thick. There is a 20X47X14 bearing that is easy to get, and that spacer(s) might be a better route. Any thoughts on this approach?

YES. Any of those would WORK now if one will work better or worse is a better question, one that I have no answer for.

[jmnjrpa]

Do not use a drill! Use REAMERS and a good drill press or mill would be preferable. You really should find a good machinist to help out. I have been machining for 40 years but I do not think I would try to remove .08 with a drill press. Clamping problems and alignment will suffer. You will have to step up 3 to 4 steps with reamers. High dollar! I do not believe you will save any money. If the hole is not true then clamp pressure will not be good and may allow failure. NOT GOOD. A machinist would use a boring head and bore it +.002/.003 over.

Good luck.