Tuesday, January 26, 2010

Delta 14" Bandsaw Rebuild



Note:  I'm writing this for those who can benefit from what I learned, not for those who know nothing about bandsaws.  Descriptions are brief, if this is not your thing, this isn't for you!

About 15 years ago, I picked up what I thought was a nice Delta 14" bandsaw with an enclosed stand, 1HP motor, mobile base, and rip fence.  When I got it back home and put it together, I was surprised how much it shook, wobbled, and vibrated, not to mention how noisy it was.  I never enjoyed using it, so it never got much use.  Over the years I've gotten quite jealous of photos and stories of how nice it is to work on a bandsaw, so I finally got fed up, and about a month ago I started researching how to improve the saw so I could use it.  Two weeks ago, armed with info and confident I could tame the monster, I jumped into the project head first.

First up:  The blade tensioner is a notorious patience-tester and hard on the hands:


So I decided to add a small crank and handle to the knob to make it easier and more comfortable to turn:
Took a scrap piece of steel from a hinge and cut it down for the lever, then added a bolt and bored out a dowel for a handle - looked pretty promising:



However, in practice, the handle would bind on the lever and stick.  I tried nylon washers to reduce the friction, then realized what was wrong:  the bolt needs to be fastened securely to the lever, that way the bolt cannot move and thus the handle will not bind.  But even after I did that it didn't work all that well.  That's when I tried a small brass nipple, which was a winner, except it rattled on the bolt when the saw ran.  So I slipped an old stiffened-up rubber washer beneath the nipple, and put a couple wraps of paper over the bolt, slipped the nipple on and I had a very functional and simple handle that works GREAT.  I drilled and screwed into the top of the original knob, which is aluminum:


OK, that took 10X longer than I anticipated but I learned a lot and haven't had to buy anything yet - all made from things I've had on hand.

Now to the real deal.  I tore the entire machine down:  removed the guards, wheels, blade tensioning assembly, and took the saw off the base.  My first goal was to solidify the base and make it as stable and stiff as possible.  Here's the sequence:

  • bolt the stand to the mobile base (it was just sitting in the base before, and I noticed it moving around within the base while running)


  • bolt on a piece of 3/4 melamine faced particle board to the top of the base - the thin sheet metal on top of the base flexes too easily, adding to instability.  The melamine stiffened this up a lot.


  • get longer bolts and bolt the saw firmly to the stand.

The result:

 

With only the motor and lower wheel turning, I could tell a difference.  Now that I had a stable platform to work from, I began looking for vibration in every moving part, starting with the motor.  I took the belt back off.  The motor shaft spun very freely, with no wobble and play, so it seemed like the motor bearings were solid.  With the belt off and no sheave (the thing that looks like a pulley wheel) on the shaft, the motor ran very smoothly.  However, the slight motor vibration could be felt throughout the stand and saw and you could hear a harmonic buzz.  So I loosened the motor mount bolts and slid some pipe insulation between the motor frame and mounts (taking a trick from building computers in which I never solid-mount a moving part to the case without isolation).  This did 3 things:  took up any slight "slop" in the base to mount connection, deleted the harmonic buzz, and provided a wee bit of isolation.  I could feel and hear the difference - much less vibration throughout the saw.  Here's the idea:



Next up, level the motor so the shaft is horizontal, and align the motor sheave with the saw sheave, ideally with both sheaves very close to the motor and saw bearings.  The problem is, the saw sheave was on the very outside end of the shaft, which applies a lot of leverage to the bearings.  Not good:


I couldn't move the motor in the proper direction to correct this, so I had to move the saw over and re-drill the stand:


Then I re-bolted the saw to the stand, and moved on to the next moving part, namely the lower wheel, shaft, and bearings.  These bearings have a snug friction fit, but do not require any tools to remove/replace.  I cleaned up the bearing casting with worn scotchbrite.  The tolerances between the shaft and inner ring of the bearings is extremely tight, so any little nick, burr, or dirty spot on the shaft creates trouble fitting the bearing to the shaft.  I had to sand a couple burrs back with 150 sandpaper that I made being careless removing the first bearing from the shaft, then as a general remedy I lightly touched up the entire shaft with a piece of worn scotchbrite.  I got tired of handling the wheels, which have sharp ridges from the casting, so I took after both wheels with a variety of bastard files to smooth them out a bit.  While I was at it, I tapped out the upper wheel bearing and replaced them with new bearings.  The stock bearings were NTN, and it only costs $2 each to move up to better quality bearings from SKF or whateve your local bearing house carries:


















A note about bearings:  you have many options, but when you say "sealed" bearings, it's not clear if you want "shielded", which have a metal cap between the inner and outer bearing sleeves (left, below), or "sealed" which have a rubber seal that is usually black (right, below).  Shielded bearings have a bit less friction (you can feel the difference by hand) but allow small particles, water, etc. into the bearings.  You can tell from the number of the bearing whether it's shielded vs. sealed:  6204 ZZ or 6204 2Z vs. 6204 2RS.  The ZZ or 2Z means shields both sides, whereas 2RS means 2 rubber seals.  On my saw, the stock upper bearings were shielded, and the bottom bearings were sealed.  This makes some sense, because the bottom bearings get lots of dust and grime, while the upper bearings get relatively little dust.  However, given that no matter what, dust prevails and is not nice to bearings, I used sealed bearings all the way around.


So, with a stable base and the parts all cleaned up, I re-installed the lower shaft, bearings, wheel, and once again aligned the motor and saw sheaves, and placed the belt on the machine.  I ran the saw without the lower wheel to see how much belt slap I had and it seemed a bit excessive and a hair noisy, and the machine vibrated just a bit more than I was happy with.  Since I was using a Power Twist belt, I took a link out of the belt and tested again.  It still went on quite easily, and had a bit less slap, noise, and vibration.  I removed another link and it fit snugly, and had yet less belt slap, noise, and vibration.  Taking off another link made the belt seem a wee bit too tight, and indeed it vibrated and hummed more than the last iteration, so I put the last link back in.  Running the saw, there was only juuuuuust barely more vibration than with the motor alone. 

Satisfied, it was time for the big ticket item of balancing the wheels.  You can see from the pictures above that I had drilled out the rim of one wheel (a LOT).  That was from my first attempt to statically balance the upper wheel, which was actually quite successful (and illustrated how hopelessly unbalanced it was to start with).  However, from the Old Woodworking Machines site, I learned how to dynamically balance the wheels while they're in place on the saw.  Here's where I started to splurge, by purchasing a dial indicator and magnetic base, which I've always wanted anyway.....  You set the dial indicator on a solid piece of equipment (in my case, the infeed table of my jointer), and set the dial onto the table or some other surface of the machine.  Then I clipped on a large metal paper clip to the wheel in a random spot, recorded the amount of vibration by reading the dial on the dial indicator, then moved the clip 90 degrees and so on, getting a vibration reading all around the wheel at 4 cardinal locatioins.  Then I found the largest reading and experimented placing the clip to either side to find the maximum vibration.  This indicates the heavy spot of the wheel.  This is all done with the tire on the wheel (if you need a new tire, replace it before doing this procedure!).  The first time I had readings of .004 and .0045, indicating that the lower wheel was quite balanced.  I drilled a couple light holes, retested, and it was equal all around the wheel.  Done!  Now running the saw with the lower wheel, I noticed less vibration than before the balancing, and again a wee incremental bit of additional vibration above just running the motor and shaft, but it was still only barely discernable, and a nickel stood on edge on the table (a very common vibration benchmark) was nearly undisturbed, so I was satisfied.  The saw was running very smoothly and felt stable and solid.  I repeated the procedure with the top wheel on and a band in place until the top wheel was balanced.  Again, a bit more vibration, but much of the vibration was removed by very slightly tweaking the tracking and tension of the blade until the blade ran the quietest and with the least amount of side-to-side play.  In fact, from playing with this, I quickly realized this is the best way to tension the blade anyway - play around until the blade is the quietest and vibrates the least.  One thing that bothered me was the wobble of the lower wheel - and it measured .018 inch, which seemed excessive, but the saw was running very smoothly nonetheless, so I moved on.  Seemed Very promising!





Next up, the final tweak to the functional tuning:  Trimming the tires to run perfectly round.  This stumped me for awhile, because the comlicated, clumbsy contraption I've seen in an old Fine Woodworking magazine wasn't simple nor elegant enough for me.  Then I started thinking of a razor blade and a way to mount it that might work.  First time out of the chute, I had a winner.  Pictures are in order, it's a bit hard to explain:



For the upper wheel (top left pic), I clamped the jig to the fence, and placed a shim under the outboard part of the jig, which serves to feed the razor blade further into the wheel.  Turn the wheel counterclockwise -the other way provides too aggressive of a cut and the razor blade jams into the wheel, leaving a nasty cut in the wheel.  My wheels are crowned, so I put a bit of an angle on the razor blade and only really concerned myself with the front side of the crown.  Further, since my cast aluminum wheels are soft, I just milled the rim of the wheel down with the tire - the razor blade made remarkably quick work of this all.  A similar setup, only this time not even requiring a clamp, secured the jig between the upper back guard and table for the lower wheel.  On the lower wheel, I turned clockwise.  Then I put a band back on to the saw and ran it - the blade ran perfectly straight, no side-to side wobble, and the vibration was incrementally improved once again!  At this point, I was really excited - my saw was running like a top, making a very satisfying low drone of a well-tuned machine!

Next, I cut an old ski-tuning brush with stiff short bristles down to size and mounted the brushes to the top and bottom wheels.  In both cases I mounted them to some 2-hole corner angles I had and creatively utilized existing fasteners/mounts on the saw.  These make a big difference - I was surprised how well these keep the wheels clean, and now consider them a necessity:



Next up, dust collection.  I had the factory sheet metal upgrade that was sold awhile back, but it was big, clumbsy, and was always in the way when I needed to change blades or tune the lower guides.  I couldn't find anything convincing on the web made commercially.  Finally, I just decided to run the hose straight to the blade below the table, like this:



Then I squared the table to the band, squared the fence to the table, and cut some 3 1/2" black walnut with results I would have only dreamed of before:

















I now have a very satisfying machine to use and even more satisfying, the process of re-building and tuning it took away all the intimidation and I realize how simple of a machine this really is.  My woodworking career just turned a major corner!


6 comments:

Dag said...

Thank you for this inspiring piece of instruction! I have a 1940-ish (I think...) Rockwell-Delta 14" bandsaw with variable speed belt transmission, which is noisy, vibrates and tracks badly, and needs a similar rebuild. I know it is a good old saw, but it has been well-used over the decades, but not well maintained. Now I have something of a roadmap and some great ideas to follow--wish me luck!

Paul Bouchard said...

I have an old Beaver 3300 that I need to replace the wheel bearings on and this was a huge help in figuring out how to do it. I'll be taking the rotor to my local garage (instead of busting it by trying to do it myself).

Peter Hartman said...

Thanks for the post. I am picking up an older delta tomorrow. ( with out the brotherhood logo)

JPK300 said...

Thank you so much for this post. I just picked up an older Delta bandsaw and am struggling with vibrations on it. I'm going to go through my bandsaw using these techniques.

Thanks again!!!!!

callahan 26 said...

OMG! I also picked up an old delta about 8 years ago. Same story as yours - loud, vibration, pain to use, poor cutting. Have spent last two weeks trying to tune the thing but not getting acceptable results. (Problem seems to be with pulley, motor, lower wheel shaft/bearings) This is a GREAT tutorial on how and where to focus my efforts.

Have been looking online for a few days, especially for lower wheel shaft info. So glad you posted this! Thank you, thank you, thank you!

Chris Lehr said...

Thanks for posting and sharing this. I have a Rockwell/Delta in need of some time and effort, this helps a lot.